Prenatal and perinatal infections: report on a WHO meeting, Graz, 24–26 May 1982

11 World Health Organization 18- f ~ Regional Office for Europe ~ . ~ Copenhagen ~ rl EURO Reports and Studies 93 Prenatal and perinatal infections Report on a WHO meeting Graz, 24-26 May 1982 ICP/ BVM 0 16 ISBN 92 890 1259 5 © Wo rld Health Organiza t ion 1985 Pu blications of the Wo rld Health Organizatio n enjoy copy right protectio n in acco rdance with the provisio ns of Protoco l 2 of the Universa l Copyrigh t Conven- tion. For rights of reproduction o r t rans la tio n, in part o r in toto. of publicatio ns issued by th e WHO Regional Office for Europe applica ti o n sho uld be made to th e Regio na l Office fo r Europe, Scherfigsvej 8, DK-2100 Cope nh age n 0 , Denm ark. The Regiona l Office welcomes such a ppli ca ti ons. The designations empl oyed a nd th e prese nt a ti on of th e material in this publi- cation do not imply th e expressio n of a ny op ini o n whatsoever o n the pa rt of the Sec reta ri a t of the Wo rld Hea lth Organizatio n co nce rnin g the lega l status of any country, terr ito ry, ci ty o r a rea o r of its a uth o riti es, o r co nce rnin g th e delimit a ti on of its fro ntie rs o r bounda ries. The menti o n of spec ific co mpa ni es o r of ce rta in ma nufacturers' products does no t imply th a t the y are endo rsed o r reco mm ended by th e Wo rld Hea lth Organiz- ation in prefere nce to o th ers o f a si mil a r na tu re th a t a re no t menti o ned . Erro rs and o missions exce pted, the names o f propriet a ry products a re di stin gui shed by initia l capital lette rs. The views expressed in this publicatio n are th ose of the participa nts in the meeting a nd d o not necessa rily represe nt the decisio ns o r the stated po li cy of the World Hea lth Orga ni za ti on. PRI NTED IN DENMARK ISSN 0250-8710 CONTENTS Page Introduction .. ...... .. ...................... .. ................. ....... ........................ .. .. . Morphology and mechanisms of prenatal and perinatal viral infections - G.S. Kistler ........................................................................... 3 The relative importance of vira l infections in congenital defects - U. Krech ... . . . . . . .. . . . . . .. . . . ... ......................... . The relative importance of other infections Bacteria l infecti ons - H.P.R. Seeliger .... .... .......... ...................... . Chlamydiae - G.M. Antal ..... ..... .... . . Toxoplasmosis - H. As pock ....................................................... . 17 29 29 39 43 Urina ry tract infections - K. Menzel, N. Dziambor & A. Lemmer ...... ... ... ....... ... .. 52 Detecting prenatal and perinatal infecti ons ............................... ... ....... .. .... .. 55 Labora to ry diagnosis by traditional and new tec hniques - G. Enders 55 Rapid diagnosti c techniques in Spain - M. C. Echevarria, J.M. Echevarria, A. Llticer, A. Tellez & R. Najera .. .. ... .. ..... .. ........ .. .. .... .. ... ...... ... .. 85 Screening fo r total serum lgM in the German Democra tic Republ ic - K. Menzel, A. Lemmer & M. Linke .... ......... .... .... .. ...... .................. . ... ............. ..... ... 94 The shortcomings of epidemiological stud ies: an example from Finland - G. Granroth ................ .. ......... ... ... ...... ... ........ .... ... ... .. .... . ,............. 96 The long-term consequences of prenatal and perinatal infections ..... .. 105 Subnormality - A. Bove ... .. .. ... ... ..... .... .... .. .... ... .. ... .... ..... ..... .. .. .... ... .. .... ..... .... ... ... I05 Mental ret a rdatio n - J.H. Henderson ...... ... . 116 Prevention and control of prenatal and perinatal viral infections - I.A . Dudgeon.. ..... .. ... .. .. .. . ........... ..... ............ ............ ....... ... .. 121 Conclusions and recommendations.. .. ... ...... .... .. .... .. .... .. ... .. .. ... ... .. .. .... ....... ... .. 129 Annex I. Participants.. ... ... ..... ......... ..... .. ... ..... .... ......... .. .......... ....... .. ..... ... ..... .. ... 132 Summaries in French, German and Russian ............. ............. ..... .. .... ....... ... 135 Introduction A Working Group on Prenatal and Perinatal Infections was held in Graz, Aus tri a, from 24 to 26 May 1982. The members of the Working Group included an embryologist, epidemiologists and microbiologists. The pur- pose of the meeting was to assess current knowledge and identify remain- ing problems relating to the etiology, laboratory diagnosis, epidemio logy and, where appropriate, prevention and treatment of such infections, thereby enabling recommendations to be made to health authorities in different countries. Some of the data presented by members oft he Working Group and the discussion arising from it is likely to be relevant not only to the more rational use of existing expertise and health service resources in providing methods for preventing int rauterine and perinatal infections, but a lso to the identificat io n of future avenues of research. Although it is often ass umed that prenatal and perinatal infections present a relatively sma ll problem numerically, the number of children damaged by such infections may well have been considerably underestimated. Thus, it must be appreci- ated tha t many maternal infections, for example those induced by cy to- megalovirus and Toxoplasma gondii, are subcl inica l o r induce relatively minor no nspecific symptoms and therefore go unrecognized and a re not investigated. Infants at nsk may appea r healthy at birth, but prolonged follow- up studies show that such infants may subsequentl y be found to have developmental anomalies of varyi ng severity, including mental retar- dation. It may be difficult or even impossible, howeve r, to confi rm a diagnosis of pre-or per in a tal infection by laborato ry tests by the time that these children are assessed in later infancy or childhood. This report therefore emphasizes the importance of ca rrying out addi tional prospec- tive enquiries, s ince these may lead to a better assessment of the incidence a nd long-term effects of pre- and perinatal infect ions. Many of the laboratory techniques described during the meeting, pa r- ticularly enzyme immun oassays th at have recently been developed to detec t sero logica l res po nses to in fec ti o ns, in pa rti cul a r spec ific lgM res po nses, have considera ble po tenti a l a s tes ts o f enhanced se nsiti vit y a nd reliabilit y with whi ch to di ag nose ma tern a l infec ti o ns a nd mo nit o r infant s in hig h-ri s k ca tego ri es. Neverthe less, it is esse nti a l th a t c linicia ns, epidemi- o logis ts a nd la bora to ry wo rk ers a re made awa re of the pit fa ll s o f these rece ntl y deve lo ped techni q ues, pa rti cul a rl y as co mm ercia l kit s a re now ava il a bl e; in the ha nds of in ex peri enced wo rk ers they may so metim es p rovide erro neo us di ag noses. Wh a teve r the ac tu a l frequency o f pre- o r perina ta l infec ti o ns, the re ca n be no do ub t tha t in fa nt s wh o a re da m aged as a result of the m prese nt co nside rab le econo mic a nd socia l p rob lems, sin ce they o ft en require li fe- lo ng ca re. A multidi sc iplina ry a pproac h to the p ro ble ms o f eti o logy, preventi o n a nd ma nage ment o f pre- a nd perin a ta l infec ti o ns is th erefo re esse nti a l. T hi s repo rt hig hlig ht s ma ny as pec ts o f thi s a pproac h. 2 Morphology and mechanisms of prenatal and perinatal viral infections G.S. Kistler In pathophysiological terms, most postnatally acqu ired primary vira l infections in immunocompetent persons take a simi lar course regardless of the virus involved. The principal features of such infections are summar- ized in Table 1. It must be app reciated. however, that a lthough the virus may be tra nsmitted to th e co ncep tus, viraem ia usually precedes ma tern a l clinical sym pto ms suc h as fever or ras h, and that many maternal vira l infections may be subclin ica l. Almos t a ll human vira l diseases have been reported during pregnancy a nd adverse effects on the fetus, such as spon taneous abo rtio n and developmental anomalies. have been attributed to a number of them (1). Nevertheless, a clear-cut and consis ten t etio logica l rela ti o n- ship between ma ternal virus infecti on and developmental a no ma lies has been established o nl y fo r a few vi ruses, the best known being cytomegalo- virus and rubell a vi rus. The gesta tional age a t which ma terna l infec ti on occurs, the virus itself a nd it s " infec ti vi ty", as well as the severit y of ma ternal infectio n a re so me of the factors tha t may innu ence the outcome o f a virus infection in pregnancy. The possible consequences for the unborn child of a viral infecti o n in it s mother are outlined in Table 2. Indirect effects on the fetus of a viral infection in the mother Spontaneous abortion, stillbirth and neo- or perinatal mortality, which may result from various viral diseases in pregnancy (for example va ricella, innuenza, meas les, mumps o r po liovira l fever) are probably not , in most cases, due to a vira l invas io n of the fetus. Nevertheless, the precise patho- genic mechanisms involved in inducing such deleterious effects in the fetus remain to be clearly defined. There see ms, however, to be a correlation between the severity of maternal infection and the likelihood of an adverse effect on the concept us . Prolo nged fever, the impairment of cardiac and/o r pulmonary functions resulting in ma ternal (and fetal) hypoxaemia , 3 Table 1. Pathogenesis of a postnatal viral 1nfect1 on with vIraemIa in an Immunocompetent organism • 1nfect1on of epithelial cells of the inner or outer body surface • virus replication and spread to neighbouring cel ls • plasma or ce llul ar v1 raem1a/ lymphogen1c spreada • onset of the immune response with inflammatory rea c ti on • ··neu tral1zat1on" of the 1nfect1 on • repair o f the epithelial damage • 1n some instances. persistence of the 1nfect1on 1n various cell types (e .g . herpesv1ruses) a Viraem,a precedes cl,n,cal symptoms Table 2 . Possible consequences of a viral infection in pregnancy • no placental or fetal 1nfect1on no ef fect on the development of the unborn secondary effects on the development of the unborn (fever. hypoxaemia. 1mpa1rment of homeostasis) • placental 1nfec t1on without fetal infection secondary effects on the unborn (malnutrition. 1mpa1red gas exchange. toxi c ity) • pla cen tal infection with fetal 1nfect1on dea th and resorption of th e emb ryo spontaneous abor ti on 4 still birth prema tur ity conge nital defects. congeni tal ove rt disease anomalies 1n infancy. childhood. adulthood a no ma lies in ac id -base, elec tro lyte and body nui d ba lances, a nd the tra ns- placenta l tra nspo rt of tox ic byproducts may be so me of the fac tors res po n- sibl e fo r feta l d a mage. Nevert heless, inap pro pri a te o r inse nsitive tech- niques fo r de tec t ing the virus invo lved may be respons ible for the fa ilure to detect a viru s in the p roducts of concept ion. Routes of virus transmission in prenatal infections A number o f viruses may be t ra nsmitted to the fe tus a nd the ro utes by which thi s may occur a re presen ted in Table 3. T ab le 3. Routes of v i rus transm ission in prenatal 1nfect 1ons via blood pla sma via white and red blood ce ll s (cellular v1raem1a) • As cending 1nfect1on via cervix uteri and fetal membranes persistent 1nfec t1on o f the myometr1al ep1 th el1um • Descending infec tion - via oviduc ts • Tran smura l 1nfect1on - via myometr1um and dec1dua • Vertica l virus transmis sion - via egg o r sperm cel ls Transp/acental passage of the virus This is the mos t co mm o n way in which viruses a re tra nsmitt ed fr o m mo ther to fetus. During the course o f ma ny ma terna l vi ra l di seases , the initia l local infectio n is fo ll owed by the entrance of the virus into sma ll blood vessels and lympha tics . Viraemia usua ll y begins seve ra l days before 5 the o nse t of clinica l sy mpto ms. Virae mia se ldo m persists fo r more than o ne o r two weeks, since spec ific a nt iviral neut ra li zing a ntibodies deve lo p sho rtl y a ft er the o nse t o f sy mptoms. Alth o ugh the mec ha nism by which a n in fec tious vi rus reaches the bloodstrea m has no t yet been full y elucida ted, endothe li a l ce ll s may, witho ut being ha rmed themselves, ta ke up a virus by micropinocytos is, tra nspo rt it with in ves icles th ro ugh the cy to plas m a nd then d ischa rge it into the circula ti o n. Viruses ma y a lso in fect endo thelia l ce ll s a nd multiply within their nucleus or cyto plas m (for insta nce, vacc inia virus; see Ref. 2). T he dea th o f endotheli a l ce ll s may lead to a simultaneous release o f a la rge num ber o f infecti o us virus pa rticl es int o the circula t ion. A virus tha t has ga ined access to the vascu la r sys tem may reach th e cho ri o ni c o r place nta l villi tha t ba the in ma terna l bl ood . Mos t viruses a re a ppa rentl y t ra nspo rt ed free within the b lood plas ma, and a re no t a tt ac hed to a nd d o no t li e within ce llula r elements. Wh ereas so me ea rli er in ves- tiga tors (3, 4) ass umed th a t vi ruses could cross cho rio nic o r place nta l villi a nd leave them intact, it is now accepted tha t in genera l placenta l in fec ti o n preced es fe ta l infectio n (5- 8). The resulting vira l place ntiti s may va ry in seve rit y a nd ca n be foca l o r ex tend over la rge r a reas of the place nta . It is usua ll y assoc ia ted with infiltra tes consisting o f mono nuclea r ce ll s. Pl a- ce nta l villi often d ispl ay signs of ac ute foca l nec rosis a nd a re fo und to be covered with fibrin depos its (F ig. 1). In so me viral infec ti ons, inclusio n bod ies may be o bse rved within the cy to tro pho blas tic a nd syncy tio- t ro ph o bl as ti c ce ll s of th e villi . In a dditi o n , th e st ro ma l blood vesse ls o f the villi often a ppea r dil a ted a nd hyperae mic. These pa th o logica l cha nges leave littl e do ubt tha t th e cho rio nic a nd placenta l villi a re th e entry po int for the virus a nd tha t they a re directl y in vo lved in the in fec tious p rocess. In ves tiga tio ns (9) have sho wn tha t ma terna l blood ce lls ca n cross the place nta l ba rri er either active ly or by pass ive tra nspo rt. Since a n in fec tious virus (o r it s geno me) is o ft en present within lymphocy tes, neutrophili c gra nulocy tes a nd macro ph ages (1 0. 11) a nd since pa rticl es may a lso be a tt ached to the membra ne of e ryth rocy tes, ce llul a r viraemi a is a no ther mecha nism by which a virus may cross the placenta. In th ese pres uma bly rare insta nces, o ne wo uld expect the placenta l lesio ns desc ribed above to be less pro no unced o r even a bsent , in spite o f a fet a l infecti o n. Ascending infection A vira l in fec ti o n of th e epithelium of the ce rvix uteri (the p roto types a re huma n (a lpha) herpesvirus 2 a nd cyto megalo virus) may spread to the dec idua a nd to the feta l membranes. Likewise, a pre-exi sting , persistent viral infectio n o f the endometrial epithelium (12) may directly infect an embryo o r even be transmitted to the rapidl y growing tropho blast of a n impla nting blas tocyst during the ve ry first da ys of develo pment. 6 Fi g. 1. Viral placent1t1s. rubella-infected human fetus -.J Note swollen and necrob,ot,c v,11, ( @ ) and infiltrates cons ,st,ng predom,nantly o f mononuclear cells (m) Other routes of infection Whereas descending infecti o ns (via the oviducts) and transmural infec- tions (via the myometrium a nd decidua) are ways in which some bacterial infections are transmitted to the fetus, their practical importance in viral infections is probably negligible. The vertica l transmission of a virus that persists as an integrated genome in germ cells must, however, be con- sidered as another possible way in which infection oft he fetus occ urs in the earliest stages of its deve lopment. Since the vira l ge nome replicates during mitosis, all daughter cells arising from an infect ed , fertilized ovum will carry the virus. Whether such a mecha nism of virus transmiss ion is oper- at ive in ma n remains to be elucidated. Conceivably, thi s type o f infection could play a role in spontaneous abortions at ear ly stages of embryonic development. Embryonic and fetal reactions to a viral infection Histopathological and virological studies indicate that both the embryo and the young fetus are extremely susceptible to the widespread dissemina- tion of a virus as well as to its destructive effects. This sensitivity can be explained by a number of factors that have been found to be markedly different in mature and immature fetuses. Thus, rapidly dividing and metabolically highly active embryonic cells form an optimal living sub- strate for the replication of viruses, but an immature fetus has a larger number of ti ss ues with specific cell membrane receptors for viruses than does a mature one (13). In the embryo and young fetus, both the ce llular and humoral defence mechanisms are underdeveloped, the bone marrow and the primary and secondary lymphatic organs being still in the process of differentiation and growth (Table 4). Although interferon synthesis has been detected by about the I 0th week of gestational life (14) and lgM production around the I Ith to 13th week, the fetal immune system is unlikely to be fully operative until the 15th or 16th week of gestation. Physiological and morphological barriers , such as the blood/brain barrier, are less effective in immature organisms than in mature ones (15). It is therefore not surprising that there are striking differences between the response of an embryonic or early fetal orga nism to a viral infection and that of a peri- or postnatal individual (Table 5). Morphological findings in virus-infected embryos Affected embryonic cells (single or in groups) are usually observed in most tissues , independent of the virus involved. The predominant histopatho- logical changes, often present in a number of different organs, are summar- ized in Table 6. 8 Table 4. Biological characteristics of the normal human embryo • metabolically and mitot1cally highly active cells • efficient mononuclear phagocyt1c system • deficient cellular and humoral defence mechanisms: maturation of T-lymphocytes 9th week mature lymphocytes and granulocytes In peripheral blood onset of interferon synthesis production of lgM 10th week 10th week 1 1th-13th week • sensItIve phases of organogenesIs differ from organ to organ : central nervous system and sensory organs d1fferent1ate throughout gestation • functional deficiency of morphological and physiological barriers. e.g . blood/brain barrier Table 5 . Characteristics of human embryonic virus infections • extensive virus replication at site of entry (usually the chorionic villi) • virus d1sseminat1on into most of the organs (no viral organotropism) • persistence of v1rus-produc1ng cell clones throughout gestation (chronic infection) • absence of inflammatory reactions of the postnatal type • primary and secondary effects of infection. including cell death. mitotic inhibition. interferon synthesis. toxic products of cell destruction • postnata l shedding of virus (excretions. secretions) 9 Table 6. Histopathology of human embryonic vi ral 1nfect1ons • necrob1ot1c cells (single or in groups) 1n most organs • focal . per1vascular bleedings • focal loss of vascular endothelial ce lls • intravascular thrombi and embol1 • intra - and extravascular mult1nucleated giant cells • nuclear and/or cytoplasmic viral 1nclus1on bodies • h1stopatholog1cal picture more o r less independent of virus type • absence of inflammatory reactions of the postnatal typ e • predominance of vascular damage Multifocal, perivascular bleedings of variable degree (Fig. 2) are pres- ent in the vascular system including the endocardium, while focal swelling, necrosis and focal absence of endothelial cells are also noted (Fig. 3). The lumen of blood vessels often contains thrombi amd emboli that consist of pyknotic nuclei, ce ll debris and an eosinophilic, fibrillar material, the origin of which is unknown (Fig. 3 and 4). Nuclear and/or cytoplasmic viral inclusion bodies may be detected in epithelial and other ce ll s of an embryo infected by a virus of the herpesvirus or poxvirus groups. Gross external malformations are not usually seen (5,8). The occurrence of uniform histopathological changes in a variety of prenatal, embryonic infections (for example those caused by herpes- viruses, cytomega lovirus , influenza viruses, mumps virus, rubella virus and hepatitis virus A) points to a common pathogenic factor that seems to be operative in most virus-infected embryos. It has been assumed that the genera lized vascular damage of virus- infected human embryos and fetuses represe nt this common factor, and that the pathological changes of endothelial cells are due to extensive multiplication of the virus within these cells, resulting in cell necrosis and the shedding of both virus particles and dead cells into the bloodstream (5). Virus iso lation studies, however , have shown that there is only a low concentra ti o n of virus particles in the organ homogenates of affected em bryos and fetuses. Thus , in the tissues of rubella virus-infected unborns, 10 virus titres seldom exceed values of l-l.51og 10TCID5/ml. In addition, immunofluorescence studies in such tissues have revealed virus-specific antigens in on ly a sma ll minority of the fetal cells . Finally, extensive ultrastructural investigations on a variety of organs from rubella virus- infected embryos and young fetuses have led to the surprising results that mature rubella virus particles have never been detected in endothelial or other ce ll s, despite successful virus isolation in tissue cultures. However, the cytoplasm of a variable number of endothelial cells contains tubulo- reticular structures (TRS) (8) . Similar if not identical structures have a lso been observed in endothelial cells of embryos with viro logically proven cytomegalovirus infection (Fig. 5 and 6) . TRS consisting of branching tubules of different lengths and with a diameter of 18-30 nm have been described within dilated endoplasmic reticulum cisternae of endothe li al and other cells in a variety of conditions. Thus, TRS have been observed in cells of apparently normal human and animal postnatal individuals, and in cells of patients with autoimmune diseases (such as systemic lupus erythematosus), malignant tumours (leu- kaemias, carcinomas and sarcomas) and viral infections caused by either DNA or RNA viruses ( 16.17). Whereas earlier investigators regarded TRS as viral or virus-like in nature, it is now widely accepted that these struc- tures represent focal proliferations of endoplasmic reticulum membranes and that they reflect a cellular response to a broad range of stimuli (17) . Gyorkey et al. (18) reported recently on the occurrence of abundant TRS in the lymphocytes, hepatocytes and vascular endothelial cells of a patient with acute Kaposi's sarcoma who died from cytomegalovirus viraemia, Pneumocystis carinii pneumonia and vira l hepatitis. The patient harboured, in addition, Giardia lamblia and Entamoeba spp. in his int es- tines and Cryptococcus neoformans in his lungs and in some of his lymph nodes. Interferon blood levels were very high. Moss et al. (19) found aggregates similar to TRS in the endoplasmic reticulum of hepatocytes of newborn mice that had been injected with partly or highly purified mouse interferon preparations. Finally, in an in vitro study (20) interferon was found to induce the formation of TRS in cultured lymphoid cells. These findings are further evidence that TRS, observed in a variety of cells in human embryos and fetuses infected in utero by rubella virus or ot her virus types, may be an effect of endogenously produced interferon. In rubella virus-infected embryos and fetuses, endothe li al cells display- ing TRS undergo necrosis, characterized by mitochondrial swelling and the loss of cytoplasmic ground substance. The widespread endothelial damage observed in these tissues may be related to the presence of these structures (8) but their exact role remains to be clarifed. In adults affected by autoimmune diseases, the inflammatory processes regularly observed (e.g. myocarditis, encephalitis, glomerulonephritis) are assumed to be caused by the accompanying immune-complex vasculitis. It 11 Fig . 2-6 Source · Kistler, G . (8) 12 is therefore tempting to speculate that similar mechanisms could be responsible for the endothelial cell damage present in rubella virus- infected embryos and fetuses. The absence of overt inflammatory reactions in the fetus could thereby be explained by the lack of mature effector cells of the immunological defence sys tem. Other reactions of the virus-infected embryo and fetus A virus-induced interferon synthesis in embryonic and/or fetal cells has also to be considered as a possible cause of the multiple pathological changes observed in infants who have experienced intrauterine viral infec- tion. Besides it s antiviral activity, interferon is also known to be a potent inhibitor of mitosis (and therefore of growth). A long-lasting interferon synthesis in response to a persistent viral infectio n could have profound effects on a large number of precursor and differentiating cells. Thus, the inhibition of mitosis during the organogenic phase could be responsi ble not only for a substantially reduced overall number of cells (which in turn would explain the often observed reduced body weight at term) but also for such developmental anomalies as cardiovascular malformations. The pathological changes observed in neonates infected in utero at an early stage of gestation may be caused by the combined action of a number of detrimental factors and are summarized in Table 7. Clinical and histopathological findings in newborn infants with an embryonic viral infection Infants who survive intrauterine viral infections are often born with a number of clear-cut stigmata that point to a long-lasting prenatal patho- logical process (Table 8). Occasionally, however , the sequelae of a prenatal Fig. 2. Focal haemorrhage in the cerebellum of a rubella-infected fetus. The damaged blood vessel in the centre of the focus is indicated by the arrow. Fig. 3. Cerebral vcnule of a rubella-infected fetus, containing a thrombus that consists of pyknotic nuclei, cell debris and fibrill a r material. The arrows point to a region of the blood vessel that is denuded of endothelial cells. Fig. 4. Em bolus consisting of cell debris and fibrillar material in an apparently normal fraction of a myocardial venule of a rubella-infected fetus. Fig. 5. Tubulo reticular complex (arrowed) within a n endothelial cell (En) of a blood vessel of a placental villus. Rubella-infected fetus . Lu = vascular lumen . Fig. 6. Tubuloreticular complex (arrowed) in an endothelial cell (En) of a liver sinusoid. Di = space of Disse, Fi = fibroblast, Ee = erythrocyte. 13 Table 7 . Pathogenesis of prenatal . v1rus-1nduced damage Developm en tal anomalies 1n c lud1ng malformations may be ca used by the combined ac tion of : • prolonged vi rus repl1cat1on 1n precursor cells causing substantial cell death or permanent loss of ce lls • 1nh1b1t1on of mitosis by viru s-induced interferon resulting 1n overall reduction of the total cell mass • prol onged trophi c damage to tissues and organs as a result of generalized vascu lar lesions • toxi c damage to tissues and organs by byproducts of virus replication and cellular necrob1os1s H1 stopatholog1cal findings point to vascula r damage as a predominant factor in the pathogenesis Table 8. Clinical and histopathological findings in newborn infants who have expe rienced intrauterine infection in early gestation • prematurity or reduced birth weight at term • developmental anomalies in the ca rdiovascular system. ce ntral nervous system. skele ton . etc . • lgM ant1bod1es 1n cord serum • florid peri - and postnatal 1nfect1on with inflammation such as adenitis. encephalitis. hepatitis. pneumonia • prolonged virus shedding (excretions. secretions) • " late onset disease" • anomalies in childhood . adolescence or adulthood such as Juvenile diabetes and possibly malignant tumours 14 infection may become evident only as a late onset disease that is probably mediated via immune complexes. In addition, some infants with appar- ently normal hearing in infancy may later be found to have seve re percep- tive deafness (21). This has been shown in rubella virus-infected babies (22). Forrest et al. (23) noted a high frequency of diabetes mellitus in young adults with congenital rubella. Finally, a possi ble association between virus infections in pregnancy and malignant disease in the child or yo ung adult has been discussed by a number of investigators (24-26), but the validity of these retrospective studies remains to be confirmed. Characteristics of perinatal viral infections The method by which a virus is transmitted from mother to infant peri- natally is similar to that in embryonic or fetal infections (plasmic or cellular viraemia; ascending, descending or transmural viral passage). Infection may often be facilitated by degenerative changes and focal leaks in the placental villi. When viral infection occurs in older fetuses or newborn infants, however , it is usually accompanied by an acute inflam- matory reaction similar to that observed in postnatally acquired infec- tions, as the immune system has reached a high degree of maturation . Since the number of cells bearing specific virus receptors is usually less than at earlier stages of development , virus dissemination is in most instances less widespread. On the other hand, pronounced inflammatory reactions ma y themselves have adverse effects that are not observed in embryos or young fetuses (Table 9). Table 9 . Characteristics of human perinatal virus infections • extensive or moderate virus replication • often l1m1ted virus dissemination . depending on virus type and organotrop1sm • inflammatory reactions of th e postnatal type due to a well developed immune system (in part) • primary and secondary effects of infection often more pronounced th an 1n infancy anrl childhood • prolonged shedd ing o f the virus 15 References I . Remington, J.S. & Klein, J.O. Infectious diseases of the fetus and newborn infant. Philadelphia , W.B . Saunders, 1976. 2. Wisniewski, M. & Kistler, G.S. In: Intrauterine infections. A Ciba Foundation Symposium. Amsterdam, Elsevier, 1973, p. I 8. 3. Baskin, J.L. & Soule, E.H. American journal of diseases of children, 80 : 10(1950). 4. Kibrick, S. & Benirschke, K. Pediatrics, 22: 857 ( 1958). 5. Tondury, G. Embryopathien: Ober die Wirkungsweise (Infektionsweg und Pathogenese) van Viren au/ den mensch/ichen Keimling. Berlin, Springer, 1962. 6. Kistler, G.S. & Gertsch, R. Deutsche medizinische Wochenschrift, 23: 1254 ( 1970). 7. Tondury, G. & Kistler, G.S. Zeitschrift fur Prii ventivmedizin, 18: 45 ( I 973). 8. Kistler, G.S. Beitriige zur Pathologie, 155: IOI (1975). 9. Loke, Y.W. Immunology and immunopathology of the human fo etal- maternal interaction. Amsterdam , Elsevier, I 978 , p. 7 I. I 0. Gresser, I. & Lang, D. Progress in medical virology, 8: 62 ( 1966). 11. Sommerville, R.G. Progress in medical virology, 10: 398 (1968). 12. Dehner, L.P. & Askin, F.B. Obstetrics and gynecology, 45: 211 ( 1975). 13 . Holland, J.J. Virology, 15: 312 (1961). 14. Banatvala, J.E. et al. In: Intrauterin e infections. A Ciba Foundation Symposium. Amsterdam, Elsevier, 1973, p. 77 . 15 . Eichenwald, H.F & Shinefield, H.R. Advances in pediatrics, 12: 249 ( 1962). 16. Schaff, Z. et al. Journal of the National Cancer Institute, 51 : 293 ( 1973). 17 . Uzman, B.G. et al. Laboratory investiga tion, 24: 492 ( I 971 ). 18. Gyorkey, F. et al. Lancet, 2: 984 (1982). I 9. Moss, J. et al. Lancet, 2: I 277 ( 1982). 20. Rich, S.A. Science, 213: 772 (1981). 21. Peckham, C.S. Archives of diseases in childhood, 47: 571 ( 1972). 22 . Marshall, W.C. In: Intrauterine infections. A Ciba Foundation Sym- posium. Amsterdam, Elsevier, 1973, p. 8. 23 . Forrest, J.M. et al. Lancet, 2: 332 (1971). 24. Fedrick, J. & Alberman, E.D. British medical journal, 2: 485 (1972). 25 . Bithell, J .F. et al. British medical journal, 1: 706 ( I 973). 26. Vianna, N.J. & Polan, A.K. American journal of epidemiology, 103: 32 I (1976). 16 The relative importance of viral infections in congenital defects U. Krech To assess the relative importance of virus infection in inducing congenital defects, perinatal infection or late onset disease, it is important to obtain the following information through prospective studies. I. The prevalence of virus-specific antibodies among women of childbearing age. This will assist in an evaluation of the risks of infection during pregnancy. 2. The incidence of infection with specific viruses in pregnant women at each stage of gestation . 3. The ratio of clinically apparent to inapparent infection at birth and among children with proven prenatal or perinatal infection during their development. 4. Information relating to the teratogenic potential of different virus strains. For instance, rubella vaccine strains have apparently only a very low pathogenicity, since the rate of congenital defects induced after vacci- nation of rubella-susceptible mothers is either very low or non-existent when compared with the effect induced by naturally acquired rubella virus strains. All the above information must be based not only on clinical features but also on laboratory data. Unfortunately complete data are not yet available and any estimate of the importance of virus infection in congenital defects must therefore be only an approximation . The data that are available at present are summed up in Table I and they suggest that the following general conclusions may be drawn . When compared with such other causes as radiation, toxic or 17 .... 00 Table 1. The re lat ive im portance of vi ra l infect io ns in co ngen ita l defects Antibody carriers Calculated 1nc1dence of Congenital defects Ratio of 1nfect1on Virus 1nfect1on during pregnancy In the childbearing age per 100000 births per 100000 births to defect Rubella 80-95% 4-2000 4-20 1nterep1dem1c Up to 10% of (exceptionally under 80%) 200 epidemic infection 1nfec11on in 1st month of gestation. 1 5% 1nfect1on in 2nd month of gestation. 25% 1nfect1on in 3rd month of gestation. 1 7% 1nfect1on In 4th month of gestation. 6% infection In 5th month of gestation. 1. 7% Human !beta) 40-100% 500- 2500 0-50 Up to 6% of herpesvirus 5 (primary and reactivated primary infection cannot be d1fferent1ated) Mumps >90% 80-100 Rare Measles >90% 4-6 Rare Lymphocyt1c < 1% Rare choriomen1ng1t1s Human (alpha) >90% 10-20 In early pregnancy Rare <0 1% herpesvirus 3 Infection less than 4 days before 30-50% ,n delivery Is rare 1 5 cases recorded Human (alpha) 60-90% Primary or reactivated herpesvirus 1 or 2 prenatal 1nfect1ons Rare Perinatal 1nfect1ons 2-16 cases Hepat1t1s B virus 10- 100% 1 0-1 00 for primary 1nfect1on 1 00% for chronic carrier genetic disorders, the incidence of congenital defects induced by infection is low. They are most likely to account for less than 109c of all congenital defects. On the other hand, it seems that most children delivered of mothers who have experienced virus infection in pregnancy are apparently healthy at birth. The proportion developing disease as a result of intrauterine infection is unknown. The follow ing conclusions may, however, be reached when available data are analysed. I. Viral infections are responsible for only a small proportion of congenital defects. 2. Only a sma ll proportion of infections occurring during pregnancy will affect the development of the fetus. The ri sk of intrauterine infection causing congenital defects depends on: the proportion of susceptible women of childbearing age; the frequency of infection during pregnancy and the gestationa l age at which it occurs; the virulence of the virus involved. Rubella virus The most complete data for an estimate of the relative importance of a particular viral infection in inducing congenital defects are those for rubella virus . To assess the risk of primary rubella virus infection during pregnancy, extensive sero-epidemiological studies have been conducted in many parts of the world among women of childbear ing age. Although the results of some of these studies are not strict ly comparable, since different sero logical techniques were emp loyed, a remarkably uniform distribution of rubella virus antibodies in va rious age groups in different countries throughout five continents has been found. Thus in the 17- to 22-year age group, antibodies were detected in at least 80% of 12 of the 25 study groups and in over 707c: in the 20 participating countr ies. There are, however, a few geographical areas in which the percentage of pregnant women with antibodies drops below 50% (l-3). The risk of rubella-acq uired congenital defects in rubella-susceptible women is clearly related to the time of maternal infectio n (4,5). This o bse rvation is confirmed in a ll studies, a lth ough the propo rtion of affected children may vary considerably from one st ud y to another. This va ri a tion ma y be at least partl y explained by the different methods by which studies were conducted and also reflects the often sma ll numbers of patien ts available for assessment. In retrospective s tudies, as would be expected, the incidence is considerably higher than in prospective ones. Perhaps 19 strains of rubella virus vary in their capacity to induce congenita l defects (6, 7). It is therefore difficult to provide precise figures as to the number of children born with rubella-induced congenital defects, but it is estimated that during the pandemic that swept through Europe in 1962 and then spread to the United States and lasted until 1965, "many thousands" of ch ildren were born with multiple handicaps as a result of intrauterine infection by the rubella virus. In the Uni ted States more than 33 000 pregnancies were affected: 13410 women had therapeutic abortions a nd 20 000 children were born with congenital defects. Human (beta) herpesvirus 5 (cytomegalovirus) Rubella virus and human (beta) herpesvirus 5 are the viruses mos t fre- quently isolated from newborn infants. The prevalence of huma n (beta) herpesvirus 5 antibodies among women of ch ildbearing age in different areas of the world varies from 40% to 100% (8). The high incidence of fe ta l infection in highly immune populations suggests that the intrauterine transmission of the virus also results from the reactivation ofa latent vi rus (9). This concept is also supported by reports on cytomega lovirus infection in consecutive pregnancies (10.11). Analysis of these accum ulated data from a number of studies indicates that although human (beta) herpes- virus 5 reactivitation may result in virus transmission to the fetus, it is unlikel y to result in developmental anomalies (/2). Fetal damage is most likely to occur after primary maternal infection. Nevertheless, it is difficult to evaluate the effects of maternal primary or reactivated infections re- liably as neither is usually associated with the development of clinical features in the mother. It is estimated, however, that in socioeconomically highly developed areas about 50% of women of childbearing age have human (beta) herpesvirus 5 antibodies and that the rate of congeni ta lly acquired infection is somew here between 0.5% and l % . The proportion of chi ldren who are clinical ly affected is between 0.05% and 0.1 % (/3). In underdeveloped areas the proportion of congenitally acq uired infections is we ll above l %, but there is no evidence so far that these infections contrib- ute towards the development of congen ital anomalies. Human (alpha) herpesvirus 3 Very few women reach childbearing age without acquiring huma n (alpha) herpes virus 3 a ntibod ies; perhaps less than 5% of women of childbearing age arc suscept ible to human (alpha) herpesv irus 3 infec tion . Nevertheless, a lth ough there is no question that congenital defects can result from huma n (alpha) herpesv irus 3 infection of the mother, only 15 well documented cases have been described in the literature. Prospective 20 studies conducted in the Federal Repu blic of Germany (Enders, G. per- sonal communication, 1982), the United Kingdo m (14) a nd the United Sta tes (15) sugges t tha t the risk of human (alpha) herpesv irus 3 infecti on during pregnancy ca using congen it a l defects is very mu ch smal ler tha n for rubell a virus o r cytomega lov irus infection. Hum an (a lpha) herpesvirus 3 is difficult to iso lat e from products of co ncepti o n a nd the evidence tha t th e fetus escapes in mos t cases is derived from sero log ical follow-up s tudies in newbo rn infant s wh ose moth ers acq uired huma n (alpha) herpesvirus 3 in pregnan cy . Such studi es have fa il ed to detec t perina tal antibody in hea lth y infants a t ri sk. Very occas io na ll y huma n (alpha) herpesvirus 3 ma y be reactivated during pregna ncy but the re are o nl y two repo rt ed cases in which th e fetus has been affected in thi s way (16, 17). Hum a n (a lpha) herpesv irus 3 infec ti o n be longs to a group of viral infec tio ns that, when acq uired shortl y before delivery a nd transmitted to the newborn, may cause a n unu suall y severe di sease in the newbo rn . The fa ta lit y ra te in th e new bo rn following ma terna l infectio n with huma n (alpha) herpesv irus 3 less than four da ys before delivery is between 30% and 50% (1 8), but negligible when acquired m ore tha n four days before delivery. Huma n (a lpha) herpesv iruses I a nd 2, polioviruses, coxsackiev irus and o ther enteroviruses may also ca use severe disease if acquired sho rtl y befo re o r after delivery . Human (alpha) herpesviruses 1 and 2 Infec tion with hum an (alpha) herpesvi ru ses I a nd 2 during pregnancy, pa rti cul a rl y if prim a ry, may lead to infection of the fetus a nd thi s is proba bl y in co mpa tibl e wi th con tinu ed life. Abortion may occur a nd hum a n (alpha) herpesviruses I a nd 2 may be iso la ted from th e p roduc ts of co nce pti o n . Intraute rine infection may occur shortly before delivery a nd may res ult in infa nt s bein g delivered with ev ide nce o f di sease induced by human (a lph a ) herpesv iru ses I a nd 2. Such cases are, ho w- eve r, ex tremely rare. Mo re co mm o nl y, infec ti o n with human (alpha) herpesv iru s I is acquired eith er during passage throu gh th e birth ca na l, o r pos tn a ta ll y fr o m maternal les io ns or fr o m th e les ions of a tt endin g perso nn e l. During pregnancy as ma ny as I% of pregna nt women exc ret e human (a lpha) herpesv iru ses I a nd 2 but thi s ma y increase to I 0% befo re te rm. The rate declines gradu a ll y pos t-pa rtum . Reac ti va ti o n o f hum a n (alpha) he rpes viruses I a nd 2 during pregna ncy may res ult from a s imil a r mec ha ni s m as with hum a n (beta) herpesv iru s 5. Th e g rea tes t risk to th e newborn infant is from primary ma ternal genital lesions prese nt at birth , although a mother with a hi s tory o f rec urrent herpes viral vulvovaginitis, cervicitis or urethriti s may still tran s mit a seve re and so metimes fatal infection to the infant. Studies in the United Kingdom a nd the United States (19) s uggest tha t 2 to 16 cases of perina tal infecti o n with hum a n 21 (alpha) herpes viruses I and 2 occur for every 100 000 births, and that this is associated with a high mortality rate. Measles virus The incidence of measles virus infection during pregnancy seems to be much lower than for rubella virus, mumps virus, and human (alpha) herpesvirus 3 (20, 21). Thus even if measles virus infection in pregnancy caused congenital defects, the association would be difficult to detect. In addition, it must be appreciated that a diagnosis on clinical grounds alone may be difficult to make. There is some information from the virgin soil epidemics in Greenland (22) in which data were collected retrospectively, but without laboratory confirmation. The abortion rate following measles virus infection in the first trimester was about 30%. Among 300 live births there were 8 with severe congenital defects and 10 later had hearing defects . Hepatitis B virus Until the diagnosis of viral hepatitis 8 could be confirmed by laboratory procedures, the incidence of hepatitis 8 virus infection in children was considered to be negligible or non-existent, being based on clinical evi- dence alo ne. With the introduction of laboratory procedures for the detec- tion of hepatitis 8 virus markers, however, this concept has changed markedly. Thus, a lthough there are very few children with clinical hepa- titis 8 virus infection, the incidence of hepatitis 8 surface antigen (HBsAg) carriers in this age group may be high in certain geographical areas, part icularly in sub-Saharan Africa and South-East Asia. Although some infa, its may be infected in utero as a result of maternal viraemia, caused either by an acute viral hepatitis B attack or the carrier state, the majority of infants are infected either peri- or postnatally by close contact with mothers carrying the HBsAg in their blood (23). The chance of transmis- sion seems to be particularly high if the mother has recently experienced an acute viral hepatitis B infection or is an HBsAg carrier but is also HBeAg positive. Follow-up studies have shown that most infected children develop normally, although in a very few cases where liver biopsy has been carried out, histological evidence of chronic persistent viral hepatitis B infection may be present. Of particular relevance is the finding that persistent HBsAg carriage may be a predisposing factor in the development of primary liver cell carcinoma (hepatoma), one of the commonest malignant tumours in parts of the world where HBsAg carrier rates are high. It is possible that the acquisition of infection perinatally predisposes to the subsequent devel- opment of primary liver cell cancer, although it is probable that other environmental factors also play a role. 22 There is very little information on the role of hepatitis A virus 10 inducing congenital infections . The laboratory diagnosis of hepatitis A virus has only recently become generally available. Perhaps further infor- mation on the risks associated with maternal infection will be available before long. Mumps virus A diagnosis of mumps virus infection during pregnancy is quite frequently made, although the clinical diagnosis cannot always be confirmed by laboratory data . Mumps virus infection seems to occur more frequently during pregnancy than infection with measles virus or human (alpha) herpesvirus 3. The spontaneous abortion rate following mumps virus infection in the first trimester may be greater than 30% , which is well above the "average abortion rate" (24) . Mumps virus has been incriminated in congenital defects more often than any other virus except rubella virus. Nevertheless, the evidence linking mumps virus infection in pregnancy with endocardial fibroblasts in the newborn is unconvincing. Maternal infection shortly before delivery may result in the development of severe disease in the newborn (Krech , U., unpublished data). Lymphocytic choriomeningitis Sporadic infection with lymphocytic choriomeningitis virus during preg- nancy may lead to congenital defects . This view is supported by experi- mental work in monkeys. On the other hand, infections with the virus in humans are a rare event (25) and therefore even rarer in pregnancy. Smallpox vaccination during pregnancy (26) Primary smallpox vaccination in pregnancy could result in a perinatal fetal infection, spontaneous abortion or the delivery, often premature, of an infant with multiple and active vaccinial lesions. Since smallpox has been eradicated from the globe, however, there is no need to vaccinate anyone, let alone a pregnant woman, against smallpox. From all the above data it becomes quite evident that a precise evalu- ation of the relative importance of virus infection in congenital abnor- malities is at present not possible. Results from laboratory controlled prospective studies are available only on a limited scale for rubella virus. Only very few newborns of infected mothers are followed up to detect the possible late appearance of clinical symptoms connected with prenatal infections. 23 Etiology of prenatal and perinatal infections in Canada During 1978, 32 infants were either stillborn or died within the first 7 days of delivery, as a result of maternal infections. In 1976, in the first 4 weeks after delivery, 3168 infants died as a result of infection, including parasitic disease. In 1977, an analysis of records of admission to hospital during the first year of life showed that 78 infants were admitted with streptococcal infections, 39 with human (alpha) herpesvirus I and 2 infections, 279 with rubella virus infection, 6 with hepatitis A virus infection, 16 with human coxsackievirus infections and 699 with an assortment of other viral infec- tions. In addition 12 infants were admitted with toxoplasmosis, 10 with congenital toxoplasmosis, 8 with gonococcal infection and 10 with other sexually transmitted diseases . Table 2 provides data obtained from laboratory returns on pre- and perinatally acquired virus infection and Table 3 on viral infections in children under 6 months of age. Tab le 2. Number of cases of pre- and per1natally acquired v irus infections in Canada In 1980 and 1 981 Year Agent 1980 1981 Human (beta) herpesv1rus 5 19 10 Rub ella virus 21 3 Hum an (alpha) he rpesv1ruses 1 and 2 4 3 Human rotav1rus 3 0 Respiratory syncy t1 al virus 0 Adenov1ruses 4 2 Para1nfluenza v iru s 2 1 0 Human ec hov1rus 2 0 Total 53 19 No reliable estimate is available on the total number of congenital rubella cases occurring annually in Canada. The congenital anomalies surveillance programme involves 5 out of 10 Canadian provinces, and this programme reported 9 cases in 1973 , 13 in 1974, and 33 in 1975. In 1978, 24 Table 3. Number of cases of virus infections acquired by infants under six months of age in Canada in 1980 and 1981 Year Agent 1980 Human rotavirus 386 Adenoviruses 182 Human coxsackievirus B 43 Human echoviruses 54 Astrovirus 38 Parainfluenza viruses 38 Respiratory syncytial virus 35 Rubella virus 32 Human (alpha) herpesviruses I. 2 and 3 26 Pol 1ovi ruses 23 Human (beta) herpesvirus 5 55 Total 912 1981 295 142 44 38 20 24 28 5 26 16 27 665 rubella virus infection was confirmed in 39 pregnant women from 6 Can- adian provinces. Statistics Canada reported 29 cases in 1979, 15 cases in 1980 and no cases in 1981 . There are only a few reported cases of ophthalmia neonatorum reported in Canada, and this suggests that preventive measures are being widely and effectively taken. There were 9 cases in 1979, 14 in l 980, and 8 in 1981 according to Statistics Canada. Using the congenital anomalies reporting system, which receives statis- tics from 10 Canadian provinces, Table 4 shows selected congenital infec- tions depicted as cases and rates per 100 000 live births, for the period 1973 to 1979. Clearly, this is an incomplete picture and probably serves to demonstrate the lack of adequate and comprehensive reporting in this country. 25 N °' Tab le 4. Number of cases and rates per 100 000 live births 8 of selected congenital infections in six Canadian provinces. 1973-1979b Year of birth 1973 1974 1975 1976 Myocobacrenum ruberculos,s 0 0 0 0 Human (alpha) herpesv1rus 3 0 0 1 0 ((J4) Human (alpha) herpesv1ruses 1 and 2 0 0 0 0 Rubella 9 13 33 8 (4 2) (6 0) (14 8) (3 6) Human (beta) herpesv1rus 5 2 3 4 3 (0 9) (14) (1 8) (1 4) Plasmod,um spp . 0 0 0 0 Treponema pal/1dum 7 5 4 5 (3 2 ) (2 3) (1 8) (2 3) Toxoplasma gond11 1 4 4 4 (0 5) (1 8) (1 8) (1 8) a Rate per 100 000 live births g,ven 1n parentheses b 1 973-1976 based on 5 provinces (New Brunswrck. Ontario. Manit oba. Alberta and Br1t1sh Columbia) 1977 - 1979 based on 6 provinces (Prince Edward Island added) 1977 1978 1979 0 0 0 0 0 0 0 0 1 (0 4) 5 5 12 (2 2) (2 2) (5 3) 2 6 3 (0 9) (2 7) (1 3) 0 0 0 4 1 2 (1 8) (0 4) (0 9) 2 1 1 (0 9) (0 4) (0 4) References I. Horstmann, D.M. Rubella: the challenge of its control. Journal of infectious diseases, 123: 610 ( 1971 ). 2. Rawls, W.E. et al. WHO collaborative study on the sero-epidemiology of rubella. Bulletin of the World Health Organization, 37: 79 ( 1967). 3. Witte, J.J. et al. The epidemiology of rubella. Atlanta, National Com- municable Disease Center, Health Services and Mental Health Administration, 1969. 4. Alford, C.A. et al. Virologic and serologic studies o n human produc- tions of concept ion after maternal rubella. New England journal of medicine, 271: 1275 (1964). 5. Alford, C.A., Jr. Congenital rubella. A review of the virologic and serologic phenomena occurring after maternal rubella in the first trimester. Southern medical journal, 59: 745 ( 1966). 6. Kono, R. Antigenic structures of American and Japanese rubella virus stra ins and experimental vertical transmission of rubella virus in rab- bits. In: International Symposium on Rubella Vaccines, London, 1968. Basie, S. Karger, 1969 (Symposia Series in Immunological Standard- ization, Vol. II), pp. 195-204. 7. Kono, R. et al. Experimental vertica l transmission of rubella virus in rabbits. Lancet, I: 343-34 7 ( 1969). 8. Krech, U. & Tobin, J. A collaborative study of cytomegalovirus anti- bodies in mothers and young children in 19 countries. Bulletin of the World Health Organization, 59: 605-610 (1981). 9. Schopfer, K. et al. Congenital cytomegalovirus infection in newborn infants of mothers infected before pregnancy . Archives of diseases in childhood, 53: 536-539 ( 1978). I 0 . Embil, J.A. et al. Congenital cytomegalovirus infection in two siblings from consecutive pregnancies. Journal of pediatrics, 77: 417-421 ( 1970). 11 . Krech, U. et al. Congenital cytomega lovirus infection in sib lings from consecutive pregnancies. Helvetica paediatrica acta, 26: 355-362 ( 1971 ). 12. Stagno, S. et al. Congenital cytomega lovirus infection . The relative importance of primary and recurrent maternal infections. New Eng- land journal of medicine, 16: 945-949 ( 1982). 13 . Krech, U.H. et al. Cytomegalovirus infections of man. Basie, S. Karger, 1971. 14. Bradford Hill, A. et al. Virus diseases in pregnancy and congenita l defects. British journal of preventive and social medicine, 12: 1-7 ( 1978). 15 . Siegel, M. Congenital malformations following chickenpox, measles, mumps and hepatitis. Journal of the American Medical Association, 226: 1521-1524 (1973). 27 16. Duehr, P. Herpes zos ter as a ca use o f congenit a l ca ta ract. American journal of ophthalmology, 30: 157- 16 1 (1955) . 17 . Webster, M.H. & Smith, C.S. Congenit a l abno rma lit ies and ma terna l zos ter. British medical journal, 2: 11 93 ( 1977). 18. Meyers, J.D. Co ngenit a l va ri ce ll a in term in fa nt s. Ri sk reconsidered . Journal of infec tious diseases, 129: 2 15 ( 1974 ). 19. Nahmias, A. et al. In fec ti o n o f the newbo rn with Herpesv irus ho minis. In: Schulma nn , I. , ed . Advan ces in pediatrics. C hicago, Yea r Book M edica l Publishers, 1970, pp. 185-226. 20. Sever, J. & White, I.R. Int ra uterine vira l infec ti o ns. Annual review of medicine, 19: 4 7 1 ( 1968). 2 1. Siegel, M. & Fuerst, H.T. Low birth we ight a nd ma terna l virus di s- eases. A prospective s tud y of ru be ll a, meas les, mum ps, chickenpox, a nd hepa titis. Journal of the American Medical Association, 197: 88 ( 1966). 22. Christensen, P.F. et al. A n epi de mic of meas les in southern G reenland. 195 1. Measles in virgi n soi l. II. The ep idemic proper. Acta medica scandinavica, 144: 430 ( 1953). 23. Crumpacker, C.S. H epa titi s. In: Rem ingto n , J. S. & Kl ei n, J .O., ed . Infec tious diseases of the fe tus and newborn infant. Philadelphia, W .B. Sa unders, 1976, pp. 492-520. 24. Siegel , M. et al. Compa ra ti ve fe ta l mort a lit y in ma terna l virus di s- eases. A p ros pect ive study o n rubell a, measles, mumps, chichnpox a nd hepa titis. New England journal of medicine, 274: 768 ( 1966). 25. Ackermann, R. et al. Pra na ta le In fe kti o n mit dem Virus de r lympho- zy ta ren C ho rio meningiti s. Beric ht uber 2 Fa lle. Deutsche medizin ische Wochenschrift, 99: 629 ( 1974) . 26. Smallpox and vaccinia. In: H a nshaw, J .B. & Dudgeo n, J .A., ed . Viral diseases of the fe tus and newborn. Philadelphia, W .B. Saunders, 1978. 28 The relative importance of other infections Bacterial infections - H.P.R. Seeliger Since the early days of bacteriology, various bacterial species have been shown to induce severe and often fatal disease both prenatally and peri- natally . On ly four species of bacteria are discussed here since current evidence suggests that they represent the principal bacterial pathogens that induce prenatal and immediate postnatal disease. The effect of maternal infection by other sexually transm itted agents such as Mycoplasma hominis and Ureaplasma urealy ticum remains to be established conclusively. Some studies suggest, however, that such organisms may be associated with spontaneous abortion or stillbirth . Neisseria gonorrhoeae N. gonorrhoeae was the first bacterial species known to cause ophthalmia neonatorum , gonococcal blennorrhoea, conjunctivitis and/or keratitis. The infection is usually transmi tted during delivery. Within three days an acute suppurative conjunctivitis develops, the purulent exudate rapidly spreading from the conjunctiva to the cornea. If not rapidly and ade- quately treated, other parts of the eye may be affected. Antibiotic treat- ment cures most infections but delay in therapy or inadequate treatment may cause permanent visual disturbance or lead to complete blindness. Recognition may be delayed if the initial symptoms are mild or develop after departure from hospital. Recent evidence indicates that if membranes rupture prematurely, gonococcal eye infections can be acquired prenatally (/),cervical gonococci infecting the infant before delivery. This could also provide an explanation for the observation that gonococcal ophthalmia can occur after Caesarian section (2,3). Before penicillin became generally available, the most frequently used prophylactic was that of Crede (i.e. the instillation of one drop of I% silver nitrate solution into each eye of the newborn infant immediately after delivery) . This method is still used in some cou ntries, being a safe and 29 reliable method of preventing neonatal gonococcal ophthalmia. This method will fail, however, if infection has already taken place before delivery ( 1). In view oft he increasing recognition of inclusion conjunctivitis caused by chlamydia( infection of the newborn, I% erythromycin has recently been recommended in the United States as an alternative method for the prophylaxis of neonatal ophthalmia, since both gonococci and chlamydiae are sensitive to this antibiotic. Several retrospective investigations have demonstrated a relationship between maternal gonococcal infection and an increased incidence of prematurity, low birthweight and fetal wastage . Whether this association indicates a direct effect of gonococcal infection or reflects compound co-factors is not clear. Because of the high gonococcal infection rate ( 1-5%) found in preg- nant women in some countries, prenatal screening using culture tech- niques is not being recommended in the United States. Considerably lower rates in Europe (about 0.3-1 % ) have led some authors (4) to think that this is not likely to be cost-effective and to recommend instead that screening should be limited to certain groups regarded as being at high risk of infection: women with premarital and extramarital conceptions women with a husband away from home for long periods drug addicts women who are known to be promiscuous women with trichomoniasis (50% of cases of gonorrhoea are accompanied by trichomoniasis (5)). To prevent prenatal or perinatal infection 1t 1s important to detect maternal infection in the third trimester of pregnancy even if the patient is asymptomatic. Only in the acute typical case, however, can a diagnosis be reliably made by means of a Gram stain. Thus, whenever possible, this procedure should be supplemented by attempts to culture and determine the antibiotic sensitivity of the organism. It is important to stress that cultural methods are of the utmost importance in establishing a diagnosis in patients who are asymptomatic or have minimal symptoms. Treatment should always include the often asymptomatic sex partner and should be followed by appropriate culture tests to ascertain bacteriological cure. The excellent results obtained in prophylaxis and treatment with adequate doses of penicillin or related compounds may be jeopardized by the appearance of beta-lactamase producing gonococci (6, 7). 30 Treponema pal/idum One of the principal agents of prenatal and perinatal damage was recog- nized when Hoffman & Schaudinn first detected Treponema pa/lidum in tissue sections of the liver and other organs of newborn infants with congenital abnormalities. It was only after the advent of chemotherapy, however, that the fetus and newborn infant could survive without develop- ing the severe and often multiple stigmata of congenitally acquired syphi- lis, often the result of unrecognized maternal infection . Congenital syphilis is transmitted via the placental barrier from the infected mother to the fetus . This may result in abortion during early pregnancy. In later pregnancy (usually after the fifth month) fetal tissues are invaded and this may lead to fetal death , abortion or stillbirth. Never- theless, some infected infants are delivered at term . The diagnosis of syphilis during pregnancy must depend on the out- come of serological tests for the presence of antibodies against T. pallidum in the mother during pregnancy and/or in the newborn infant. Steps have been taken by public health services in many countries to detect syphilitic infection among pregnant women by appropriate serological screening. This may be supplemented by the serological screening of couples before marriage, an obligatory measure in some countries and a recommended procedure in others. Prenatal routine care, compulsory in some countries and strongly recommended in others, offers an opportunity free of cost to reduce the incidence of congenital syphilis. Such programmes may include unmarried mothers as well as those who may have acquired syphilis since the pre- marriage tests were carried out. Appropriate serological diagnostic procedures are a prerequisite for the recognition of syphilis during pregnancy. The high degree of specificity of modern serological reactions now makes it possible to diagnose such infections reliably. The great value of the Treponema pallidum haemagglut- ination assay (TPHA) test was recently confirmed in Lower Saxony (8). Twelve fresh syphilitic infections, five of them among foreign mothers not aware of their disease, were detected when 25 000 serum specimens were screened. For screening purposes, the cardiolipin Venereal Disease Research Laboratories (VDRL) test, the rapid plasma reagin (RPR) test and the TPHA test are recommended . Sera in which the results of the VDRL and TPHA tests do not agree should be tested by the fluorescent treponemal antibody absorption (FT A-ABS) test. A microhaemagglutina- tion assay for T. pallidum (MHA-TP), a micromethod of the TPHA test , is now being used extensively in laboratories. For detailed references to treponemal serology, the recent report of a WHO Scientific Group on Treponemal Infections may be consulted (9) as may a number of articles by Luger and others (10-12) . 31 If such methods are used routinely no case of congenitally acquired syphilis should escape attention. Such screening procedures are only avail- able, however, for a small proportion of the world population; indeed, the vast majority of pregnant women never experience any prenatal care. The diagnosis of congenital syphilis in the newborn may be com- plicated by the transplacental transmission of maternal treponemal IgG antibodies to the fetus. These antibodies will decline in the absence of an active infection over a 3-6 month period. A persistent treponemal IgG response therefore indicates congenital syphilis. T. pa/lidum-specific IgM antibodies are present in congenitally acquired infection and a diagnosis may readily be made by detecting the class of T. pa/lidum-specific antibodies . It is now possible to diagnose congenital syphilis with great accuracy by detection of specific IgM antibodies in serum specimens (IgM FT A- ABS test). If by this method antibodies cannot be demonstrated in pla- cental blood, congenital syphilis can usually be excluded. In a few in- stances, however, such antibodies may originate from the sera of reactive mothers and may thus lead to confusing results. For differential diagnosis in such cases, the use of the l 9S IgM FT A-ABS test in conjunction with the solid phase haemadsorption test may be of value in solving this diagnostic dilemma. The use of these tests is still confined to a few research labora- tories. A positive result is considered as proof of active congenital syphilis, but a negative result does not necessarily exclude prenatal infection . Nega- tive tests should therefore always be followed by repeat tests at 1-2 week intervals, and antibody levels should be evaluated on the basis of quantita- tive serology. In such cases a careful X-ray examination should also be made to detect the bone lesions characteristic of syphilis. Penicillin treatment in adequate dosage and over the necessary period of time will usually protect the fetus at any stage of its development against intrauterine syphilitic infection. It will also cure an established infection due to T. pallidum (13) . Consequently the chances of recognizing and treating congenital syphilis may be regarded as very promising in those areas where appropriate medical facilities exist. Streptococcus agalactiae Group B beta-haemolytic streptococci consist of a number of closely related bio- and serotypes that may be referred to collectively as Strepto- coccus agalactiae. The importance of this group of organisms in inducing abortion and perinea( infection was not realized until the reports of Fry (14) in the United Kingdom and Hill & Butler (15) in Australia were published in 1938 and 1940. Their importance is now recognized world- wide and has been extensively reviewed by Patterson & Hafeez (16) and Mi.iller (17). 32 The accu mul ,1 tcd da ta prese nted in th ese re views as well as data obta ined fro m studies carried o ut by Ba ker ( 18) show tha t th ere a re two distinct clini ca l sv ndromcs that occur in infant s. I. Early onse t di sease occurs in newborn infants usuall y less tha n 5 days (maximum IO clays ) of age. eona ta l infec ti o n is acq uired by d irec t trans- 111iss ion from th .: ma ternal birth ca nal to th e newborn infant . which may be infected \'ia th e mouth o r by minut e epithe li al les io n, . Res pirat o ry distress. se pticaemia anJ meningitis ma y occur and there is a high mortalit y rate. Ea rl y onse t disease is often associa ted with o bstetri ca l complications. 2. Late onse t disease occurs in o lder infants. after the 10th day of lif"e . This is clini ca ll y charac teri zed primarily by a purul en t lcpt o meningiti s. Most cases an: ob\·iousl y nosocomia l in or igin. Both forms of disease ha vc a hi gh case fa t:11 it y ra tc varying from I 5' i to 80' i . The total incidence among ne\\"born infants is esti mat ed to be bct\\'een0.1 1 i and0.Q I<; ofli\'cbirths. The mag11itudc of the problem is exemplified by the co mprehensive repo rt of I Iood ct a l. ( /9) from the Charil~ l Iosp it al in Ne\1 Orleam. Co mplicati ons suc h as abo rti on. perinatal death. prcmaturit) or severe illness wit h ultimat e recovery were no ted in 56 o ut of 151 infants (37 .0' ; ) born of moth ers fro m \\'hom a beta-haemolytic streptococcal age nt was iso lated in ,pcc im cm ob tained from the genital tract. These o rga nisms we re present in th e gen it a l trac t o f 5-6q o r pregnant women with no e\·idencc or disease. a figure cor respo nding close ly with that obtained by Kexel (20) in the Federa l Rep ublic of Germany . A somew hat higher rate ( 12.4 '7c ) was found in the ge nital discharge o f wo men a fter de livery (21). On the o ther hand. Hood ct a l. ( /9) sho \\'ed that S . agalacriae \\'a s re- cove red fro m 16 out of 66 pa ti ent s (24 .2'"i) wh o expcricnced marked difficulties durin g pregnan cy. Scrotypc I 1/1 has been fo und to be th e ca use o f over S0''i of menin gea l infecti ons in ea rl:,- o nse t di sease. a nd in mo re tha n 95'1 in late onse t di case ( / 7) . Other scro types may a lso cause neonata l infecti ons a nd death. Proph ylax is in the hospi ta l environm ent is difficult a nd requires a high standard of genera l and personal hygiene in delivery suit es a nd o n obstet- rical wards. Th e rou tin e examination of specim ens from all wome n at term would be o f va lu e in identifyi ng newbo rn infa nts at high ri sk. but the cos ts of such a sc reen in g programme make it unrea li stic. Rou tine proph ylac tic adminis tration of penicillin o r ampicillin in ade- quate doses is a matter of conjecture and debate . Such " th era py" ma y be cn\'i saged o r even reco mmended ( / 7) when S. agalacriae is detected pre- o r post na tall y in th e moth er' s gen ital tract o r o ral cavity o r in so me part o f'the infa nt' s skin o r mucous membran es. On th e o ther ha nd. this view is no t 33 shared by Patterson & Hafeez (16) who state: "Recommendation for the chemotherapeutic treatment of colonized infants should be viewed in light of the disparity between risk of colonization and risk of disease. Potentially 99 per cent of colonized infants would be placed on penicillin prophylaxis unnecessarily. In addition, there are no data from any control led study that show association between antibiot ic therapy and decreasing incidence of group B disease". Listeria monocytogenes Listeria monocytogenes was first fully described as an organism by Murray et al. (22). Its role in infections of the newborn can be traced back, however, to the 19th century where the disease was recognized by pathol- ogists under terms such as "pseudotuberculosis of infants" , " epticaemia due to argentophile rods" or "miliary liver necrosis". The role of L. monocytogenes in nature and its habitat are not clear, but virulent li steriae have been cultivated from faeca l specimens, soil, and decaying vegetab les and plants. Physiological characteristics suggest that the microorganisms might be included in the psychrophilic to mesophilic so il bacteria . This is certainly true for the closely related but a pathogenic species, Listeria innocua. L. monocytogenes will infect a wide spectrum of an imals, and under natural conditions more than 50 animal species may be infected. A variety of dangerous clinical syndromes may occur in man, particularly in the newborn and immunosuppressed. Seven international symposia have dealt with listeriosis since 1957. A full account of human and animal listeriosis is given by Seeliger (23) and by Gray & Killinger (24). Infections during pregnancy may frequently result in transmission from placental tissue to the fetus or neonate either during intrauterine life or during delivery via the infected birth canal. After delivery, newborn infants may be infected by contaminated environmental sources. Maternal infection and its attendant risk to the fetus and newborn infant are comprehensively dealt with by Seeliger & Finger (25) . Despite there being a large number of animal reservoirs of L. mono- cytogenes, the epidemiology of human listeriosis remains obscure, with a few exceptiona l cases such as transmission of the agent to a veterinarian who ha ndled aborting domestic animals (26). Small numbers of L. mono- cytogenes may be excreted transiently in the faeces of healthy , pregnant women, from which the organism may be cultivated by the specia l co ld enrichment method. The technique takes many days to complete, however, by which time the infant has already been delivered. On the other hand, consistent failure to detect an exogenous source of L. monocytogenes points to hidden endogenous niches of the o rganism. L. monocytogenes has been culti vated from throat swabs and faecal specimens of midwives (27). 34 From th e ex perience in western and central Europe it a ppea rs that L. monocytogenes-induced fetal death, abortion or premature delivery of septic newbo rn infants occurs ma inl y in late spring a nd th ro ugh o ut the summer . Alth o ugh mos t cases occ ur sporadicall y, a series of o utbreak s occurred in 1949 in th e district of Halle in the German Democratic Republi c. A large outbreak of li s terios is occ ured in 1981 in the Atlantic provinces of Canada. A case was defined as a mo ther a nd/or chi ld with a positive cu lture for L. monocy togenes in the natal period or an adult no n-pregnant individual with positive bl ood o r cerebrospinal nuid. Between April and August 198 1, there were 29 confirmed a nd 3 probable cases in Nova Scotia. Twenty-five were perinatal cases a nd included 3 apparently hea lth y mothers from whom a placental isolate of L. monocy 10genes was made. Perinatal illness was characterized by an innuenza-like illness in the mother followed by spo nta neous abo rtio n , sti llbirth or delivery of a li ve but serio usly ill infant. There were 17 infants born alive, 5 of whom did not survive (case fa ta lit y rate, 29.4% ). There were 3 stillbirth s a nd 5 spontaneous abo rti ons. The ove ra ll fatality ra te from perinatal ca uses was, therefore, 52% . During the sa me peri od in New Brunswick, 7 cases occ urred: of 6 newborn infants, 5 survived. In Prince Edwa rd Isla nd , 4 cases occ urred, a nd there was I stillbirth . In Newfoundla nd there were 2 cases , o ne of whom was a neo nate who survived. This Canadian o utbreak was investigated thoroughly a nd a foodborne mec ha ni sm of spread was incrimina ted. The agent, of the sa me sero- type (4b), was isolated from patients a nd from cabbage fertilized with sheep manure. Little do ubt ex ists tha t L. monocytogenes ca n spread in obs tetrica l wards. Preve nti on of fetal o r neo natal infection may o nl y be ac hieved by making a di ag nosis in the pregna nt mo ther. The clinical features are nonspecific, however , consisting of an acute febri le innuenza- like illness; pyelonephritis has also bee n reported. Such clinical features are probabl y the result of a Listeria bacteraemia, and if they occur in the last trimester may result in septic abortion o r the delivery o f an infec ted infant. On the o ther hand, alt hough most fetal infections occ ur when maternal L. monocytogenes infection is acqui red a fter th e fifth month of pregnancy, documented cases have occurred in which infect ion has been acqu ired in the third a nd fo urth month s. L. monocy togenes is considered to be the cause of 1-6% of aborti ons in various parts of Europe a nd ma y also induce recurrent a bo rtio ns. This may be the result o f maternal reinfection from necrotic placental ti ssues which in turn lead s to a listeria bacteraemia and abortion. Nevertheless, the interva l between maternal infec tion a nd in- vasion of the fetu s has no t been determined . L. monocytogenes may be detected in cord blood, lochia, ti ssue obtained by curettage, cervical mucus, urine a nd placental tissues. The possible pathways responsible for intrauterine infection are shown in Fig. I. 35 w °' Fig . 1. The rou te of 1nfec t1on in d1aplacen tally tra nsmi tt ed l1steriosis Haematogenous spread Placenta membranes / lntervillous space '-..... Chorionic villi Source: Muller, G. Geburcsh,lfe und Frauenhedkunde. 16 505 ( 1956) Amnion and chorion laeve t A mniot ic flu id Fetal uri ne Fetus Intestinal tract Lungs Kidneys ♦ Organs The clinica l diagnosis in the newborn infant is made with difficulty, but respiratory distress, cyanosis, refusal to drink , vom iting, convulsions, soft whimpering, early meconium evacuation and symptoms ascribed to "aspiration of amniotic fluid" may be observed. A characteristic feature , however, is the presence of roseoles and small focal granulomas on the posterior pharyngeal wall and skin. The centra l nervous system may also be involved, and this is assoc iated with a high mortality rate and is probably the result of a ge neralized infection. On clinical grounds, however, it cannot be differentiated from other bacterial infections of the meninges, for example those caused by S. agalactiae. The mortality rate among untreated infants or in those in whom it is delayed is approximate ly 70% . Laboratory diagnosis may be achieved by microscopic (Gram stain) examinat ion and the cu lture of meconium, placenta and genital secretions. L. monocytogenes is present in the cerebrospinal fluid of infants with meningitis and it should a lways be exam ined at the slightest suspicion oflisteriosis. Patients should be treated with high doses of ampici llin . Prenatal and perinatal L. monocytogenes infection is notifiable accord- ing to the public health regulations of the Federal Republic of Germany as well as in some other countries. Cases presenting with central nervous system manifestations are also notifiable. In that context, it is regarded by public health authorities as simi lar to S. agalactiae, whereas N. gonor- rhoeae and T. pa//idum are notifiable under legislation to prevent the spread of sexually transmitted diseases. In contrast with gonorrhoea and syphi li s, for which there are estab li shed methods of prevention, however, the reporting of infection by L. monocytogenes or S. agalactiae cannot result in con trol measures. Nevertheless, notification is of value in assess- ing the relative importance of these infections as a cause of fetal wastage and perinea! disease. References I . Thompson, T.R. et al. Journal of the American Medical Association, 228: 186 (1974) . 2. Rees, E. et al. British journal of venereal diseases, 53: 173 ( 1977). 3. Strand, C.L. & Araneo, V .A. Sexually transmitted diseases, 6: 77 ( 1979). 4. Sparks, R.A. et al. British journal of venereal diseases, 51: 110 ( 1975). 5. Catterall, R.D. British journal of hospital medicine, 3: 55 ( 1970). 6. Percival, A. et al. lancet, 2: 1379 (1976). 7. Pang, R. et al. British medical journal, 1: 380 (1979). 8. Sander, J. & Niehaus, C. Deutsche medizinische Wochenschrift, 105: 358 ( 1980). 9. WHO Technical Report Series, No. 674, 1982 (Treponemal infections: report of a WHO Scientific Group). 37 10. Luger, A. Bulletin of the World Health Organization, 59: 647 (1981). 11. Luger, A. & Catterall, R.D. Syphilis immunology. In: Koning, G.W., ed. Dermatologie in Praxis und Klinik. Stuttgart, G. Thieme, 1981. 12. Luger, A. et al. British journal of venereal diseases, 57: 188 ( 1981 ). 13. Thomas, E.W. Bulletin of the World Health Organizdtion, 18: 269 (1949). 14. Fry, R.M. Lancet, 1: 199 (1938). 15. Hill, A.M. & Butler, H.M. Medical journal of Australia, 1: 293 ( 1940). 16. Patterson, M.J. & El Batool Hafeez, A. Bacteriological reviews, 40: 774 ( 1976). 17. Muller, H.E. Hygiene & medicine, 5: 339 (1980). 18. Baker, C.J. Journal of infectious diseases, 136: 137 (1977). 19. Hood, M. et al. American journal of obstetrics and gynecology, 82: 809 ( 1961). 20. Kexel, G. Zeitschrift fur Hygiene, 151: 336 ( 1965). 21. Kexel, G. & Beck, K.J. Geburrshi/fe und Frauenheilkunde, 25: 1077 ( 1965). 22. Murray, E.G.D. et al. Journal of pathology and bacteriology, 29: 407 ( 1926). 23. Seeliger, H.P.R. Listeriosis. Basie, Karger, 1961. 24. Gray, M.L. & Killinger, A.H. Bacteriological reviews, 30: 309 (1966). 25. Seeliger, H.P.R. & Finger, H. Listeriosis. In: Remington, J .S. & Klein, J .O. Infectious diseases of the fetus and newborn infant. Philadelphia, W.B. Saunders, 1976. 26 . Boysen-M0ller, J. Human listeriosis. Diagnostic, epidemiological and clinical studies. Copenhagen, Costers Bogtrykkeri, 1972. 27 . Ortel, S. Zentralblatt fur Bakteriologie, Parasitenkunde, Infektions- krankheiten und Hygiene, 217: 41 ( 1971 ). 38 Chlamydiae - G.M. Antal Chlamydia trachomatis is an obligate intracellular parasite that has devel- oped a complicated method of replication and is incapable of surviving outside the host cell. It has no known animal reservoir . It often produces subclinical or chronic infections (1 ,2) . The C. trachomatis serotypes A, B, Ba and C are invariably associated with classical endemic trachoma in rural areas of developing countries, while serotypes D to K and L 1, L2 and L), are predominantly sexually transmitted in industrialized communities(]). The organisms are found in the squamous-columnar epithelial cells of the endocervical canal and the urethra . C. trachomatis causes 30-50% of all cases of nongonococcal urethritis and is a leading cause of mucopuru- lent urethritis and endocervicitis; pelvic inflammatory disease and epidi- dymitis (4) are potential complications and they seem to occur at different rates in different countries. A number of other anatomical sites are known to be infected by C. trachomatis. Many genital tract infections with C. trachomatis are asymptomatic. Screening studies on healthy women in San Francisco suggest that nearly 50% of chlamydia! infections are asymptomatic; likewise, men may have asymptomatic chlamydia! infections of the urethra and most of them will be found to have asymptomatic pyuria. Maternal infection C. trachomatis has recently been recognized as one of the most prevalent genital pathogens in pregnant women, the organism having been detected in 2-12% of women in various studies in the United States (5-7). In Europe, a recent study in Landskrona, Sweden, showed that C. tracho- matis was present in specimens from 23 (8. 7%) out of 273 women (8) six and seven days after delivery. The isolation rate is comparable to that obtained in a group of women at less than 19 weeks of gestation in Seattle (9) (see Table I). The prospective study in Seattle reported a significant association between cervical chlamydia! infection in the first half of pregnancy and perinatal mortality. Stillbirth and neonatal death occurred for 6 (33%) of the 18 women with C. trachomatis and 8 (3 %) of the 238 uninfected women. The ri sk of perinatal death was 10 to 18 times higher for infected women than for matched noninfected women. Congenital human (beta) herpesvirus 5 infection or congenital anomalies were found in none of the six cases associated with chlamydia! infection, but in three of the eight cases not associated with chlamydia! infection. If these findings are corroborated by other studies, and if they can be attributed to infection itself, maternal infection must be sought not only at term but also in early pregnancy. 39 Table 1. Pos1t1ve ch lamydia! isolation rate In postpartum women in Sweden and In pregnant women In the United States in each age rang e Landskron a. Sweden Sea ttl e. USA Ag e (years) No. of No. o f No. of No. of ,nfect, ons women % 1nfect1ons women % found tested found tes ted :::a; 19 6 25 24 4 11 49 22 4 20-24 11 108 10 .2 6 107 5.6 ~25 6 140 4 3 11 2 09 To tal 23 273 8 7 18 268 6 7 Among pregnant women, chlamydial infection occurs more commonly in younger persons, the unmarried, during first pregnancies, and among mothers of low socioeconom ic sta tus. The lower prevalence of genital chlamydial infection among older age groups suggests that such infections a re often se lf-limiting and/or that these women are less sexually active. Perinatal morbidity An infant passing through the infected birth canal provides a sensitive indicator of chlamydial infection of the cervix. A number of prospective serological studies have shown that 60-70% of exposed infants acquire infection (5-7). Approximately 25-50% of exposed infants will develop conjunctival infection and conjunctivitis (5-7), most of which presents during the first two weeks of life. If left untreated, the infection usually runs a prolonged but often benign and self-limiting course, lasting from a few weeks to months. In Europe, data on chlamydial infection of the eye are derived from a prospective study conducted in Lands krona in Sweden, which showed that 22% of infants delivered of infected mothers had chlamydiae isolated fro m the conjunctiva (8). None developed conjunctivitis, however, probably because of early treatment. In the United Kingdom , a study conducted in London in 1965 showed that C. trachomatis caused ophtha lmia neo- natorum 5 to 6 times more often than did Neisseria gonorrhoeae ( 10); since then, chlamydial infection has increased in frequency . 40 Studies carried out in the United States suggest that approximately 10-20% of infants born to chlamydia-infected mothers will develop pneu- monia and 30-50% of pneumonias in the first six months of life are caused by C. trachomatis ( 1, I 1, I 2). A few cases of chlamydia-associated pneumonia have also been reported from Europe (in Sweden (13) , France (14) and the United King- dom (15)). Beem & Saxon (16) described a clinically distinct syndrome of chlamydia! lung infection in newborn infants characterized by a gradually worsening, pertussis-like cough that generally starts at 3 to 4 weeks of age. It constitutes a series of closely spaced staccato coughs but without the inspiratory whoop. Clinical findings are usually sparse. This syndrome is not usually seen in children over six months of age. Virtually all chlamydia! pneumonia is accompanied by high levels of IgM antichlamydial antibody titres (] 7). A retrospective study of 37 seropositive children and matched controls showed a statistically significant difference only for the history of pneu- monia ( I 8). Nine out of 13 cases in the seropositive group occurred during the first year of life, compared to none in the control group. References I. Schachter, J. Chlamydia! infections. New England journal of medicine, 298: 428-435; 490-495; 540-549 ( 1978). 2. Taylor-Robinson, D.H. & Thomas, B.J. The role of Chlamydia tracho- matis in genital tract and associated diseases. Journal of clinical pathol- ogy, 33: 205-233 ( 1980). 3. Grayston, J.T. & Wang, S.P. New knowledge of chlamydiae and the diseases they cause. Journal of infectious diseases, 132: 87-105 ( 1975). 4. Berger, D.E. et al. Chlamydia trachomatis as a cause of ideopathic epididymitis. New England journal of medicine, 298: 301-304 ( 1978). 5. Chandler, J.W. et al. Ophthalmia neonatorum associated with ma- ternal chlamydia! infections . Transactions of the American Academy of Ophthalmology and Otolaryngology, 83: 302-308 ( 1979). 6. Hammerschlag, M.R. et al. Prospective study of maternal and infantile infection with Chlamydia trachomatis. Pediatrics, 64: 142-148 (1979). 7. Schachter, J. et al. Prospective study of chlamydia! infection in neo- nates. Lancet, 2: 377-380 ( 1979). 8. Mardh, P.A. et al. Colonisation of pregnant women and puerpural women and neonates with Chlamydia trachomatis. British journal of venereal diseases, S6: 96-100 (1980). 9. Martin, D.H. et al. Prematurity and perinatal mortality in pregnancies complicated by maternal Chlamydia trachomatis infections. Journal of the American Medical Association, 247: 1585-1588 ( 1982). 41 10. Dunlop, E.M.C. et al. Genital infection in association with TRIC virus infection of the eye. III. Clinical and other findings. Preliminary report. British journal of venereal diseases. 40: 33-42 (1964). 11 . Hammerschlag, M.R. Chlamydia! pneumonia in infants . New England journal of medicine, 298: 1083 ( I 978). 12. Harrison, H.R. et al. Chlamydia trachomatis infant pneumonias; com- parison with matched controls and other infant pneumonitis. New England journal of medicine, 298: 702 ( 1978). 13. Hallberg, A. et al. Pneumonia associated with Chlamydia trachomatis infection in an infant. Acta paediatrica scandinavica, 68: 765-767 ( 1979). 14. Couvreur, J. et al. Pneumonie interstitielle a Chlamydia trachomatis chez le nourrisson. Nouvelle presse medica/e, 9: 3245 ( 1980). 15. Dunlop, E.M.C. et al. Subclinical pneumonia due to serotype D-K of Chlamydia trachomatis. British journal of venereal diseases, 56: 337 ( 1980). 16. Beem, M.O. & Saxon, E.M. Respiratory tract colonization and a dis- tinctive pneumonia syndrome in infants infected with Chlamydia tra- chomatis. New England journal of medicine, 293: 306-310 (1977). 17. Schachter, J. & Dawson, C.R. Chlamydia! infections, a world wide problem : epidemiology and implications for trachoma therapy. Sex- ually transmitted diseases, 8(Suppl.): 167-174 (1981). 18. Black, S.B. et al. Serological evidence of chlamydia! infection in child- ren. Journal of pediatrics, 98: 65-67 ( 1981 ). 42 Toxoplasmosis - H. Aspock The first convincing evidence of congenitally acquired toxoplasmosis in man was demonstrated in 1939 by Wolf, Cowen and Paige. a This led to a worldwide intensification of research on toxoplasmosis, the organism, Toxoplasma gondii, having been described as early as 1908. More than 10000 papers and almost 100 monographs have been published on toxo- plasmosis in man and animals and today many of the questions relating to the biology and medical importance of T. gondii have been clarified. There are sti ll a number of unsolved problems, however, and considerable con- troversy st ill surrou nds certain aspects of congenitally acquired infection. There follows an outline of the current understanding of the nature of congenitally acquired toxoplasmosis, which provides a framework for preventive measures. T. gondii has a worldwide distribution and is one of the most prevalent parasites among mammals and birds. The whole lifecycle can , however, be completed only in the intestine of the cat (and other species of Felidae). Only these animals excrete the oocytes that become infective after a period of maturation that lasts for a few days. Under favourable conditions oocytes remain viab le for many months. All other animals and birds, as well as man, may be infected by T. gondii, but once ingested the parasite multiplies only asexually by endogenesis, forming pseudocysts and cysts, and will not be excreted. Raw or insufficiently cooked meat may contain such cysts. Pork and to a lesser extent beef and mutton are the most important source of human infection. In addition, infection with oocytes by direct or indirect contact with cats may play a considerable role, particularly in rural regions and under conditions of poor hygiene. Besides these two sources and apart from laboratory-acquired infection there are probably no other methods by which humans may acquire the infection after birth. Owing to the wide distribution of T. gondii, human infections are very frequent, alt hough there are considerable regional differences. Infection rates increase gradually with age: in general, in Europe, about 20% of 20-year-olds and 50% of 50-year-olds have antibodies. Postnatally acquired infection is often subclinical, a lthough mild or moderate nonspecific fea- tures such as fever, malaise or headache may occur. In some cases a marked and persistent lymphadenopathy may occur, particularly involv- ing the cervical lymph nodes. Severe and let ha l cases oftoxoplasmosis may rarely occur, but almost exclusively in immunologically compromised individuals. a Wolf, A. et al. Toxoplasmic encephalomyelitis. Tra11sac1io11s ofrhe American Neurolog- ical Associalion, 65: 76-79 ( I 939). 43 Infection with T. gondii results in an immunity that probably persists for life. It has been suggested that immunity is maintained by means of antigenic stimulus derived from the central nervous system and in the musculature . If acquired during pregnancy, a primary infection may infect the developing fetus, the pathogen spreading from a focus in the placenta . Such infection may occur at any time during pregnancy, but it is generally accepted that the risk is much lower at an early stage than at a later stage of pregnancy. The risk of fetal infection as a result of a primary maternal infection is about 50% (about 15% in the first trimester and about 70% in the third trimester) . Factors such as gestational age, strain virulence, and individual immunological factors may influence the course of maternal and fetal infection. During very early gestational life, abortion is a lmost invariable. If the infection occurs later, particularly in the second half of pregnancy, fetal damage may result, but the classical triad of features in prenatal T. gondii infection (hydrocephalus, intracerebral calcifications, chorioretinitis) is relatively rare. Nevertheless, clinically inconspicuous lesions present a t the time of delivery may later become apparent, particu- larly the development of ophthalmological complications or mental retardation. It is important to emphasize that previously acquired infection, result- ing in immunity, precludes infection in pregnancy. Thus only during a single pregnancy may a woman be at risk of acquiring T. gondii. Recurrent abortion due to the presence of latent toxoplasmosis, although often discussed in the literature, does not occur. The diagnosis of infections with T. gondii is based almost exclusively on serological findings. The most important and conclusive tests are the complement fixation test (CFT), the indirect immunofluorescent tests (IIFT) including the Sabin-Feldman test (SFT) and the fluorescent anti- body test (FAT), and the II FT for the detection of specific IgM antibodies. The enzyme-linked immunosorbent assay (ELISA) may also be of value. The indirect haemagglutination test (IHA) provides conclusive evidence of infection only under certain circumstances. The appearance of different antibodies, detectable by different tests, is correlated with the time of infection in Fig. I. The dynamics of different antibody responses can be determined precisely so that in most cases a combination of tests employ- ing paired sera provides evidence of whether antibodies are of recent origin, and the fetus is at risk , or are longstanding. Immediate therapy at the very beginning of primary infection during pregnancy may prevent fetal infection or at least reduce the risk of severe damage. A combination therapy of pyrimethamine and sulfametoxydia- zine is the treatment of choice. Pyrimethamine may be teratogenic, how- ever, and it should not be given during the first trimester of the pregnancy (or even up to the 20th week). In the early stage of pregnancy spiramycin may be used, but its effectiveness is open to some doubt. 44 SFT IHA 16000 4000 1000 CFT 320 160 256 80 40 64 20 10 16 5 Fig . 1 . Dynamics of titres in several tests after infection with Toxoplasma gondi, 2 3 4 5 6 7 Months 8 9 10 11 12 Source.· Aspock, H. D,e D,agnos t ,k der Toxoplasma- lnfekt,onen M edmn,sche Labora1orwm. 33. 240-24 7 ( 1980) Obligatory surveillance for the prevention of toxoplasmosis has been carried out in Austria since 1975. Fig. 2 is a simplified diagram of the procedures and the consequences depending on the results of the tests. The basic tests are the FAT and SFT on the one hand and the CFT on the other. If the results are inconclusive, attempts to detect specific IgM antibodies are made. As soon as possible after the beginning of pregnancy, every woman is tested by FAT or SFT. If she proves to be serologically negative, she is tested again in the second and third trimester to determine whether she has become infected between tests. Seronegative pregnant women are strongly advised not to eat raw meat and to avoid any contact with cats. If a low or moderate titre is found in the first test, i.e. I: 16, I: 64 or I: 256, a second blood sample is taken about ten days later to determine whether the titre has remained stable and low thus reflecting a previous and therefore harmless infection, or risen significantly as a result of a primary infection. Sera with titres of~ I : 1000 should be re-tested and tested by other methods such as CFT. Recent infection is suspected to have occurred if the first blood sample, when tested by FAT or SFT, has an exceedingly high titre (i.e. I : 4000 or higher) or if a seroconversion from negative to positive occurs with a significant rise in titre between paired serum samples. A 45 .i,. °' Fig. 2. Simplified diagram of toxoplasmosis surveillance during pregnancy in Austria FAT negative • repeat In 2nd and 3rd trimester \ First test at the beg1nn1ng of pregnancy FAT 1 : 16. 1 : 64 or1 : 256 • repeat after 1 0 days for verif1cat1on of old 1nfect1on l FAT : 1000 .. 1mmed1ate repeat usrng FAT. lgM-FAT. SFT. CFT I In case of suspIcIon of primary rnfectron ------1►► therapy FAT ;;;1 : 4000 suspicion of primary 1nfectron therapy • 1mmedrate repeat using FAT. lgM-FAT. SFT. CFT i dec1s1on to continue or end therapy significant rise must be at least over 16-fold, i.e. two steps in the test system. To save time, treatment is started immediately, although the suspicion may prove unjustified in many cases. Further tests are then carried out, includ- ing another FAT and SFT, or CFT, tests for the detection of specific lgM antibodies (FAT or ELISA) as well as IHA. The data that will be obtained from these tests will , in most cases, definitely clarify whether the antibodies are longstanding or recently acquired and whether the therapy has to be continued or stopped. Under this surveillance scheme, more than 70 000 pregnant women have been examined in the laboratory of the Institute of Hygiene at the University of Vienna over the past 7 years (Table 1 ). All the data collected have been computerized and they show that about half (53 %) of all pregnant women were seronegative at the beginning of their pregnancies and therefore susceptible (Table 2). About half had antibodies, thus being protected as most of these resulted from previous infections: the frequency of primary infections in pregnancy in eastern Austria for the period 1975-1982 was 0.27% (Table 3). The surveillance for the prevention of toxoplasmosis during pregnancy has proved to be a success. This may be concluded from the fact that the incidence of congenitally acquired T gondii in Austria has decreased significantly during the past few years: it is probably 50-70% lower than it was about 10 years ago. Nevertheless further reduction is possible and might be achieved by the following action. 1. Intensification of health education programmes for pregnant women to ensure that strict prophylaxis measures are used. Tab le 1. Number of women tested and tests made. 1 October 1975 - 30 April 1982 Pregnant women tested 70 893 Blood samples taken 102 500 Tests carr ied out FAT 102 500 CFT 7 774 SFT 3 598 lgM -FAT 2 050 ELISA -6000 IHA -6000 47 Table 2 . FAT t itr es in first tests on 70 893 pregnant women. 1 October 1975 - 30 April 1982 Titre No . of women % Nega t ive 38 056 53 .68 Total pos iti ve: 32 837 46.32 : 16 16 74 2 23 .62 : 64 12 89 4 18.18 : 256 2 574 3 63 : 1 000 534 0 75 : 4 000 89 0 .13 . 16 000 4 0 .0 1 2. More reliable test systems and better preventive measures insti- tuted by gynaecologists and pregnant women themselves . 3. More rapid serological test systems, so that every serum sample can be put through at least one of the basic tests within one day. 4. The introduction of more reliable methods for the detection of specific lgM antibodies. Unfortunately tests carried out at present (mainly IIFT and ELISA) often yield false positive results (even when a false positive result due to the presence of a rheumatic factor has been ruled out). In addition, false negative results may occur although much more rarely. The immunosorbent agglutination assay may prove to be an ef- ficient alternative test. By using tests to detect a specific IgM, a diagnosis of primary infections with low titres (i.e. at the very onset of infection) could be made by the examination of only a single serum sample. 48 Table 3 . The frequency of primary infection. 1 October 197 5 - 30 April 1 982 Pregnant w om en tested Suspici on o f recent 1nfect1on Recent 1nfec t1 on verified 70 89 3 1 000 191 (027 %) A striking improvement in toxoplasmosis survei ll ance cou ld be ach ieved by testing all women before pregnancy, for example during the last school-year or at the time of marriage, as is done in France. Simple tests co uld thus be used that would be insufficiently sensitive for su rvei l- lance during pregnancy (e .g. the IHA or some of the commercia l ELISA kits). Serological examination during pregnancy could then be restricted to seronegative women. Finally, a ttention should be drawn to another problem. Although women as well as gynaecologists a nd o bstetricians are concerned about toxoplasmosis during pregnancy, their interest tends to wane after the delivery of an apparen tl y healthy child. Nevertheless, even if, when a primary maternal infection has occurred the newborn infant is apparently normal, it is essential that carefu l serological studies be carried out in infancy, so that it can be determined whether the antibody pers ists or declines and further information can be obtained relating to the progress of the child developmentally. There are still many questions relating to such asymptomatic cases. Although more information is needed, it may be possible to co ll ect this only by enforcing legis lation to guarantee a careful medical fo llow-up of children at risk. Bibliography Aspock, H. Die Diagnostik der Toxoplasma-Infekt ionen . Medizinische laboratorium, 33: 240-24 7 ( 1980). Aspock, H. et al. Aktuelle Probleme der Serodiagnostik im Rahmen der Toxoplasmose-Oberwachung wahrend der Schwangerschaft. Mittei- lungen der osterreichischen Gesel/schaft fiir Tropenmedizin, 3: 20-25 ( 1981 ). Broadbent, E.J. et al. Screening for toxoplasmosis in pregna ncy. Journal of clinical pathology, 34: 659-664 ( 198 1 ). Couvreur, J. et al. La toxo plas mose congenitale traitee. Resultats cliniques et bio logiques. Anna/es de pediatrie, 27: 647-652 ( I 980). Desmonts, G. T oxoplasmose et interruption de grossesse. Journal de gyneco/ogie, obstetrique et biologie de la reproduction, 5: 645-649 ( I 976). Desmonts, G. & Couvreur, J. Congenital toxoplasmosis . A prospective study of 378 pregnancies. New England journal of medicine, 290: 1110- 1116 {1974). Desmonts, G. & Couvreur, J. Congenital toxoplasmosis. A prospective study of the offspring of 542 women who acquired toxoplasmosis during pregnancy . In: Perinatal medicine, 6th European Congress, Vienna, 1978. Stuttgart , G. Thieme, 1979, pp. 51-60. 49 Desmonts, G. et al. Immunoglobulin M-immunosorbent agglutination assay for diagnosis of infectious diseases: diagnosis of acute congenital and acquired Toxoplasma infections. Journal of clinical microbiology, 14: 486-491 ( 1981 ). Devroede, M. et al. Congenital toxoplasmosis: late appearance of retinal lesions after treatment. A eta paediatrica scandinavica, 68: 76 I-762 ( 1979). Flamm, H. & Aspock, H. Die Toxoplasmose-Uberwachung der Schwanger- schaft in bsterreich - Ergebnisse und Probleme. Piidiatrie und Grenz- gebiete, 20: 27-34 (1981). Flamm, H. et al. Die Toxoplasmose-Untersuchung von Schwangeren und Neugeborenen . dsterreichische llrztezeitung, 30: 15-17 ( 1975). Hengst, P. Zur Haufigkeit und Entwicklung der Kinder mit la tenter kon- nataler Toxoplasma gondii-Infektion. Angewandte Parasitologie, 20: 216-221 (1980). Lambotte, R. Toxoplasmose congenitale. Journal de gynecologie, obste- trique et biologie de la reproduction, 5: 265-269 ( 1976). Migliorini, A. & Ragazzini, G. Studio clinico-statistico sulla toxoplasmosi in gravidanze nel Valdarno Aretino [Clinical/statistical study on toxo- plasmosis during pregnancy in Valdarno Aretino]. Minerva ginecolog- ica, 33: 321-324 (1981) . Miller, E.-C. Die Bedeutung der Toxoplasmose fur die Schwangerschaft. Zeitschrift fur iirztliche Fortbildung, 71: I 097-110 I ( 1977). Naot, Y. et al. IgM enzyme-linked immunosorbent assay test for the diagnosis of congenital Toxoplasma infection. Journal of pediatrics, 98: 32-36 (1981). Piekarski, G. Die Toxoplasmose. Infektionswege, Diagnostik und thera- peutische Konsequenzen. Medizinische Welt, 28: 1582-1586 (1977). Piekarski, G. Die Toxoplasmose. Infektionswege, Diagnostik, therapeu- tische Konsequenzen. Gyniikologie, 10: 9-14 (1977). DeRoever-Bonnet, H. et al. Follow-up of children with congenital Toxo- plasma infections and chi ldren who became serologically negative after I year of age, all born in 1964-1965. In: Perinatal medicine, 6th Euro- pean Congress, Vienna, 1978. Stuttgart, G. Thieme, 1979, pp. 61-75. Schmelz, P. Toxoplasmose und Schwangerschaft, Bio Med, 4: 21-25 ( 1979). Schmelz, P. Toxoplasmose und Schwangerschaft. Bio Med, 5: 14-1 8 ( 1979). Stray-Pedersen, B. Toxoplasmosis in Norway. In: Perinatal medicine, 6th European Congress, Vienna, 1978. Stuttgart, G. Thieme, 1979, pp. 89-91. Stray-Pedersen, B. A prospective study of acquired toxoplasmosis among 8043 pregnant women in the Oslo area. American journal of obstetrics and gynecology, 136: 399-406 ( 1980). 50 Stray-Pedersen, B. Infants potentially at risk for congenital toxoplasmosis. American journal of diseases of children, 134: 638-642 ( 1980). Stray-Pedersen, B. & Lorentzen-Styr, A.-M. Uterine Toxoplasma infection and repeated abortions. American journal of obstetrics and gynecology, 128: 716-721 (1977). Stray-Pedersen, B. & Lorentzen-Styr, A.-M. The prevalence of Toxo- plasma ant ibodies among 11 736 pregnant women in Norway. Scan- dinavian journal of infectious diseases, 11 : 159- I 65 ( 1979). Stray-Pedersen, B. & Lorentzen-Styr, A.M. Epidemiological aspects of Toxoplasma infections among women in Norway. Acta obstetrica gynecologica scandinavica, 59: 323-326 ( 1980). Stray-Pedersen, B. et al. Estimations of the incidences of Toxoplasma infections among pregnant women from d ifferent areas in Norway. Scandinavian journal of infectious diseases, 11 : 247-252 ( 1979). Thalhammer, 0. Die Toxoplasmose. Untersuchung von Schwangeren und Neugeborenen. Wiener klinische Wochenschrift, 87: 676-68 I ( I 975). Thalhammer, 0. Int roductory remarks - some problems surrounding prevention of prenatal Toxoplasma in fection. In: Perinatal medicine, 6th European Congress, Vienna, 1978. Stuttgart , G. Thieme, 1979, pp. 44-51. Thalhammer, 0. Toxoplasmose in der Schwangerschaft. Milleilungen der osterreichischen Sanitiitsverwaltung, 81 : 124-127 ( 1980). Thalhammer, 0. Toxoplasmose. Deutsche medizinische Wochenschrift, 106: 1051-1053 (1981). Thalhammer, 0. & Heller-Szollosy, E. Erfahrungen mit routinemaBigem Toxoplasma-Screening bei Schwangeren zwecks Verhutung angeborener Toxoplasmose. Eine prospektive Un tersuchu ng. Wiener klinische Wochenschrift, 91: 20-25 ( 1979). Werner, H. et al. Interpretation des Nachweises spezifischer lgM-A nti- korper bei Verdacht auf konnatale Toxoplasmose. Zeitschrift fiir Geburtshilfe und Perinatologie, 180: 438-444 ( 1976). Wildfiihr, G. & Wildfiihr, W. Toxoplasmose. Jena, VEB Gustav Fischer, 1975. Williams, H. Toxoplasmosis in the perinatal period. Postgraduate medical journal, 53: 614-617 (1977). Wilson, C.B. & Remington, J.S . Preventio n of congenital toxoplasmosis. A viewpoint from a laboratory in the United States. In: Perinatal medi- cine, 6th European Congress, Vienna, 1978. Stuttgart, G. Thieme, 1979, pp. 76-89. Wilson, C.B. et al. Development of adverse sequelae in children born with sub- clinical congenital Toxoplasma infectio n. Pediatrics, 66: 767-774 ( 1980). Wolf, A. et al. Toxoplasmic encephalomyelitis. Transactions of the Ameri- can Neurological Association, 65: 76-79 ( 1939). Zamora, A. Toxoplasmose et grossesse. Soins gynecologiques et obstetricaux et la puericulture, 3: 15-17 ( 1981 ). 51 Urinary tract infections - K. Menzel, N. Dziambor & A. Lemmer It has been known for a long time that urinary tract infection during pregnancy tends to affect the outcome of the pregnancy ( 1-4) and may lead to pyelonephritis in the mother. Increased prematurity and perinatal mor- tality may be the consequences of apparent infections during pregnancy. Observations in the United States have shown that asymptomatic bacteri- uria is present in 4-7% of all pregnant women (3,5) and that if it is not treated, there is a 20% risk of developing pyelonephritis. If, however, bacteriuria is successfully treated the risk of developing pyelonephritis decreases to exceedingly low levels. Similar observations have been made by a number of other authors (6-9). Hypotrophic newborn infants are found more often in women with bacteriuria than in healthy women (4, 10-15). The rate of stillbirth and prematurity is also higher in untreated or unsuccessfully treated pregnant women with bacteriuria and pyelonephritis than in comparable control groups (4, 7, 16-18). The rate of congenita l anomalies is greater in children born to women with bacteriuria and pyelonephritis than in children born to healthy mothers (7, 19). In a study conducted in Erfurt in 1978 and 1979 by the authors, the following results were obtained. Among 7871 pregnant women, 326 (4.14%) had a pathological urine analysis. Of these 326 women, 252 showed symp- toms of pyelonephritis (77.3%), 68 women showed asymptomatic bacteri- uria (20.9%) and 6 had chronic pyelonephritis before pregnancy. Congenital anomalies, such as the defective development of the heart, face or extremities, were present in 1.32% of chi ldren born to mothers without urinary tract infections whereas defects were present in 4.3% of children born to women with infections of the urinary tract during pregnancy. These data clearly indicate that urinary tract infection during preg- nancy is a severe complication for the mother and for the newborn. References 1. Dodds, G. Bacteriuria in pregnancy, labour and the puerperium. Jour- nal of obstetrics and gynaecology of the British Commonwealth, 38: 773 (1931). 2. Dodds, G. The immediate and remote prognosis of pyelonephritis of pregnancy. Journal of obstetrics and gynaecology of the British Com- monwealth, 39: 46 ( 1932). 3. Norden, C.W. & Kass, E.H. Bacteriuria of pregnancy - a critical appraisal. Annual review of medicine, 19: 431 (1968). 52 4. Savage, W.E. et al. Demographic and prognostic characteristics of bacteriuria in pregnancy. Medicine, 46: 385 ( 1967). 5. Kincaid-Smith, P. Bacteriuria in pregnancy. In: Kass, E.H. Progress in pyelonephritis. Philadelphia, Davies, I 965, p. 11 . 6. Le Blanc, A.L. & McGanity, W.J. The impact of bacteriuria in preg- nancy - a survey of 1300 pregnant patients. Texas reports on biology and medicine. 22: 336 ( 1965). 7. Kincaid-Smith, P. & Bullen, M. Bacteriuria in pregnancy. Lancet, 1: 395 ( 1965). 8. Little, P.J. The incidence of urinary infection in 5000 pregnant women. Lancet, 2: 925 (1965). 9. Condie, A.P. et al. Complications of bacteriuria in pregnancy. In: O'Grady, F. & Brumfitt , W. Urinary tract infection. Oxford, Oxford University Press, 1968, p. 148. IO. Bryant, R. et al. Asymptomatic bacteriuria in pregnancy and its associ- ation with prematurity. Journal of laboratory and clinical medicine, 63: 224 (1964). l l. Stuart, N.L. et al. Bacteriuria pre-eclamptic toxemia, and prematurity. In: Kass, E.H. Progress in pyelonephritis. Philadelphia, Davies, I 965, p. 45. 12. Henderson, M. & Reinke, W.A. The rela tionship between bacteriuria and prematurity. In: Kass, E.H. Progress in pye/onephritis. Philadel- phia, Davies, 1965, p. 27. 13. Wilson, M.G. et al. Effect of bacteriuria on the foetus. New England journal of medicine, 274: l l l 5 ( 1966). 14. O'Grady, E. & Brumfitt, W. Urinary tract infection. Oxford, Oxford University Press, I 968. 15. Whalley, P. Bacteriuria of pregnancy. American journal of obstetrics and gynecology, 97: 723 ( 1967). 16. Kass, E.H. Pyelonephritis and bacteriuria: a major problem in preven- tive medicine. Annals of internal medicine, 56: 46 (I 962). 17. Brumfitt, W. et al. Bacteriuria in pregnancy, with reference to prema- turity and long-term effects on the mother. In: Symposium on pyelo- nephritis, Edinburgh. 1966. 18. Elder, H.A. et al. The natural history of asymptomatic bacteriuria during pregnancy: the effect of tetracycline on the clinical course and the outcome of pregnancy. American journal of obstetrics and gynecol- ogy, 111: 441 (1971). 19. Kass, E.H. Infection and maternal and infant capabilities. In: Kretch- mer, H. & Hasselmeyer, E.G. Horizons in perinatal research. New York, John Wiley & Sons, 1974, p. 30. 53 Detecting prenatal and perinatal infections Laboratory diagnosis by traditional and new techniques - G. Enders The viral and other microbial agents that can affect the product of concep- tion and the health of the newborn or the infant, by prenatal, perinatal and early postnatal infection, are summarized under the American acronym "TORCH" in Table I. For assessing the importance and frequency of these infections during pregnancy and in the newborn, as well as for preventing the infection in the mother and its effect on the infant, an etiological laboratory diagnosis is essential. In recent years traditional methods of diagnosing an infection by demonstrating the agent and/or by detecting the immune response to the infection by serology have been enhanced in sensitivity and speed, and even replaced by new techniques. Table 2 lists some traditional as well as newly devised techniques for demonstrating an infectious agent (viral, bacterial, chlamydia[, myco- plasmal, protozoa[, spirochaetal or mycosal). The most important ad- vance, allowing a direct, rapid and practical assay of the agents' presence in the body fluids or tissues, is probably the detection of antigen by enzyme- linked immunosorbent assay (ELISA). This technique is likely to be improved in its specificity by the incorporation of monoclonal antibodies in the test system. This will enable investigators to distinguish between closely related species and the various antigens of agents present during different stages of infection. To evaluate the specificity and clinical value of some of the new and more sensitive techniques, isolation of the agent by conventional methods is the yardstick against which they must be com- pared. In many instances, however, appropriate clinical specimens are not available, or techniques for detection of the specific agent remain to be developed or are still very time-consuming. Under these circumstances, the detection of a specific immune response in serum or body fluids (such as in the cerebrospinal fluid) may be the only way of making a diagnosis. Such 55 Vt °' T = Toxoplasma gondii Table 1. TORCH : viral and other microbial agents that can cause prenatal. perinatal and postnatal infections 0 = Other infectious microorganisms: Proven : Human (alpha) herpesvirus 3. coxsackievirus B. hepat it is B virus. hepat it is A virus. chlamyd iae. Listeria mono- cytogenes. Neisseria gonorrhoeae. Treponema pallidum. Plasmodium spp .. Trypanosoma cruzi. Group B beta- haemolytic streptococc i. Suspected: Mumps virus. measles virus . Hypothetical : Influenza A virus. lymphocytic choriomeningitis virus. human papilloma virus._ human (gamma) herpesvirus 4 . R = Rubella virus C = Cytomegalovirus (human (beta) herpesvirus 5) H = Human (alpha) herpesviruses 1 and 2 V, --..J Table 2. Techniques for detecting infectious agents in body fluids and tissues Approach Indirect Isolation Identification of isolates Direct Cytology Histology Structure Ant igen Local ant ibody Nucleic ac id 8 New techn1Ques . Technique Tissue-culture. egg. laboratory animal. co-cultivation with perm issive cells 8 or helper virus8 Bacteriological media. laboratory animal Neutralization test. HAI . CF. IFL. CIE. precipitation. agglutination, C'-IF. 8 immune peroxidase stain.8 ELISA8 Smears : colour-IF-staining. Cell pathology ELM. 8 lmmune-ELM 8 CIE. IF, RIA.8 ELISA.8 immunosorbent assay:8 enzyme- Ii nked / haem adsorption/ passive haemagglutination Chemical luminescence.8 time resolved fluoro-immuno- assay (TR-FIA).8 gas-chromatography8 lgM. lgA. lgG: ELISA, 8 RIA8 Hybridization.8 molecular cloning in bacteria8 Agent Viral Others Viral and others Viral and others Viral and others Viral Viral and others Bacterial Viral Viral methods are also required to determine the immune status of pregnant women exposed to organisms that are a risk to the fetus or newborn infant as well as for sero-epidemiological purposes, such as for studies on anti- body prevalence and changes resulting from vaccination programmes. The serological diagnosis of acute primary or reactivated infection is usually made by demonstrating seroconversion, i.e. a significant increase in antibodies between serum samples taken in the acute and later stages of infection. Nevertheless, the recognition that specific IgM and IgA are markers of acute primary or sometimes reactivated postnatal infection, as well as of pre- and perinatally acquired infection, now permits a more rapid serodiagnosis on only a single serum sample. As with antigen detec- tion, ELISA tests are evolving as a most promising tool for detecting not only the IgM and IgA antibodies as markers for recent postnatal or pre- and perinatally acquired infection, but also specific IgG antibodies. This provides evidence of previous infection in early infancy, or the pres- ence of maternal antibodies. Perhaps even more valuable for detecting specific lgM and lgA antibodies with respect to specificity is the direct anti-IgM or lgA solid phase immunoassays using enzyme-labelled anti- gens (ELA) rather than labelled antibodies (]). Alternatively non- sensitized or antigensensitized erythrocytes may be used as an indicator, tests being read by haemagglutin~tion (solid phase immunosorbent test (SPIT)) (2) or haemadsorption (haemadsorption immunosorbent test (HIT)) (3). The yardstick for class-specific IgM antibody demon- stration is the sucrose density gradient centrifugation method. It is essen- tial, however, that rapidly developing new methods for IgM class-specific antibodies be compared carefully with existing methods before being introduced for routine diagnosis (see Table 3). For serodiagnosis of acute or recent infection, specific IgM antibody detection should be carried out in conjunction with tests for detecting IgG antibody. This is often carried out by such conventional techniques as the complement fixation (CF) test and the haemagglutination inhibi- tion (HAI) test. A combination of tests is often necessary to give reliable information about the stage, or duration, of the infection. To determine immunity status, the most sensitive and specific method for detecting previously acquired antibodies should be chosen. Although radioimmuno- assays and enzyme-immunoassays are very sensitive for such infections as rubella, measles and influenza, the standard HAI test and the newer haemolysis in gel (HIG) or single radial haemolysis (SRH) test are also satisfactory. Diagnostic kits, though valuable in the hands of the experienced, may lead to erroneous results, particularly in small laboratories without sero- logical experience and without recourse to adequate clinical information with which to interpret the results of investigation. 58 Vl '° Table 3 . Techniques for detecting the immune response to viral and other microbial infections Approach Te chn ique Detection of antibody r ise in 2 sera CF. HAI . passive HI . neutral1zat1on test Prec1p1 tat1 on. agglutination . lys1s tests IF. RIA. ELISA Detection of lgM . lgA. class-specific antibodies 1n the first serum sample8 Detection of cell-mediated immunity (CMI) a New approach Separation : density gradient centrifugation (SDG) Gel filtration . S DG-ch romatography Redu ction of: lgM : lgG: lgG + lgA 2· ME -ethanol protein A prec1p1tat1on w ith ant1-lgG 1mmunosorpt1on with CPG Enzyme-linked rad101mmunoassay Indirect (ELISA. RIA) lmmunosorbent assays : ant1-lgM . ant1 -lgA HIT-SPIT= with haemagglutinat1on or haemadsorpt1on ELA with peroxidase-l abelled antigen Skin tests with purified antigens Macrophage-in h1b1t1on test (MIF) Lymphocyte transformation test (L TF) Agent Viral Others Viral Viral and others Viral /others Viral / o thers V iral / others Rubella virus The assessment of rubella virus infection in pregnancy is carried out only by serological methods, and a great number of techniques and their modifications, including commercial kits, have become available in the last few years for this purpose. The standard method for detecting and measuring rubella antibodies is still the HAI test and, for detecting IgM antibodies, HAI after sucrose density gradient centrifugation (SDG). In addition to making a rapid diagnosis by this method, specific lgM antibo- dies may be detected by indirect ELISA, radioimmunoassay (RIA) and SPIT, or a modification of it, HIT. The specificity and sensitivity of these new techniques are still controversial (4,5). The more regularly such tests are used in laboratories , however, the more the pitfalls of the individual tests will become known and the better will be the interpretation of results. Rubella-specific lgM antibodies are shown to be present in acute rubella from 5 and 30 days after the onset of symptoms by all the methods discussed above. Before and after this time, however, these techniques differ in their capacity to detect a low concentration oflgM antibodies. In a comparison carried out in Stuttgart, ELISA proved more sensitive than either SDG or HIT in detecting low concentrations of rubella IgM anti- bodies present in the early and later stages of infection (5,6). Since rheumatoid factor is sometimes present in pregnancy and may be a source of false positive results in indirect RIA and ELISA, it must be removed before testing. The SDG, although considered to be among the most reliable IgM methods, may vary in its sensitivity as a result of minor day-to-day variations in technique. This is particularly likely to be a problem when high IgG antibody titres are present, often long after the onset of infection. Thus, when only a single serum sample is available which has been collected very early or late, the most sensitive lgM antibody assay test, such as ELISA, should be used. A specific lgM response, but of short duration and at a low level , may occur if persons with vaccine-induced immunity are exposed to rubella virus. Rubella-specific lgM responses may also be detected after rubella vaccina- tion, but although of a lower concentration they may be of a longer duration than those following naturally acquired infection (>2-6 months) (5, 7). When rubella-specific lgM antibody tests are required to rule out an acute or recent rubella virus infection in pregnancy, the HAI, HIG, or other methods for detecting IgG antibody should always be used in paral- lel. The HAI test as used since 1968 has been a very sensitive test for detecting antibodies early after onset of exanthem. The HAI and ELISA Enzygnost detect antibodies in 40% of patients between the first and third day and in 90-100% of patients 4-6 days after the onset of exanthem. The HIG and the Rubazyme IgG test as well as the FIAX test detect immune responses reliably, but can do so regularly only more than 11-20 days after 60 onset of exanthem (5). All tests are therefore suitable for testing immunity status more than 21 days after natural infection and more than 4 weeks after vaccinat io n. In cases of suspected rubella virus infection in pregnancy it is prudent to confirm borderline positive IgM results by repeat testing and to obtain a second serum sample, if possible, to demonstrate a significant antibody rise in addition to IgM antibodies by HAI and HIG (5,6). Since there is no single test that is completely free of pitfalls , a combi- nation oflgM and IgG antibody tests is recommended before a decision is made to terminate a pregnancy. The risk of fetal damage in maternal rubella virus infection is closely related to the gestational age at which maternal infection occurred. As there is little risk of severe fetal damage after the 16th week of gestation, the termination of a pregnancy is usually not recommended following maternal infection after the 16th week of pregnancy (6) . In addition to using serological techniques, attempts should be made to isolate the rubella virus from infants who may have congenitally acquired rubella virus infection . The virus may be recovered from throat secretions, cerebrospinal fluid, urine and, if the infant dies , from various organs. This may best be achieved by inoculation of the specimens into primary Cerco- pithecus aethiops monkey kidney cell cultures or into continuous rabbit kidney cultures (RKIJ) or rabbit cornea cells (SIRC). The presence of rubella virus is identified by interference with echovirus 11 in monkey kidney cell cultures and by a cytopathic effect and indirect immuno- fluorescence in RK 13 or SIRC cell cultures. It is possible that the technique of immunoperoxidase staining of inoculated baby hamster kidney cells may shorten the isolation and identi- fication procedure from 15-20 days to 4-6 days (8). Nevertheless, confirma- tory reports are still needed. The serological diagnosis employing serum or cerebrospinal fluids may be made by techniques used in the diagnosis of postnatally acquired infection. There may be some difficulty, however, in demonstrating the presence of specific IgM antibodies in the cord blood, this being due to materno-fetal transfusion of IgM antibodies. The presence of rubella- specific lgM antibody findings in the cord blood should therefore be confirmed by testing a blood sample obtained during the first to second week of life. In a study at the Virological-Medical Diagnostic Institute in Stuttgart on 375 children with rubella embryopathy over the last 10 years, rubella-specific IgM was found to be present , employing the SDG tech- nique, between the first week of life and 3 months of age in 90-95% of cases, but between the ages of 6 and 18 months this declined from 60% to 15%. More recently, these results have been compared with other tech- niques, including ELISA Enzygnost with Rubazyme M, the SPIT and the HIT (4,5) . 61 Currently available methods and their use in the diagnosis of congeni- tally acquired rubella virus infection, and the combination of tests for the diagnosis of acute , previous and prenatally acquired infection are shown in Tables 4 and 5. Human (beta) herpesvirus 5 The methods for the laboratory diagnosis of cytomegalovirus infection are summarized in Table 6. The detection of human (beta) herpesvirus 5 in body fluids and tissue is still best achieved by isolation in human diploid cell cultures (9). It may take from 4 to 20 days for a cytopathic effect to become visible, depending on the amount of virus in the specimen. This time interval may be shortened by using the anti-complement immuno- fluorescence (ACIF) test (10) which detects early human (beta) herpes- virus 5-specific nuclear antigen. The direct detection of antigen in urine using various ELISA test systems has so far not been very successful, possibly because of the lack of a suitable hyperimmune serum directed against human (beta) herpesvirus 5 envelope antigen to serve as a "catch- ing antibody". The use of monoclonal antibodies may, however, overcome this problem. Although the detection of cytomegalovirus infection by electron microscopy is rapid , a diagnosis is likely to be achieved by this method only when a very large number(> 106/ml) of virus particles are present. Furthermore, this technique is not routinely available. For the detection of human (beta) herpesvirus 5 antibodies, the CF test is reliable provided a potent antigen is used, but it is less sensitive and less rapid than the passive haemagglutination (PHA) test or the indirect ELISA. Indirect immunofluorescence, the reading of which may bring in a subjective element, is no longer used. For measuring antibody levels, the PHA test and ELISA are over I 0-50 times more sensitive than the CF test. The best, most specific and rapid assay for human (beta) herpesvirus 5-specific lgM antibody is the recently introduced anti-lgM immuno- sorbent assay, using a peroxidase-labelled antigen (ELA) ( 1). With indirect ELISA or RIA, however, high levels of lgM-type rheumatoid factor may give rise to false positive results. Adsorption of sera with latex reagents or anti-lgG will effectively remove rheumatoid factor, and should be carried out before testing. Rheumatoid factor may be estimated quantitatively by means of a newly introduced indirect ELISA. For determining the lgM or lgG antibody levels, endpoint titration of serial four-fold dilutions is performed. To simplify the daily test pro- cedure, various methods for examining a single serum dilution in conjunc- tion with a standard curve have been suggested ( 11). The ratio method of multiples of normal activity (Mona) is a direct attempt to correct the deficiencies of reporting raw adsorbence data from an assay of a single serum dilution. Nevertheless, exact measurement, especially of low and 62 0\ \;.) Methods HAI HIGl._ SR'§ ELISA Kit: Enzygnost Rubazyme (Abbott) lgM tests SDG ELISA Kit: Enzygnost Rubazyme (Abbott) Solid phase anti-lgM im munosorbents : SPlrl_ HIT_J Table 4 . Rubella virus serology Va lue Standard Very sensitive Simple Cheap Rap id Sens it ive Simple Standard Rapid . simple. more sensitive than SDG Cheap Problems Specificity in low dilutions Negative early in acute rubella and post-vaccination Expensive Negative early in acute rubella and post-vaccination Time-consuming. expensive Expensive Rheumatoid factor : specificity of low borderline value Some wrong positive results Time-consuming . less sensitive than ELISA and SDG. sub1ect1ve read ing of test results 0\ ~ Table 5. Laboratory diagnosis of rubella virus infection Diagnostic problems Acute and recent infection in pregnancy Immunity status : > 2 weeks post-i nfection > 4 weeks post-vaccination Verification of borderline HAI titres ( 1 : < 3 2) Prenatal infection: cord blood later blood Embryopathy Test combination HAI + HIG + lgM ELISA SRH JEnzygnost8w ~ ubazyme (Abbott) Repeat testing of borderline and posItIve results ~ ~tor HAI (ELISA also possible) HAI + HIG (ELISA also poss ible) SRH Total lgM + lgA HAI + lgM ELIS Af_ HAI + HIG + lgM ELIS~ or ELISA _jE nzyg nost8w 7!l ubazyme (Abbott) Serology plus vi rus 1solat1on attempts from throat. ur ine. cerebrospinal fluid . t issues 0\ VI - Table 6. Methods for laboratory diagnosis of cytomegalovirus infection Methods Virus isolat ion in cell culture with complement AC / IF Viral antigen Viral structure Serology: CF PHA ELISA lgM antibody ELISA ELA Rheumatoid factor (lgM) ELISA ELISA ELM Test results ava ilable 4-20 days 2- 4 days 6 hours 3- 6 hours 2 days 3 hours 6 hours 6 hours 6 hours 6 hours Comments Standard. slow Exper imental stage Exclus ive. rap id Standard Rapid. simp le Rapid. expens ive Cave rheumatoid factor Specif ic. simple. reproduc ible. rap id Quantitat ive. speci f ic. rapid extremely high antibody concentrations, cannot be achieved. This is due to the lack of a linear relationship between extinction values and antibody titres. A more exact measurement can be made, however, using two optimal working dilutions selected from the beginning and the end of the relative linear part of the standard curve. This approach is applicable to both the indirect ELISA and the direct ELA (12). The immunofluor- escence test for measuring lgM antibodies has never been very sensitive and specific with whole serum, particularly in the presence of higher lgG titres, on account both of the IgG binding on the Fe-receptor of cells and of the rheumatoid factor. This test is therefore used less today, and results obtained with it in the past may sometimes be incorrect. Testing for cell-mediated immunity (CMI), an important factor in immunocomprom- ised patients who may experience reactivated cytomegalovirus infection, may be performed by means of the macrophage inhibition factor (MIF) test or lymphocyte transformation factor (L TF) test. These tests are still used more for research purposes. The nucleic acid hybridization technique is a new research tool that will be useful for strain identification and epidemiological studies. Postnatally acquired primary cytomegalovirus infections are usually subclinical or produce ill-defined and clinically often mild symptoms. Appropriate specimens for virus isolations, such as urine, are therefore only infrequently collected. This is even more so in reactivated infection during pregnancy. A diagnosis is therefore more often made serologically. Among newborn infants with suspected congenital cytomegalovirus infec- tion, the differential diagnosis of prenatal, perinatal or early postnatal infection is dependent on the detection of viruria during the first week of life. The laboratory diagnosis of cytomegalovirus infection in pregnancy and in the newborn infant is summarized in Table 7. Human (alpha) herpesviruses 1 and 2 The rapid laboratory diagnosis of human (alpha) herpesviruses I and 2 in pregnant women shortly before delivery is important in preventing virus transmission to the neonate. Methods by which a diagnosis may be rapidly made are summarized in Table 8. By far the best way to detect human (alpha) herpesviruses l and 2 in the genital tract, throat secretions, urine, vesicle fluid, skin lesions, brain biopsy and autopsy tissues is by virus isolation in cell culture. Cytopathic effect usually appears in 1-3 days. Type-specific identification can be performed reliably with the cell culture isolate in six hours using a four- layer sandwich ELISA (13), and by RIA. Direct and indirect immuno- fluorescence and immune peroxidase staining are less specific, and the kinetic neutralization test more time-consuming. 66 °' -..J Diagnostic problems Acute primary infection in pregnancy Acute reactivated infection in pregnancy Prenatal intrauterine infection Perinatal. early postnatal infection Table 7. Laboratory diagnosis of cytomegalovirus infection Methods Virus isolation from urine, throat. in diploid cells Serology: CF. PHA. ELISA. lgG lgM: ELA (ELISA) Rheumatoid factor-ELISA titre Virus isolation from urine. cervix Serology: CF, PHA, ELISA. lgG lgM : ELA (ELISA) Virus isolation from urine before 6 days of age Serology: CF. PHA, ELISA. lgG lgM: ELA (ELISA) Rheumatoid factor-ELISA titre Virus isolation from urine after 1 0 days of age Serology: CF. PHA. ELISA. lgG lgM : ELA (ELISA) Rheumatoid factor-ELISA titre Expected results Human (beta) herpesvirus 5 positive in low titres Antibody conversion lgM-AB rise with slow decline Rise to high levels Human (beta) herpesvirus 5 positive in low titres Antibody rise lgM-AB negative to low-positive Human (beta) herpesvirus 5 positive in high titres for long periods Antibody rise lgM-AB often rise and persist up to 6-8 months Low levels Human (beta) herpesvirus 5 positive in medium to high titres Ant ibody rise lgM-AB rise May be elevated °' CXl Methods Virus isolat ion: (not from cerebrospinal fluid) Bra in biopsy: Antigen test : Serology: Cerebrospinal fluid" Serum Table 8 . Rapid laboratory diagnosis of infection with human (alpha) herpesvirus 1 and 2 with 1st specimen cell culture with identification cell culture ELM- immune sandwich-ELISA antigen detection identificat ion of isolates RIA. ELISA: lgM . lgA. lgG ELA: lgM . lgA RIA. ELISA: lgM. lgA. lgG ELA: lgM. lgA Resu lts + presumptive + spec ific +\.specific. rapid j . (+) less sensitive than isolation· + rap id. specific (+) indicative (+) (+) slow rise (+) indicative Available 1-3 days 3-4 days 2-4 days 4-6 hours 6 hours 6 hours 6 hours 6 hours 8 C aut,on must be exercised because of the poss1bd1ty of 1mmunoglobul1n leakage through the blood / cerebrospmal fluid barrier . and heterotyp1c react,on with human (alpha) herpesv,rus 3 lgG antibody. Clinical specimens may be used to detect hu man (alpha) herpesvirus I and 2 antigens and for typing purposes. In addition, evidence of asympto- matic herpetic cervicitis in the pregnant woman may be obtained by means of the Papanicolaou (Pap) smear. This has a sensitivity approximately two thirds that of the viral cultures (14). Direct immunofluorescence has also been employed for direct detection and typing of human (alpha) herpes- virus I and 2 in vesicle fluid , skin scrapings and brain biopsy impression smears. Non-specific fluorescence may nevertheless make the interpreta- tion of resu lts obta ined with immunofluorescence tests difficult. The imm une peroxidase staining method, which is under investigation, re- quires no specialized equipment. For antigen detection the four-layer sandwich ELISA has already proved useful but is somewhat less sensitive than virus isolation in cell culture. It is, however, an excellent and rapid method for typing isolates. Since there is an understandable reluctance to perform brain biopsy in newborn infants who may have herpesviral encephalitis, diagnosis may be attempted by non-invasive techniques. One method is to demonstrate human (alpha) herpesvirus I and 2 antibody synthesis in the central nervous system by detecting it in the cerebrospinal fluid ( 15). Class-specific IgM, IgA and IgG antibody responses may be demonstrated by the in- direct ELISA and RIA and for IgM and IgA antibodies by direct ELA ( 16). In advanced stages of encephalitis, nonspecific leakage of immunoglobulin thro ugh a damaged blood/brain barrier may occur but this can be recog- nized by determining albumin levels as well as by comparing the titres of human (a lpha) herpesvirus I and 2 antibody in serum and in the cerebro- spinal fl uid with those of such other virus antibodies as rubella virus, human (alpha) herpes virus 3 and measles vi rus. The results of the cerebro- spinal flu id/albumin and seru m/album in ratios a re expressed as the antibody index. In the case of the use of reference ant ibody, the se rum/ cerebrospinal fluid ratio fo r human (alpha) herpesvirus I and 2 must be four times lower than tha t for the reference antibody ratio. The inclusion of the human (alpha) herpesvirus 3 reference serum/cerebrospinal fluid ratio is important since a heterotypic reactivity between human (alpha) herpesviruses I and 2, and human (alpha) herpesvirus 3 may exist in the presence of human (alpha) herpesvirus 3 maternal antibody titre in the newborn (1 7) . The drawback with the serological detect ion of lgM and IgG anti- bodies in cerebrospina l fluid as well as in the serum by ELISA or RIA and ELA is the fact th at in primary human (alpha) herpesvirus I and 2, as well as in the other herpesvirus infections, antibody development is rather slow. Therefore, serological diagnosis with a specimen taken in the first seven days of disease may not be successful. Consecutive cerebrospinal fluid and 69 serum samples have to be investigated to confirm the presumptive diag- nosis (16-18). Significant titre rises oflgM and lgG antibody in the ELISA and RIA tests, as well as in the old CF test, only become evident between the first and second week after the onset of the disease. There also exist types of ELISA (19) and RIA that can be used to differentiate between human (alpha) herpesvirus I and 2 antibodies (20.21). With an indirect RIA using envelope-, capsid- and cell-extract protein antigens, a distinction can be made between primary and recurrent human (alpha) herpesvirus I and 2 infection in addition to measuring the class-specific lgM and lgG responses to the various antigens. This and the nucleic acid hybridization technique for strain identification and epidemi- ological studies are at present still of interest for research. The laboratory diagnosis of human (alpha) herpesvirus l and 2 under various circumstances is summarized in Table 9. Human (alpha) herpesvirus 3 Only 5% of women of childbearing age are still seronegative and suscep- tible to human (alpha) herpesvirus 3 infection (12,22), but it is recognized that maternal infection in early pregnancy may result in congenital defects, while infection occurring in the last four days of gestation or up to 48 hours after delivery can result in severe and often fatal generalized infection of the newborn (23). The rapid and specific serodiagnosis of infection and immunity status to human (alpha) herpesvirus 3 in pregnancy is important, particularly because after exposure varicella may be prevented or attenuated in suscep- tible pregnant women or neonates by the administration of high titred zoster immunoglobulin. This should be given 48-72 hours after exposure, or as soon as possible after birth. The conventional and newer methods for virus and antibody detection are shown in Table 10. Human (alpha) herpesvirus 3 or its antigens can rapidly be detected in vesicle fluid by counter current immunelectrophoresis (CIE) or electron microscopy (24). As vesicle fluid may not be available, a serological diagnosis is often made. Conventional methods such as the CF test, the neutralization test, the immune adherence haemagglutination (IAHA) test and fluorescent antibody test to membrane antigen (FAMA) mainly detect virus-specific lgG antibody responses. Two serum samples taken 6-8 days apart are required to demonstrate a significant rise in antibody titre, but if there is a delay in obtaining the first serum sample, the antibody titre may already have reached its maximum level. A heterotypic antibody rise to human (alpha) herpesvirus l and 2 may occur in 20% of patients with pre-existing human (alpha) herpes virus I and 2 antibody ( I 2). Since the 70 --.J Table 9 . Laboratory diagnosis of infection with human (alpha) herpesvi rus 1 and 2 Diagnostic problems Primary 1nfect1on in pregnancy Reactivated infection In preg- nancy before delivery Neonatal infection In new- born (seps1s-encephal1t1s) Immunity status in childbearing age Methods Virus 1solat1on from lesions. throat. urine in cell cul ture Identifi ca tion : ELISA. neutralization test Serology: CF. ELISA. lgG lgM antibody: ELA (ELISA) Virus 1solat1on from lesions. cervix. urine Serology: CF. ELISA. lgG lgM antibody: ELA (ELISA) Virus 1solat1on from lesions 1f present. from brain biopsy including immune-ELM Serology: cerebrospinal flu1dl ELISA. ELA serum J Serology (CF) ELISA Viral excretors in childbearing Virus isolation: age from cervix from urine Expected results Human (alpha) herpesv1rus 1 or 2 posItIve Antibody rises slowly lgM antibodies are of low titre Human (alpha) herpesv1rus 1 or 2 : posItIve Human (alpha) herpesv1rus 1 and 2 : may be posItIve Antibody seldom rises lgM antibody seldom posItIve Human (alpha) herpesv1rus 1 and 2 : posItIve Human (alpha) herpesv1rus 1 and 2 : posItIve lgM. lgA. lgG. human (alpha) herpesv1rus 1 and 2 antibodies posItIve and human (alpha) herpesv1rus 3 antibodies should be tested too 90% posIt Ive for human (alpha) herpesv,rus 1 and 2 30-60% posItIve for human (alpha) herpesv1rus 1 and 2 20 - 40% human (alpha) herpesv,rus 2 0 .95% posItIve 0 .50% pos,t,ve --..J N Table 10. Methods for laboratory diagnosis of human (alpha) herpesvirus 3 infection Methods Virus 1solat1on in cell culture with complement immunofluorescence Viral antigen (in vesicle fluid) : 1mmunofluorescence staining ELM counter-immunelectrophoresis Serology: CF IAHA FAMA ELISA lgM-lgA antibody: ELISA ELA Test results available 3-14 days 2- 8 days 6 hours 3 hours 2 hours 2 days 1 day 6 hours 6 hours 6 hours 6 hours Comments Slow. laborious Less specific Exclusive Rapid. spec1f1c Standard. laborious Simple. spec1f1c1ty7 Difficult Rapid. easy Rapid . sensItIve. cave rheumatoid factor Rapid . specific. sensItIve presence of virus-specific IgM and IgA antibodies is indicative of either a primary human (alpha) herpesvi rus 3 infection or a reactivation in the form of zoster, their detection is of major importance in obtaining a specific rapid serodiagnosis, for only a single sample is required ( 12) . Although IgM antibody may be detected by immunofluorescence and FAMA, these tests are less reliable than a recently introduced commer- cially available indirect enzyme-linked immunosorbent assay (Enzygnost). This test detects human (alpha) herpesvirus 3-specific IgM , IgA and IgG antibodies using a single microtitre plate. False positive IgM results may be caused by the presence of rheumatoid factor, which occurs commonly in pregnancy (25), but adsorption of sera with latex reagent or anti-IgG before testing will obviate this problem. Occasionally very high IgG anti- body concentrations may reduce the sensitivity of the virus-specific IgM response. This may be overcome by labelling the antigen, a technique for detecting virus-specific IgM responses to cytomegalovirus infection (1) and human (alpha) herpesvirus I and 2 (16). Nevertheless, occasionally even this ELA test can give rise to false positive results, because of the presence of antinuclear antibodies directed against antigens consisting of semi-purified virus preparations from human diploid cell cultures, con- taining cell nuclei . This problem may be avoided by the addition of uninfected control antigen to the labelled antigen before use. If sera are collected from 3-5 days to 2-4 months after the onset ofvaricella or zoster, a diagnosis may readily be established by means of detecting class-specific antibodies. The test may be completed in 6 hours. Table 11 illustrates the use of some of the tests described above in the diagnosis of human (alpha) herpesvirus 3 infection or reactivation. Mumps virus Although the association between mumps virus infection in pregnancy and endocardial fibroelastosis (EFE) and hydrocephalus in the newborn has never been clearly established, requests for the laboratory diagnosis of mumps virus infection in pregnancy or the immunity status of patients exposed to mumps are quite often made (22). In most countries, however, mumps is rare in pregnancy, since most women of childbearing age are already immune. According to a sensitive ELISA IgG test 97.2% of women of childbearing age in the Federal Republic of Germany possess mumps virus IgG antibodies at a mean level of 1 : 776 (range I : 128-4096). By CF, however, which is an insensitive test for determining mumps virus immun- ity, only 42% were positive, the mean titre being only I: 4 (22). Most cases of clinically diagnosed mumps virus infection in pregnancy turn out to be wrongly diagnosed when investigated virologically. Formerly, 73 -..J ~ Table 11. Laboratory diagnosis of human (alpha} herpesvirus 3 infection Diagnostic problems Acute primary infection in pregnancy Acute reactivated infection in pregnancy Neonatal 1nfect1on Suspected congenital infection Immunity status in pregnancy Methods Virus detection in vesicle fluid with CIE Serology: (CF). ELISA. lgG lgM antibody: ELISA. ELA Virus detection In vesicle fluid with CIE Serology: CF. ELISA. lgG lgM + lgA ant1bod1es : ELISA. ELA Virus detection in vesicle fluid with CIE Serology: CF. ELISA. lgG lgM + lgA ant1bod1es: ELISA. ELA Virus isolation in cell culture and from vesicle fluid with CIE Serology: CF. ELISA. lgG lgM + lgA antibodies: ELISA. ELA Serology: ELISA. lgG Expected results Human (alpha) herpesv1rus 3 posItIve in 2 hours Antibody rise slowly. heterotyp1c rise of human (alpha) herpesv1rus 1 and 2 antibod ies lgM antibody posItIve >3 days post-1nfect1on Human (alpha) herpesv1rus 3 pos11Ive In 2 hours Antibody rise slowly, heterotypic rise of human (alpha) herpesv1rus 1 and 2 antibodies Pos1t1ve > 4 days post-eruption Human (alpha) herpesv1rus 3 posItIve in 2 hours Antibody rise slowly. heterotyp1c rise of human (alpha) herpesv1rus 1 and 2 antibodies Pos1t1ve > 3 days post-infection Should be positive. but so far such reports lacking Persisting antibodies Pos1t1ve in cord blood and later - 9 5% seroposItIve a virological diagnosis was made by virus isolation from throat secretions, urine, or, if meningitis occurred, the cerebrospinal fluid. With the avail- ability of the indirect ELISA IgG and IgM test (22) as well as a direct haemadsorption immunosorbent test for IgM and IgA antibodies (26), a serological diagnosis can now be made reliably and rapidly, in a similar way to that previously described for human (alpha) herpesvirus 3. Mumps virus IgG antibody persists for life and may occasionally but rarely be boosted by reinfection, or more commonly by infection with parainfluenza viruses (27). Because of cross-reactivity with parainfluenza viruses, the presence of low levels of antibody may justify the administration of mumps virus immunoglobulin to protect pregnant women who have been exposed to mumps. Measles virus The consequences of measles infection in pregnancy are principally known from reports of the virgin soil epidemics in Greenland in 1951-1962 (22,28). In Europe up to 98% of women of childbearing age possess naturally acquired measles virus antibody when sera are tested by a sensitive ELISA IgG test (22). These antibodies confer protection, even when present at a low level. In the United States and the German Democratic Republic natural immunity has been replaced over the last 10-12 years by vaccine- induced immunity, both countries having effective legislation to enforce vaccination. Antibody titres after vaccination are lower than after natural infection and tend to decline further in the absence of a circulating measles virus. With the projected eradication of the measles virus in the United States, antibody screening (29) and immunoglobulin prophylaxis may be requested to protect pregnant women who are seronegative and who travel abroad or come into contact with imported measles virus. Except in remote areas of the world, measles virus infection in preg- nancy is rare. As virus isolation in cell culture is difficult and the direct detection of viral antigen in throat secretions depends on obtaining speci- mens before or at the onset of exanthem, a diagnosis is usually made serologically. A rapid diagnosis by indirect ELISA may be made if the first serum sample taken from 3-4 days to 1-2 months after the onset of illness is positive for measles virus-specific IgM and IgG antibodies (22). Enteroviruses There are 23 coxsackieviruses A, 6 coxsackieviruses B, 30 echoviruses and 3 polioviruses. Enterovirus infections are mostly transmitted perinatally 75 via the mother or via other infected infants. Laboratory diagnosis is often difficult, however, partly because there are a large number of distinct serotypes and not all enteroviruses grow in cell cultures. Thus such viruses as some of the coxsackievirus A serotypes can only be isolated in suckling mice, and this is a somewhat cumbersome procedure. In addition, sero- logical procedures are of limited value since heterotypic antibody re- sponses may occur. Such responses may be greater than homotypic ones. In serological studies, microneutralization tests are usually used but they are time-consuming and complicated. Although the isolation of entero- viruses from the cerebrospinal fluid or the brain is significant, their iso- lation from stools or pharyngeal secretions does not necessarily mean that these viruses are the cause of the disease, since apparently healthy persons may also excrete enteroviruses. An ELISA technique for detecting and serotyping human rotavirus in faecal specimens has been adapted for detecting coxsackieviruses A (30) and although some cross-reactivity occurs, this technique may have considerable potential for the diagnosis of coxsackievirus A infections. A similar technique has also been developed for detecting coxsackievirus B infections (31); this technique may also soon be available for the detection of echoviruses. Coxsackievirus B-specific IgG and IgM responses may also be detected by ELISA (31-33). Coxsackie- virus B-specific IgG and neutralizing antibody responses do not correlate. Heterotypic responses are likely to occur but this may not be of major importance clinically. The use of purified antigens or monoclonal anti- bodies in such test systems may obviate such problems. As in other congenitally acquired infections, enteroviruses can be isolated from body fluids, or at post-mortem from different organs, since such infections are usually generalized. In addition a virus-specific lgM response may be detected and neutralizing antibodies will persist beyond 6 months of age. The laboratory diagnosis of enterovi rus infections is summarized in Table 12. Conclusion Table 13 summarizes the methods of choice for making a laboratory diagnosis of virus infections that are known to cause pre- or perinatal infection as well as those that are, as yet, less clearly associated with such infections. Table 13 also includes techniques for the laboratory diagnosis of perinatal hepatitis B virus infection as well as virus infections that may be spread nosocomially in the neonatal period, such as human rotavirus, respiratory syncytial virus and parainfluenza virus 3 infections. The techniques in the laboratory diagnosis of other microbial agents that may cause pre- or perinatal or early postnatal infections are summar- ized in Table 14. 76 -..J -..J Table 1 2. Etiological laboratory diagnosis of enterovirus infections (coxsackieviruses A and B. echoviruses. polioviruses) Methods Virus isolation cerebrosp inal fluid tissue stool Identification Virus isolat ion plus serology for isolate and prototype Ant igen test Serology only: Antibody rise 2 serum samples lgM (lgA) antibody lgM . lgG antibody lgM. lgA ant ibody Cell culture Baby mice Cell culture Baby mice Results + specific + (+) 1nd1cative + specif ic Microneutral1zat1on test+ (lgM test) + opt imal Neutralization in baby mice ELISA (+) In progress tM icroneu tral ization test CF-ant igen pools CF-single types ~E coxsackieviruses B ELISA: purified virions crude antigen ELA: peroxidase-labelled serotype antigen + {heterotyp ic reactions among ± all 62 enteroviruses ± (+) homotypic and heterotyp1c (+) homotypic heterotyp1c in progress Available 5- B days 3-1 0 days 12-20 days 6 hours 5- 6 days 1 - 2 days 2 days 6 hours 6 hours -.l Table 13 . Laboratory diagnosis of pre- and perinatal and early postnatal viral infection and disease 00 Serology Agent antibody Infection Isolation Antigen Old SDG New Old ELM New CF / HI / IF / NT RIA-ELISA-ELA tissue cult ure IF -RIA-ELISA Rubella virus 1BIJcFIH1GI l lgM! lgG ll gM I (+ ) - (+ ) Cytomegalov1rus [£)1HA lgGjlgMI ill (+) + (+ ) Human (alpha) herpes- @£) -FAMA-IF llgG + lgMI ~ virus 3 (+) - + Human (alpha) herpes- CF lgG~ El ill + virus 1 and 2 type-spec. Human (gamma) herpes- IF -lgG-lgM - pgM - lgA! virus 4 EA-EBNA Mumps virus CF I1gG+lgMI (+ ) Measles virus ~ lgG+~ (+) - [±] Influenza vi ruses @ -HI lgG +jlgMI + - E] Para1nfluenza viruses @HI-NT lgG~ + [±] Respiratory syncyt1al virus CF-NT I1 gG + lgA I + + -..J '° Table 13 (contd) Human cocksack1e- vIruses Echov,ruses Hepat1t1s virus B Hepat,t,s v,rus A Human rotav,rus Lymphocyt,c chor,o- men,ngItIs Arbov,ruses (CEE) Polyomav,ruses JC and BK virus Papillomav,ruses §)CF CF CF ~ -NT ~ IF ~NT HI CF. NT. IF-lgM :ii=! 1--" ,- -- -- -, M ethod of choice L _______ _; No routine :igM] 1----4 Ant, HBs Ant, HBc-lgM j Ant, HAV-lgMI r1gG-lgM: L _______ J ~l gG-lgM: ~----- -' 11gG-lgMI m (+) + - ± + ± + - ~ M, ce + tissue cul ture Baby m,ce + tissue culture (+) + r Hybr,d,zat,on : I cytology : molecular clo n- : + : ,ng,nE'col : __________ J 00 0 Table 14. Laboratory diagnos is of pre- or per in atal or early postnatal microbial in fect ion Serology Agent Infection ant ibody Culture Smears Ant igen --- Toxoplasma gond11 !CF. IHA. IF. lgG + lgM Mice ~ sal :TC1sA7 -----~ ELISA lgG lgM n ELA IM Treponema pa/1,dum Rabbit !Dark- I ~TTis~ - ___ , testicle Plasmod,um fatc,parum I IF IHA!@~~~-~~~ Th1ck-th1n i{LISA: i.:,;.; __ .;i blood films Pneumocysr,s carm11 CiLIEl Im . suppr . Stain [fill -- rats tolu1d-gomom :ELISA: '---·~ L,stena monocytogenes CF agg l. 1m. d1ff. I±l Gram IF IH / O. IV stain Group B beta-haemolyt ic Aggi. prec1 pt. ltA- [ciE-:ELisA: streptococci la . I b. le. II . Ill Opson. test ~-----~---, o I camp-test 1Gas- I I I bac1tr . disc Lc~r.9~~t':g...'. a_P~'.!'..: 00 Table 14 (contd) Ne,sser,a gonorrhoeae Chlamydia uachomaus (serotypes D - K) M ycoplasma pneumon,ae M ycoplasma homm,s Ureaplasma urealyt,cum M ethod of choice CF ~R IA (purified ; :p ilus antigen) : CF r~1,<:;_r9-l F .c. I gG_.:l_g~; CIE ~ Growth -1nh 1b1t !ELISA lgG/lgM! ;----7 ._ ___ --~ No routine yet +-ox1dase act1v1ty suga r fermat. FA-. co-aggl. M c-coy-cells yolc sac + [±] Gram FA-stain !ELI SA! CIE CIE References l. Schmitz, H. et al. Detection of lgM-antibodies to cytomegalovirus (CMV) using an enzyme labelled antigen (ELA). Journal of general virology, SO: 59-68 ( 1980). 2. Krech, U. & Wilhelm, J.A. Immunosorbent technique for rubella lgM. Journal of general virology, 44: 281-286 ( 1979). 3. van Logt, J.M. et al. Hemadsorption immunosorbent techniques for determination of rubella immunoglobulin M-antibody. Journal of clini- cal microbiology, 3: 410-415 (l 981 ). 4. Enders, G. Virusdiagnostische Befunde und ihre Interpretation: Roteln . In: Virusdiagnostikfiir Klinik und Praxis. Marburg, H. Spiess, Deu tsches Griines Kreuz, 1980. 5. Enders, G. et al. Comparison of various serological methods and diagnostic kits for the assessment of acute , recent and previous rubella infection and immunity. Journal of infectious diseases (in press). 6. Enders, G. Rotelnembryopathie noch heute? Geburtshi/fe und Frauen- heilkunde, 42: 403-413 ( 1982). 7. Banatvala, J.E. et al. Specific IgM-responses after rubella vaccination : potential application following inadvertent vaccination during preg- nancy. British medical journal, 2: 1263-1264 ( 1977). 8. Schmidt, N.J. et al. Application of immuno peroxidase staining to more rapid detection and identification of rubella virus isolates. Jour- nal of clinical microbiology, 13 (4): 627-630 (1981). 9. Krech, U. Diagnostik und Pravention pra- und perinataler Virusinfek- tionen , Cytomegalie und andere Herpesviren. In: Der prii- und peri- natale Virusinfekt. Marburg, H. Spiess, Deutsches Griines Kreuz, 1982, pp. 95-106. IO. Stagno, S. et al. Comparative study of diagnostic procedures for congenital Cytomegalo-virus infection. Pediatrics, 65 (2): 251-257 (1980). 11. de Savigny, D. & Voller, A. The communication of Elisa data from laboratory to clinician. Journal of immunoassay, 1 (1): 105-128 ( 1980). 12. Enders, G. Serodiagnosis of varicella-zoster virus infection in preg- nancy and standardisation of the Elisa lgG and lgM antibody test. Journal of biological standardization, S2: 221-236 ( 1982). 13. Vestergard, B.F. & Jensen, 0. Diagnosis and typing of herpes simplex virus in clinical specimens by the enzyme-linked immunosorbent assay (Elisa). In: Proceedings from the International Conference on Human Herpes-viruses, Atlanta, Georgia, USA, 1980. 14. Nahmias, A.J. The laboratory diagnosis of herpes simplex virus. Pros- pects and problems. In: Diagnostic horizons. Microbiological Associ- ation, 1977. 82 15. McCallum, F.O. et al. Antibodies to Herpes simplex virus in the cerebro- spinal fluid of patients with herpes encephalitis. Journal of medical microbiology, 7: 325-331 ( 1974). 16. Klapper, P.E. et al. Rapid noninvasive diagnosis of herpes encephalitis. Lancet, 2: 607-609 (1981). 17. Enders, G. et al. Noninvasive laboratory diagnosis of neonatal Herpes simplex type I and 2 infection with encephalitis (in preparation). 18. Vaan Loon, A.M. et al. Diagnosis of herpes encephalitis by Elisa. Lancet, 2: 1228 (1981). 19. Vestergard, B.F. & Grauballe, P.C. Elisa for herpes simplex virus (HSV) type specific antibody in human sera using HSV type I and type 2 polyspecific antigens blocked with type heterologous rabbit antibodies. Acta pathologica microbiologica scandinavica, B, 87: 261- 263 (1979). 20. Kalimo, K.O. et al. Solid phase radioimmunoassay of herpes simplex virus IgG and lgM antibodies . Journal of immunological methods, 14: 183-195 (1977). 21. Kalimo, K.O. et al. Solid phase radioimmunoassay of human immuno- globulin Mand immunoglobulin G antibodies against herpes simplex virus type I capsid, envelope and excreted antigens. Infection and immunity, 15: 883-889 (1977). 22. Enders, G. Diagnostik und Pravention pra- und perinataler Virus- infektionen: Varizellen - Mumps - Masern. In: Der prii- und perinatale Virusinfekt. Marburg, H. Spiess, Deutsches Griines Kreuz, 1981. 23. Hanshaw, J.B. & Dudgeon, J.A., ed. Viral diseases of the fetus and the newborn. Philadelphia, W.B. Saunders, 1978. 24. Frey, H.M. et al. Rapid diagnosis ofvaricella zoster virus infections by counter current immunoelectrophoresis. Journal of infectious diseases, 143: 274-280 (1981). 25. Almeida, J. & Enders, G. The presence of antiglobulin (lgM-Rheuma- factor) in different clinical conditions: acute infections and pregnancy. Journal of general virology (in press). 26. Van Logt, J.T.H. et al. Hemadsorption immunosorbent technique for determination of mumps immunoglobulin M antibody. Journal of clinical microbiology, 15: 82-86 ( 1982). 27. Enders, G. Paramyxoviruses : parainfluenza, mumps, measles, respir- atory syncytial virus. In: Baron, S., ed. Medical microbiology. Menlo Park , CA, Addison-Wesley Medical Division, 1982. 28. Jespersen, C.S. et al. Measles as a cause of fetal defects: a retrospective study of ten measles epidemics in Greenland. Acta paediatrica scandi- navica, 66: 367-377 (1977). 29. Centers for Disease Control. Measles prevention . Morbidity and mor- tality weekly report, May 7, 1982. 83 30. Yolken, R.H. & Torsch, V.M. Enzyme-linked immunosorbent assay for detection and identification of Coxsackieviruses A . Infection and immunity, 31: 742-750 (1981). 31 . Katze, M.G. & Crowell, R.L. Immunological studies of the group B Coxsackie viruses by the sandwich enzyme-linked immunosorbent assay (Elisa) and immunoprecipitation . Journal of general virology, 50: 357-367 ( 1980). 32. Katze, M.G. & Crowell, R.L. Indirect enzyme-linked immunosorbent assay (Elisa) for the detection of Coxsackievirus group B antibodies. Journal of general virology, 48: 225-229 ( 1980). 33. EI-Hagrassy, M.M.O. et al. Coxsackie B virus-specific lgM responses in patients with cardiac and other diseases. Lancet, 2: 1160-1162 ( 1980). 84 Rapid diagnostic techniques in Spain - M. C. Echevarria, J.M. Echevarria, A. L/(icer, A. Tellez & R. Najera Some of the methods discussed in the previous section were used in a large-scale study in Madrid, carried out by the authors over a four-year period from 1978 to I 981, on 11 478 women who had been exposed to rubella virus. They were screened using the haemagglutination inhi- bition test (HAI) and if seroconversion occurred, the sera were tested by rubella virus-specific lgM by HAI following sucrose density gradi- ent (SDG) fractionation. There were 156 conversions, and tests to detect rubella virus-specific lgM suggested that 112 of these infections were primary. Fig. I shows that rubella virus infection occurred during the early summer months , particularly in 1978, the year in which there was an extensive epidemic. A diagnosis of prenatal and perinatal infections was made by virus isolation or by demonstrating the presence of a virus by electron mi- croscopy (for cytomegalovirus infection in urine samples). A serological diagnosis was established by demonstrating virus-specific IgM or the persistence of antibody beyond the time at which maternal antibodies would be expected to decline to undetectable levels. Table I shows that rubella and human (beta) herpesvirus 5 were the most frequently encountered viruses inducing such infections and that a large number of infants were born with congenitally acquired rubella virus infection in the winter after the extensive 1978 epidemic. The clinical features of congenitally acquired cytomegalovirus and rubella virus infec- tions are shown in Tables 2 and 3 and those of some of the less frequently encountered infections in Table 4. Fig. 2 shows that two commercially available kits for detecting rubella virus-specific lgM by ELISA produced results that correlated well and Fig. 3 confirms that both these ELISA test systems, which were of com- parable sensitivity, also correlated well with the conventional SDG method for detecting rubella virus-specific lgM. It must, however, be stressed that rubella virus-specific lgM antibody results should not be considered in isolation but be interpreted in conjunc- tion with other rubella virus antibody tests as well as clinical data. Only when consideration is given to all these factors should a decision be made as to whether a pregnancy should be terminated because of maternal rubella virus infection. Laboratories with limited resources and experience should be encour- aged to have their results confirmed by reference laboratories with more extensive experience in rubella virus serology. 85 00 °' Fig . 1. Reported cases of rubella and the number of seroco_nversions detected in Madr id. 1978-1981 (monthly data) 2800 2500 2000 § l3 ~ 1500 0 a. ~ 0 0 z 1000 500 400 300 200 100 !' 0 ·~ 20 ii! "O 15 C: tl 8 ~ 10 ~ .. ~ "O 5 0 0 z >-; S O N .. C: ~ FMAMJJASON~JFMAMJJASONDJFMAMJJASOND 1979 1980 1981 JFMAMJJASON~JFMAMJJASOND 1980 1981 00 --..l Year of birth 1978 1979 1980 1981 Total No. of No. of cadsesd pos1t1ves stu 1e 93 21 258 44 338 65 437 35 1126 165 Table 1. Congenital infections laboratory results Rubella Cytomegalo- Toxoplasma Human (alpha) Human (alpha) Coxsackie- virus virus gond11 herpesv1rus herpesv1rus 3 viruses 1 and 2 Total % Total % Total % Total % Total % Total % 10 48 6 28 .7 5 4 7 25 57 13 29 4 9 1 2 1 2 12 18 39 60 8 12 2 3 3 4 7 20 19 54 6 1 7 1 2 54 33 77 43 24 14 4 2 4 2 1 0 .6 88 Table 2 . The clinical features found in 74 cases of congenitally acquired cytomegalovirus infection Clinical feature Hepatosplenomegaly with purpura Central nervous system Microcephaly Encephalopathy Psychomotor retardation Cerebral calcifications Hydrocephalus Intrauter ine growth retardation Liver abnormalities Hepatomegaly (asymptomatic) Hepatitis Biliary atresia Respiratory disease Pneumon1tis Broncopneumonia Ocular defects M icrophthalmia Retinopathy Uveitis Maculopathy Cardiovascular disease Deafness Normal infant No. of manifestations 46 44 16 11 9 6 2 20 20 3 16 6 4 2 4 4 2 Table 3 . The clinical features found in 45 cases of congenitally acquired rubella virus infection Clinical feature Cardiovascular disease Ocular defects Cataracts Pigmentary retinop athy Choro1ditis Deafness Osteopathy Central nervous system Mi crocepha ly Psychomotor retardation Aseptic meningItIs Hydrocephalus Intrauterine growth retardat ion Hepatosplenomegaly with purpura Rash Physi cal malformations Normal infant No. of manifestations 20 17 15 5 3 19 6 9 2 2 18 6 2 5 2 89 \0 0 Table 4. The clinical features found in some other neonatal infections Infection Human (alpha) herpesvirus 1 and 2 Human (alpha) herpesvirus 3 Coxsack1evirus 8 5 Echovirus 2 5 No. of cases 6 3 Cl1n1cal feature Rash Central nervous system Encephalitis M enI ngoencephal1t1s C horio ret ini tis Papuloves1cular rash Without symptoms Feve r Diarrhoea M enI ngoencephalitis Respiratory distress Aseptic meningItIs No. of manifestations 3 2 2 (zoster immune globulin) \0 Fig. 2 . Detect ion of rubella virus-specif ic lgM . using two ELISA kits . showing good correlat ion 3.0 1.0 rr 2.0 I I ~ I~ lq; 0.5 ~- <l + cut-off + 1.0 0.2 0 Ir,, I I ; ... , 'i[, ,, .., i----ir------------------ ,,, . , I / I , I *1 I I I < B - +f+ + l'I 11 1,1 1.,!.1 3 6<1 12B HAl -lgM ----+ E LISA -lgM rubella 61 cases 92 sera * - Behring •- - -Abbott l,Q N Fig . 3 . Evolution of lgM antibodies in 53 cases of clinical rubella ELISA }: _:j • t"l~~;~:~t---------~=~~:_~_.:'.~'._.=_ ____________ ___ HAI 1~...--- _ ,-- 1.4 - 52/52 c .... I HJOlll ELISA lgM + lgG + HAI lgM + Total + 3.0 1.0 2.0 0.5 --, -- ' ' ' _/,,-,, / V / .......... ----- HAI-SDG ------ ELISA Behring ________ ELISA Abbott / :.:'"'-----,, "• .. ', I ............ .. ',, .......... ... ' ... ,t i ·~ I ',, .... ,, ' .. .... .... , 1/ ...... :] I~ ,.o 0.2 ,, + .. .... ll-------_ limit _______ ____ L ________________________ ~,- ---.c---- --- ' • l ·-......... :, I I I I 40 JJ 16 < B ~ 0 0 <I,) <( :,: ]_ ___i_ --1L __a_ ____.s__ ---5._ -----~-----__,No. of cases 0 10 15 20 25 30 40 50 Days post-exanthema -----+ Although enzyme immunoassay kits probably provide the best method for the detection of rubella virus-specific IgM antibody, care must be taken to exclude nonspecific reactions that may be caused by different anti- globulins. It may be necessary to use other techniques to exclude positive or negative results . The use of kits by those who are inexperienced in rubella virus serology must be strongly discouraged. 93 Screening for total serum IgM in the German Democratic Republic - K. Menzel, A. Lemmer & M. Linke Since the fetus may respond to an antigenic stimulus in utero as early as the 20th week of gestation, an estimation of total IgM concentration during the first week of life may be a useful, though nonspecific , way of obtaining evidence of an intrauterine or perinatal infection. Recent studies carried out at the Erfurt Medical School among healthy infants born at term showed that the normal range of IgM concentrations was 3-27 mg/100 ml (mean 9.5 mg/100 ml) (Fig. I). Four infants with prenatally acquired infections such as rubella virus, Listeria monocy to- genes and Toxop/asma gondii had markedly increased concentrations, and infants with respiratory distress syndrome and transient hyperbilirubin- aemia also had raised levels (Table I) . On the other hand, as the infants with hyperbilirubinaemia did not have haemolytic disease, infection could not be ruled out as a cause of the raised serum IgM in this group of infants. Mean serum lgM concentrations were only slightly raised in premature eutrophic and full-term hypotrophic infants. 94 Fig . 1. lgM level in healthy newborn babies . Statistical ca lculation made with the help of the Wil coxon test 30 111.._ - 25 20 15 9 .5 10 5 2JL _ mg/ 100ml • • • ---- - ---- ---- • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • ·••:••·· • • • • • • -. -.- ........... ..--.-. • • •• • • • •••• • • • • • • • • • • • • • • • • •• • • • -------L-.-- - 97.5% 2.5% \D Vl Table 1. lgM levels in newborn babies with various infections or conditions lgM (mg/ 100 ml) Infection or condition No. of babies x(50%) x(2 .5%) Manifest prenatal 1nfect1on 4 99.4 28. 7 Respiratory distress syndrome 23 14 .3 10.2 Hyperbil 1rub1naem1a 11 38 .6 9 .3 Pre-term eutroph1c newborn babies 19 14 .5 2.5 Full-term hypotrophic newborn babies 1 7 23 .9 9.7 x(97 5%) 348 .8 45 .9 76 .0 43 .9 42.4 The shortcomings of epidemiological studies: an example from Finland - G. Granroth Congenital defects may be caused not only by prenatal and perinatal infectious diseases such as rubella or toxoplasmosis , but also by hundreds of other factors that are suspected of having teratogenic properties(]). On the other hand no primary detection of a single teratogen has as yet been achieved in epidemiological studies (2). At the same time some 70% of all congenital defects are still without a causal explanation (3). All the know- ledge in this field originates from primary clinical observations, that have then been confirmed by epidemiological means. The development of such methods nevertheless seems to be useful in the characterization of defects and the detection of risk indicators. There are three possible explanations for the shortcomings of epidemiological investigations (4). I. The majority of congenital defects are genetically determined and environmental factors do not play a major role (5) . 2. Environmental teratogens act , on the whole , additively with other exogenous and genetic factors and studies become confounded by too many unknowns. 3. Present methods are inadequate and their power insufficient to detect causative factors with low expressivity. This chapter focuses on the second and third alternatives and discusses the capacities and limitations of epidemiological studies to detect tera- togens . It draws heavily on the writings of Dr Lauri Saxen (6), one of the initiators of the Finnish Register of Congenital Malformations in the early 1960s, and several investigations and studies based on the material col- lected by this Register will be used as examples. There are three ways of approaching epidemiological investigations: surveillance, cohort studies and case-control investigations. Population surveillance The main functions of medical registers are the continuous monitoring of the population to detect significant changes in the incidence of the disease under surveillance and the subsequent search for genetic or environmental clues as to the causes of these changes. In teratology such variations in time and space have often been reported and their background factors dis- cussed, but no conclusive associations with causative factors have been detected (7) . Several retrospective studies on the epidemiology of the 96 thalidomide embryopathy are exceptions and show a clear temporal corre- lation between the typical defect and the sale of thalidomide. The con- clusion is that surveillance studies could lead to the detection of new teratogens when a previously rare defect increases rapidly. This is true when a teratogen spreads rapidly in a large population and its terato- genicity is high. Furthermore, the effect of the teratogen should be specific and lead to easily recognizable defects while information on the exposure should be available and reliable . This is an ideal situation, however, and is extremely rare in practice. In most studies the situation is the opposite because most teratogens seem to have a variety of effects depending on the dose and time of exposure as well as on the fetal genotype. In addition most new teratogens spread slowly, only a small and scattered population is exposed, reliable exposure data are rare, and the study tends to be con- founded by other factors. A surveillance study of the severe influenza epidemic in 1957 in Finland is an example of the difficulties encountered in this kind of investigation. A cohort of 6000 children in Helsinki was followed up to the age of IO and all defects were recorded . The children were grouped according to the stage of their intrauterine development at the time of the short epidemic in October 1957, but only those who were in the stage of organogenesis ("the sensitive period") were considered study cases, and these were compared to the rest of the material. A significant association between defects of the central nervous system and the influenza epidemic was detected (8). Since it was assumed that the study situation would be confounded, however, information on drug consumption during the epidemic was collected and compared to the sales figures of the pharmacies in the study area for four consecutive years. The comparison showed a clear association between the epidemic and the increased sale (and consumption) of drugs. Thus the study situation was confounded and did not justify drawing final conclu- sions until additional results were available. Case-control studies An adequately designed and functioning register can collect valuable material for case-control studies and this has become the main function of the Finnish Register of Congenital Malformations. It was started in 1963, and its matched pair system has been in use since 1964. The organization of the Register has been described in detail elsewhere (9) and only a brief summary will be provided here. The reporting of all congenital malformations during the first year of life is compulsory in Finland and the incidence of malformation is about 1.3%. This figure, however, suffers from several shortcomings in the reporting and detection of malformations. Together with skeletal defects and oral clefts, defects of the central nervous system form a special group of 97 "indicator defects" , for which extensive information is collected through maternity welfare centres. A network of such centres covers the entire country. After the first report of a defect to the Register, the corresponding centre is informed and a multitude of data on the case and a control are obtained by interviews with the mothers and from antenatal and hospital records. Death certificates are also used for complementary information. The control for each case is the mother whose delivery preceded that of the study mother in the same maternity welfare district. Several analytical studies have been performed on the material and, as expected, many associations between maternal histories and various defects have been detected. For example, in a detailed investigation of 710 mothers of children with central nervous system defects it was possible to pick out more than 20 different associations (10-12). As can be seen from Table I, several of these were maternal diseases and the consumption of drugs during pregnancy. In a more limited study with only a few associations, all of them could have been relevant. The great number of associations found in a large study is unrealistic and leads to a general rule for epidemiological investigations : they have many limitations and pitfalls, and the results should be evaluated critically before any conclusions as to their biological significance can be made. The main limiting factors to be considered are the following: lack of and inadequate definition of defects, chance correlations, maternal memory bias, and confounding factors . These major drawbacks and the way they should be handled are discussed briefly below and examples of efforts to eliminate them are given . Detection and classification of cases The Finnish Register of Congenital Malformations, like many similar surveillance systems, suffers from both an under-reporting of cases and a low detection rate of malformations. It is estimated that only some 70% of malformations detected during the perinatal period are notified and, in particular, that minor defects remain unreported. In addition, even in the best obstetric wards with specialized personnel, one third of malfor- mations remain undetected at birth and immediately thereafter (13, 14). The use of indicator defects may somewhat lessen the drawbacks caused by low reporting and detection rates. Such indicator defects should be clearly defined, easily detectable at birth and severe enough to attract the clin- ician's attention. Chance correlations Chance or nonsense correlations are unavoidable in studies where a great number of variables is analysed without hypothesis. There are basically two ways of recognizing such false associations: by repeating the study on 98 Table 1. Factors found to be si gnificant ly assoc ia ted with defects of the ce ntral nervous sys tem in pr ima ry inves ti ga t io ns In fl ue nza Threa tened abo rt io n Depr ess ive state Nausea Toxaemia o f pr eg nancy Diab etes mell1t us Rh eum ato id arthr1t 1s An t1pyret1c and ant1 -1n flammatory analgesics Sa l1cyl ates Pyr azo lones/ a nil , nes Euphor1s t1 c analges ics Sym pathom 1met1cs Source Granroth, G. et al. ( 10- 12) Psyc ho pha rm ace u t 1ca Is Ant1 neur1t1 cs Barbitur ates Ant1b1ot1cs and chemo the rapeut ics Sulfona mides Peni c illin s Co ug h medic in es M aternal age over 3 4 ye ar s Birth o rder 4 or above Pr ev ious st 111 b1 rth s Pr ev ious abo rti o ns Pr evious c h ild ren wi t h any de fec t Previous c h ild ren wi th cen tral ne r- vous sys tem de fec t independent ma teri al and by co nsidering th e bi ologica l mea ning of the associa ti o ns detected . Fo r exa mple, a n assoc ia tio n between any potentia l teratoge n during the second trim es ter a nd cl eft pala te is not mea ningful since the pa la ta l shelves fu se much ea rl ier. Maternal memory bias Whenever use is made o f retros pecti ve data based o n materna l interviews aft er delive ry, memo ry bias sho uld se rio usly be ta ken into account as both the mother a nd often a lso the interviewer a re aware of the pregna ncy outcome. Mothers o f defective children may more effectively sea rch fo r " ca usa ti ve" factors during their ea rl y pregna ncy than mothers o f hea lthy babies who have usuall y lost interest in their pregnancy. This bias has neve r been co nclusive ly shown, but in repea ted interviews it has been shown that materna l hi stories do cha nge during pregna ncy (1 5) . The o nl y way to eliminate completely the poss ibility o f maternal memo ry bi as is by a prospecti ve study: pros pecti ve in the sense that a ll information is co ll ected befo re the pregna ncy o utcome is known . On the other ha nd , the co ll ecti on o f eno ugh d efec ti ve cases fo r a mea ningful a nalysis o f this kind requires a n eno rm o us study gro up , a nd thi s means th at data must be co ll ected fro m ma ny hospit a ls and by a great number o f 99 people. To standardize the collection of all these data is not easy. Moreover such studies are slow, laborious and expensive, which makes them unreal- istic in many countries. For testing some relatively simple hypotheses, prospective studies will certainly prove useful. Despite the retrospective design of a study, as in our surveillance programme, much prospectively collected data are available. The ante- natal clinics routinely record data on maternal conditions and diseases, all prescribed drugs and all physical examinations and laboratory tests per- formed. Minor complications and illnesses as well as non-prescribed drugs are incompletely recorded, however, and the information is mostly based on retrospective interviews and hence subject to maternal memory bias. One way to deal with this problem is to select the controls from among mothers of children with similar defects. To investigate the relative signifi- cance of maternal influenza and the consumption of drugs in teratogenesis , a special study was devised (16). From the files of the Register, 80 mothers of defective children were found with a history of influenza during the first trimester of pregnancy. The controls were selected from among mothers of children with similar central nervous system defects but no history of influenza. Risk ratios were calculated by comparison with their matched pair controls. There was a clear trend for the risk ratios to be considerably higher in the influenza group , suggesting that the disease acts either like a teratogen or as an additive factor to drug-induced malformation. In a later study another type of control material was used ( 12). Since polydactyly is presumed to be a defect with a predominantly hereditary background which is less affected by exogenous factors, children with this defect were considered an adequate control for the group with central nervous system defects . A comparison of the risk ratios of the factors listed in Table 2 in the two groups of children showed that many of these factors were also associated with polydactyly. Whether this can be explained entirely by maternal memory bias or whether these factors may, in fact , be associated with polydactyly cannot be decided, but the first explanation seems more plausible. When the factors associated with both types of defect were eliminated from the list, the number of risk indicators for defects of the central nervous system was reduced to five. Interestingly enough, the much discussed salicylate showed very much the same risk ratios in both groups. Conj ounding factors In a non-experimental study, the most serious limitation is due to con- founding factors, i.e . factors linked to the actual causative agents and hence also to the defects. By combining different types of linkage and adding several confounding factors to a hypothetical situation one can design any kind of study. 100 Table 2. Risk factors during first trimester of pregnancy that are significantly associated with defects of the central nervous system Tim e of 1nvest1gat1on Primary Aft er el 1m1nat1on of factors associated with poly - dac tyly and co ntrol for memory bias Source Granroth, G. / 12) Diseases and co nd1t1 ons Influenza Threatened abortion Depressive state Toxaemia of pregnancy Diabetes mell1tus Influenza Depressive state Toxaemia of pregnancy Diabetes mell1tus Drugs Analgesi cs Sympathom1met1cs Barb iturates Cough med1 c1nes In sulin Cough medicines The Register data showed that both cleft palate and defects of the central nervous system were associated with maternal age and were most frequent in the group of mothers over 40 years of age. This result was obtained by comparing the age distribution of these mothers with that for the whole country, but it could not be shown by comparison with the matched pairs because the older age groups were also over-represented in the time/area matched controls. This was revealed by the observation that both defects were most frequent in a certain district of the country, the province of North Karelia, where the average age of the mothers was higher than in the rest of the country and the oldest age group was especially large. This example also shows the dangers of over-matching. If, as is only rarely the case, confounding factors cannot be eliminated by the design of the study, the correct way to deal with the problem is by multivariate analysis (17). The many associations between risk indicators and defects of the central nervous system in the study mentioned on page 98 (Table I) have also been exposed to such an analysis (18). A complicated network of associations was detected among various factors and multivariate analysis was then performed. The factors that remained as risk indicators after this analysis are presented in Table 3. Just one example of the results dealing with maternal influenza and its confounding factors is pointed out. It seems that influenza "explains" the associations between defects of the central nervous system and many drugs, and should itself be considered a potential teratogen . Salicylates show an additive effect. IOI Table 3. Risk indicators for defects of th e central nervous system that remain aft er multivariate analysis High maternal age Mult1par1ty with previous stillbirths and children with defects . especially of the cen tr al nervous system Maternal influenza. especially 1n combination with the intake of sal1cyla tes Source Granroth, G. et al. ( 1 BJ Conclusion Although all epidemiological studies have se rious limitations and sources of error, they have to be performed because at present they constitute the only direct a pproach to the unsolved problem of human teratogenesis . The development of reliable monitoring systems, the use of indicator defects , and the critical collection of maternal histories will soon lead to an accumu- lation of valuable data ready for use in retrospective studies whenever a n exogenous factor is suspected of being teratogenic . Moreover, prospective studies and detail ed preplanned studies are highly recommended as they serve as warning systems for detecting new teratogens. International col- laborative studies should also be encouraged. The results of such epidemiological studies can lead to two types of practical consequence. On rare occasions a single teratogen may be found and consequently eliminated. Drugs as well as many chemicals can easily be withdrawn from the market or a warning issued, although this should be done only after thorough consideration. For instance, there are good reasons to believe that certain untreated diseases are more deleterious to the developing fetus than medication . Another way to proceed from epidemiological data to practical measures might , in the future, be the delineation of risk groups, i.e. the definition and characterization of those mothers who have an increased risk of delivering a defective child. In these cases, various diagnostic and preventive measures already available could be considered, and ultimately the pregnancy could be terminated. References I. Shepard, T.H. Catalog of teratogenic agents. Baltimore, The Johns Hopkins University Press, 1973. 2. Miller, J.B. Some epidemiological aspects of teratogen detection . Mutation research, 33: 45 ( 1975). 102 3. Wilson, J.G. Environment and birth defects. New York, Academic Press, 1973. 4. Saxen, L. Drugs and teratogenesis. In: Colombo, F. et al., ed. Epidemi- ological evaluation of drugs. Amsterdam, Elsevier/North-Holland Biomedica l Press, 1977, p. 267. 5. Roberts, C.J. & Powell , R.G. Interrelation of the common congenital malformations. Lancet, 2: 848 ( 1975). 6. Saxen, L. Epidemiological studies for detection of teratogens. In: Infante, P.F. & Legator, M.S., ed. Proceedings of a workshop on method- ology for assessing reproductive hazards in the workplace. Cincinnati, National Institute for Occupational Safety and Health , 1980, p. 357. 7. MacMahon, B. & Yen, S. Unrecogn ized epidemic ofanencephaly and spina bifida. Lancet, l : 31 ( 1971 ). 8. Hakosalo, J. & Saxen, L. Influenza epidemic and congenital defects. Lancet, 2: 1346 ( 1971 ). 9. Saxen, L. et al. The use of a matched pair register in studies on selected conge nital malformations. American journal of epidemiology, 100: 297 (1974). 10. Granroth, G. et al. Defects of the centra l nervous system in Finland: I. Variations in time and space , sex distribution and parental age. British journal of preventive and social medicine, 31 : 164 ( 1977). 11 . Granroth, G. et al. Defects of the central nervous system in Finland: II. Birth order, ou tcome of previous pregnancies and family history. Teratology, 17: 213 ( 1978). 12. Granroth, G. Defects of the central nervous system in Finland: III. Diseases a nd drugs in pregnancy. Early human development, 2: 147 ( 1978). 13. Hakosalo, J. Cumulative detection rates of congenital malformations in a ten-year follow-up study . Acta pathologica, microbiologica et embryologica scandinavica, A, 242(Suppl.): I ( 1973). 14. Klemetti, A. Definition of congenital malformations and detection of associations with maternal factors. Early human development, 1: 117 ( 1977). 15 . Klemetti, A. & Saxen, L. Prospective versus retrospective approach in the search for environmental causes of malformations . American jour- nal of public health, 57: 2071 ( 1967). 16. Karkinen-Jiiiiskelliinen, M. & Saxen, L. Maternal influenza, drug con- sumption, and congenital defects of the central nervous system . Amer- ican journal of obstetrics and gynecology, 118: 815 ( 1974). 17 . Heinonen, O.P. et al. Birth defects and drugs in pregnancy. Littleton , MA , Publishing Sciences Group, 1977. 18. Granroth, G. et al. Defects of the central nervous system in Finland: V. Multivariate analysis of risk indicators. International journal of epi- demiology, 7: 30 I ( 1978). 103 The long-term consequences of prenatal and perinatal infections Subnormality - A. Bove An evaluation of the handicaps caused by prenatal and perinatal infections has been attempted only for certain intrauterine infections, rubella virus, cytomegalovirus and Toxoplasma gondii, and the difficulties encountered in such studies are great. In retrospective studies, the tests indicating a suspicion of an intrauterine infection should be discussed in terms of the clinical and serological criteria to be used. It will often be difficult to constitute a real control group. In prospective studies, there is a problem of poor yield even though these studies tend to be more rigorous. One has to follow a large number of pregnancies with good biological criteria of infection and it is necessary to follow the newborn for several years . It has been easier to carry out these prospective studies in the case of congenital rubella virus infection following large epidemics, such as the one in the United States in 1963-1964. The most common causes of abnormality are auditory and ocular defects, mental retardation and to a lesser degree cardiac malformations. It is above all in congenital rubella virus infection that the frequency of these types of subnormality has been evaluated. The Rubella Project in New York, for instance, has followed about 400 children born with congenital rubella during the rubella epidemic in 1964. The distribution of the major clinical manifestations found among these children is shown in Fig. I. Besides the auditory defects found either as a consequence of congeni- tal rubella virus infection or cytomegalovirus infection diagnosed at birth, audi tory defects that could be connected with past, unnoticed intrauterine infections have been the subject of various retrospective serological studies. One such study took place over a four-year period, 1972-1975. An unselec- ted sample of 568 children under four years of age, referred for assessment to the Nuffield Hearing and Speech Centre in London, were tested for rubella antibody. Since rubella antibody is seldom acquired before the age of four years as a result of postnatal infection, the presence of antibody in these children suggested that they must have undergone congenital rubella virus infection. The results are shown in Fig. 2. 105 Fig . 1 . MaJor clinical manifestations of congenital rubella virus infection. 1964-1974 Hearing loss Cataract glaucoma Heart disease Mental retardation 0 20 Source. Cooper, L.Z. /I) 40 60 80 Per cent involved 87% 271 337 ~ E ., ·;: "' a. ... 0 338 :;; .J:l E :, z 292 100 Fig . 2 . Rubella antibody status of children aged 6 mon ths to 4 yea rs Total number tested 568 ~ ~ 349 219 Sensor1neural deafness / ~ No sensori neu ral deafness /~ Seropos1tive Seronegat1ve Seroposit1ve• Seronega t1 veb 83 (23 8%) 266 (76 2%) 19 (87%) 200 (91 3%) a Conductive deafness (6). congenital heart defect and developmental delay ( 1 ). family h1s1ory of deaf- ness ( 1). maternal history of rubella and / or con tact ( 1). referred for speech delay and/or suspected hearing loss bu! hearing normal ( 10) b Conductive deafness (47). developmental delay (22). cleft palate (5) , menmg1 t1s (4), family history of deafness (28). maternal history of rubella con tact and /or rash (25). referred for speech delay or suspected hearing loss but hearing normal (69) . Source: Peckham, C .S. et al. /2) . 106 Another study of children attending the Nuffield Centre between 1967 and 1975 compared some 250 norma l children with 147 deaf children with and without a hi story of congenital rubella virus infect ion. The results (Fig. 3) show tha t the incidence of rubella antibody in the deaf children is much mo re marked than in the no rm al children. What is of spec ial interest is that the deaf children with no histo ry of congenita l rubella virus infec- tion nevertheless have a much higher incidence of rubella antibody than the normal ch ildren. This suggests that it would be wrong to classify them a ll as sufferi ng from "deafness of unknown etio logy" and about 25 % could be due to rubella after a ll , and thus be preventable. Fig . 3 . Comparison of rubella antibody sta tu s in three groups of children. aged 6 months to 6 years 100 > -0 _g 75 ·;::; C "' £ -~ ., g 50 ., -0 ·;; C: .... C ~ ~ 25 a.. 10 0 -Pm Source 78 19 9 + 25 ·=· + :• ::: 23:: •:•:•:•:•:• ::::::::::: 6 months-2 years 105 27 36 :!3~::: 13 2-4 years 4-6 years Control group normal ch ildren who are not deaf and do not have co ngenital rubella . No history: chil dren who are deaf and have no history of congeni tal rubella Maternal rub ella . deaf child ren with congeni tal rubella Dudgeon, J .A. /3) 107 Mental retardation The estimation of mild forms of mental retardation will have to take into account both the consequences of the sensory defects the child has (in congenital rubella virus infection, for example) and the socioeconomic milieu of the family (in cytomegalovirus infections in particular). A number of studies were made of congenital rubella virus infection , following the epidemic of 1963-1964. As part of the Rubella Project in New York, for instance, Chess, Fernandez and Korn found that of 2 IO children aged 8-9 years with congenital rubella virus infection, about half had no psychiatric disorder and 26% ( or 54 children) were mentally retarded (see Fig. 4). The degree of retardation among those 54 children is shown in Fig. 5. As part oft he same project, Appell, Cooper and Fedun studied the Fig . 4 . Psyc hiatri c diagnoses among 210 children with congenital rubella virus infection No psychiatric disorder Mental retardation Reactive behaviour disorder Cerebral dysfunction Autism Neurotic behaviour disorder Other 0 10 20 30 40 Per cent involved Source: Cooper, L.Z. /I/. 108 50 Fig . 5. Degree of retardation 1n 54 children with congenital rubella virus infection Unspecified Borderline Mild Moderate Severe Profound 0 4 8 12 16 20 Number of children Source Cooper. LZ. (I) 24 appropriateness of the education being received by 269 children with congenital rubella virus infection . At two years of age, only 59% who needed special education had been referred for such service, though by the age of three this had risen to 94%. Nevertheless the education being received by 50% of them was grossly inadequate. The type of education being received at the ages of five and ten is shown in Fig. 6. It was estimated that while at five years old, 28 % of children were inappropriately placed, by ten years of age this had dropped to 17%. For cytomegalovirus infections, prospective surveys have shown the importance of socioeconomic factors. Hanshaw and his colleagues in the Department of Pediatrics at the University of Rochester School of Medi- cine and Dentistry, carried out a study on 44 children with positive specific IgM antibody located through tests at the Strong Memorial Hospital between 1967 and 1970. These children were compared with equal 109 Fig. 6. Educational placement of 269 child ren with confirmed congenital rubella virus infection at age 5 and 1 0 years Regular school 9% ____ _,10% Regular school (resource help) ■ 5 years old 0 10 years old Special education (single handicaps) Special education 38% t---------------J 36% 16% 17% (multiple handicaps),__ ______ _, I nsti tuti on No placement 10 Source Cooper. LZ. ( I J 11% 11% 20 30 Per cent 40 50 numbers of matched control and random control children between the ages of 3.5 and 7 years. Table I shows the relationship between social class (using the Hollingshead system), IQ and cord human (beta) herpesvirus 5 IgM antibody, and Fig. 7 the relationship between school failure, cord human (beta) herpesvirus 5 lgM antibody and social class. This shows that all the school failures in the antibody-positive group were from classes IV and V even though 32% of all failures were from social classes 1-111. Economic aspects The cost of caring for these children handicapped as a result of intrauterine infections has been estimated for congenital infections with rubella virus by Schoenbaum, Hyde, Bartoshesky and Crampton (Table 2). These 110 ..... ..... Ho ll ingshead soc ial c lass I, II . Ill IV. V Table 1. Relationsh ip between soc ial class. 10. and cord human (beta) herpesv1rus 5 lgM antibody• Human (beta) he rpesv1rus 5 lgM pos ItIve 1 2 2 6 ( 14) 93 .3 (30) M atch ed co ntrol 129 3 ( 15) 102 .7( 2 9) a Figu res in pa ren theses are the numbers tested Source · Hanshaw, J .B. et al. (4 ) Rand om cont ro l 124 1 (29) 109 7 ( 15) "O t; ~ ~ (1) .0 E :, z □ □ 0 ■ Fig 7 Relationship between school failure. co rd human (beta) herpesv1rus 5 lgM an tib ody. and soc ial class 22 20 18 16 14 12 10 8 6 4 2 II Ill Social class (Hollingshead) Human (beta) herpesvirus 5 lgM positive Human (beta) herpesvirus 5 lgM negative matched control ~ iman (beta) herpesvirus 5 lgM negative random con trol School failure IV Source Hanshaw, J .B. et al. /41 V ca lcula ti o ns a re based on th e expenses fo r ch ildren with ca tarac t , deafness, card iac ma lfo rmations and mental retardati o n , and show the percentage of children each requirin g a pa rti cu lar type of ca re. This es tim a te was made in 1976 as pa rt of a cost-benefit a na lys is o f rubell a vaccination po li cies . A similar es timat e was made in 1978 by Wilson a nd Re min gto n fo r th e 3300 children bo rn eac h yea r in the Unit ed States with congenital infec ti o n with Toxop/asma gondii (Table 3). II 2 - -w Table 2. Est imated frequency of utilization of special services by persons with congenital infection with rubella virus and estimated duration and cost of these services Special services Newborn nursery - excess care Schooling for moderately deaf Schooling for totally deaf Heart surgery Cataract surgery Schooling for blind Schooling for moderately retarded School Ing/ 1 nstIt utI0 na I IzatI on for severely retarded Source: Schoenbaum. S.C. et al /5). Ut1l1zat1on (%) 20.0 23 .5 24 .5 35 .0 20.0 1.4 8.4 30.0 Duration and cost 5 days· intensive care+ an extra 5 days = S 1350 / hosp1tal1zat1on S8000 / yr X 10 yr S8000 / yr X 20yr s 5000 / operat,on S 1000/ operation X 2 eyes. 1 5% failure resulting In blindness S 7 500/yr X 20 yr+ S l 500/yr X 40 yr for special equipment S3500 / yr X 15yr S7800 / yr X 40yr - -.j:>. Table 3. Cost of special services for the 3300 child ren born in the United States each year with congenital infection with Toxoplasma gondii Service required Aid to totall yb disabled Special schooling and equip- ment for functionally blind Special schooling for moderately visually handicapped Special schooling for moderately retarded Institutional or state-supported foster care for severely retarded Strab1smus surgery Yearly ophthalmologic follow-up care Total Uti I 1zat1on• (%) 8.4 8.4 4.2 13.6 12.2 12 .5 46 .0 Cost of service S5 952 /yr X 44yrs = S261 881 S 7 500 /yr X 20 yrs + S 1 500 /yr X 40yrs = S210000 $7 500 /yr X 1 0yrs = $ 75 000 S3 500 /yr X 15yrs = S 52 500 $ 7 800 /yr X 40 yrs = S 31 2 000 S 1 000/procedure S 45 /yrX20yrs=S 900 a Based on mst1tut1onal or foster care for approximately 8% of children with severe disease at birth (6) Total cost for children born each year with con- genitally acquired Toxoplasma gond11 S 72 593 400 s 58 212 000 S 10 395 000 S 23 562 000 $125611200 S 412 500 S 1 366 200 $292152300 Average cost per child with con- genitally acquired Toxoplasma gond11 $21998 S 1 7 640 S 3 150 $ 7140 S38 064 s 12 5 S 414 S88 531 b The one blrnd child without severe retardation and one of the three moderately retarded chil dren have been considered unable to work (see 5) Source: Wilson. C .B. & Remington, J .S . (7) . References I. Cooper, L.Z. Congenital rubella in the United States. In: Krugman, S. & Gershon, A.A., ed. Infections of the fetus and the newborn infant. New York, Alan. R. Liss , 1975, p. I. 2. Peckham, C.S. et al. Congenital rubella deafness: a preventable dis- ease. Lancet, l : 258-260 ( 1979). 3. Dudgeon, J.A. Congenital rubella in the United Kingdom, In: Krug- man, S. & Gershon, A.A ., ed. Infections of the fetus and the newborn infant. New York, Alan R. Liss, I 975, p. 23. 4. Hanshaw, J.B. et al. CNS sequelae of congenital cytomegalovirus infection. In: Krugman , S. & Gershon, A.A., ed. Infections of the fetus and the newborn infant. New York, Alan R. Liss , 1975, p. 47. 5. Schoenbaum, S.C. et al. Benefit-cost analysis of rubella vaccination policy. New England journal of medicine. 294: 306 ( 1976). 6. Desmonts, G. & Couvreur, J. Congenital toxoplasmosis. A prospective study of the offspring of 542 women who acquired toxoplasmosis during pregnancy. In: Perinatal medicine. 6th European Congress, Vienna. 1978. Stuttgart, Thieme, 1979, pp. 51-60. 7. Wilson, C.B. & Remington, J.S. Prevention of congenital toxo- plasmosis . A viewpoint from a laboratory in the United States. In: Perinatal medicine. 6th European Congress. Vienna, 1978. Stuttgart, Thieme, 1979, pp. 76-89. 115 Mental retardation - J.H. Henderson As the term is used today, mental retardation involves two essentia l components: first , intellectual functioning significantly below average; second, marked impairment in the ability to adapt to the demands of society. It is now widely agreed that both intellectual functioning and adaptive behaviour must be impaired before a person can be regarded as mentally retarded. Neither low intelligence nor impaired adaptive behav- iour alone is sufficient. Moreover, it is important to use objective measures of intellectual functioning and of adaptive behaviour as far as possible; no one should be regarded as mentally retarded merely on the basis of a subjective impression or as a result of the use of inappropriate criteria or measures . Mental retardation is not a disease, and not a syndrome, but comprises a wide range of conditions that may be determined by many factors of a biological, psychological or social nature. From the point of view of etiology, it is reasonable to divide mental retardation, at least actuarially, into two large groups: the more severe grades of retardation, where it may be easier to identify a single causative factor largely responsible for the dysfunction of the brain; and mild retardation which is best considered as determined (at least in a large number of cases) by many interacting factors, both biological and social. Epidemiology Severe mental retardation is uncommon: it occurs in about 3-4 per 1000 of the population of older school age in deve loped countries. The great majority of severely handicapped children remain more or less dependent throughout life. There is at least suggestive evidence that within certain limits social adjustment be considerably improved in many individual cases. The overwhelming importance of biological factors, both genetic and environmental, in the etiology of severe mental retardation is well established. Mild mental retardation is much more common. Using an IQ range of about 50-70 as a criterion, it affects about 25 per I 000 school-age children in developed countries. The frequency of this category , however, changes markedly with age and many mildly retarded children later function more or less inconspicuously in the adult community. Their adaptation is largely influenced by historical , social and economic factors (tradition, family structure, attitudes towards the retarded within a given society, and man- power balance) as well as by individual delayed maturation. It should be stressed that the IQ range suggested must not be applied too rigidly. For example, a number of people within that range do not act like handicapped persons, while a proportion above it may do so. Nor is the IQ necessarily 116 constant during the individual' s lifespan . So, too , with social adaptation: a maladjusted adolescent may, a decade later , have become a normal citizen. The following section on mental retardation is taken from Mental disorders: Glossary and guide to their classification in accordance with the ninth revision of the international classification of diseases (pp. 56-5 7) published in 1978 by the World Hea lth Organization, Geneva. Mental retardation (317-3 19) A co nditi o n of arres ted or incomplete development of mind which is especially characterized by subn orma lit y of intelligence. The coding sho uld be made on the individual's curren t level of functioning without regard to its nature or ca usati on - suc h as psychosis, cu ltural deprivation, Down's syndrome, etc. Where th ere is a specific cogn iti ve handicap - such as in speech - th e fo ur-digit coding should be based on assessmen ts of cogni ti on outside the area of specific handicap. The assessment of intellectual level shou ld be based on whatever information is available, including clinical evidence, adaptive behav- iour a nd psychometric findings. The IQ levels given are based on a test with a mean of 100 and a standard deviation of 15 - suc h as the Wechsler scales. They are provided o nl y as a gu ide and shou ld not be applied rigidly. Mental retardation often involves psychiatric disturbances and may often develop as a result of some physical disease or injury. In these cases, an additiona l code or codes should be used to identify an associated conditi on, psychiatric or physical. The Impairment and Handicap codes shou ld also be co nsu lt ed. 317 Mild mental retardation Feeble-minded High-grade defect Moron IQ 50-70 Mild mental subnormalit y 318 318.0 Imbecile IQ 35-49 318.1 IQ 20-34 318.2 Idiocy IQ under 20 Other-specified mental retardation Moderate mental retardation Moderate mental subnormality S evere mental retardation Severe mental subnormality Profound mental retardation Profound ment a l subn ormality 3 I 9 Unspecified mental retarda tio n Mental deficiency NOS Mental subnormality NOS This arrangement for coding mental retardation remains unsatisfactory. The medical information given in this coding is quite inadequate in cover- age and has proved to be most unreliable . 117 The desirability of a multi-axial approach to classification is being studied and promoted in the current WHO mental health programme. The axes selected refer to the clinical psychiatric syndrome, to the intellectual level, to associated medical conditions and to psychological or social factors that might be important in etiology. Prevention The ultimate goal in combating mental retardation is prevention. Today it is possible to consider many areas in which multidisciplinary action would result in a reduction in the numbers of the retarded. The means already exist both to prevent mental retardation on a substantial scale and to improve the levels of functioning of most of the retarded. Advances in understanding the causative factors of mental handicap form a basis for effective prevention and many such factors have been identified. These include viral and bacterial infections, parasitic invasions of the mother, fetus and child, physical trauma to the mother or child, other physical and chemical agents (such as lead and mercury poisoning), disorders of metabolism and nutrition, genetic factors involving both single and multiple gene action, chromosomal abnormal ities , develop- mental anomalies and conditions due to diseases or unknown influences that result in a lowering of the level of intellectual function and adaptive behaviour. It may be assumed that in developing countries the ro le of infection and malnutrition will be far greater than in industrial ones. Within industrial countries their effect may be greater among the poor than among the privileged. The last decade has seen major advances in the understanding of prenatal infection and its role in mental retardation. At the same time, immunization is now possible against two of the major hazards: rubella virus infection in the prenatal period, and measles virus infection in early childhood . At present, some children still become mentally retarded fo l- lowing a rubella or measles epidemic, although prevention is possible fo r whole populations. There is little doubt that cytomegalovirus infection and Toxoplasma gondii infection are also responsible for cases of mental retardation, and preventive measures can soon be anticipated on the basis of this new knowledge. Bibliography Mental retardation: prevention, amelioration and service delivery. Brussels, Joint Commission on International Aspects of Mental Retardation, 1980. 118 Sinha, S.K. Progress in prevention of mental retardation caused by viral infections. In: Mittler, P., ed. Frontiers of knowledge in mental retar- dation, Vol. II. Baltimore, University Park Press , 1981. Blomquist, H.K. Mental retardation in children: an epidemiological and aetiological study of mentally retarded children born 1959-/970 in a northern Swedish county. Umea University, Medical dissertations, 1982 (New series No. 76). 119 Prevention and control of prenatal and perinatal viral infections J.A. Dudgeon The general approach to the prevention and control of infections caused by rubella virus, human (beta) herpesvirus 5, hepatitis virus B and human (alpha) herpesviruses 1, 2 and 3, which may constitute a hazard during the early stages of pregnancy and the perinatal period, depends on the natural history and pathogenesis of each individual infection. Methods of preven- tion and control can be considered under three main headings. 1. Specific prophylaxis active immunization with a vaccine passive immunization by means of antisera or immunoglobulin a combination of passive and active immunization 2. Specific treatment - antiviral chemotherapy 3. General or nonspecific procedures. As a general working rule specific prophylaxis by active immunization is the method that should be adopted when a large section of a population is at risk, whereas passive immunization, antiviral therapy or nonspecific measures are more applicable to the individual case after exposure (see Table 1). The decision to adopt one or other method of prevention and control will depend on the natural history of the infection and the number of fetuses and newborn babies likely to be exposed to such infections. For example, Table 2 shows the incidence of major defects in children, deter- mined at or soon after birth, following a clinically diagnosed virus infec- tion in pregnancy. This prospective study carried out in the United King- dom in 1950-1952 (1) showed that maternal rubella virus infection in 121 Table 1. Methods of prevention and control of infections Cause Rubella virus Human (beta) herpesv1rus 5 Hepat1t1s virus B Human (alpha) herpesv1ruses 1 and 2 Human (alpha) herpesv1rus 3 Method Active ImmunIzatIon prior to pregnancy Passive ImmunIzatIon In selected cases Nii at present Passive ImmunIzatIon of babies at birth (Possibly also active ImmunizatIon In the neonatal period) Obstetric care with regard to delivery Ant1v1ral c hemotherapy Passive ImmunIzatIon with zoster 1mmunoglobul1n Ant1v1ral chemotherapy with ac1clov1r Table 2. Virus infections in pregnancy in the United Kingdom . 1950-1952 Infection Rub ella virus Mumps virus Human (alpha) herpesv1rus 3 Measles virus Human polio viruses Influ enza viruses Total Controls Source: Manson, H.M . et al . /I/. 122 No. of live births 547 487 288 99 30 163 1614 5469 Per ce ntage with maior defects 6 .8 2.3 2. 7 .1 3.3 3.2 4 .2 2.3 pregnancy was the only infection that constituted a serious hazard to the fetus. The incidence of major defects after measles virus infection in pregnancy was also greater than in the control group, but the type of defect was variable and showed no consistent pattern as is so invariably observed after cases of maternal rubella virus infection. Following this and a later prospective study of the effects of rubella virus infection (2), it appeared that two major factors determined the risk to the fetus: the gestational age at which the infection was acquired and the immune status of the mother. Initially, the period of greatest risk was thought to be the first trimester, but on closer analysis it has become evident during the past two decades that there is a risk to the fetus following infection between the 16th and 20th week. Furthermore, experience has shown that in the case of rubella virus the effect on the fetus is almost entirely related to primary infections acquired in pregnancy. Although a few documented cases of congenital rubella virus infection have been reported, following the re-exposure or re-infection of a pregnant woman considered to be immune from a previous natural infection, these cases are so rare that they are not considered to be a risk of any significance. Subsequent studies on the effect of rubella and other virus infections in pregnancy have emphasized the importance of the overall effect on the fetus and newborn infant. Major defects or malformations are of special consequence, but these should not detract from the importance of other forms of damage resulting from a disseminated virus infection occurring at a critical stage in the fetal or perinatal period (3). An attempt has been made to assess the incidence of fetal infections from various sources in the United Kingdom (Table 3). It is evident that the size of the problem in terms of total fetal damage is small and although it is probable that there are geographical differences, certainly in respect of hepatitis B virus infec- tion, it is unlikely that the overall figure will vary greatly from the figures quoted. Methods of prevention and control can now be considered for the five virus infections mentioned in Table 3. Rubella virus infection The reason for the specific prophylaxis of maternal rubella virus infection is the prevention of congenital defects. Congenital rubella is a preventable disease. This can be accomplished by active immunization and to a lesser extent by the use of immunoglobulin. Live attenuated rubella vaccines were licensed for use in many countries over a decade ago and since then extensive experience has been gained regarding their usage. It is now abundantly clear that congenital rubella occurs as the conse- quence of primary maternal rubella virus infection in pregnancy. The few reported cases of congenital rubella following re-infection can be dis- counted for practical purposes. It is also evident that vaccine-induced 123 Table 3 . In cidence of prenatal and perinatal infections in the United Kingdom Infection Congenital rubella virus 1nfect1on Congeni tal cytomegalov1rus 1nfect1on Congenital hepa t1t1s B vi ru s 1n fec t1 on j Perinatal human (alpha) herpesv1 ru s 1 and 2 infection Perinatal human (alpha) herp esv1 ru s 3 1nfec t1 on a Based on a birth rate of 800 000 per annum Rate per 1000 live bi rt hs 0 .25-0 .3 0.5 -1 .0 Too rare to es timat e Source: lnrrautenne 1nfec11ons. A Ciba Foundation Symposium /4) Expected number per annuma 200 -250 400-800 rubella is usually a very mild but, more important, non-transmissible disease. The vaccine can therefore be administered to susceptible individ- uals without risk of spread. As the vaccine virus is potentially teratogenic, however, it must not be administered to anyone who is pregnant or is likely to become so within the next three months. The object of vaccination is to induce immunity prior to the age of childbearing. This can mean inducing a high degree of herd immunity by means of mass vaccination of all children of both sexes from the second year of life to puberty, with a secondary emphasis on the vaccination of known susceptible adolescent or adult females, especially those at special risk such as doctors, nurses, teachers and hospital personnel. This is the method used in the United States and many parts of Canada. By contrast, in the United Kingdom and some European countries, girls aged 10-13 are the main target group for select ive immunization but there is increasing emphasis on adult females known to be susceptible as the result of a blood test. Postpartum vacci- nation is actively encouraged. Neither scheme has yet achieved the desired effect, although in the United States there has been a significant change in the epidemicity of natural rubella and of the reported incidence of congenital rubella (5). On the other hand rubella continues to occur in adolescent and young adults according to the Morbidity and mortality weekly report put out by the Centers for Disease Control in Atlanta. In the United Kingdom, the acceptance rate of vaccine amongst schoolgirls aged 11-14 (the starting age 124 has recently been reduced to the 10th year of life) is around 75-85% , and the incidence of congenital rubella has shown little change. A far greater degree of urgency is the need for vaccination in all countries. This may well be achieved in the United States by means of the strategy of using com- bined measles-mumps-rubella vaccine and requiring a vaccination certifi- cate before school entry. It is unlikely that such a scheme would work in the United Kingdom, so what can be done to improve the situation? A far greater awareness of the need for vaccination must somehow be instilled into the public and the professions. Prevention is an educational as well as a health problem. Alternatively, one could employ a two-dose schedule. The first dose given at 15-18 months of age and the second at school entry as recommended by Krugman (6) and more recently by Taranger in Sweden (7) . Such a procedure in practice means that immuniz- ation would have to be carried out by means of a combined measles-mumps- rubella or measles-rubella vaccine. In some countries such as the United Kingdom, where the uptake of measles vaccine is no greater than 80% , it is a matter of opinion whether such a theoretical exercise would work in practice. Passive immunization The prophylactic effect of immunoglobulin in the prevention of rubella following contact with a rubella virus in pregnancy is controversial. It is generally accepted, however, that it should only be used in special circum- stances where the termination of a pregnancy is not desired and where careful laboratory assessments can be made (8) . Cytomegalovirus infection The need to prevent cytomegalovirus infection in pregnancy is much greater than with rubella virus infection. Cytomegalovirus infection, par- ticularly a primary infection, is a major environmental cause of mental retardation. Live attenuated vaccines have been developed but at present none have been licensed for use in these particular circumstances. One of the main reasons for lack of progress is the fear that a live human (beta) herpesvirus 5 vaccine might be oncogenic. Apart from the fact that human (beta) herpesvirus 5 is a DNA virus there is no other reason why it should be oncogenic, but the risk remains and until the matter is resolved, or an alternative vaccine developed possibly by means of genetic manipulation, little progress is envisaged (8). Hepatitis B virus infections Initial reports on the effect of maternal hepatitis 8 virus infections on the fetus have almost certainly underestimated the risk to the fetus (8). More 125 recent ev idence suggests that the risk of the fetus contracting chronic or chronic-active hepatitis is considerable following ac ute hepatiti s B in pregnancy. There is evidence that passive immuniza tion has some effect in reduc- ing the incidence of antigenaemia (9), but whether such a protective effect will be long-lasting is difficult to say with certainty. Another possible approach to prophylaxis in such cases is combined passive-ac ti ve immuniz- ation using both immune globulin and hepatitis B vacc ine. Human (alpha) herpesvirus 1 and 2 infection Nahmias et al. ( 10) have carefully reviewed the effect of human (alpha) herpesvirus I a nd 2 infections on the newborn . No specific prophylactic measure is avai lable and the question that remains is whether delivery sho uld be by the vaginal route or by Caesarian section if there is evidence of herpesvira l vulvovaginitis, cervicitis or urethritis in the mother. It now seems to be genera lly agreed that if there has been a hi story of such disease in the mother in a previous pregnancy, delivery should be by the vaginal route. If the disease was present early in pregnancy, but subsequent tests show that it has cleared up, then vaginal delivery should still be carried out. Special attention should be paid to the newborn and if any evidence of herpetic lesions appears then appropriate antiviral chemotherapy should be inst" uted \\ ithout delay. If the disease is active a t the time of parturi- tion, deli , ·ry ~hou ld be by Caesarian section and immediate treatment wi th antiviral chemotherapy should be instituted with the most effective compound available . A newly licensed compound, ac iclovir , shows con- siderable promise in such circu mstances. In this type of case it is important to carry out a Caesarian section befo re rupture of the membranes. Human (alpha) herpesvirus 3 infection A small number of cases of congenital infection with human (alpha) herpesvirus 3 have been reported, but they are so rare that no special measures a rc recommended (8). There is evidence that the risk of a dissem- inated infection is greatest when maternal infection occurs four days or less before delivery. Details from a study of 50 cases in New York State are shown in Ta ble 4. Although prophylaxis with zoster immunoglobulin is recommended in a dosage of 0.6 ml per kg of body weight , prevention ca nno t be guaranteed . In severe cases or in those that come into the severe at-risk group, antiviral chemotherapy is also recommended. It is concl uded that as the risk to the fetus following infection with human (alpha) herpesvirus 3 in early pregnancy is so rare, termination of pregna ncy is not recommended. In view of the undoubted existence of a few cases of congenital infection with human (alpha) herpesvirus 3 which 126 Table 4 . Th e effect of maternal infection with human (alpha) herpesvirus 3 near term on the fetus ( 50 cases) Onset M atern al infec ti on five or more days before delivery and baby·s infection at four days old or less Maternal infec t ion four days or less before delivery and baby"s infection at five to ten days o ld Source: Gershon, A .A . ( 1 1 ). Effect 2 7 out of 2 7 su rvived 16 out of 23 survived. 7 died of dissemina ted var1cella. 2 had seve re disease but survived can lead to gross deformities, however, it is recommended that steps be taken by the referring physician to determine whether fetal damage has occurred, where appropriate, by the use of modern techniques now avail- able in some centres . In the case of maternal infection close to term , the newborn baby should be given zoster immunoglobulin in the appropriate dose, as indi- cated above, immediately on delivery. As zoster immunoglobulin is not freely available, it should be confined to cases where maternal infection has occurred within five days of delivery. References I. Manson, M.M. et al. Rubella and other virus infections during preg- nancy. Rep. Pub. Hlth. Med. Subj. No. 110. 2. Lundstrom, R. Rubella during pregnancy. A follow-up study of child- ren born after an epidemic of rubella in Sweden, 1951 . With additional investigations on prophylaxis and treatment of maternal rubella . Acta paediatrica, 51 (Suppl. 133): 1-110 (1962). 3. Dudgeon, J .A. Infective causes of human malformations. British medi- cal bulletin. 32: 77-83 ( 1976). 4. Intrauterine infections. A Ciba Foundation Symposium. Amsterdam, Elsevier, I 973. 5. Rubella surveillance, /976-1978. Atlanta, Centers for Disease Control, 1980. 127 6. Krugman, S. Rubella immunization. Present status and future perspec- tives. Pediatrics. 65: 1174 ( 1980). 7. Taranger, J. Vaccination programme for eradication of measles, mumps and rubella. Lancet, 1: 915-916 (1982). 8. Hanshaw, J.B. & Dudgeon, J .A. Viral diseases of the fetus and newborn. Philadelphia, W.B. Saunders, 1978. 9. Beasley, R.P. et al. Hepatitis B IG efficacy in the interruption of perinatal hepatitis B carrier state. Lancet, 2: 388-392 ( I 981 ). 10. Nahmias, A.J. et al. Simplex infections in the fetus and newborn. In: Infections of the fetus and the newborn infant. New York, Alan R. Liss, 1975, p. 63 . 11. Gershon, A.A. Varicella in mother and infant : problems old and new. 128 In: Infections of the fetus and the newborn infant. New York, Alan R. Liss, 1975, pp. 79-95. Conclusions and recommendations 1. The control of prenatally and perinatally acquired infections should be encouraged and promoted through a number of activities. The health organizations of individual nations shou ld take appropriate measures such as the institution of surveillance, case finding, serological surveys, vaccina- tion regimens and evaluation programmes . 2. The long-term follow-up of children with evidence of prenatal and perinatal infection should be' carried out from birth up to school entry and even longer, if possible. 3. Infections due to Group B streptococci (Streptococcus agalactiae) and to Listeria monocytogenes may be acquired during intrauterine life or during or shortly after delivery. A fatal o utcome is common, but can be prevented by the use of appropriate and rapid diagnostic procedures that should be followed at the sligh test clinical suspicion of infectio n so that appropriate chemotherapy can be instituted without delay. 4. In numerical terms, primary cytomegalovirus infection is a more common cause of congenitally acquired disease than primary rubella virus infection and results in severe handicap inducing mental subnormality and sensorineural deafness . Attenuated vaccines, that have been employed in recent trials, are unlikely to be licensed for the general prophylaxis of susceptible healthy women on account of their potential oncogenicity. Alternative vaccines, such as a subunit vaccine, need to be developed. The continuous epidemiological surveillance of cytomegalovirus infection in different populations is also required. 5. All possible steps should be taken to eradicate congenital rubella by employing a vaccinat ion programme appropriate to the health organiz- ation of individual nations . 129 6. Women may inadvertently receive rubella vaccine shortly before or after conception. In this event, routine termination of pregnancy is not recommended but the final decision must rest with the vaccinated woman and those caring for her during her pregnancy. A large number of rubella virus-susceptible and pregnant women have been given attenuated rubella vaccine and chosen to go to term. Although no rubella virus-induced defects have been reported in newborn infants, the number of pregnant women given one of the now commonly used vaccines (RA27 /3 vaccine strain) is too small to completely exclude possible adverse effects. Each country should therefore carry out follow-up studies on women inadvert- ently vaccinated in pregnancy. 7. There is concern over the apparently increasing incidence of herpes- viral genital infections as they may induce perinatal infections with a high risk of perinatal morbidity and mortality. Since it is recognized that vaginal delivery is not always advisable, it is recommended that women with a history of infection and/or suspicious symptoms be tested for virus shedding a few days prior to term. If human (alpha) herpesvirus I or 2 is present, delivery by Caesarian section should be considered. 8. As the laboratory diagnosis of Toxoplasma gondii infection is very often carried out unsatisfactorily, evidence of infection is often detected too late for therapy to be effective. Appropriate steps should be taken to ensure a high standard of technical performance in adequately equipped laboratories. As prenatal Toxoplasma gondii infection often becomes overt in early or late childhood, long-term follow-up studies including careful clinical and parasitological tests should be carried out. If during screening in the early stages of pregnancy, a patient is found to be seronegative, repeated serological examination for Toxoplasma antibodies (monthly to bimonthly) are recommended to detect seroconversion. 9. Although the number of cases of congenital syphilis has declined markedly in Europe, there is little evidence of a decreasing incidence of sexually acquired disease. The continued routine screening for syphilis is strongly recommended. 10. Rapid and appropriate laboratory procedures should be encouraged to implement prophylactic measures, document infections during preg- nancy and in the newborn and, where appropriate, provide treatment. Specific IgM antibody determination is the best method for establishing pre- and perinatal infection. Particular care should be taken to exclude nonspecific reactions. Results should be interpreted in conjunction with other serological tests as well as clinical data before advising whether or not a pregnancy should be terminated. It is recommended that any such 130 decision be based o n th e highest quality of clinical and labora tory diag- nos is and be consistent with nati o nal legislati o n. The poss ible widespread use of commercially available kit s by inexperienced perso nnel is viewed with considerable app rehensi on. 11 . While the commitment to a multisec toral approach for the after-care and long-term rehabilitation of children with congenital defects is endorsed, the extremely high costs of maintaining such services further emphasizes the importance of preventive measures . 12. Further epidemiological monitoring of congenital defects among newborn infants is recommended to clarify existing and detect new associ- ations with infections in pregnancy . Such studies should preferably be prospective studies but if this is not possible, well designed retrospective ones are a satisfactory alternative. Because the number of infants with various defects induced by infectious agents may be small, collaborative studies should be encouraged. 131 Annex 1 PARTICIPANTS Temporary advisers 132 Professor H. As pock, Head of the Department of Medical Parasitology at the Institute of Hygiene, University of Vienna, Austria Dr J.E. Banatvala, Professor of Clinical Virology, Head of Depart- ment of Virology, St Thomas's Hospital and Medical School, Lon- don, United Kingdom (Rapporteur) Professor A. Boue, Institut national de la Sante et de la Recherche medicale, Groupe de Recherches de Biologie prenatale, Paris, France Dr A.J. Clayton,a Director-General, Laboratory Centre for Disease Control, Health and Welfare Canada, Ottawa, Ontario, Canada Professor J .A. Dudgeon, Emeritus Professor of Microbiology, Univer- sity of London, United Kingdom Professor G. Enders, Chief, Virological-Medical Diagnostic Institute, Stuttgart, Federal Republic of Germany Dr G. Granroth, Pathologist, Department of Pathology, Central Hos- pital of Vasa, Finland Professor G. Kistler, Chief Medical Officer, The City of Zurich, Switzerland Professor U. Krech, Institute of Medical Microbiology, St Gallen, Switzerland a Participation expenses not paid by WHO. Professo r K. Menzel, Childrens' Clinic oft he Erfurt Medica l Academy, German Dem ocratic Republic Professo r J .R. Mose, Direc to r , In stitute of H ygiene, Uni vers it y of Graz, Austria Dr R. Najera, Alberto Alcocer 44, Madrid, Spain Dr H.P.R. Seeliger, Institute for Hygiene a nd Microbiol ogy, Wi.ir z- burg, Federal Republic of Germany Dr V.M. Syde lni kova, Head of the Department of the All-Union Centre for Materna l and Child Hea lth, c/o Ministry of Health of the USSR, Moscow, USS R World Health Organization Headquarters Dr G.M . Antal, Bacterial and Venereal Infecti ons Dr T. Bektimirov, Virus Diseases Dr R.J. Guidotti, Maternal and Child Health Regional Office fo r Europe Dr J.H . Henderson , Regiona l Officer fo r Mental Hea lth Dr B. Velimi rov ic, Regio nal Officer for Communicable Diseases 133 RESUME Introduction Le groupe de travail, reunissant un embryologiste , des epidemiologistes et des microbiologistes, avait ete organise pour faire le point des connais- sances acquises et des problemes qui subsistent concernant l'etiologie, Jes diagnostics de laboratoire, l'epidemiologie et, le cas echeant, la prevention et le traitement des infections prenatales et perinatales. Le but recherche consistait a formuler des recommandations pour Jes soumettre aux services de sante nationaux . II est prevu que Jes observations du groupe presenteront de l'interet non pas seulement du point de vue d'une utilisation rationnalisee des compe- tences et ressources sanitaires existantes aux fins de la prevention des infec- tions, mais aussi de !'identification des voies que pourrait ulterieurement emprunter la recherche. Si l'on admet souvent que Jes infections constituent un probleme relativement mineur sur le plan quantitatif, le nombre d'enfants qui, de ce fait , subissent des dommages , pourrait bien avoir ete sous-estime de fac;on notable . De nombreuses infections dont sont atteintes Jes meres, et notamment celles qui sont provoquees par des cytomegalovirus et Toxo- plasma gondii, ne relevent pas directement de la clinique ou produisent des sympt6mes relativement mineurs et non specifiques, d'ou ii s'ensuit qu 'elles passent inaperc;ues et ne font J'objet d'aucune etude . Les enfants sujets a risque peuvent etre apparemment en bonne sante a la naissance , alors que des etudes de suivi sur de longues periodes montrent qu'ils peuvent ulterieurement presenter des anomalies du developpement de severite variable , y compris l'arrieration men tale. Neanmoins, la confirmation du diagnostic des infec- tions, grace a des tests de laboratoire au COUTS d'evaluation realises a un stade plus tardif de la premiere ou de la seconde enfance , peut se reveler difficile voire impossible. En consequence, le groupe a souligne combien ii irnporte de proceder a de nouvelles enquetes prospectives dans la mesure ou elles per- mettent de mieux apprecier !'incidence et Jes effets a long terme des infections. Nombreuses sont Jes methodes de laboratoire, decrites au cours de la reunion - et l'on citera notamment Jes irnmunotests enzymatiques de depis- tage des reponses serologiques aux infections, et particulierement Jes reponses specifiques aux irnmunoglobulines (M) - qui presentent un interet potentiel considerable quant aux possibilites d'accroitre la sensibilite et la fiabilite des diagnostics d'infection maternelle et le monitorage des nourrissons a haut 135 risque. 11 est neanmoins essentiel que les cliniciens, les epidemiologistes et les specialistes de laboratoire voient leur attention attiree sur les ecueils possibles des lors que l'on fait usage de ces methodes. En particulier, des trousses sont desormais disponibles dans le commerce et risqueraient de provoquer des diagnostics errones si elles etaient utilisees par des mains inexperirnentees. Quelle que soit la frequence des infections, il ne saurait etre douteux que les nourrissons, qui subissent un dommage de leur fait , posent un pro- bleme economique et social considerable , car ils necessitent souvent des soins durant toute une vie . Une approche multidisciplinaire aux problemes relatifs a l'etiologie , a la prevention et a la gestion des infections s'irnpose done. Conclusions et recommandations 1. 11 conviendrait de promouvoir la lutte contre les infections prenatales et perinatales par un certain nombre d 'activites. 11 est notamment recommande que les services de sante de chaque pays entreprennent les actions voulues telles que !'institution d'une surveillance, d'un depistage, d'enquetes sero- logiques, de plans de vaccination et de programmes d'evaluation . 2. 11 est vivement recommande de suivre pendant de longues periodes les enfants portant la marque des infections. Si possible, il conviendrait de les suivre de la naissance a leur entree a l'ecole, voire plus longtemps. 3. Les infections dues aux streptocoques du groupe B (Streptococcus aga/actiae) et a Listeria monocytogenes peuvent etre acquises pendant la gestation , pendant l'accouchement ou irnmediatement apres. Une issue fatale n'est pas rare, encore qu'elle puisse etre evitee grace a des actes de dia- gnostic appropries et rapides, qu'il conviendrait de pratiquer des le moindre sympt6me clinique d'infection, de fa~on a pouvoir entreprendre sans retard la chimiotherapie necessaire. 4 . Quantitativement , les infections primaires a cytomegalovirus constituent une cause plus frequente d'infections congenitales que la rubeole primaire et entrainent des incapacites severes engendrant une arrieration mentale et une surdite de perception . Les vaccins attenues employes dans des essais qui ont recemment fait l'objet de communications n'ont que peu de chances d'etre agrees pour la prophylaxie generalisee des femmes en bonne sante mais vulnerables, compte tenu de l'oncogenicite potentielle de ces vaccins. 11 irnporte de mettre au point d'autres substances, notamment un vaccin constitue de fractions antigeniques. 11 faut aussi proceder a la surveillance continue SUI le plan epidemiologie des infections a cytomegalovirus. 5. 11 faudrait prendre toutes les mesures possibles pour eradiquer la rubeole congenitale par un programme de vaccination tenant compte de !'organisation sanitaire de chaque pays. 136 6. Les femmes peuvent etre accidentellement vaccinees contre la rubeole peu avant ou peu apres la conception. Dans ce cas, le groupe, ne recomman- dant pas une interruption de grossesse systematique , considere que la decision definitive doit incomber a la femme vaccinee et a ceux qui la suivent pendant sa grossesse. Un grand nombre de femmes enceintes vulnerables a la rubeole ont re9u des vaccins attenues contre la rubeole et ont choisi de mener leur grossesse a terme . Bien qu 'aucune lesion provoquee par la rubeole n 'ait ete signalee chez les nouveau-nes , le nombre de femmes enceintes ayant ete vac- cinees avec Jes souches le plus communement employees (vaccin RA2713) demeure trop limite pour que !'on puisse exclure completement toute possibi- lite d'effets negatifs. Chaque pays devrait proceder a des etudes de suivi sur Jes femmes ayant ete par inadvertance vaccinees en cours de grossesse . 7. L'incidence apparemment croissante de !'herpes genital, qui peut entrainer des infections perinatales associees a un risque eleve de morbidite et de mortalite perinatales, est preoccupante. Vu que l'accouchement par le vagin n'est pas toujours indique, ii est recommande que Jes femmes, pre- sentant des antecedents d'herpes genital et/ou des sympt6mes suspects, devraient etre sournises a un test de depistage de virus quelques jours avant le terme . En cas de resultat positif, ii conviendrait d'envisager un accouche- ment par cesarienne. 8. Le diagnostic de laboratoire de la toxoplasmose est tres souvent insatis- faisant et la presence de !'infection est souvent constatee trop tard pour que la therapie soit efficace . II est recommmande que Jes mesures appropriees soient prises pour assurer une performance tres elevee des laboratoires sur le plan technique; ces laboratoires doivent etre convenablement equipes. La toxoplasmose prenatale se declare souvent dans la premiere enfance ou plus tardivement; aussi conviendrait-il de proceder a des etudes de suivi a long terme comportant des epreuves cliniques et parasitologiques effectuees avec soin. Si, aux stades precoces de la grossesse, Jes examens serologiques sont negatifs , on recommande qu'ils soient repetes tous Jes mois a tous Jes deux mois, afin de determiner Jes anticorps dans le toxoplasme pour que la seroconversion ne passe pas inaper9ue . 9. Bien que le nombre d 'episodes de syphilis congenitale ait decline de fa9on marquee en Europe , ii n'est pas prouve que !'incidence des maladies acquises par Jes voies sexuelles decroisse . En consequence, ii est vivement recommande de poursuivre Jes campagnes regulieres de depistage de ces maladies . 10. II faudrait encourager le recours a des protocoles de laboratoire rapides et appropries en vue d'une action prophylactique, de la mise en evidence de la presence d 'infections en cours de grossesse et chez le nouveau-ne et , 137 le cas echeant, du traitement. Determiner des anticorps specifiques des immunoglobulines M constitue la meilieure fa~on d 'identifier !es infections prenatales et perinatales. II faudrait s'efforcer tout particulierement d'exclure Jes reactions non specifiques. Cependant, Jes resultats devraient etre inter- pretes en liaison avec d'autres observations serologiques et avec des donnees cliniques avant qu'il soit conseille a une patiente d'interrompre ou non une grossesse. Toute decision de cet ordre devrait etre fondee sur des diagnostics cliniques et de laboratoire de la qualite la plus elevee possible , dans le respect des legislations nationales . Le groupe prevoit la possibilite d'une large utilisa- tion de trousses disponibles dans le commerce par un personnel inexperimente et s'en preoccupe vivement. 11 . Le groupe s'assoc ie au choix de l'approche multisectorieUe envers la convalescence et la readaptation a long terme d 'enfants atteints de deficiences congenitales. Le emit extremement eleve du maintien de ces services est un aspect important qui fait ressortir d 'autant plus la necessite de prendre des mesures preventives appropriees . 12. Le groupe recommande la surveillance continue sur le plan epidemio- logique des malformations congenitales chez Jes nourrissons afin de preciser Jes connaissances acquises et de determiner des associations nouvelles avec Jes infections en cours de grossesse. Les etudes a cette fin devraient , de prefe- rence, etre prospectives mais , a defaut , des analyses retrospectives bien con~ues peuvent egalement donner satisfaction . Le nombre de nourrissons presentant des lesions variees , produites par des agents infectieux , n 'est peut-etre pas tres eleve ; aussi conviendrait-il d 'encourager Jes etudes en collaboration. 138 ZUSAMMENF ASSUNG Einleitung Die Arbeitsgruppe, an der ein Embryologe, Epidemiologen und Mikrobiolo- gen teilnahmen, war zum Zwecke der Erorterung des gegenwartigen Wissens- stands und der noch bestehenden Probleme auf den Gebieten A tiologie, Laboratoriumsdiagnostik, Epidemiologie sowie ggf. Pravention und Behand- lung von pranatalen und perinatalen Infektionen und mit dem Ziel , Empfeh- lungen an die nationalen Gesundheitsbehorden abzugeben, einberufen worden. Die Ergebnisse der Tagung di.irften nicht nur for eine sinnvollere Umset- zung bzw . Inanspruchnahme der vorhandenen Fachkenntnisse und Gesund- heitsressourcen hinsichtlich der Vorbeugung gegen derartige Infektionen, sondem auch for die ktinftige Ausrichtung der Forschung von Wert sein. Ob- gleich diese lnfektionen oft for ein numerisch relativ unbedeutendes Problem gehalten werden , ist die Zahl der dadurch geschadigten Kinder bisher wahr- scheinlich bedeutend unterschatzt worden . Viele lnfektionen bei Miittem, wie beispielsweise die durch den Zytomegalievirus und Toxoplasma gondii hervor- gerufenen lnfektionen , sind subklinisch oder zeigen verhaltnismaBig geringe, nichtspezifische Symptome und bleiben deshalb unerkannt und unbehandelt. Die gefahrdeten Kinder konnen bei der Geburt gesund wirken,jedoch wurde in langeren Folgestudien aufgezeigt, daB bei ihnen spater unterschiedlich sch were En twicklungsstbrungen bzw. Anomalien, einschlieBlich geistiger Retardierung, auftreten konnen . Es diirfte jedoch schwierig oder sogar unmog- lich sein, durch Labortests bei der Feststellung einer Anomalie im spateren Sauglings- oder Kindesalter die Diagnose derartiger lnfektionen zu bestatigen . Deshalb hob die Arbeitsgruppe die Wichtigkeit der Durchfohrung zusatzlicher Prospektivuntersuchungen hervor, well dadurch eine bessere Beurteilung der Haufigkeit und der langfristigen Auswirkungen dieser Infektionen moglich sein konnte . Viele der im Verlauf der Tagung beschriebenen Laborverfahren , wie bei- spielsweise die kiirzlich entwickelten Enzym-Immunoassays zur Feststellung serologischer Reaktionen, um Infektionen zu diagnostizieren, vor allem spezi- fische IgM-Spiegel, sind besonders geeignet in bezug auf eine groBere Sensi- bilitat und Zuverlassigkeit bei der Diagnose von Infektionen der Mutter und der Dberwachung der Kinder stark gefahrdeter Gruppen . Trotzdem ist es 139 unbedingt erforderlich , Kliniker, Epidemiologen und Laborpersonal auf mog- liche Irrtiimer bei der Anwendung dieser Methoden aufmerksam zu machen. Vor allem sind jetzt Testsets auf dem Markt , die in der Hand von unerfah- renem Personal zu Fehldiagnosen fohren kbnnen. Wie haufig oder selten diese Infektionen auch immer auftreten mbgen , so gibt es keinen Zweifel dari.iber, d~ Kinder mit daraus resultierenden Scha- digungen ein betrachtliches wirtschaftliches und gesellschaftliches Problem darstellen , da oft eine lebenslange pflegerische Betreuung erforderlich ist. Des- halb besteht die Notwendigkeit einer multidisziplinaren Verfahrensweise in bezug auf die A tiologie , Verhiitung und Behandlung derartiger Infektionen . SchluEfolgerungen und Empfehlungen I. Die Bekampfung von pranatalen und perinatalen Infektionen sollte <lurch eine Reihe von M~nahmen unterstiitzt werden. Es wird empfohlen, d~ die Gesundheitsorganisationen eines jeden Landes geeignete Schritte untemeh- men, wie z.B . die Ausarbeitung von Programmen zur Oberwachung und Friih- erkennung sowie for serologische Erhebungen, lmpfvorschriften und Aus- wertungen. 2. Bei Kindem mit nachgewiesenen lnfektionen wird eine langfristige Nach- betreuung dringend empfohlen, die vom Zeitpunkt der Geburt bis zur Ein- schulung oder, wenn moglich, sogar noch langer dauem sollte. 3. Durch Streptokokken der Gruppe B (Streptococcus agalactiae) und Listeria monocytogenes verursachte Infektionen kbnnen wahrend der Schwan- gerschaft oder bei bzw. kurz nach der Entbindung erworben werden. 1hr Ver- lauf ist for gewbhnlich tbdlich, was jedoch <lurch eine richtige und rasche Diagnose bereits bei dem geringsten klinischen Verdacht einer Infektion und die unverziigliche Einleitung der erforderlichen Chemotherapie verhindert werden kann. 4. ZahlenmaEig ausgedriickt ist eine kongenital erworbene K.rankheit hau- figer die Folge der primaren Zytomegalievirus-lnfektion als der primaren Rubeola und sie fiihrt zu schweren Behinderungen, einschlieBlich geistiger Subnormalitat und sensorisch-neuraler Taubheit. Wegen der potentiellen Onkogenizitat ist es unwahrscheinlich, d~ die in kiirzlich beschriebenen Ver- suchen verwendeten abgeschwachten Zytomegalievirus-lmpfstoffe zur allge- meinen Prophylaxe zugelassen werden. Die Entwicklung von altemativen Substanzen, wie beispielsweise Spalt-lmpfstoffe, ist erforderlich. Ferner besteht die Notwendigkeit einer fortlaufenden epidemiologischen Oberwa- chung der Zytomegalievirus-Infektionen bei den verschiedenen Bevblkerungs- gruppen. 140 5. Es soil ten samtliche moglichen Schritte zur Ausrottung der kongenitalen Roteln mittels eines dem Gesundheitswesen eines jeden Landes angepaEten lmpfprograrnms eingeleitet werden . 6 . Eine Frau kann unbeabsichtigt kurz vor oder nach Beginn einer Schwan- gerschaft Rubeola-Vakzine erhalten. In diesem Fall empfiehlt die Arbeits- gruppe keinen routinemiilligen Schwangerschaftsabbruch ; vielmehr sollte die endgilltige Entscheidung der geimpften Frau und den sie wahrend der Schwangerschaft betreuenden Medizinem uberlassen bleiben. Viele roteln- verdachtige Schwangere haben abgeschwachte Rubeola-Vakzine erhalten und sich entschieden , <las Kind auszutragen. Obgleich keine <lurch Roteln verur- sachte Schadigungen bei den Neugeborenen bekanntgeworden sind, ist die Anzahl der schwangeren Frauen, bei denen einer der nunmehr ublicherweise verwendeten lmpfstoffe (RA2713 Vakzine-Starnm) verwendet wurde , noch zu gering, um die Moglichkeit von schadlichen Nebenwirkungen vollstandig ausschlieBen zu konnen. Deshalb soil ten in jedem Land bei Frauen , die verse- hentlich wahrend einer Schwangerschaft geimpft wurden, weitere Dberwa- chungen vorgenommen werden . 7 . Die offensichtlich zunehmende Haufigkeit von Herpes genitalis , der for perinatale lnfektionen mit einem hohen perinatalen Morbiditats- und Mortali- tatsrisiko ursachlich sein kann , ist besorgniserregend. Angesichts der Erkennt- nis, daB eine vaginale Entbindung nicht immer ratsarn ist, wird empfohlen, daE Frauen mit festgestelltem Herpes genitalis und/oder verdachtigen Sym- ptomen einige Tage vor der Niederkunft auf Virusausscheidungen untersucht werden. Wenn der Virus nachgewiesen wird, sollte eine Entbindung <lurch Kaiserschnitt in Erwagung gezogen werden . 8. Da die Diagnose der Toxoplasmose <lurch Labortests haufig unbefriedi- gend ist, erfolgt der Nachweis einer Infektion oft zu spat for eine wirksarne Therapie . Es wird empfohlen, geeignete MaEnahmen zu treffen, um eine tech- nisch hochqualifizierte Verfahrensweise in entsprechend ausgeriisteten Labo- ratorien sicherzustellen. Da eine pranatale Toxoplasmose sich haufig erst im fruhen oder fortgeschrittenen Kindesalter manifestiert , sollten langfristige F ollow-up-Studien einschlieBlich sorgfal tiger klinischer und parasitologischer Tests vorgenommen werden . 1st <lurch eine Screening-Untersuchung im Fruh- stadium einer Schwangerschaft bei einer Patientin der seronegative Nachweis erbracht worden, empfiehlt es sich, weitere serologische Untersuchungen auf Toxoplasmose-Antiki:irper (im Abstand von einem Monat oder zwei Monaten) zur Feststellung einer Serokonversion durchzufiihren. 9 . Obgleich die Fa.lie von konnataler Syphilis in Europa deutlich zuruckge- gangen sind, gibt es kaum Beweise for ein geringeres Auftreten der <lurch Sexualkontakt erworbenen Krankheit. Desha.lb wird nachdrucklich angeraten, die routinemiilligen Screening-Untersuchungen auf Syphilis fortzusetzen . 141 10. Rasche und geeignete Laborverfahren sollten gefordert werden , um pro- phylaktische MaBnahmen durchzufiiluen , lnfektionen wahrend der Schwan- gerschaft und beim Neugeborenen nachzuweisen und erforderlichenfalls eine Behandlung in dje Wege zu leiten . Die beste Methode zur Feststellung von pranatalen und perinatalen Infektionen ist ein IgM-Spiegel. Es sollte besonders sorgfaltig vorgegangen werden , um nichtspezifische Reaktionen auszuschlie- Ben . Jedoch sollten vor dem Anraten eines Schwangerschaftsabbruchs die Er- gebnisse zusammen mH den Ubrigen serologischen Befunden und klinischen Oaten ausgewertet werden . Es wird nahegelegt, jede derartige Entscheidung auf der Grundlage hochqualifizierter K.linik- und Laboriliagnosen und nach MaBgabe der gesetzlichen Vorschriften eines Landes zu treffen. Die Arbeits- gruppe beftirchtet ilie Moglichkeit einer allgemeinen Anwendung von handels- Ublichen Testsets <lurch unerfahrenes Personal . 11 . Die Arbeitsgruppe beftirwortet zwar ein multisektorielles Vorgehen in bezug auf dje Nachsorge und Langzeitrehabilitation bei Kindern rrut ange- borenen Defekten , jedoch sind die auBerordentlich hohen Kosten fiir die Auf- rechterhaltung ilieser Dienste ein wichtiger Gesichtspunkt, der die Notwen- digkeit geeigneter VorbeugungsmaBnahmen deuilich macht. 12. Die Arbeitsgruppe empfiehlt eine weitere epiderruologische Oberwachung der angeborenen Defekte bei Neugeborenen , um den jetzigen Wissensstand zu erhellen und neue Zusammenhange rrut lnfektionen wahrend der Schwanger- schaft zu erkennen . Dabei sollte es sich vorzugsweise um Prospektivstudien handeln ; wenn iliese Moglichkeit allerdings nicht gegeben sei , waren gut kon- zipierte Retrospektivauswertungen eine zufriedenstellende Alternative. Da dje Zahl der Kleinbnder rrut irgendwelchen <lurch infektiose Erreger verur- sachten Schailigungen gering sein dtirfte, sollten Kollaborativstudien gefordert werden . 142 PE3DME Bae.netrne ,UaHHoe coaematrne, a oaoore KOToooro noHHRJIH y"aCTHe 3MOO1-1onor, 3nH.DeMHonorn H MHKooOHonorn, 6blno CO3BaHO .IlJlR o6cyJLlleHHR HaKonneHHblX 3HaHH:H H HeDE!llleHHblX nooOneM B oOnacTH 3THOnorHH, naOooaTODHOH .nHarHOCTHKH, 3nH.IleMHOnorHH, a TaK~e nOO<!>IUlaKTHKH H neqeHHR noeHaTa.TlbHblX H neoHHaTaJlbHblX HH4)eKI..IHH B Ue.JlRX oa3oa0on<H DeKOMeH..DallHH .IlJlR ooraHOB 3.IlDaBOOXoaHeHHR B CToaHax. O~.naeTCR, qTO De3YJlbTaTbl oaOOI'bl I'oyrnn,J no3BO11RT He TOJlbKO .noOH'I'bCR Oonee DaUHOHaJlbHOro HcnOJlb3OBaHHR HMet0111erOCR om,rra H oecyocoB 3.IlDaBOOXoaHeHHR, HanDaBnReMhlX Ha nOO<!>IUlaKTHKY 3THX HH(j)eKUHH, HOH onoe.neJIHTb HOBble nyTH Hayt{Hl,IX HCCne.noaaHH:H. XOTR Heoe.IlKO nonaraJOT, qT() B KOJIH'iecTBeHHOM OTHOllleHHH noe- H neoHHaTaJlbHble HH4)eKUHH He noe.ncraBJlRIOT coOoA ce0be3HOH nooOneMhl, qyfCJlO neTeH, ncx:Toa.IlaBIIIHX OT HH<t,eKI..IHH, CKooee acero He.nooueHHBaeTCR. MHorne MaTeDHHCKHe HH<l>eKUHH, Bbl3blBaeMhle, B qaCTHCX:TH, UHTOMeranOBHDYCOM H Toxoplasma gondii, HCX:RT cyOK.TIHHlNecKHH xaoaKTeo H COITDOBO~.IlalOTCR coaBHHTeJlbHO He3HatDITeJlbHb!MH H HecneuH<t,HqecKHMH CHMnTOMaMH, B CBR3H C qeM OHH He acer.na oacnO3HaJOTCR H HCcne.lIYJOTCR. TIO.IlBeO~eHHble OHCKY .neTH MOryT nDH DOJLlleHHH Ka3a'I'bCR OTHCX:HTeJlbHO 3.IlOOOBblMH, O.IlHaKO ncx:nellYl()111ee llTTHTeJlbHoe Hccne.noaaHHe Heoe.m<o BblRBnReT y HHX HaoymeHHe Da3BHTHR Da3JIHqHOH CTeneHH TRJteCTH, B TOM qHCne 3a.neD~KY YMCTBeHHoro Da3BHTHR. 8 xone HCcne.nOBaHHR B Oonee nO3.IlHHe neOHO.Ilhl .neTCTBa OblBaeT cno~HO IUlH .nai,;e HeBO3MO:ni0 no.IlTBeD.IlHTb .nHarHO3 m>e- H neOHHaTaJlbHOH HH4)eKUHH C nOMOlllb~ naOooaTODHblX TeCTOB. 8 3TOH CBR3H roynna no.nqeOKHYna BaJICHCX:Tb nooaeneHHR nonoJTHHTeJlbHblX 1-1ccne.noaaHH:H, ncx:KOJlbKY OHH MOryT .llaTb B03MOllHCX:Tb nyqme oueHHTb oacnDOCToaHeHHOCTb 3THX HH4)eKI..IHH H HX .nonrcx:oo\flible ncx:ne.ncTBHR. MHOrHe H3 naOooaTODHblX MeTO.IlOB, oxaoaKTeOH3OBaHHblX BO BDeMR COBelllaHHR, K noHMeoy, He.IlaBHO oa3oaooraHlible 3H.IIHM- HMMYHO-COo0eHTHble TecTbl, B qaCTHCX:TH cneUH<!>HqecKHe, .Il1lR 143 Bb!RBJleHHR IgW-ai-rnrren, ~Jla.aaJOT ll!HPOKHMH noreHUHaJlb- HblMH B03MOllHOCTRMH, o()ecnetD-fBalOlllHMH noBNll!eHHYJO TOtJHOCTb H HallellHOCTb .llHan-lOCTHKH MaTeDHHCKHX KH<l>eKIDIH H MOHHTODHHra COCTORHHR 3.IIODOBbR .oe-rei:i roynnbl BblCOKOro DHCKa. TeM He MeHee BcUrHO, '-IT~ KJlHHHI..IHCTbl, 3nH.DeMHOJ10rH H na6ooaTODHble oaOoTHHKH OblJJH nOCTaBJleHbl B H3BecTHOCTb O nDOOneMax, KO'rOOble MOI,'T 0031-!HKHYTb nPH HcnOJlb30BaHHH BblilleynOMRHYTbIX MeTO.llOB. B "<IaCTHOCTH, B nDO.Ilalle HaxO.llRTCR IlHarHOCTH"<IecKHe TecT-Ha6oDbl, HCnOJlb30BaHHe KOTOOblX HeoriblTHblMH DafuTHHKaMH MO;'leT nDHBecTH K CXIIH6o'lHblM De3YJlbTaTaM. KaKOH Obi HH Ob1J1a "<IaCTOTa B03HHKHOBeHHR neDKHaTaJlbHblX KH<l>a<UHH, HeT COMHeHHR B TOM, 'ITO nooaxeHHble HMH .DeTH Bbl3b1Ba!OT ce0be3Hble COUHa.1lbH<r3KOHOMH'-leCKHe ~JleMbl, T.K. BO MHOrHX CJlY'laRX OHH Toe0yxrr nOMOIIIH B Te"<IeHHe BCeH ;'IH3HH. n03TOMY ~JleMbl 3THOnonrn H mxxt>HJ1aKTHKH nDe- H neoKHa- TaJibHOH KH<l>eKUHH, a TaK;,;e 6o0b0bl C HHMH H~XOilHMO oemaTb Ha OCHOBe Me;'IJIHUHCUHTI.llHHaDHoro no.nxo.oa. BblBO.llbl H DeKOMeH.DaUHH 1. H~XOilHMO Da3BHBaTb .neRTeJlbHOCTb no Oc)[)bOe C noeHaTaJibHblMH H neDHHaTaJlbHblMH m-t<t>eKUHRMH. Y'IDeUeHHRM 3.llDaBOOXDaHeHHR B KallOH CTDaHe DeKOMeH.IIYeTCR nDOBO.llHTb COOTBeTCTBYIOll1He MeDOllOHRTHR, B "<IaCTHOCTH, ooraHH30Bb1BaTb Ha.a30D, Bb!RBJleHHe cnyqaeB KH<l>eKIDIH, ceDOJlOnf'.lecKHe HCCJle.DOBaHHR, MeDOnPHRTHR no BaKUHHal.lHH H nDOT"Da.MMbl oueHJ<H. 2. HaCTORTeJlbHO DeKOMeH.IIYeTCR nDOBOilHTb .llOJII"OCDOtJHoe oOcne.nOBaHHe .DeTeH C nOH3HaKaMH neoKHaTaJibHblX KH<l>eKUHH. TaKHe oOcne.DOBaHHR uenecoo0oa3HO nDOBOilHTb, Ha"<IHHaR C MOMeHTa 00;,t.DeHHR oeOeHJ<a BnJlO'rb .no ero nOCTYnJleHHR B lllKOJlY HJ1H no B03MO;'IHOCTH .nage B Te"<IeHHe 6onee nDO.DOJl;'IHTeJlbHOro neDHO.Da BoeMeHH. 3. HH~HUHPOBaHHe CTDenTOKOKKaMH roynnbl B (stre ptococcus aga lactiae) H Listeria monocytogenes MO;'leT nDOH30HTH B neOHO.D BHYTPHYTD00Horo Da3BHTHR, BO BoeMR HJIH CDa3Y nocne DO.DOB. CMeD'I'eJlbHh!A HCXO.D B TaKHX CJlY'laRX - Heoe.mcoe a IgM HMM-yHorn~nKH M (noHMe"<IaHHe neoeBOmHKa) 144 RBneHHe, H H30egaTb ero MOgHO C nOMOIIIbm YCKODeHHhlX mra.rHOCTINecKHX nooue.llYD, K KOTOOblM CJle.IIYeT noHOeraTb nPH MaJlE!HllleM K~ecKOM no.II03PeHHH Ha HHc!)eKUHJO C TeM, '-ITOObl He3aMe.IIJ1HTeTlbHO HCOOJtb30BaTb CDe.IlCTBa XHMHOTeoanHH. 4. B KomNecTBeHHOM OTHaJ!eHHH nePBH'IHaR UHTOMeranOBHPYC- HaR HHc!)eKUHR - Oonee \.laCTaR nPH\.IHHa BPOMeHHblX 3aOoneBaHHH, '-leM nePBH'IHaR KDaCHYXa. TIPH\.leM neoBoe H3 ynOMRHYThlX HHC!>eKUHOHHblX 3aOoneBaHHH Bb13blBaeT Oonee cePbe3Hl,le HaoymeHHR, nPHBO.IlRIIIHe K YMCTBeHHOH OTCTanOCTH H rnyxOTe, CBR3aHHOH C HaPY111eHHeM ceHCOPHoro HeHPOHa. ATTeHYHPOBaHHble BaKUHHhl, De3YJtbTaThl Hcr10Jtb30BaHHR KOTODblX B xo.ne HCilhlTaHHH ~IJlH He.IIaBHO onHCaHl,I, BPR.Il J1H Oy.nyT o.noOoeHbl .IIJLFI ueneH oCi!JleH mxxf>HJ!aKTHKH 3aOoneBaHHR CDe.IlH 3.IlOPOBblX BOCTTOHHM\.IHBblX ~eHIIlHH H3-3a B03MO~Horo PHCKa B03HHKHOBeHHR onyxoneH. HeoOXO.IIHMO Da3DaOOTaTb aJtbTePHaTHBHble noenaoaThl, HanoHMeP, BaKIDIHbl H3 cyO'.oe.llHHlfil BHPyca. HeoOXO.IIHMO OCYllleCTBJIRTb HenDePhlBHblH 9mUieMHOnOnNecKHH Ha.n30P 3a LIHTOMeranoBHPYC- 1-lblMH HHc!)eKUHRMH CDeIDi Da3JlH\.IHblX roynn HaceneHHR. s. HeoOXO.IIHMO nPHHRTb Bee MePbl .IIJLFI IMKBH.IIal.lHH BPOUeHHOH KDaCHYXH rryreM nooBe.IIeHHR nPOI"PaMMbl BaKI..IHHauHH C Yt.leTOM cneUH<t,HKH CHCTeMbl 3.IIPaBOOXDaHeHHR KaMOH CTOaHl,I. 6. JF:eHIIlHHbl MOryT ObiTb Cny\.laHHO BaKl.lliHHIX)BaHbl nPOTHB KDaCHYXH He3a.nonro H.11H coa3y ~e nocne 3at.1aTHR. I'oynna C\.IHTaeT, '-!TO B TaKHX Cny'-laRX He cne.nyeT noePblBaTb OeoeMeHHOCTb, 0.IIHaKO OKOH\.laTeJtbHOe PE!llleHHe .IlOJDIHa nPHHRTb caMa geHIIIHHa H nePCOHaJt, HaOmo.II~IIIHH 3a COCTORHHeM ee 3.IlOPOBbR B nePHO.Il OeoeMeHHOCTH. 3Ha\.lHTeTlbHOMY '-!Hcny BOCllOHHM\.IHBblX K KDaCHYXe r.>eoeMeHHl,IX ~eHll!HH OblJlH c.nenaHbl nPHBHBKH nPOTHB 9TOro 3aOoneBaHHR C HCnOJtb30BaHHeM aTTeHYHIX)BaHHblX BaKUHH, H OHH De!IIHJIH He nDePblBaTb OeoeMeHHOCTb. H XOTR y HOBOPOMeHHblX He oono OTMet.1eHO KaKHX-JlH(')o OTKnoHeHHH, Bb13blBaeMblx KDaCHYXOH, \.IHCJIO OeoeMeHHblX ~eHMHH, KOTOPblM .nenaJTHCb nPHBHBKH O.IIHOH H3 ll!HPOKO noHMeHReMblX B HaCTORlllee BDeMR BaKUHH (BaKUHHHblH IIITaMM RA2713), CJIHIIIKOM He3Ha\.lHTeJtbHO, t.lTOObl c nonHOH yBeDeHHOCTbm MOgHo OblnO Obi rOBOPHTb O Oe30llaCHOCTH TaKHX nPHBHBOK. 8 CBR3H C 9THM B CTDaHax HeoOXOIDiMO nPO.IIOJlllHTb HCCne.noBaHHR COCTORHWi 3.IlOPOBbR geHIJlHH, KOTOPblM cny\.laHHO c.nena.Hbl nPHBHBKH nDOTHB KDaCHYXH B nePHO.Il OeoeMeHHOCTH. 145 7. 8bl3b1BaeT 03aOoqeHHOCTb HaOmo.naBMHHCR DOCT CJJY'{aeB 3aOoneaaHHR reHHTaJlbHblM reonecoM, 'ITO MOlKeT OPHBecTH K B03HHJ<HOBeHHIO neoHHaTaJlbHblX HHc!)eKUHH, C03.llaJ0lllHX 0oJlblllYIO onacHOCTb 3aOoneBaeMOCTH H CMePTHOCTH. Yt{HTblBaR, 'iTO ec-reCTBeHHoe PO.llOPa3DeMeHHe noKa3aHO He BO Bcex CJTYt{a.RX, coe.IIH llteHIIIHH, neoeOo.neBMHX reHHTaJtbHblM reonecoM H/HJTH HMelOlllHX OPH31-laKH B03MOlltHOr'O 3aOoneBaHHR, OOKOMeH.nyeTCR nPOBO.IIHTb oOcJJe.nOBaHHe Ha Bbl.A:BJleHHe BHOyca 3a HecKOJlbKO .llHeH .no 00.llOB. TTPH BbL"IBJleHlffl BHPYCHOO HHc!)eKUHH cne.nyeT yqecTb B03MOlltHOCTb KecaoeBa ceqel-!WI. 8. TTOCKOJibKY O'iel-tb qacro naOooaTOPHblH .IIHarH03 TOKCOOJJa3M03a He .QaeT Y.llOBJleTBOPHTeJlbHblX 003YJibTaTOB, HaJIH'IHe HH(!)eKUHH Heoem<o yCTaHaBJIHBaeTCR CJlHlllKOM 003.llHO ,IlJ1R nPOBe.lleHHR 3¢ct)eKTHBHOro Jle'ieHHR. PeKOMeH.llOBaHO OPHHJITb COOTBeTCTBYIDl!lHe MePbl no oOecneqe1-1Hro BblCOKOr'O TeXHH'iecKOr'O YPOBHR oaOoTbi .naOooaTOPHH H OCHaCTHTb HX ,IlJ1R 3TOr'O COOTBeTCTBYIDlllHM oOooy.noaaHHeM. TTOCKOJlbKY noeHaTaJlbHblH TOKCOnJla3M03 t{aCTO 000.A:BmteTC.A: B oaHHeM HJlH 003.llHeM nePHO.lle .lleTCTBa, cne.nyeT 000.llOJIJltHTb .llOJirOCPO'iHble HC'CJle.llOBaHWI, BKJDO'ia.R TlllaTeJibHO CnJlaHHPOBaHHble K.TlliHHt{ecKHe H naoa3HTOJlCr nNecKHe TeC'I'bl. EcJIH ceoononNecKoe oOcne.noBaHHe Ha oaHHHX CTa.IIIDIX OeoeMeHHOCTH .naeT OTPHUaTeJlbHblH 003YJlbTaT, OOKOMeH- .nyeTCR nOBTOOHTb 3TO oOcnellOBaHHe .I1JlR BblJiBJleHwt aHTHTeJI K TOKCOOJJa3Me (o.mrn-.ima Pa3a B Mec.A:u), 'iTO 003BOJlHT oOHaoy- lltHTb cePOKOHBePCHro. 9. XOT.A: 'iHCJlO cnyqaeB BPOlltlleHHoro cKt>HJIHCa B EBoone 3Ha'iHTe.rtbHO yMeHl,IJIHJlOCb, HeT COOTBeTCTBYKIJIHX .naHHblX, 00.IJTBePlltllaJOlllHX yMeHbnieHHe 3aOoJleBaeMOCTH CH¢HJlHCOM B 003YJlbTaTe nonoooro KOHTaKTa. BBHllY 3TOr'O HaCTO.A:TeJlbHO OOKOMeH.IIYeTC.A: H B .llaJtbHeHll!eM 00000.IIHTb PYTHHHblH CKPHHHHr Ha Bbl.A:BJleHHe CH¢HJJHca. 10. HeoOXO.llHMO cnocoOcTBOBaTb BHe.noeHHJO Ha.nnelKal!lHX YCKOOOHHhlX .naOooaTOPHblX nooue.nyo, HeoOXO.IIHMhlX ,IlJ1R nooBe- .neHwt noo¢H.naKTH'iecKHX MePOOPH.A:THH, Bel]eHHID llOKYMeHTal.IHH 0 CJlyqa.RX HH¢HUHPOBaHHR B nePHOll OeoeMeHHOCTH H HOBOPOlltlleH- HOCTH H OOH Heo0XO.IIHMOCTH ooraHH30BblBaTb neqeHHe t">oJlbHblX. Onoe.neneHHe cneUH¢H'iecKHX IgM aHTHTen - HaHJTYt{l]!HH MeTO.Il Bbl.A:BJleHwt noe- H neoHHaTaJtbHblX Hli<l>eKUHH. Oco6oe BHHMaHHe Heo0XO.IIHMO Y.IleJJHTb TOMY, 'iToObl HCKJDO'iHTb B03MOlKHOCTb HecneUH¢H'iecKHX oeaKUHH. 0.llHaKO .no BblHeceHH.A: DeKOMeH.IIaUHH 0 TOM, cne.nyeT JlH naUHeHTY noePblBaTb OeoeMeHHOCTb HJlH HeT, 146 Heo0XO.IIHMO mx,aHaJIH3HDOBaTb KJJHH},fl.{ecKHe .DaHHble C yqeTOM Pe3YJlbTaTOB DDYI"HX ceOQJ1or1NecKHX Hccne.noBaHHH. no.no6Hoe oemeHHe PeKOMeH,llyeTCR nPHHHMaTb Ha OCHOBaHHH KJJHHHqeC!<HX H .naOooaTOPHhlX .IIHarH03OB, nOCTaeneHHhlX Ha CaMOM BblCOKOM YOOBHe, C yqeTOM IleHCTBYJ()lllero B CToaHe 3aKOHO.DaTe.TlbCTBa. YtraCTHHKH PoYffilhl Bbloa3Hml ce0be3HYK> o6ecnoKoeHHOCTb B CBR3H C B03MO;i(HOCTb~ MHDOKOI"O HCOOJlb3OBaHHR ceOHHHO BbITTYCKaeMhlX TecT-HaOoooe neocoHanoM, He HM~MHM nOCTaTO'IBOro OllblTa. 11. XOTR y<1aCTHHKH PDYTIIlbl H nOJITBeO.IIHnH CBOIO nPHBeOlleH- HOCTb MHOI'OCeKTODaJtbHOMY nonxony, HCOOJlb3YeMOMY nPH oKa3amrn nOMOIIIH BY3nooaBmma.KllllHM H nooeeneHHH nonrocoo'IHhlx oeaOHm!TauHOHHhlX MePOnPlfflTHH coenH ne-reA C BOOJIIleHHblMH .nel)eKTaMH, noeneJlbHO BblCOKHe oacxo~ Ha IleRTeJlbHOCTb 9THX Cnyl'(O RBJlRJOTCR BallHhlM <!)aKTOOOM, nonTBeOll.llalOOUiM HeoOXOnH- MOCTb ocyMeC'I'B.11eHHR COOTBeTCT~DlHX nocxt>HJlaKnf'leCKHX MeOOnPHRTHH. 12. roynna DeKOMeH.OOBa.Jla nPOilOJlllHTb 9TTH.JleMHOnor!NecKHH MOHHTOOHHr BPOJIIleHHhlX nel)eKTOB y HOBOOOJIIleHHhlX nnR YTO'IBeHJffl HM€!0111HXCR ceeneHKH H BYRBnelilffl HX HOBYX accoUHailHA C HH<l>eKUHRMH B nePHOil OeoeMeHHOCTH. TaKHe HCCne.OOBalilffl IlOJimibl noemiO'ITHTeJlbHO HOCHTb neocneKTHBHhlH xaOaKTeo, HO ecnH 9TO HeB03MO;ir;HO, XOOOIIIO nooeeneHHhlH oeTPOCneKTHBHhlA aHanH3 Mor Ol>I noencTaBHTb coOc:,A YilOB.11eTBOOHTeJlbHhlH BaOHaHT aJlbTeOHaTHBHOro oemeHHR. Cnenye-r CTHMYJIHOOBaTb nooeeneHHe COBMeCTHblX HccnenOBaHHH BBH.ItY Toro, 'ITO 'IHCnO ne-reA C Da3JllNHblMH Bbl3BaHHhlMH HK!)eKUHRMH ne<PeKTaMH MOlleT OYTb He0oJlbMHM, 147 lllllll lllll lllll lllll~llllllllllllllllll llllllll H n n o 1 3 7 ? n H