Performance and potency of tetanus toxoid: implications for eliminating neonatal tetanus.

Reviews/Analyses Performance and potency of tetanus toxoid: implications for eliminating neonatal tetanus V. Dietz,' J.B. Milstien,2 F. van Loon,1 S. Cochi,1,3 & J. Bennett4 Neonatal tetanus (NT) is a major cause of mortality in developing countries, with over 400000 deaths esti- mated to occur annually. WHO has adopted the goal of eliminating NT worldwide, and a major strategy for its prevention is the administration of at least two properly spaced doses of tetanus toxoid (TT) to women of childbearing age in high-risk areas to protect passively their newborns at birth. In certain countries the locally produced TT vaccine has been shown to be subpotent, while other countries have reported NT among infants born to vaccinated women. An extensive review of production and quality control procedures was carried out between 1993 and 1995 in 8 of22 TT-producing countries that also report NT cases, with a more superficial assessment being carried out in the remaining 14 countries. Only 4 of the 22 countries have a functioning national control authority to monitor TT production and vaccine quality. A total of 80 TT lots from 21 manufacturers in 14 of the 22 NT-reporting countries were tested for potency. Of these, 15 lots from eight manufacturers in seven countries had potency values below WHO requirements. TTpotency can also be compromised by improper vaccine handling. To eliminate neonatal tetanus worldwide requires assurance that all doses of TT meet WHO production and quality requirements and that the field effectiveness of TT is monitored through systematic NT case investigations and assessment of coverage. Introduction In 1989 the World Health Assembly adopted a reso- lution calling for the global elimination of neonatal tetanus (NT) as a public health problem by 1995.a In this respect, one of the key strategies identified by WHO is the achievement and maintenance, among 1 Medical epidemiologist, National Immunization Program, Centers for Disease Control and Prevention, Atlanta, GA, USA. Requests for reprints should be sent to Dr Dietz at the following address: Division of Parasitic Diseases, Centers for Disease Con- trol and Prevention, MS: F22, 4770 Buford Highway NE, Atlanta, GA 30341-3724, USA. 2 Scientist, Vaccine Supply and Quality, Global Programme for Vaccines and Immunization, World Health Organization, Geneva, Switzerland. 3Chief, Polio Eradication Activity, National Immunization Pro- gram, Centers for Disease Control and Prevention, Atlanta, GA, USA. 4Director for Scientific Affairs, Task Force for Child Survival and Development, Atlanta, GA, USA. a Resolution 42.32. Handbook of resolutions and decisions of the World Health Assembly and the Executive Board, vol. III, third edit. (1985-92). Geneva, World Health Organization, 1993. Reprint No. 5746 women of childbearing age in high-risk areas, of high coverage levels for at least two doses of tetanus toxoid (TT).b Studies have demonstrated that two doses of properly spaced TT protect the offspring of female recipients against NT (1-3). To be maximally effective, the first two doses of TT (TT1 and TT2) should be given at intervals of at least 4 weeks, with the ideal interval between YT2 and birth being at least 4 weeks (4, 5).C In addition, the interval be- tween TT1 and YT2 and that between TT2 and the delivery of the child are critical to ensuring good immunogenicity (6). The proportion of women whose titres reach a protective level becomes greater with increasing intervals between lT1 and TT2 and between TT2 and birth. WHO estimates that in 1994, b Revised plan of action for neonatal tetanus elimination. Unpub- lished document WHO/EPI/GEN/94.4, 1994 (available upon re- quest from Global Programme for Vaccines and Immunization, World Health Organization, 1211 Geneva 27, Switzerland). c Galazka A. Immunization of pregnant women against tetanus. Unpublished WHO document EPI/GEN/83/5, 1983 (available upon request from Global Programme for Vaccines and Immunization, World Health Organization, 1211 Geneva 27, Switzerland). Bulletin of the World Health Organization, 1996, 74 (6): 619-628 619 V. Dietz et al. a total of 46% of pregnant women had received two doses of TT (7).d To eliminate NT globally, not only must TT coverage levels be raised but the vaccine must be safe and potent. There have been reports of NT oc- curring in infants born to women who had received at least two doses of TT (8, 9), causing some workers to question its potency (8, 10). However, as with any vaccine, some seroconversion failures are to be ex- pected, even when it is administered at the appropri- ate intervals. Thus, case reports of NT among infants born to vaccinated women do not necessarily imply that TT is not effective. Nevertheless, monitoring case reports of NT and the vaccination status of the mothers concerned is important since this provides information about the field performance of TT. Therefore, reports of NT cases in infants born to vaccinated women (i.e. potential TT failures) must be evaluated to determine whether the observed number of failures is greater than expected and whether the doses were properly spaced. If either of these is found to be the case, further evaluation of TT production and effectiveness would be warranted. Field epidemiological studies in Bangladesh and in selected areas of Pakistan have demonstrated that the effectivenesse of two doses of TT in preventing NT was <50% (11, 12). In Bangladesh, subsequent inspection of production facilities revealed multiple deficiencies in TT production, and locally produced TT was determined to be subpotent by two WHO reference laboratories (11). In Pakistan, UNICEF- supplied YT collected from the field in 1992 in selected areas and nationally produced TT (used for the army but not in the national Expanded Pro- gramme on Immunization (EPI)) were tested for po- tency. The potency of the UNICEF-supplied YT collected from the field several years after the obser- vations on TT field effectiveness was borderline, al- though the potency of the vaccine at release had been adequate, and the potency of the nationally produced TT was subpotent.! In this article we review two aspects of monitor- ing the performance of TT: published reports of NT cases from a review of the literature that included d Expanded Programme on Immunization information system. Unpublished document WHO/EPI/CEIS/95.2, 1995 (available upon request from Global Programme for Vaccines and Immuniza- tion, World Health Organization, World Health Organization, 1211 Geneva 27, Switzerland). e Vaccine efficacy refers to the ability of a vaccine to protect against disease under ideal circumstances. Vaccine effectiveness refers to the ability of a vaccine to protect against disease under field conditions. Thus, in this article effectiveness and not efficacy is generally discussed. information on the vaccination status of the mothers of infants who developed NT; and currently avail- able information on the quality of TT manufactured in countries where NT cases still occur. Sources of data Data on vaccination coverage, reported disease inci- dence, and NT estimated deaths were obtained from the EPI Information System, Global Programme for Vaccines and Immunizations, WHO, Geneva.d Information on TT production procedures, quality, and potency were obtained through the Task Force on Situation Analysis (TFSA) of the Children's Vac- cine Initiative. The TFSA has identified as high pri- ority for production assessments, vaccine-producing countries that also report NT cases. Manufacturers in 10 of these countries (Bangladesh, Brazil, Egypt, India, Indonesia, Islamic Republic of Iran, Mexico, Pakistan, Philippines, and South Africa) have received formal on-site assessments of vaccine supply sources by teams of experts fielded by the TFSA.9 In addition, several of these countries have been identified by WHO/EPI as being high priority for accelerated activities for NT elimination since they accounted for over 80% of the estimated global NT deaths in 1994. The 12 EPI priority countries are as follows: Angola, Bangladesh, China, Ethio- pia, Ghana, India, Indonesia, Nigeria, Pakistan, Somalia, Sudan, and Zaire. In the course of TFSA site visits to TT manufac- turers, production and control processes are ana- lysed. Also evaluated is the extent of overseeing provided by national control authorities, if any, on TT manufacturing, compliance with good manufac- turing practice, the proportion of TT lots that fail quality control and potency testing, the capacity and ability of the manufacturer to meet production tar- gets, production costs per dose, technical aspects of vaccine production (e.g., sterility and purity) and compliance with WHO norms. If possible, samples of TT are randomly selected from representative production lots for potency testing of the tetanus component at WHO Collaborating Laboratories. Potency testing involves an assessment of protection against a lethal challenge of tetanus toxin in mice or I Unpublished WHO document EPI/TECHCOM/WP/93.33, 1993 (available upon request from Global Programme for Vaccines and Immunization, World Health Organization, 1211 Geneva 27, Switzerland). 9 Reports of these assessments are available from Global Pro- gramme for Vaccines and Immunization, Vaccine Supply and Quality Unit, World Health Organization, 1211 Geneva 27, Switzerland. 620 WHO Bulletin OMS. Vol 74 1996 Performance and potency of tetanus toxoid guinea-pigs, standardized against the International Standard for Tetanus Toxoid (13). A potency value of at least 40 International Units (IU) per human dose is specified by WHO requirements (or ¢'60IU per human dose if the potency of the tetanus compo- nent is measured in the presence of the pertussis component in mice), and lots possessing this mini- mum potency level are classified as having passed (13). During other TFSA site visits at which com- plete assessments are not carried out, vaccine pro- duction and control procedures are inventoried. A detailed summary of the evaluations completed to date by the TFSA has been prepared (14), and in this article we report on the results of potency testing of TT produced in countries that also report NT cases. Results Performance failures Case reports of NT in infants of women "adequately" immunized with TT. Two doses of TT given at the appropriate interval should result in protection against NT in at least 90% of recipients (5).hProtec- tion is defined as a tetanus antitoxin level >'0.01 IU/ ml in in-vivo neutralization tests or a level ¢0.1 IU/ ml in in-vitro tests (15, 16). However, NT cases have been reported in infants born to vaccinated mothers (8-10,17-20). For example, in Nigeria, 17 (33%) of 52 mothers of infants with NT had a history of immu- nization with TT (8), and 12 of these 17 mothers were said to have received adequate doses of TT- defined as two doses given during pregnancy, with the last dose being given 2 weeks before delivery; however, a 2-week interval between TT2 and birth may not be optimal. In addition, information about the interval between TT1 and TT2 was not provided, precluding determination of whether the doses were properly spaced. Thus, a statement about TT per- formance cannot be made with any confidence. In India, a study of 30 cases of NT showed that 17% of the children's mothers were partially, and 13% completely immunized (9); however, no infor- mation was provided on the vaccination intervals. An investigation of an NT case series in Madras, India, found that 20% of 61 NT cases in 1987-88 and 32% of 19 NT cases in 1989 were born to mothers who had received at least two prior doses of TT (17). Information on the timing of these doses was pro- vided for only 6 of the 18 cases; of these six mothers, hSee footnote c, p. 619. only two appeared to have an adequate interval be- tween doses, while a further two had received all three doses during the ninth month of pregnancy. In another report from India, 2% of 100 NT cases in infants studied were born to mothers who had re- ceived adequate doses of TT, although neither the timing nor the intervals between doses were re- ported (18). Finally, a study of 385 NT cases in Delhi, India, found that 6% were born to mothers reported to be fully immunized (19). Thus, although poten- tially an important source of information for moni- toring TT effectiveness, the data provided in case reports are often insufficient to permit a reliable evaluation of the field performance of TT, since the vaccination intervals are often not specified. Retrospective investigations using medical records are often unable to ascertain the immuniza- tion history of the mothers of infants who develop NT. For example, in a retrospective review of over 400 hospitalized NT infants in Nigeria, the investiga- tors were able to obtain information from the hospi- tal records on the immunization status of only 8 mothers (20). In Uganda, 228 hospitalized NT in- fants were retrospectively investigated but the im- munization status was ascertained for only a few cases because of lack of information (21). Effectiveness of 17 in the field setting Field estimates of TT vaccine effectiveness can also be biased if the interval between doses is not verified. For example, a case-control study of risk factors for 26 NT cases showed that the protection conferred by TT vaccine against NT was less than optimal, with a vaccine effectiveness of only 70% (95% confidence interval (CI) = 52-100%) (22). Information was not available on 13 of the 26 NT cases; of the remaining 13 cases, 10 were incompletely vaccinated, i.e. they had received <2 doses. Moreover, the intervals were not stated for the three NT cases born to women who had received ¢'2 doses. Had even one of the three vaccinated women received improperly spaced doses, TT vaccination would have been protective against NT with an estimated vaccine effectiveness of 82%, and 92% if two of the cases had been misclassified. In a previous review, reports of TT failures were also noted (16); these were attributed to inaccurate immunization histories or improperly spaced doses. Some reports of TT failures were attributed to poor maternal immune response, inadequate placental transfer, or excessive toxin exposure; the potential role of these factors is discussed elsewhere (23). Thus, because information about the timing of doses may not be available, reports of NT in infants born to vaccinated women, or TT failures, although po- WHO Bulletin OMS. Vol 74 1996 621 V. Dietz et al. tentially very useful, are often difficult to evalu- ate; caution should be exercised in interpreting them. TT failures have therefore not been well docu- mented by data from case investigations; in-the-field effectiveness studies of TT provide a better basis for evaluation. We could identify no recent field evalua- tions other than those conducted in Bangladesh (11) and Pakistan (12). In 1990, a population-based sur- vey involving 60 clusters was undertaken in Punjab Province, Pakistan (12). Of the nearly 24000 live births during the year preceding the survey, 229 were considered to be NT cases. The mothers of 15 of these infants had histories of appropriate TT immu- nization as defined above, while two additional mothers met all criteria except that the intervals be- tween TT1 and TT2 were less than 1 month (26 days and 27 days). Only 5 of the 15 mothers had card documentation of vaccination status. Relative to mothers who reported that they had never received a TT dose (NT incidence = 11.2 per 1000) the inci- dence of NT among live births involving mothers who fully met criteria for appropriate immunization by history (NT incidence = 5.9 per 1000) or with card documentation (8.5 per 1000) suggested a vaccine effectiveness of <50% by either history or card. A subsequent case-control study also confirmed inad- equate protection from the UNICEF-supplied TT in use at that time. Tetanus toxoid coverage was ap- proximately 11%. Data from non-Punjab regions of Pakistan pro- vided contrasting information. Surveys of live births from April 1987 to May 1991 documented only three NT deaths among 1519 live births (NT incidence, 1.97 per 1000) to mothers with card documentation of appropriate TT vaccination with UNICEF- procured 1T, while 324 deaths occurred among 27368 live births (NT incidence, 11.8 per 1000) to mothers who had never received a dose of TT. This corresponds to an in-the-field estimated effective- ness of 83.3%. Furthermore, the risk of NT death among infants whose mothers reported receipt of only one dose of TT one or more months before delivery (NT incidence; 2.4 per 1000) was substan- tially less than that among unimmunized mothers (P = 0.04, Fisher's single-tailed test), with an estima- ted efficacy (80%) similar to that of those appropri- ately immunized; 409 of these 414 women (includ- ing the mothers of the cases) received their single dose of TT less than 20 months before delivery. These findings are consistent with earlier observa- tions that deaths among 4-14-day-old infants were reduced equally by either one or two doses of TT within the first 20 months following vaccination (25). It should be noted that the NT incidence among infants of mothers who had card documentation of appropriate immunization was significantly higher in Punjab than elsewhere in Pakistan (8.5 per 1000 versus 1.95 per 1000, resp.; P = 0.042, Fisher's two-tail test), but that the incidences for the in- fants of unimmunized mothers were similar (NT incidence, 11.2 per 1000 versus 11.8 per 1000, resp.; P = 0.60). Finally, NT cases have been reported in infants born with satisfactory levels of laboratory-confirmed tetanus antitoxin (26). Global TT production and quality control At least 63 manufacturers in 42 countries produce TT (Table 1) (14). Neonatal tetanus cases are re- ported in 22 of these countries (with 33 TT produc- tion facilities) (Table 2), and these 22 countries accounted for 74% of all NT cases reported to WHO in 1994, and for 59% of NT deaths estimated in 1994 (WHO, EPI Information System, 1994), as well as accounting for 88 million births each year, i.e. 71 % of the 124 million births estimated to occur each year in developing countries (27). In 1994, an estimated 219.4 million doses of TT were produced in these 22 countries. Five TT-producing countries (China, In- dia, Pakistan, Bangladesh, and Indonesia) also had the greatest numbers of estimated NT deaths in 1994. These five countries (with 53% of the total number of births in developing countries each year) have a TT production capacity in excess of 150 mil- lion doses. Although 42 countries produce TT, only the nine manufacturers (in Australia, Canada, France, Germany, Hungary, India, Italy, Yugoslavia (Serbia and Montenegro), and Switzerland) that supply the vaccine to UNICEF are formally permitted to state that their products meet WHO requirements and are acceptable for purchase by U.N. agencies (14). This does not necessarily imply that TT produced else- where is of poor quality; for example, the USA and the United Kingdom have independent, competent national control authorities and review processes, and tetanus toxoid produced there is of acceptable quality. Only four of the 22 countries that both produce TT and report NT cases (Brazil, India, Indonesia, and Mexico) possess fully functioning national con- trol authorities, leaving the women and children in the remaining countries with no guarantee of quality for the locally produced vaccines they receive. In the five countries with the highest estimated number of NT deaths, and which also produce TT, only two have functioning national control authorities; in reality, the situation in these countries may be even worse. Tests carried out on 80 TT lots produced by 21 manufacturers in 14 of the 22 countries found WHO Bulletin OMS. Vol 74 1996622 Table 1: Countries reporting production of tetanus toxoid, by WHO regiona Eastern South-East Western Africa Americas Mediterranean Asia Pacific Europe South Africa Argentina Egypt Bangladesh Australia Austria Brazil Islamic Democratic People's China Bulgaria Canada Republic Republic of Korea Japan Croatia Chile of Iran India Philippines Czech Republic Colombia Jordan Indonesia Republic of Korea Denmark Cuba Myanmar Viet Nam Finland Ecuador Thailand France Mexico Germany Uruguay Hungary USA Israel Venezuela Italy Netherlands Poland Romania Russian Federation Switzerland Turkey United Kingdom Yugoslavia (Serbia and Montenegro) a See ref. 14. Jordan recently notified WHO that it no longer produces vaccines; Pakistan has produced TT in the past, but it is now blending and filling imported TT bulk. Table 2: TT-producing countries reporting cases of NT, ranked by the esti- mated number of NT deaths in 1994, and the number of reported NT cases by year, 1992-94 Reported number of NT cases in: Estimated NT deaths in 1994 1992 1993 1994 China 84000 NRa NR NR India 83000 4010 4438 2226 Pakistan 49000 1 737 1 685 1677 Bangladesh 26000 588 720 834 Indonesia 26000 807 638 NR Islamic Republic of Iran 5900 18 12 17 Myanmar 4900 111 105 NR South Africa 4700 13 16 7 Egypt 4600 1830 1277 998 Philippines 4000 347 343 NR Viet Nam 3600 187 333 NR Turkey 1300 29 46 NR Thailand 1000 120 67 62 Brazil 80 229 218 76 Mexico 60 137 97 63 Democratic People's 40 3 0 NR Republic of Korea Colombia 25 100 71 25 Argentina 10 13 6 9 Ecuador 10 71 81 13 Venezuela 10 29 27 13 Jordan 5 6 6 5 Chile 1 3 1 1 Subtotal 298241 (59)b 10388 (64) 10187 (67) 6026 (74) Global total 509000 16316 15258 8157 a NR = Report not received. b Figures in parentheses are the % of the global total. WHO Bulletin OMS. Vol 74 1996 623 V. Dietz et al. suboptimal potency, i.e. batch failure, in 15 lots (19%) from eight (38%) manufacturers in seven (50%) countries, including India (Table 3). The low potency results from Indian vaccine produced by some (but not all) manufacturers are linked to the potency testing requirements in India, which differ from those of WHO; although all Indian vaccines meet Indian Pharmacopoeia requirements, this does not necessarily imply compliance with WHO levels. Table 3: Potency testing results for tetanus toxoid (TT) in 14 countries, July 1995 Country Bangladesh Brazil Year 1993 1994 1992 China (manufacturer): A B C D E Colombia Democratic People's Republic of Korea (DTP)b India (manufacturer): A B C D Indonesia Islamic Republic of Iran (DT)b Mexico Pakistanc Manufactured, from seed Imported bulk 1995 1995 1995 1995 1995 1992 1994 1992 1993 1993 1993 1995 1994 1994 1992 1991 1994 1995 No. of lots 10 5 3 2 1 2 1 3 2 2 3 5 3 2 3 3 3 2 2 Philippines DTPb 1993 2 TT 1995 5 South Africa (DTP)b 1993 3 1994 3 Venezuela 1992 3 Viet Nam 1993 5 Total - 80 a Passed implies potency test results 240 IU per human dose. bThe tetanus component was tested. Results of TT potency testing: Passeda Failed 10 - 5 - 3 - 2 12 2 2 3 2 3 2 3 3 2 3 2 4 3 3 3 4 65 (71)d 1 1 2 15 (19) WHO Bulletin OMS. Vol 74 1996 c Production of TT from seed has ceased in Pakistan; the national manufacturer is currently blending and filling imported bulk vaccine. d Figures in parentheses are percentages. 624 Performance and potency of tetanus toxoid Discussion Field studies suggest that cases of NT are being born to women who have received at least two doses of TT. Investigations conducted by the TFSA have con- firmed that 7T produced in several countries is subpotent. Case reports ofNT are, however, difficult to interpret since the interval between doses and timing of vaccination is not often clear. Country reports, which represent a crucial source of informa- tion about TT performance, are therefore not be- ing used to their full potential to monitor TF performance. The findings in Pakistan and the unconfirmed reports of NT cases in vaccinated women in other countries emphasize the need to standardize investi- gation of NT cases.i Standard guidelines must be used to evaluate such reports.i National authorities can expect the occurrence of more NT cases among infants born to vaccinated women, since (even with highly effective but imperfect vaccine) as 7T cover- age increases a greater proportion of NT cases will occur among infants born to fully vaccinated women (28). For example, at a vaccine effectiveness of 80%, managers would expect to see 6% of NT cases born to vaccinated women if coverage with TT2 were 25%, and 17% if coverage were 50% (Table 4). In addition, since the effectiveness of two properly spaced doses of TT in unvaccinated women should be greater than 90%, -10% of vaccinated women may be at risk for primary vaccine failure, i.e. are nonresponders. Since not all 7T doses are properly spaced (i.e. the intervals are too short) nor are all newborns born during the period of protection from the last valid TT dose, additional NT cases will be born to vaccinated women. Furthermore, vaccinated mothers of infants with NT may self-select for notifi- cation through their health care utilization behav- iour or may live closer to health facilities and be more likely to use services and thus be detected; this would produce a bias towards detection of NT cases born to vaccinated rather than unvaccinated women. Since vaccine effectiveness depends on both the pro- portion of cases born to vaccinated (PCV) and to unvaccinated women, and because PCV would be artificially inflated, such trends would underestimate Guidelines for investigating suspected cases of neonatal tetanus. Unpublished document WHO/EPI/TRAM/93.3, 1993 (available upon request from Global Programme for Vaccines and Immuniza- tion, World Health Organization, 1211 Geneva 27, Switzerland). i Detection of tetanus toxoid vaccine failure and response in the field. Unpublished WHO document EPI/TECHCOMMWP/93.34, 1993 (available upon request from Global Programme for Vaccines and Immunization, World Health Organization, 1211 Geneva 27, Switzerland). Table 4: Expected proportion of NT cases among in- fants born to mothers vaccinated with two or more doses of properly spaced TT, by vaccine effectiveness (VE) level % of cases at VE level of:a Coverage (%) 80% 90% 20 5 2 25 6 3 30 8 4 35 10 5 40 12 6 45 14 8 50 17 9 55 20 11 60 23 13 65 27 16 70 32 19 75 38 23 80 44 29 90 64 47 100 100 100 a PCV = PPV- (PPV x VE) 1- (PPV x VE) where PCV = proportion of NT cases born to vaccinated wo- men; PPV = proportion of the population vaccinated (coverage with BTT2); and VE = vaccine effectiveness. Zero doses are com- pared to at least 2 TT doses. See ref. 24, 28, 34. TT vaccine efficacy. Thus, managers may face in- creasing numbers of reports of NT cases involving infants born to vaccinated women, due not to TT failure but to improved detection ofNT cases. A case of NT involving an infant born to a "vaccinated" mother cannot necessarily be regarded as a TT fail- ure until comprehensive investigation of the prior vaccination history permits such a conclusion to be reached. Reports of NT cases must be investigated to confirm the clinical diagnosis; confirm both the number and dates of TT doses received, ideally from a vaccination card or from health facility records; and assure that unimmunized mothers are immu- nized, in view of the substantially higher risk of addi- tional NT cases among infants born to mothers with a prior history of an infant with NT (29). Only an NT case that meets the standard WHO case definition,k who was born to a woman with at least two properly- spaced doses of TT should be considered a TT fail- ure. Thus, the TT status of mothers of NT cases must k An infant with normal suck-and-cry during the first 2 days of life, who develops the inability to suck between days 3 and 28 of life and who develops either stiffness, convulsions or both. WHO Bulletin OMS. Vol 74 1996 625 V. Dietz et al. also be appropriately classified. For example, in Kenya the TT status of NT cases was classified as follows: protected with documentation; protected without documentation; unprotected; uncertain; and unknown (30). Caution must be exercised in deter- mining the TT vaccination status of women without card documentation, although some workers have suggested that maternal history of the number of doses received correlates well with antibody levels (31); also, similar efficacy estimates were obtained when either histories or card documentation were used to assess effectiveness in the Pakistan study cited above. The proportion of "toxoid failures" among a sequential sample of NT cases, e.g. the most recent 30 cases reported, should be updated as new cases are investigated. This information, along with TT coverage data, should then be used to assess periodi- cally vaccine effectiveness using formulas for vaccine effectiveness (Table 4). The magnitude of further investigative efforts should be guided by these as- sessments, with indications of unsatisfactory ef- fectiveness prompting more vigorous efforts. In ad- dition, the WHO Steering Committee on Epidemi- ology and Field Research has been evaluating the use of tetanus serological methods to assist in evalua- ting potential toxoid failures (32). Once a TT failure is detected, a review of the cold chain in the district should be conducted to determine whether the vaccine has been exposed to excessive cold or heat. Any women in the commu- nity who are found to be unvaccinated should be vaccinated. In addition, reviews of admission regis- tries in health care facilities in the district (with emphasis on hospitals) can be used to identify other NT cases that may represent TT failures. Finally, national authorities should consult WHO and UNICEF as to the appropriateness of conducting TT potency testing, reviewing production facilities and carrying out any special studies, such as sero- prevalence investigations, to identify the reasons for the magnitude of poor TT effectiveness. The unexpected and unacceptable performance of TT in the Punjab, Pakistan, has not been satisfac- torily explained. Exposure of Punjab neonates to greater concentrations of tetanus toxins does not seem likely, in view of the closely similar rates for unimmunized mothers in all regions. Babies of ap- propriately immunized mothers had a significantly higher risk of NT in Punjab than elsewhere; thus the toxoid used in Punjab appeared to be significantly less immunogenic. This may have arisen for the fol- lowing reasons: the TT that was supplied to the area was intrinsically less potent; the TT was mishandled in the field; or differences in response of mothers because of host factors such as malaria. The first two of these factors provided the most likely explana- tions. Additional studies of in-the-field effectiveness of TT in preventing NT in the developing world are clearly needed. In 1992, Member States at the Forty-fifth World Health Assembly resolved to use only vaccines that met WHO requirements in their immunization pro- grammes and to include this requirement in their immunization plans' and requested that the Director- General obtain information on steps to assure vac- cine quality, and specifically requested information on efforts to ensure that countries establish infra- structures for the quality assurance of tetanus tox- oid and poliovirus vaccine. In addition, in 1980 the EPI Global Advisory Group recommended that WHO develop strategies for vaccine quality control (33). The findings in Bangladesh (11) and the data we have presented illustrate the urgency in review- ing production facilities in countries that both produce 1T and report NT cases. However, it is important to note that evaluating production and ensuring that a vaccine is safe and effective involve more than just potency testing. Although it is reas- suring that a vaccine has been shown to be potent, vaccines can vary in potency between lots and even within the same lot. Thus, a positive potency test result, i.e. potent vaccine, must be interpreted with some caution; however, a negative result, i.e., subpotent vaccine, demonstrates a production prob- lem. By 1997 the WHO Global Programme for Vaccines and Immunization is planning to have com- pleted follow-up evaluations of manufacturing and control processes in the 22 countries with reported NT cases and local production of TT. Priority has been given to those countries with demonstrated potency problems in their locally produced TT and which continue to report NT cases. All countries will be encouraged to develop some form of national control authority for vaccines, with all vaccine- producing countries receiving priority support for such development. It is clear that the 1995 goal of NT elimination has not been achieved because vaccination coverage is unacceptably low, especially in high-risk areas, and thousands of cases continue to occur. Intensified im- munization of women in these high-risk districts is urgently required. In addition, many countries with NT cases are using 1T vaccine that is of unknown quality and of uncertain in-the-field effectiveness. Major efforts must be made to raise coverage and ensure that all Ti is of good quality, that vaccines Handbook of resolutions and decisions of the World Health As- sembly and the Executive Board, vol. III, third edit. (1985-92). Geneva, World Health Organization, 1993. 626 WHO Bulletin OMS. Vol 74 1996 Performance and potency of tetanus toxoid are handled appropriately, and that suspected TT failures are properly investigated and classified ap- propriately either as true vaccine failure or occurring as a consequence of improper timing or interval of TT administration. Additional studies are needed to understand more fully the factors that may affect the immune response to TT. The monitoring of TT effectiveness is crucial and must be based on the following: accurate surveillance, including case in- vestigation to ascertain vaccination history to mini- mize misclassification of vaccination status as well as selection bias in reporting NT cases; in-the-field performance estimates based on coverage and the proportion of toxoid failures among recent NT cases; and ongoing monitoring of TT potency and production. Unfortunately, current NT case repor- ting may not be adequate enough to conclude con- fidently that a problem does or does not exist. However, failure to act now to monitor TT effec- tiveness could compromise the activities and efforts of many countries that are actively involved in realizing the goal of global elimination of NT as a public health problem. Resume Qualite et activite de I'anatoxine tetanique: repercussions sur 1'elimination du tetanos neonatal Le t6tanos n6onatal est une cause majeure de mortalit6 dans les pays en d6veloppement, ou il provoque chaque ann6e plus de 400000 d6ces. L'OMS a adopt6 comme objectif 1'6limination de cette maladie a 1'6chelle mondiale. La principale strat6gie de pr6vention dans les r6gions a haut risque consiste a administrer aux femmes en age de procr6er au moins deux doses d'anatoxine t6tanique convenablement espac6es afin de pro- t6ger les nouveau-n6s a la naissance par transfert d'anticorps. Des 6tudes ont toutefois montre que, dans certains pays, I'anatoxine t6tanique de pro- duction locale poss6dait une activit6 inf6rieure a la normale, tandis que dans d'autres on observait des cas de t6tanos n6onatal chez des nouveau-n6s dont la mere avait 616 vaccin6e. Cependant, la plupart des rapports concernant ces derniers cas ne mentionnent pas le calendrier d'administration des doses d'anatoxine. Ces r6sultats, bien que tres utiles, sont souvent difficiles a 6valuer et doivent 6tre interpr6t6s avec prudence. Quoi qu'il en soit, des cas de t6tanos n6onatal chez des nouveau- n6s dont la protection antit6tanique avait 6t6 confirm6e par des analyses de laboratoire ont 616 signal6s. D'autre part, des 6tudes r6alis6es au Pakistan et au Bangladesh ont r6v6l6 que 1'effica- cit6 de l'anatoxine t6tanique etait inf6rieure a 50% chez les femmes vaccin6es. Ces rapports ont incit6 l'OMS a 6valuer les stocks d'anatoxine t6tanique dans les pays ou de tels cas sont rapportes. Cer- tains cas de t6tanos n6onatal chez les enfants dont la mere a 616 vaccin6e sont a pr6voir, car aucun vaccin n'est efficace a 100%; il est toutefois indispensable d'observer un espacement conve- nable entre les doses administr6es ainsi qu'un intervalle suffisant avant la naissance pour obtenir une efficacit6 protectrice. L'6valuation des cas de t6tanos n6onatal doit etre normalis6e pour tenir compte des ant6c6dents vaccinaux de la mere. Un examen approfondi des proc6d6s de production et de contr6le de qualit6 de l'anatoxine a 6t6 r6alis6 entre 1993 et 1995 dans huit des 22 pays produc- teurs ou sont rapport6s des cas de t6tanos n6onatal; une 6valuation plus rapide a 6t6 faite dans les 14 autres pays. Bien que 42 pays produisent de I'anatoxine t6tanique, seuls les neuf fabricants (de neuf pays) qui fournissent ce vaccin a l'UNICEF sont autoris6s a d6clarer que leurs pro- duits satisfont aux normes de l'OMS et aux criteres d'achat par les agences des Nations Unies. 11 ressort d'un inventaire des fonctions de contr6le a 1'6chelon national dans les 22 pays qui produisent de l'anatoxine t6tanique et ou sont signal6s des cas de t6tanos n6onatal, que seuls quatre d'entre eux (Br6sil, Inde, Indon6sie et Mexique) possedent des autorit6s nationales de contr6le fonctionnant con- venablement. Dans les cinq pays producteurs qui enregistrent le nombre le plus 6lev6 de deces par tetanos n6onatal, seuls deux possedent des autorit6s nationales de contr6le fonctionnelles. Dans la realit6, la situation r6gnant dans ces pays pourrait 6tre encore plus grave. Des essais d'ac- tivite ont 6t6 effectu6s sur 80 lots d'anatoxine t6tanique provenant de 21 fabricants de 14 des 22 pays concern6s. Une activit6 inf6rieure a la valeur optimale, indiquant un d6faut de qualit6, a 6t6 trouv6e dans 15 lots (19%) provenant de huit fabricants (38%) de sept pays (50%). La contr6le de l'efficacit6 de l'anatoxine t6tanique est fondamentale et doit reposer sur les points suivants: surveillance rigoureuse, avec investiga- tion des cas pour v6rifier les ant6c6dents vaccinaux (afin d'6viter des erreurs sur l'6tat vaccinal et des biais de s6lection dans la notification des cas de tetanos n6onatal); des estimations, sur le terrain, de la qualit6 du vaccin, d'apres la couverture vaccinale et la proportion d'6checs de la vaccination chez les cas r6cents de t6tanos n6onatal; la surveillance continue de I'activit6 de l'anatoxine tetanique et de ses proc6d6s de fabrication. WHO Bulletin OMS. Vol 74 1996 627 V. Dietz et al. References 1. Rahman M et al. Use of tetanus toxoid for the preven- tion of neonatal tetanus. 1. Reduction of neonatal mortality by immunization of non-pregnant and preg- nant women in rural Bangladesh. Bulletin of the World Health Organization, 1982, 60: 261-267. 2. Schofield FD, Tucker VM, Westbrook GR. Neonatal tetanus in New Guinea, effect of active immunization in pregnancy. British medical joumal, 1961, 2: 785- 789. 3. Newell KW et al. The use of toxoid for the prevention of tetanus neonatorum. Bulletin of the World Health Organization, 1966, 35: 863-871. 4. Chen ST et al. Timing of antenatal tetanus immuniza- tion for effective protection of the neonate. Bulletin of the World Health Organization, 1983, 61: 159-165. 5. Jones TS. The use of tetanus toxoid for the preven- tion of neonatal tetanus in developing countries. In: Recent advances in immunization, a bibliographic re- view. Washington, DC, Pan American Health Organi- zation, 1983 (Scientific Publication No. 451). 6. Dhillon H, Menon PS. Active immunization of women in pregnancy with two injections of adsorbed tetanus toxoid for prevention of tetanus neonatorum in Pun- jab, India. Indian journal of medical research, 1975, 63: 583-589. 7. Progress toward the global elimination of neonatal tetanus 1989-1993. Morbidity and mortality weekly report, 1994, 43(48): 885-887; and 1994, 43(48): 893-894. 8. Owa JA, Makinde 00. Neonatal tetanus in babies of women immunized with tetanus toxoid during preg- nancy. Tropical doctor, 1990, 20: 156-157. 9. Ghosh JB. Prevention of tetanus neonatorum. Indian pediatrics, 1990, 27: 210. 10. Owa JA, Makinde 00. Neonatal tetanus in babies of immunized mothers. Journal of tropical pediatrics, 1990, 36: 143-144. 11. Hlady WG et al. Neonatal tetanus in rural Bangla- desh: risk factors and toxoid efficacy. American jour- nal of public health, 1992, 82: 365-1369. 12. The Task Force for Child Survival and Develop- ment. Global 2000 child survival project on neonatal tetanus in rural Pakistan, 1988-1991. Atlanta, GA, Carter Center, April 1992. 13. WHO Expert Committee on Biological Standardiza- tion. Fortieth Report. Geneva, World Health Organiza- tion, 1989 (WHO Technical Report Series, No. 800). 14. Milstien JB et al. Global DTP manufacturing capacity and capability: status report. Vaccine, 1996, 14: 313- 320. 15. MacLennan R et al. Immunization against neonatal tetanus in New Guinea: antitoxin response of preg- nant women to adjuvant and plain toxoids. Bulletin of the World Health Organization, 1965, 32: 683-697. 16. Galazka AM. Tetanus: the immunological basis for immunization. Unpublished document WHO/EPI/ GEN/93.13, 1993. 17. Deivanayagam N, Nedunchelian K, Kamala KG. Neonatal tetanus: observations on antenatal immuni- zation, natal and immediate post-natal factors. Indian journal of pediatrics, 1991, 58: 119-122. 18. Bhat GJ, Joshi K, Kandoth W. Neonatal tetanus- a clinical study of 100 cases. Indian pediatrics, 1979, 16: 159-166. 19. Kumar H et al. Tetanus neonatorum: clini- coepidemiological profile. Indian pediatrics, 1988, 25: 1054-1058. 20. Grange AO. Neonatal tetanus in Lagos metropolis. Nigerian joumal of pediatrics, 1991, 18: 12-21. 21. Bwire R, Kawuma HJS. Hospital-based study on neonatal tetanus, Buluba Hospital, 1985-1989. Jour- nal of tropical medicine and hygiene, 1992, 95, 62- 66. 22. Cardenas Ayala VM et al. Neonatal tetanus mor- tality in Veracruz, Mexico, 1989. Bulletin of the Pan American Health Organization, 1995, 29: 116- 128. 23. Dietz VJ et al. Factors affecting the immunogenicity and potency of tetanus toxoid. Bulletin of the World Health Organization, in press. 24. Orenstein WA et al. Efficacy of pertussis vaccine. Joumal of pediatrics, 1990, 117: 508-509. 25. Black RE, Huber DH, Curlin GT. Reduction of neonatal tetanus by mass immunization of non- pregnant women: duration of protection provided by one or two doses of aluminium-absorbed tetanus toxoid. Bulletin of the World Health Organization, 1980, 58: 927-930. 26. De Moraes-Pinto Ml et al. Neonatal tetanus despite immunization and protective antitoxin antibody. Journal of infectious diseases, 1995, 171: 1076- 1077. 27. Grant JP. The state of the world's children. New York, Oxford University Press, 1994. 28. Orenstein WA et al. Field evaluation of vaccine effi- cacy. Bulletin of the World Health Organization, 1985, 63: 1055-1068. 29. Bennett J et al. Further observations on ghee as a risk factor for neonatal tetanus. International journal of epidemiology, 1995, 24: 643-647. 30. Bjerregaard P et al. Neonatal tetanus mortality in coastal Kenya: a community survey. International journal of epidemiology, 1993, 22: 163-169. 31. Vernacchio L et al. Neonatal tetanus in Peru: risk assessment with modified enzyme-linked immuno- sorbent assay and toxoid skin test. American journal of public health, 1993, 83: 1754-1756. 32. Expanded Programme on Immunization. Estimat- ing tetanus protection of women by serosurvey Burundi. Weekly epidemiological record, 1996, 71(16): 117-120. 33. Expanded Programme on Immunization. Weekly epidemiological record, 1981, 56(2): 9-12. 34. Kim-Farley R, Sokhey J. A simple screening method for field evaluation of vaccine efficacy. Joumal of com- municable diseases, 1988, 20: 32-37. 628 WHO Bulletin OMS. Vol 74 1996