Effectiveness of oral killed typhoid vaccine*

Bull. Org. inoitd. Sante 1 1971 45, 445-450 Bull. Wld Hlth Org. j Effectiveness of Oral Killed Typhoid Vaccine* C. S. CHUTTANI,1 K. PRAKASH,2 A. VERGESE,2 U. SHARMA,3 P. SINGHA4 & B. GHOSH RAY5 A controlled field trial of oral killed typhoid vaccine was carried out in Delhi, India in 1968-69. Altogether, 13 374 children below the age of 17 years were included in the study. Two comparable groups ofchildren were given either placebo or vaccine. Each tablet of vaccine contained 100 x 109 killed Salmonella typhi, and 3 tablets of vaccine or placebo were administered to each child. The vaccinated subjects werefollowed upfrom 10 June 1968 to 31 August 1969. The effectiveness of the vaccine was measured by comparing the inci- dence of typhoid fever (based only on bacteriologically positive cases) in the two groups. It wasfound that the difference in incidence ofthe disease in the two groups was not statisti- cally significant. The oral killed typhoid vaccine in the dosage schedule used in the present trial was found not to be effective against the disease. Controlled field trials with parenteral vaccines against typhoid fever conducted in Yugoslavia (Yugoslav Typhoid Commission, 1962, 1964), British Guiana (now Guyana) (Typhoid Panel, UK Department of Technical Co-operation, 1964) Poland (Polish Typhoid Committee, 1966), and the USSR (Hejfec et al., 1966), under the sponsorship of World Health Organization (Cvjetanovic & Uemura, 1965), have proved that these vaccines can provide a relatively high level of long-lasting immunity. An effective oral typhoid vaccine would however have certain advantages since repeated injections are not readily accepted by the people in developing countries. The present study was undertaken to determine the effectiveness of an oral killed typhoid vaccine by a controlled field trial. * This trial was sponsored by the Indian Council of Medi- cal Research in Collaboration with the World Health Organi- zation 1 Clinical Epidemiologist, Indian Council of Medical Re- search, New Delhi, India. Present address: Professor of Epidemiology, National Institute of Health Administration and Education, New Delhi. 2 Lecturer, Department of Microbiology, Maulana Azad Medical College, New Delhi. 3Assistant Research Officer, Indian Council of Medical Research, New Delhi. 4 Assistant Director, Statistics, National Institute of Com- municable Diseases, New Delhi. 6 Professor of Microbiology, Maulana Azad Medical College, New Delhi. AREA AND POPULATION SELECTED FOR THE TRIAL The study population consisted of children below the age of 17 years in a densely populated, low socio- economic community on the northern bank of the Jamuna River (Geeta Colony) in Delhi. The water supply was from shallow (16-20 ft; 5-6 m) tube wells with a separate hand-pump for each household. Dwellings were provided with bucket latrines and there were open drains for the removal of household waste. Each family cooked its own food. Fresh, un- pasteurized milk obtained from local dairies was boiled before use. Numerous hawkers visited the col- ony to sell cut fruits and vegetables, ice cream and, other foods. SIZE OF THE STUDY POPULATION Since no data were available on the actual incidence of typhoid in Delhi, a rough estimate of 5.0-7.5 cases per 1 000 persons per annum was made after hospital records had been examined. It was ob- served that typhoid fever occurred more often in areas without a piped water supply. On the basis of the estimated incidence of typhoid, a sample of 5 000 children per group was considered sufficient to demonstrate a statistically significant difference between the vaccinated and control groups if the effectiveness of the vaccine was at least 50 %. 2739 - 445- 446 C. S. CHUTTANI AND OTHERS The principles of controlled field trials were strictly observed. The placebo and vaccine tablets were la- belled with code letters and the double-blind test method was used. VACCINE The placebo and vaccine tablets were given the code letters " A " and " B ", respectively, and were supplied by the World Health Organization.' Each tablet coded with the letter " B " contained 100 x 109 acetone-killed Salmonella typhi. The vaccine was declared safe for use by the manufacturer.' Tablets coded with the letter " A " contained lactose. DESIGN OF THE STUDY Population census A house-to-house survey of the colony was made from 9 to 30 May 1968 by 20 research assistants. The members of each household were listed in family registers, together with data on age, sex, etc. When the census was complete, the names of all children aged 17 years and under were entered on a master sheet and numbered serially. Each family was given an identity card bearing the name and registered number ofeach child and was informed that the child- ren were entitled to free medical treatment at the cli- nic which had been specially established in Geeta Colony. (The clinic had medical officers and a staff nurse to look after sick children from the colony. All the children in the colony, irrespective of the type of disease, attended the clinic.) The total number of children recorded in the census was 16 490. Vaccination programme The children were allocated to group A or group B according to whether the final digit in their serial number was odd or even. Each child was given I tablet daily on 3 alternate days or on 3 consecutive days; as a rule, tablets were administered 1I/2-2 hours after a meal. Children who were suffering from either diarrhoea or fever at the time of vaccine administration were excluded from the list of the participants. Children who had received parenteral T.A.B. vaccine during the previous year were also excluded from the trial. The vaccine was administered during the period from 10 June to 12 August 1968. Altogether, 13 374 children were treated with tablets A or B (Table 1). 1 The vaccine was produced by the Swiss Serum and Vaccine Institute, Berne, Switzerland. Table 1. Distribution of children according to number of doses in the vaccine and control groups No. of doses T Control group T Vaccine group Total 1 359 347 706 2 176 184 360 3 6 248 6 060 12 308 total 6 783 6 591 13 374 All 3 doses could be administered to only 12 308 child- ren, 1 066 children received only 1 or 2 doses for reasons such as age below 6 months, sickness, re- fusals, and movement away from Delhi (Table 2). A vaccination card was completed for each participant. Surveillance Each participant was followed up regularly from the first day of vaccination. Each research assistant was responsible for approximately 150-200 house- holds, 60 households being visited every day by each assistant. In this way, each house was visited every fourth day. Details about the sickness of each child were entered in the family register during these visits. All sick children, irrespective of the type of disease, were referred to the clinic where they were examined and treated. Patients with fever were taken per- sonally to the clinic by the research assistants for diagnosis and treatment, and were followed up every day until they felt well. Those who had a high con- tinuous fever lasting more than 3-4 days without any signs or symptoms of another disease were investi- gated for enteric infection. A 5-ml sample of venous blood was taken from the patient, 2 ml being injected in a sterile bottle containing sodium citrate for sub- sequent culture and 3 ml being injected into an empty sterile bottle for the Widal test and clot culture. These samples were sent to the Microbiology Department of Maulana Azad Medical College, New Delhi, for further bacteriological and serological examination. Stool samples from all patients suspected of having typhoid fever were also sent to the Microbiology Department of Maulana Azad Medical College for culturing. During the first 3 months of the trial, the majority of the stool samples were collected in the first and second weeks of illness while the patients were already under treatment with chloromphenicol. Later, stool samples were collected from all patients EFFECTIVENESS OF ORAL KILLED TYPHOID VACCINE Table 2. Distribution of children according to age and the number of vaccine doses in the vaccine and control groups Age group (years) Control group 1 dose Vaccinated group 2 doses 3 doses Total 1 dose 2 doses 3 doses Total <1 28 I- 1-5 94 6-10 105 11-15 95 16-17 37 total 359 5 44 53 56 18 1/0 204 237 1 416 1 554 1 996 2024 608 6 248 2 154 2 175 663 6 783 18 10 104 36 106_ 59 1 06 59 I 198 226 1 386 1 942 97 60 1 989 22 347 19 545 184 6 060 before the chloromphenical treatment began, but no differences in isolation rates were noted. Criteria for diagnosis The following criteria were adopted for diagnosing typhoid fever: (1) a blood culture positive for S. typhi; (2) a stool culture positive for S. typhi in a clini- cally diagnosed case, with or without a positive blood culture. These two criteria were considered when a case of typhoid was diagnosed for the purpose of evaluating the effectiveness of the vaccine. Patients with a clinical diagnosis of typhoid and high or rising titres (1: 400 and above) of 0 and H antigens, but in whom S. typhi could not be isolated, were considered as presumptive cases and are pre- sented separately in the tables. made from blood cultures only while 11 were made from both blood and stool cultures. Positive stool cultLures alone were found in only 3 clinically diag- nosed cases, which also showed rising Widal titres for 0 and H antigens (Table 3). Presumptive cases An additional 82 patients had illness suggestive of typhoid fever and had either high (1:400 and above) or rising titres for 0 and H antigens on serological examination. In these patients, no organisms could be isolated in culture; they were considered as pre- sumptive cases of typhoid and were analysed separa- Table 3. Distribution of confirmed and presumptive cases of typhoid fever among the participants according to the number of vaccine doses Diagnosis group No. Control Vaccine Totalof doses group group RESULTS Over 70 000 episodes of sickness were recorded among the children in the trial during the period 10 June 1968 to 31 August 1969, and 537 patients were investigated bacteriologically and serologically for typhoid fever. Clinical follow-up of these patients re- vealed that only 283 had clinical features suggestive of this condition. Of these 283 subjects, 244 had taken one or more doses of vaccine or placebo. Culture-positive patients S. typhi were isolated from 110 of the subjects who participated in the trial; 96 of these isolations were bacteriologically confirmed cases 1 1 2 3 3 60 total serologically and clinically diagnosed (presum ptive) cases 64 2 0 44 46 1 1 2 3 2 0 3 3 3 36 40 76 total 37 45 82 l~I 1526 2 107 2 146 586 6 591 3 104 110 447 C. S. CHUTTANI AND OTHERS tely (Table 3). These presumptive cases were not in- cluded in the analysis of the effectiveness of the vaccine. The number of culture-positive patients among the participants who were given only 1 or 2 doses of vaccine or placebo was small and these subjects were excluded from further statistical analysis. Effectiveness of the vaccine Evaluation of the vaccine, based on culture-posi- tive patients with typhoid fever only (Table 4), re- vealed that 60 cases of typhoid fever occurred in the control group and 44 in the vaccine group during the period of study; that is, an incidence of 9.6 per thousand in the control and 7.3 per thousand in the vaccine group. The difference is not statistically sig- nificant (P>0.17). Table 4. Incidence of typhoid in the vaccinated and control groups from 10 June 1968 to 31 August 1969 differences in the number of bacteriologically con- firmed typhoid patients in the two groups at the end of first 6 months (22 in the control and 12 in the vaccine group) and at the end of the second 6 months (26 in the control and 14 in the vaccine group) are not significant (P>0.10, P>0.05, respectively; Table 5). A similar analysis of the total number ofconfirmed Table 5. Incidence of typhoid fever in the control and vaccinated groups analysed at 6-monthly intervals. the basis for diagnosis being bacteriologically positive blood or stool samples No. of days from vaccination to onset of fever 0-180 181-360 361 -448 Control group a Vaccinated group b Incidence No. IncidenceNo. pes NO. per Iof cases of casesthousand ,thousand 22 26 1 2 3.5 4.2 1.9 12 14 18 2.0 2.3 3.0 Control group a Diagnosis group No. Incidence of cases per Bacteriologically confirmed cases serologically and clinically diagnosed (presumptive) cases total Vaccinated group b No. Incidence of cases per thousand 60 9.6 44 ! 36 96 1 5.8 5.4 40 84 a 6 248 participants. b 6 060 participants. and presumptive cases (Table 6) also showed no significant difference in incidence in the vaccinated 7.3 and control groups. Detailed analyses based on age-specific incidence rates were made to determine whether there was any significant difference in the development of immunity 6.6 at various ages (Table 7). It was observed that the incidence of typhoid fever in the 1-5 years and 13.9 " 6 248 participants. b 6 060 participants. Similar analysis on presumptive cases of typhoid fever, i.e., clinically suggestive and serologically posi- tive but culturally negative, revealed incidences of 5.8 and 6.6 per thousand, respectively, in control and vaccine groups. This difference is also not statisti- cally significant. Combining the confirmed and pre- sumptive cases, the respective incidences are 15.4 and 13.9 per thousand, and the difference is not signifi- cant (Table 4). In order to determine whether any immunity oc- curred in the early stages of vaccination, the inci- dence of typhoid fever was compared in the control and vaccinated groups at 6-monthly intervals. The Table 6. Incidence of total typhoid fever (confirmed and presumptive cases) in the control and vaccinated groups analysed at 6-monthly intervals, the basis for diagnosis being bacteriologically positive blood or stool samples (confirmed cases) and serologically positive patients (presumptive cases) No. of days Control group a Vaccinated group b from vaccinationIniec to onset of No. p IncidenceNo fever of cases per of cases perfver ~~thousand thousand 0-180 43 6.9 29 4.8 181-360 37 5.9 28 4.6 361-448 16 2.6 27 4.5 a 6 248 participants. b 6 060 participants. 448 = EFFECTIVENESS OF ORAL KILLED TYPHOID VACCINE Table 7. Age-specific incidence of bacteriologically confirmed case of typhoid fever in the vaccine and control groups Control group Vaccine group No. No. Incidence No. vaccinated of cases Xper thousand vaccinated 204 1 416 1 996 2 024 608 6 248 1 1 6 25 4.9 11.3 12.5 14 6.9 4 6.6 60 9.6 198 1 386 1 942 1 989 545 6 060 No. Incidence of cases per thousand 0 11 17 14 2 44 0 7.9 8.8 7.0 3.7 7.3 6- 0 years age groups was 11.3 and 12.5 per thousand respectively, in the control group and 7.9 and 8.8 per thousand in the vaccine group. These differences are not significant at the 5% level (P>0.30, P>0.20, respectively). DISC USSION There was a high incidence of typhoid fever in the community tinder study. Cases of typhoid fever occurred all year round, but there was a significantly higher incidence during the period March-September (Table 8). The overall incidence of typhoid fever in participants in the trial (based on culture-positive cases only) after 448 days of the follow-up was 9.6 per thousand in the control group and 7.3 per thousand in the vaccine group. This difference is not statistically significant. Analysis of the data at 6-monthly intervals showed that the incidence of typhoid fever in the first 6 months was 3.5 and 2.0 per thousand in the control and vaccine groups, respectively and, 4.2 and 2.3 per thousand in the second 6 months of the trial. Again, the differences are not significant. During the third period of 6 months, the incidence was 1.9 and 3.0 per thousand in the control and the vaccine groups, res- pectively. The results of the trial indicate that oral killed vaccine is not effective against S. typhi infection in the dosage schedule employed. No complications or deaths due to typhoid occur- red in either of the groups, probably because the disease was detected in the very early stages and effectively treated. This was an ideal community in which to conduct a vaccination trial since the disease was highly endemic with an incidence rate Table 8. Monthly distribution of bacteriologically confirmed typhoid cases in the vaccine and control groups Period 1968 10-30 June 1-31 July 1-31 Aug. 1-30 Sept. 1-31 Oct. 1-30 Nov. 1-31 Dec. 1969 1-31 Jan. 1-28 Feb. 1-31 Mar. 1-30 Apr. 1-31 May 1-30 June 1-31 July 1-31 Aug. Control Vaccine group group 0 2 1 0 5 2 2 0 0 3 3 3 2 0 Total 0 5 13 4 2 3 o 0 0 1 2 12 6 11 4 9 5 11 2 5 13 21 449 Age group (years) < 1 1-5 6-10 11-1 5 16-1 7 total I_ 450 C. S. CHUTTANI AND OTHERS based on culture-positive cases of 9.6 per thousand in children below the age of 17 years. Although this study indicated that the oral killed vaccine was not particularly effective, a further trial is now being conducted in another community with a similar vaccine containing a greater number of bacteria per dose. The results of the second trial should soon be known. ACKNOWLEDGEMENTS The authors are grateful to Professor P. N. Wahi, Director-General, Indian Council of Medical Research, for his valuable suggestions and for permission to publish this report. Thanks are due also to Colonel B. L. Taneja for his generous help in the planning and organization of this trial, to the staff of the World Health Organization for their keen interest and help during the project, and to Dr N. G. S. Raghavan, Dr H. K. Chuttani, Dr Taskar, Dr P. Gupta for their valuable help. The authors are extremely grateful to Mr. C. S. Pande, Miss Promila Bhalla, and other members of the field staff, without whose devoted participation the study would not have been possible. They also thank the authorities of the Municipal Corporation of Delhi for giving permission to conduct the trial and for providing the necessary faci- lities. RESUME EFFICACITE D'UN VACCIN ANTITYPHOYDIQUE TUE, ADMINISTRE PAR VOIE BUCCALE On a organise un essai contr6le a Delhi (Inde) afin d'evaluer le pouvoir protecteur d'un vaccin antityphoi- dique tue administre par voie buccale. Au total, 13 374 enfants de moins de 17 ans y ont participe. Trois doses d'un vaccin renfermant 100 x 109 Salmo- iiella typhi par dose ont e donnees en 3 jours, par voie buccale, a 6060 enfants; 6248 autres (groupe temoin) ont requ 3 doses de placebo; pour diverses raisons, 1066 enfants n'ont requ qu'une ou deux doses de l'une ou l'autre preparation. La vaccination, commencee en juin 1968, a pris fin en aofit 1968 et chaque participant a e observe reguliere- ment, tous les 4 jours, jusqu'au 31 aofit 1969. Les enfants fievreux ont e examines par un medecin; en cas de fievre persistante (plus de 3-4 jours), on a pratique une hemo- culture, un examen serologique et une coproculture. Seuls les cas de fi6vre typhoide confirmes bacteriologique- ment ont e retenus pour l'analyse des resultats. Des contr6les bacteriologiques et serologiques ont e effectues chez 537 enfants fievreux dont 283 presentaient des sympt6mes evoquant la [fi&vre typhoide. On a isole S. typhi chez 110 malades, 96 fois par hemoculture seule, 11 fois par hemoculture et par coproculture, et 3 fois uniquement par coproculture. Sur ces 110 typhiques, 44 avaient requ les 3 doses de vaccin et 60 les 3 doses de placebo; l'incidence dans le ler groupe etait de 7,3 pour 1000 et dans le second de 9,6 pour 1000, la difference etant statistiquement non significative (P>0,17). Afin de s'assurer de l'existence eventuelle d'une immu- nite postvaccinale precoce, on a compare l'incidence dans les deux groupes a 6 mois d'intervalle. On n'a releve de differences significatives ni au 6e mois (P>0,10) ni au 12e mois suivant la vaccination (P>0,05). L'incidence relative aux groupes d'age 1-5 ans et 6-10 ans a atteint 11,3 et 12,5 pour 1000 dans le groupe temoin et 7,9 et 8,8 pour 1000 dans le groupe vaccine, les differences n'etant pas significatives (P>0,30 et P>0,20 respecti- vement). Ces resultats indiquent que le vaccin antityphoidique tue administre par voie buccale aux doses utilisees dans le present essai ne confere aucune protection contre la fievre typhoide. REFERENCES Cvjetanovic, B. & Uemura K. (1965) Bull. Wid Hlth Org., 32, 29-36 Hejfec, L. B., Salmin L. V., Lejtman, M. Z., Kuz'minova, M. L., Vasil'ova, A. V., Levina L. A., Bencianova, T. G., Pavlova, E. A. & Antonova A. A. (1966) Bull. Wld Hith Org., 34, 321-339 Polish Typhoid Committee (1966) Bull. Wld Hlth Org., 34, 211 Typhoid Panel, UK Department of Technical Co-operation (1964) Bull. WId Hlth Org., 30, 631-634 Yugoslav Typhoid Commission (1962) Bull. WldHlth Org., 26, 357-369 Yugoslav Typhoid Commission (1964) Bull. WldHlth Org., 30, 623