Safety of oral poliomyelitis vaccine: results of a WHO enquiry.

Bulletin of the World Health Organization, 66 (6): 739-746 (1988) ( World Health Organization 1988 Safety of oral poliomyelitis vaccine: results of a WHO enquiry K. ESTEVES1 Most ofthe 13 countries that participated in the study used trivalent oral poliovaccine while monovalent vaccines were partly used in two countries and inactivated vaccines wholly by three countries and in part by two. A total of 395 cases of acute persistent spinal paralysis was recorded from this population of 547 million over the S-year study period (1980-84), giving an incidence of0.14 cases per million per annum. The incidence varied widely between countries and not all cases were related to immunization. The number of cases among recipients of the vaccine and among contacts with vaccine recipients was less than 0. 3 per million doses ofthe vaccine distributed or administered in all but one of the 9 countries where the live vaccine was widely used. The risk was con- siderably higher in one country, with nearly S recipient cases and 6 contact cases per million doses of the vaccine administered. In both vaccineees and contacts most cases were due to poliovirus type 2 although type 3 was more common among recipient cases in the countries with a very low incidence of vaccine-associated cases. The analysis of contact cases showed the needfor immunization ofpresumed susceptible adults at the same time as their children are vaccinated. INTRODUCTION In 1969, WHO organized a collaborative study to obtain information on the possible risks associated with the use of oral poliomyelitis vaccine (OPV) and on measures to reduce these risks if they were clearly shown to exist. The findings during the first five and ten years of the study have been reported previously (1, 2) and those for the last five years of the study, 1980-84, are reported here. Thirteen countries contributed information on paralytic cases and immunization practices for these five years. Criteria for inclusion of cases All cases of spinal paralysis of acute onset, in which the paralysis was typical of poliomyelitis and persisted for more than six weeks or until death (if this occurred within the six-week period), were included. Classification of cases The cases were classified as follows: -'recipient': a case in which the illness began within ' Division of Communicable Diseases, World Health Organ- ization, 1211 Geneva 27, Switzerland. Requests for reprints should be sent to this address. 7-30 days after the patient received the vaccine; -'contact/possible contact': a case in which the patient was known to have been in contact with a vaccinee, and became ill within 7-60 days after the vaccinee had taken the vaccine; or if there was no such known contact, a case in a patient living in an area in which a mass immunization campaign had been in progress between 7 and 60 days before the onset of the disease; -'no known contact': a case in which paralysis occurred in a patient with no evidence of known contact or possible contact as defined above. FINDINGS The data received from the 13 countries partici- pating in the study in 1980-84 largely confirm earlier patterns and in particular those of 1975-79. Four countries (No. 1, 2, 6 and 12; total population, 39 million) were free from cases of acute persistent spinal paralysis (APSP) all through 1980-84, eight countries (No. 3, 4, 5, 7, 9, 10, 11 and 13; total population, 485 million) had a very low incidence of the disease, and one country (No. 8; population, 22.6 million) continued to experience more cases than any of the others. From a total population of 547 million (1983 estimate), a total of 395 cases of APSP 4934 -739- K. ESTEVES 006 0 0 9 0 0 CDto CN D 0 o O (N0 o 00 0 CD0 Coa) C0 0 (N 0 0 e C_ 0 0 0 0 CD 0 Co co ? co CD6i CD (N Co 0 C') 0co CD C' 0 CD 0 6 0 6i 0D 00 0 o~~~ cvi 0Co CN4 C- 0 0 0 0 V'- Co C Co (N O N O b N"O C4 1O 0 I*O0 0 0 0.- 0 06 6 6 6 6 6 o 6 6 6 6 6 (N CD (N S C 0 N Dnd'in 6 6 _- q* _ amo 0 _CD C'-)_ C (N CN 0 CD co(N C'(N (N(N r- 0 co CD v- co rt 0 0 D (N o 0 O N q _ a W 0 C 0 CD 0 CD w . (N 0O(N - (N D- tD 0 0 0 CO 0 CN o o T- CD 0C) o) -(N (N- N S r. ° - N 6 6 6 6 6 6 6 6 6 6 6 6 Z Z Z Z Z Z ZZ Z Z Z Z 740 a 0 E 0 C L) 0 .(_ m n C') 0 Co co CN CoY 0 aw (N CD0. LN (N (N (N CDCU CNCN CoaCCm C- 0I co 0 CD r- Oc 0 E C 00O~ 0zo C)0 oC 75 0. 0 CC 0 C C 0.-0 Q_ 0 ' a. CD L.' z CD wU CD Q- m 000?C cn Co C_ 0 u cn. _o C-o _ .r a a 6 CN Co CD o CO Co co 0 CD D coCV C') (n Co C') 0 Co LO *0 CD 0- co CD Co 0 co a- U) U1) 0 0. E co CO 0 ._ U) 0._ 0 0B 0 U) 0 0 0 . 0 . LO CD In (N (N 100. 0 0 a. D C,)an 0- 6 z 0 CD)Co 0(v) cno U _- _- aw -a Co 0 .0 (I) Co 6 z SAFETY OF ORAL POLIOVACCINE fulfilled the criteria for inclusion in the study. This represents an annual average rate of 0.14 cases per million population in the five-year period (Table 1). Country No. 8's earlier problems with the satis- factory control of OPV continued in 1980-84 and accounted for 305 of the 395 cases (77%) reported. As shown in Table 1, the annual rate in this country was 90 times higher than that of the other countries in the study, 2.71 vs. 0.03, respectively. Ten countries used oral vaccine; three of them administered 20 million doses and the remaining seven reported the distribution of 193 million doses. Relating the total number of cases classified as associated with vaccine use in the six countries relying exclusively on OPV and for which infor- mation on vaccine distribution is available shows an overall rate of 0.17 recipient cases and 0.15 contact cases per million doses, i.e., one case among recipi- ents per 5.9 million doses ofOPV distributed and one contact case per 6.7 million doses (Table 2). Countries reporting zero cases Countries No. 1, 2, 6 and 12 did not report any cases during the 1980-84 period (Table 1). Different approaches were employed to control poliomyelitis in these countries and they have all been shown to be effective. Country No. 1 (population, 15.4 million) used mono-, bi- and trivalent oral poliovaccines admini- stered in short-term campaigns. In 1980, the first dose was bivalent vaccine types 1 and 2 given to children aged 2-14 months in March, followed by one dose of monovalent type 3 vaccine in May, and a final dose of trivalent vaccine at school age. In 1981-84 the schedule was one dose of monovalent type 1 vaccine followed by bivalent types 2 and 3 vaccine. A total of 7.4 million doses were admini- stered during the 1980-84 period. Country No. 2 (population, 5.1 million) had a combined vaccination schedule of three doses of inactivated poliovaccine followed by three doses of oral vaccine administered throughout the year. Countries No. 6 and 12 (total population, 18.5 million) relied exclusively on inactivated polio- vaccine. Countries reporting a small number of cases Eight countries (No. 3, 4, 5, 7, 9, 10, 11 and 13; total population, 485 million) reported 90 cases of APSP during 1980-84;" this corresponds to an annual average incidence of 0.04 cases per million. Thirty-three cases occurred among recipients, 29 in contacts or possible contacts, and 28 cases were included in the group of 'no known contact' with the vaccine (Table 1). A few cases in the latter group could have been related to the vaccine, but the time lapse between contact with the vaccine and onset of illness was longer than that defined for vaccine- associated cases in this study. Recipient cases: Five countries (No. 3, 4, 5, 7 and 10) reported 33 cases among the recipients, and two Table 2. Number of cases of acute persistent spinal paralysis (APSP) associated 1980-84 with oral poliovaccines (OPV), Risk per million doses No. of APSP cases among: (administered/distributed) No. of doses of OPVb Countrya (millions) Recipients Contacts Recipients Contacts No. 1 7.445 A 0 0 0.00 0.00 No. 3 34.428 D 9 4 0.26 0.12 No. 4 3.958 A 1 1 0.25 0.25 No. 5 18.889 D 5 3 0.26 0.16 No. 7 19.346 D 1 0 0.05 0.00 No. 8 8.410 A 42 49 4.99 5.83 No. 9 3.562 D 0 0 0.00 0.00 No. 10 107.413 D 17 20 0.16 0.19 No.11 8.148 D 0 1 0.00 0.12 No. 3, 5, 7, 9, 10, 11 191.786 D 32 28 0.17 0.15 a Countries No. 12 and 13 used inactivated poliovaccines (IPV); country No. 2 used both IPV and OPV; country No. 6 used OPV until 1980, then IPV. b Figures refer to millions of doses administered (A) or distributed (D). 741 742 K. ESTEVES of them (No. 3 and 10) had cases in each of the five years. Of the 33 cases, 23 were in infants below six months of age, six were between 6 and 11 months old, two were aged one year, and one was two years old. This latter child was hypogammaglobulinaemic and fell ill 12 days after the third dose of oral polio- vaccine. One recipient case was in an adult, a 55- year-old man, who fell ill after one dose of trivalent oral poliovaccine (TOPV) taken before a planned trip abroad. He had not been vaccinated before. Thirty cases occurred after the first dose of TOPV, one after a second dose (a 15-months-old child), one after the third dose (the immunodeficient child mentioned above), and one case after a dose of monovalent type 3 vaccine. Poliovirus type 1 as a single virus was isolated from one case (the immunodeficient child). Serotype 3 was the most frequently isolated poliovirus: in 12 cases it was the only virus detected and in 7 additional cases it was isolated with poliovirus type 1 and type 2. Poliovirus type 2, as a single virus, was isolated from six cases and from six additional cases in mixed infections-one with types 1 and 3 and five with type 3. Almost all strains were further investigated in intra- typic differentiation tests and all those tested (two strains of poliovirus type 1, ten of type 2, and 17 of type 3) were characterized as vaccine-like. Contact cases. Five countries (No. 3, 4, 5, 10 and I 1) reported 29 cases in contacts or possible con- tacts, and one of them (No. 10) had a contact case in each of the five years. Sixteen cases were in adults aged between 21 and 54 years. Eleven of them were parents to recently vaccinated children and one ad- ditional case was a close household contact with a vaccine recipient. Three had been in contact with vaccinated children in the family (niece, nephews and, in one instance, a grandchild). Only one reported case did not have any family relation to a vaccinee or other direct contact. The most probable source of infection was via a healthy person who had been in contact with a recently vaccinated child. None of the 14 adult contact cases whose vaccine history was known had received the full course of poliomyelitis vaccine. Thirteen contact cases were in children, nine were between three weeks and eight months of age, one was three years old, and the remaining three were 8, 14 and 16 years old. Only one whose vaccine history was known had received any vaccine: a 14-year-old girl who had received one dose of TOPV in early childhood. Four of the contact cases among children occurred after direct household contact with a vaccine recipient (siblings and others), four had contact with a Data not received from country No. 4 in 1983-84. other relatives (cousins, nieces), and three with play- mates in nursery school. In two cases there was no direct contact with vaccinees; one was classified as a possible contact case which occurred during a vaccination campaign in the area. In the other case the infection was probably transmitted via a healthy babysitter whose daughter had been vaccinated recently. Poliovirus type 2 was isolated as a single virus in nine cases, type 3 in eight, and type 1 in two cases. Mixed infection with more than one virus type was diagnosed in six cases. All three serotypes were isolated in one case, and types 2 and 3 in two cases. In two additional cases of type 2 and 3 infection, only one virus type was isolated but the patient showed serological responses to that and the other poliovirus type. In one case where specimens for virus isolation were obtained two months after the onset of illness, the diagnosis was based on serological responses to all three serotypes. Except for two isolates which were not further tested, all strains were found to be vaccine-like. Cases with no known contact with the vaccine: Twenty-eight of the cases reported in the 1980-84 period were classified as not associated with oral poliovaccine. They were reported by six countries (No. 3, 5, 7, 9, 10 and 13). Eight of these cases occurred during an outbreak in country No. 13 (see below). Of the remaining 20, nine occurred in country No. 7, six in country No. 10, two each in countries No. 2 and 9, and one in country No. 5. The age of these cases ranged from three months to 39 years. Fourteen were children under four years of age, three were between seven and 17 years, and three between 20 and 39 years old. Poliovirus type 1 was diagnosed in three cases by isolation of the virus; type 2 was diagnosed by iso- lation in five cases and by serology in two, and type 3 by isolation in two and by serology in three cases. More than one virus type was isolated in three patients-all three types in one case, types 1 and 2 in one, and types 2 and 3 in one case. Vaccination history was known for all but one of the 18 cases; eleven had not received any vaccine. Four of the vaccinated cases were diagnosed as immunodeficient. Three of these immunodeficient patients fell ill within a fairly short time (33, 41 and 90 days) after TOPV administration. One case yielded vaccine-like polioviruses of all three sero- types. Isolates from the other two cases (type 2 in one, and types 2 and 3 in the other) have not been further studied. The fourth immunodeficient patient had received four doses of inactivated poliovirus vac- cine followed by three doses of oral vaccine -the last dose five years before onset. Poliovirus type 1 was isolated from this case and was characterized as a wild SAFETY OF ORAL POLIOVACCINE poliovirus by oligonucleotide mapping and by anti- genic differentiation using strain-specific sera. Two further cases had received four doses of TOPV. One yielded poliovirus type 3 and had stable high titres to types 1 and 2. The fourth dose of the vaccine had been given over a year before the onset of illness. The other case received a full course of OPV and had been given the fourth dose two days before the onset and the third dose a year earlier. No virus was isolated but stable high titres to all three serotypes were detected in paired serum samples. DISCUSSION Risk of vaccine association Sixty-two vaccine-associated cases included in the study in 1980-84 were reported by six of the eight countries with a low incidence of APSP (No. 3, 4, 5, 7, 10 and 11). The data for country No. 4 were complete only for the period 1980-82. Four countries relied exclusively on trivalent oral poliovaccine in the regular immunization programme: countries No. 3, 7 and 10 distributed the vaccine throughout the year and country No. 5 in two short- term campaigns during spring and autumn. TOPV was also used exclusively in parts of country No. 11, whereas in other parts the inactivated vaccine was used. In country No. 4, the three types of monovalent vaccine were administered throughout the year for primary immunizations and the reinforcing dose. Together these six countries distributed 192 million doses of oral poliovaccine during 1980-84. Relating this number to the 33 recipient cases and 29 contact cases gives a risk of 0.17 recipient cases and 0.15 contact cases per million doses of vaccine or one recipient case per 5.9 million and one contact case per 6.7 million doses of the vaccine (Table 2). Another way of assessing the risk of paralysis that was temporarily associated with the vaccine is by considering only the children who received primary vaccination. Of the eight countries reporting a small number of cases during 1980-84, six (No. 3, 4, 5, 7, 9 and 10) relied exclusively on OPV. In these six countries, based on an estimated overall vaccine acceptance rate of 67% in the five-year period 1980-84, nearly 24 million children would have completed the full course of primary vaccination within their first two years of life. The acceptance rates were in some cases derived from data available for less than the full five years or, when missing altogether, the calculation was based on the accept- ance rates available for the ten-year period 1970-79 (2). Among this child population of 24 million there were 32 recipient cases. Excluding three born in 1979 and 1984 and therefore not within the study period for two complete years, there were 1.7 recipient cases per million children 'at risk'. Table 3 compares the risk assessed for 1980-84 with that for 1970-79 and shows that overall the risk remained largely unchanged, around one case per million children. However, in one country (No. 3) the risk increased from less than 1 to 2.7 cases per million children who received primary vaccination. Outbreaks in country No. 13 Country No. 13, where inactivated poliovirus vaccine (IPV) was used and which reported no polio- myelitis cases for over 20 years and was free of wild polioviruses for almost as many years, experienced an important outbreak in 1984. The first indication of Table 3. Number of paralytic cases among recipients of oral poliovaccine (OPV) in children below 3 years of age, 1980-84 Risk per million No. of infants OPV acceptance Population at No. of cases in children Country <1 year (millions) ratea (%) risk (millions) born in 1980-83 1980-84 1970-79 No. 3 3.140 83 2.595 7 2.70 0.96 No. 4 0.418 98 0.410 1 2.44 3.44 No. 5 7.661 93 7.139 15 0.70 0.55 No. 7 3.428 96 3.291 0 0.00 1.57 No. 9 3.324 79 0.255 0 0.00 0.00 No. 10 17.999 57 10.259 15 1.46 0.87 Total 35.970 67 23.949 28 1.17 0.94 a Notes on acceptance rates: country No. 3 and 5, estimated on information for 3 years; country No. 7 and 10, as no information was available, the overall acceptance rate estimated for 1970-79 was used. 743 K. ESTEVES re-introduction of poliovirus in the country was in October 1984 when poliovirus type 3 was isolated from a young boy with meningitis and from several of his healthy contacts. Shortly after this, cases of para- lytic poliomyelitis appeared; in all, six cases were confirmed from October to December 1984 and one in January 1985. Retrospective investigation of cases of paralytic disease occurring before October 1984 revealed two additional cases, one with onset in August and one in September 1984. The nine cases reported during the outbreak were scattered all over the country, indicating a wide spread of the virus. This was further confirmed by numerous virus isolations from healthy persons and from sewage. Of the nine cases reported, six had received at least three doses of inactivated vaccine, all within the 10 years preceding the outbreak, and four had accepted the full immunization course of five doses. Follow-up of the cases showed three with mild paralysis, three with moderately severe sequelae, and two who were severely handicapped. One 17-year-old boy died. The transmission of wild poliovirus type 3 was interrupted by the distribution of 4.5 million doses of TOPV during a mass campaign in February-March 1985. Investigation of polioviruses isolated during the outbreak indicated some difference compared with the type 3 strains used in the vaccine and also with isolates from other countries. Investigation of sewage samples was carried out in a neighbouring country which also relied on inactivated poliovirus vaccine. Poliovirus type 3 was isolated during December 1985 and early 1986 but only from sewage samples collected in areas fre- quented by tourists (M. B6ttiger, personal communi- cation, 1986). All three poliovirus types were isolated from sewage during the TOPV campaign in country No. 13. However, no virus could be isolated from such specimens collected during the autumn 1985. Outbreaks in country No. 8 Two events marked the five-year period 1980-84 in country No. 8. Vaccination with TOPV was resumed in April 1980 after a year when only mono- valent type 1 vaccine had been available, and about twice as many children as usual were administered TOPV. There was also a higher number of vaccine- associated cases in 1980 compared with the following four years and most of the preceding 10 years of the study. The second unusual feature was the occurrence of several outbreaks associated with wild viruses, mainly types 1 and 2 in March-April 1980 and type 1 from September 1980 to mid-1982. During the five-year period 1980-84, 305 cases of APSP (78% of the total reported in the study) occurred in country No. 8 with an annual average rate of 2.7 cases per million population (range 0.7-5.5), comprising 0.4 cases per million among the recipients (range 0.2-0.7), 0.4 cases per million in contacts (range 0.2-1.1), and 1.9 cases per million in persons without known contact with the vaccine. In 1980 the vaccine was administered during two campaigns, in April-June and September-December. In 1981 and 1982 the vaccine was distributed in single campaigns from September to December. Biannual campaigns were again adopted in 1983 and 1984, the vaccine being distributed in March-May and September-November. Vaccine-associated cases. In 1980-84, a total of 8.4 million doses of oral poliovaccine were administered (7.1 million doses ofTOPV and 1.3 million of mono- valent type 1 vaccine). During this period, 91 cases were temporarily associated with poliovaccine: 42 among the recipients and 49 in contacts, all children. Related to the total number of doses administered there were 5.0 recipient cases and 5.8 contact cases per million doses. Since no vaccine-associated cases occurred during 1979 when only monovalent type 1 vaccine was used, and no vaccine-associated case has been reported with poliovirus type 1, it seems reason- able to assume that the vaccine-associated cases were all induced by the trivalent vaccine. The risk related to the number of doses of TOPV administered during 1980-84 was 5.9 recipient cases and 6.9 contact cases per million doses. Cases with 'noknown contact'. From March 1980 to September 1982, a total of 214 cases occurred which were classified as endemic. The first 41 occurred in March-April 1980. Most cases were associated with two outbreaks, one due to poliovirus type 1 and the other to type 2. Sporadic cases of types 1, 2 and 3 were also reported during these two months. Vacci- nation with TOPV was resumed in April 1980 and the incidence ofendemic cases dropped. No further cases attributable to wild poliovirus types 2 and 3 have been reported after April 1980. Starting in September 1980, epidemic cases of poliovirus type 1 occurred again, which continued throughout 1981 with a peak in September. In 1982, the incidence dropped and no case of APSP caused by non-vaccine-like polio- viruses have been reported. Of the 184 cases attributed to wild poliovirus type 1, 52 occurred during outbreaks among children below 18 months of age in homes and institutions. These communities include many children who for various reasons have not received OPV. Most (86%) of the cases occurred among such unvaccinated children. The remaining cases were in children who had not received the full course of immunizations, suggesting an extremely low rate of vaccine failure, if any. 744 SAFETY OF ORAL POLIOVACCINE CONCLUSIONS The 395 cases of acute persistent spinal paralysis during the period 1980-84 were reported from a total population of 547 million (1983 estimate) and cor- respond to an annual average rate of 0.14 cases per million, which is the same as in the first ten years of the study. However, 305 of these cases were reported from one country. The remaining 90 cases were from a total population of 524 million, i.e., an annual average rate of 0.03 cases per million, which is a big reduction compared with the rate of0.11 cases per million for the same countries in the 1970-79 period. This improvement, although shared by several countries, was most remarkable in country No. 7 where the number of cases, especially those associated with vaccine, have been declining steadily since 1973. In that year this country changed from monovalent to trivalent oral vaccine, and in 1975 the vaccination strategy was changed from short-term mass campaigns to administration of the vaccine throughout the year. In the first five-year period of the study (1970-74) a total of 139 cases were reported, in 1975-79 there were 25 cases, and in 1980-84 only ten cases. In one country (No. 8) the control of poliomyelitis continues to pose a problem. Annual averages of about five recipient cases and nearly six contact cases per million doses of oral poliovaccine were reported for 1980-84 and there were, on average, 16 vaccine- associated cases per year during the whole study period. In fact, the only year when no vaccine- associated cases occurred was in 1979 when TOPV was not available. In addition, wild polioviruses continued to circulate among pockets of poorly immunized population groups dispersed in this country. These groups are characterized by high natality and infant morbidity, poor hygiene and educational status, and a high mobility which makes it difficult for the national immunization programmes to reach them. However, with the exception of country No. 8, all other countries in the study have, in general, managed to maintain poliomyelitis control and to keep the number of vaccine-associated cases low in relation to the number of doses distributed. The number of cases among the recipients has probably been reduced as far as is possible with a vaccine derived from live, attenuated polioviruses, i.e., one case for every six million doses distributed. A further reduction in the number of contact cases should, however, be feasible. More than half of the contact cases occurred in poorly immunized adults or the parents of recently vaccinated children. The recommendation issued after the first ten years of the study, i.e, that 'parents without definite evidence of previous effective immunization should be given the vaccine at the same time as their children', therefore needs to be re-emphasized. In country No. 13, which had totally controlled poliomyelitis by the use of inactivated poliovirus vaccine, there was an outbreak caused by a slightly changed type 3 virus in 1984. As well as from the nine paralytic cases (one fatal), virus was isolated from many healthy persons. Transmission ceased after the administration of 4.5 million doses of OPV. Thus, even in a highly immunized population there is a need to continue the monitoring of polioviruses in the community. The main conclusion of this study, which extended over a 15-year period, is that oral poliovaccine continues to be one of the safest vaccines in use. The risk of vaccine-associated cases, in all but one of the participating countries, was less than one per million vaccinees, cases due to type 3 virus being the most common in both vaccine recipients and contacts. Type 2 occurred more often in contacts than in vaccinees, and type 1 was rarely implicated. Good protection was also achieved by IPV, but serious breakthroughs are possible, especially with type 3. UME INNOCUITE DU VACCIN ANTIPOLIOMYELITIQUE BUCCAL: RESULTATS D'UNE ENQUETE DE L'OMS La plupart des 13 pays ayant participe 'a cette 6tude ont employ6 un vaccin antipoliomy6litique buccal (VPO) tri- valent alors que deux pays ont utilis6 en partie des vaccins monovalents, que trois pays ont utilise exclusivement des vaccins inactives et que deux pays les ont utilises en meme temps que d'autres vaccins. On a enregistr6 au total 395 cas de paralysie sur une population de 547 millions de personnes au cours des cinq dernieres ann6es de l'etude (1980-1984), ce qui donne une incidence de 0,14 cas par million de per- sonnes et par an, identique a l'incidence observ6e au cours des dix annees pr6c6dentes (1970-1979). Cependant, 305 de ces cas ont ete signales dans un seul pays. Les 90 autres l'ont ete pour une population totale de 524 millions, ce qui represente un taux moyen annuel de 0,03 cas par million,. tres nettement inferieur au taux de 0,11 cas par million en- registredans les memes pays pendant laperiode 1970-1979. 745 746 K. ESTEVES L'amelioration a ete observee dans plusieurs pays, mais elle a e particulierement nette dans le pays NO 7 oiu le nombre de cas, notamment ceux associes au vaccin, a regulierement diminu6 depuis 1973. Cette ann6e-la, le pays en question est passe du vaccin monovalent au vaccin buccal trivalent, et en 1975, il a modifie sa strategie vaccinale, abandonnant les campagnes de vaccination de masse de courte duree au profit d'un programme de vaccination s'etendant sur toute l'annee. Le nombre de cas signales a ete de 139 entre 1970 et 1974, 25 de 1975 a 1979 et seulement 10 de 1980 a 1984. Dans un des pays (N° 8), la lutte contre la poliomy6lite continue a poser un probleme. Durant la periode 1980- 1984, il y a eu en moyenne chaque annee cinq cas de para- lysie par million de doses de vaccin buccal chez les enfants vaccines et pres de six cas chez les contacts. On a releve en moyenne 16 cas par an associ6s au vaccin sur l'ensemble de la periode d'etude. De fait, la seule ann6e au cours de laquelle on n'a enregistre aucun cas associ6 a la vaccination a ete I'annee 1979 pendant laquelle le VPO trivalent n'etait pas disponible. En outre, des virus poliomy6litiques sau- vages ont continue 'a circuler dans les ilots de population mal immunises dans diverses regions du pays. Ces groupes sont caracterises par une forte natalit6 et une forte morbidite infantile, un faible niveau d'instruction et d'hygiene et une grande mobilite qui rend difficile l'application des pro- grammes nationaux de vaccination. Toutefois, a 1'exception du pays NO 8, tous les autres pays ayant participe a l'6tude ont en g6neral reussi a contenir la poliomyelite et n'ont eu a deplorer qu'un faible nombre de cas de paralysie associes au vaccin par rapport au nombre de doses distribuees. L'incidence des cas de paralysie chez les sujets vaccines a probablement ete r6duite au niveau le plus bas qu'il soit possible d'esperer avec un vaccin prepare a partir de virus vivants attenues, soit un cas pour six millions de doses distribuees. Toutefois, il devrait etre possible de diminuer encore le nombre de cas chez les contacts. Plus de la moitie de ces cas se sont produits chez des adultes peu immunises ou chez les parents d'enfants recemment vac- cines. II faut donc insister a nouveau sur la recommandation formulde a l'issue des dix premieres anndes de l'etude, a savoir que les parents dont on n'est pas suir qu'ils soient correctement immunises doivent recevoir le vaccin en meme temps que leurs enfants. Dans le pays NO 13, oui la poliomyelite avait ete parfaite- ment maitrisee grace a un vaccin inactive, une poussee due a un virus de type 3 ldgerement modifie a eu lieu en 1984. Le virus isole chez les 9 sujets atteints de paralysie (dont l'un est decede) a ete retrouve chez de nombreux sujets en bonne sante. La transmission a cesse apres qu'on eut administre 4,5 millions de doses de VPO. Ainsi, meme lorsque la popu- lation est bien immunisee, il faut maintenir la surveillance du virus poliomyelitique dans la communaute. La principale conclusion de cette etude, qui a porte sur une periode de plus de 15 ans, est que le vaccin antipolio- myelitique buccal est toujours l'un des vaccins les plus sutrs. Dans tous les pays participants sauf un, le risque de para- lysie lie a la vaccination a ete de moins d'un cas par million de sujets vaccines, les cas dus au virus de type 3 ayant et6 les plus courants a la fois chez les sujets vaccines et chez les contacts. Le virus de type 2 a ete identifie plus souvent chez les contacts que chez les vaccines, et le virus de type I a ete rarement mis en cause. Le vaccin antipoliomyelitique inac- tivd a egalement assure une bonne protection, mais des poussees aux consequences serieuses restent possibles, en particulier avec le type 3. REFERENCES 1. The relation between acute and persisting spinal para- lysis and poliomyelitis vaccine (oral): results of aWHO enquiry. Bulletin of the World Health Organization, 53: 319-331 (1976). 2. The relation between acute persisting spinal paralysis and poliomyelitis vaccine-results of a ten-year enquiry. Bulletin of the World Health Organization, 60: 231-242 (1982).