Serological studies on cholera patients and their household contacts in Calcutta in 1968*

Bull. Org. mond. Sante 1970 43 389-399 Bull. Wld Hlth Org. J Serological Studies on Cholera Patients and their Household Contacts in Calcutta in 1968 * JOINT ICMR-GWB-WHO CHOLERA STUDY GROUP,1 CALCUTTA, INDIA Vibriocidal and agglutination tests have been performed, using a microtechnique, on 170 pairs ofsera obtained, at intervals of13-26 days, from bacteriologically proven cholera patients and their contacts, carriers and vibrio-negative contacts. Of the carriers, 44 %-46 % of those with low initial vibriocidin titres (< 1: 80) and 28 %-37% of those having high initial titres (< 1 :160) showed a 4-fold or greater rise in vibriocidal titres. Carriers and negative contacts exhibited almost similar pictures. With an increase in the number ofcarriers per household, a larger number ofnegative contacts developed significant titres in their second samples. In general, initial titres increased with age, but were highest in the 10-25-years age-group: however, 30 % of children below 10 years of age had titres > 1 : 640. The results indicated that individuals with high titres might become carriers but may not suffer from overt cholera. The investigation also showed that retrospective diagnosis of cholera infection in a highly endemic area should not depend on serology alone. Immuno-epidemiological patterns in a cholera endemic area are of great importance in the elucida- tion of the epidemiology of cholera. A serological study has, therefore, been carried out simultaneously with the cholera carrier studies in Calcutta since 1966 (Sinha et al., 1967, 1968). Blood samples were collected from cases, carriers and their Vibrio cholerae negative contacts in the community and agglutination, vibriocidal and antitoxin titres were determined. Results of the study during 1966-67 will be published in a separate communication. * Requests for reprints should be addressed to Z. Bencic, " A. Stampar " School of Public Health, Rocke- fellerove ul. 4, Zagreb, Yugoslavia. 1 This is a joint project under the auspices of the Indian Council of Medical Research (represented by the Cholera Research Centre), the Health Department of West Bengal Government and the World Health Organization. These groups were represented by the workers listed below: Indian Council of Medical Research, Cholera Research Centre: B. C. Deb, Research Officer, and D. L. Shrivastava, Emeritus Scientist, ex-Director, Cholera Research Centre. Health Department of West Bengal Government: R. Sinha, Medical Officer. Present address: Cholera Research Centre, Calcutta-16, India. World Health Organization: B. A. Freeman, Immunologist, WHO Consultant: present address-The University of Tennessee Medical Units, Memphis, Tennessee 38103, USA; and Z. Bencic, Epi- demiologist, WHO Consultant: present address-" A. Stampar" School of Public Health, Medical Faculty, Zagreb, Yugoslavia. It was observed during the earlier part of the cholera carrier studies in Calcutta (Sinha et al., 1967, 1968) that, in spite of the abundance of infection in the community, clinical cases of cholera were not very common. Accordingly an intensive investiga- tion was carried out during 1968 to determine the incidence and persistence of infection in households in which cases of cholera occurred. Serological studies were made at the same time as the epidemio- logical studies (Joint ICMR-GWB-WHO Cholera Study Group, Calcutta, India, 1970a, 1970b) to determine the immunological behaviour of the members of the households. Results of agglutination and vibriocidal tests on these sera are discussed in the present communication. MATERIALS AND METHODS When a cholera patient was admitted to the Infectious Diseases Hospital, Calcutta, and found to be Vibrio cholerae positive, blood samples were immediately collected from the other members of his household. Samples of blood were also collected from the patients within 2-5 days of their admission, except in the cases of two patients whose first serum samples were collected 8 and 13 days after their admission. 2566 - 389 390 JOINT ICMR-GWB-WHO CHOLERA STUDY GROUP TABLE I DISTRIBUTION OF PERSONS FROM WHOM SERA WERE COLLECTED BY AGE, SEX AND INFECTION STATUS _ _ Cases _ Carriers _ Negative contacts Age- No o o No. of | No ofgroup Sex No. of N o fN.o(years) persons serarNo.s paiedso.nfs areyearsbperc era coverage| bledrs sera % persons sera coveragebld tested bld tested bld tested <10 Male 3 2 66.7 9 6 66.7 14 10 71.4 Female I 1 100.0 10 7 70.0 9 7 77.8 Male 2 1 50.0 18 16 88.9 24 16 66.7 10-25 Female 1 - 0.0 13 10 76.9 47 32 68.0 Male 2 2 100.0 19 10 52.6 38 22 57.9 >25 Female 8 5 62.5 16 10 62.5 17 13 76.5 The categories of persons from whom blood was collected included (a) cases (index cholera patients), (b) carriers-members of the households from whom cholera vibrios were isolated on one or more occasions during the study but who did not show clinical symptoms of cholera and (c) negative con- tacts-members of the household from whose stools V. cholerae was not isolated despite daily sampling for a minimum period of 10 days after the onset of the index case. Some of them were sampled for a longer period but remained vibrio negative. Out of a total of 251 persons belonging to the different categories mentioned above, from whom sera were obtained, 170 were bled for a second time after an interval of about 13-26 days, except for one case and one carrier who were bled 7 and 37 days, respectively, after their initial bleedings. Only the paired sera obtained from these 170 persons are described in the present communication. The number of sera obtained from cases was, of course, very small. Venous blood was obtained in each case with the help of Vacutainers. After separation, sera were filtered through Millipore membrane filters (0.45 ,u) and aliquots (about 0.5 ml) were preserved in sealed ampoules at - 20°C. Vibriocidal tests on the sera were carried out in duplicate using the microtitration method (Benenson et al., 1968a) as this method was found advantageous in handling large numbers of sera, in requiring smaller amounts (0.05 ml) and because of its con- venience (Deb et al., 1969). Anti-cholera serum, raised in rabbits and of known vibriocidal titre, was included in each experiment as a positive control. This antiserum was prepared by hyper-immunizing rabbits with a live whole-cell suspension of V. cho- lerae, Inaba ASM Kuki. Commercial lyophilized guinea-pig complement was obtained from Messrs Markham Laboratories, Chicago, Ill., USA and was reconstituted with normal saline during the experi- ment. Complement, bacterial and sterility controls were carried out each time. Vibriocidin titres were expressed as the reciprocals of the highest dilutions of serum which lysed the test organisms. The agglutination test was also performed by the microtechnique described by Benenson et al. (1968b), with the following modifications. To pro- vide the antigen in the current study, an 18-hour surface culture on heart-infusion-agar slants was harvested in normal saline, and adjusted to a reading of 30 on the scale of the Klett-Summerson colorimeter 1 using green filter No. 54. Sera were not inactivated before the experiments. For both tests two V. cholerae strains,2 Ogawa 465 and Inaba 22463, were used. RESULTS The age- and sex-distributions of the different categories ofpersons bled once and those from whom paired sera were collected are given in Table 1. Of the 170 sera investigated in the current study, 85 belonged to males and 85 to females and the over-all coverage in the second bleeding was 67.7%. 1 Klett Manufacturing Co., New York, N.Y., USA. ' Obtained from C.R.L., Dacca, through the courtesy of Dr Y. Zinnaka. SEROLOGICAL STUDIES ON CHOLERA PATIENTS AND HOUSEHOLD CONTACTS IN CALCUTFA Replicate vibriocidal tests were performed, using the anti-cholera rabbit serum to establish the standard deviation of the microtitration method. Results of these replicate determinations are given in Table 2. A total of 66 tests was performed with each serotype. The standard deviation in the 132 tests, including both serotypes, is slightly greater than 1 cup dilution (1.09), and this indicates that the method compares favourably with most other serolo- gical tests in which serial dilutions of serum are used for end-point determinations. An analysis of these data also indicates that, with the immune rabbit serum employed, there is no significant difference between the vibriocidin titres against the two serotypes of V. cholerae. On the other hand the data shown in Table 3 may indicate that under natural conditions there might be preferential stimulation of antibody towards a single serotype. While most of the observed Ogawa/ Inaba vibriocidin titres are within plus or minus 1 dilution cup, in carriers and in negative contacts there is a small number of sera that show greater differences in titre against these two serotypes. There was a high observed correlation between the Ogawa and Inaba titres of either the first or the second sam- ples of the different categories of individuals, except for the first samples from cases. Almost similar results were obtained with the agglutination tests. TABLE 2 STANDARD DEVIATIONS OF REPLICATE VIBRIOCIDAL TESTS USING A SINGLE ANTI-CHOLERA RABBIT SERUM Number MaTest organism of titre a S.D. t 5 % ftests V. cholerae, Ogawa 66 9.57 1.20 V. cholerae, Inaba 66 9.85 0.94 0.471 1.96 Total or mean 132 9.71 1.09 a Dilution cup number showing complete lysis of the test organism (two-fold dilutions starting from 1 :20 in the first cup). The distribution of persons in the three categories according to the vibriocidal titres of their paired sera is shown in Table 4. For the purposes of this com- parison the higher of the titres against Ogawa or Inaba has been used. A good correlation may be observed between the titres of the first and second samples from the hospital cases, but for the carriers and negative contacts the observed correlation is low. A similar presentation of the distribution accord- ing to the agglutinin titres of the paired sera of the three groups can be seen in Table 5. The figures in the table indicate that the carriers showed a relatively TABLE 3 RATIO OF OGAWA/INABA VIBRIOCIDIN TITRES IN CASES, CARRIERS AND NEGATIVE CONTACTS: PERCENTAGE DISTRIBUTION a |Serum 1 Ratio: Ogawa/Inaba titres bCategory !collection <0.0625 0.125 0.25 0.5-2.0c 4.0 8.0 >16.0 Cases 1st - 82 (9) _ 18 (2) 2nd - 9 (1) 91 (10) - - 1st - 2 (1) 8 (5) 83 (49) 5 (3) - 2 (1) Carriers 2nd 2 (1) 2 (1) - 92 (55) - 2 (1) 2 (1) Negative 1st 1 (1) 1 (1) 1 (1) 83 (83) 6 (6) 3 (3) 5 (5) contacts 2nd 1 (1) - 3 (3) 87 (87) 6 (6) - 3 (3) a Percentage of total in each subject category for each serum collection. b The values ofthe correlation coefficients (r) between the Ogawa and Inabatitres of the sera: (1) Cases - 1st sample - 0.04 2nd sample - 0.73 (2) Carriers - 1st sample - 0.91 2nd sample - 0.95 (3) Negative - 1st sample - 0.47 contacts 2nd sample - 0.73 c The 0.5-2.0 dilution tube range is considered to be within the experimental error. 391 JOINT ICMR-GWB-WHO CHOLERA STUDY GROUP good response in their second samples compared with the negative contacts. Table 6 shows the vibriocidal response of the paired sera of the cases, the two categories of carriers detected before or after their first bleeding and the negative contacts, on the basis of their low or high initial titres. Since a 2-fold rise or fall in titre by the microtitration method may be considered to be within the experimental error, only 4-fold rises or falls were considered as significant changes in titre. Of the 2 cases having initial titres of 1: 80 or less, one showed a 4-fold rise in the second sample and the other did not show any change, but neither showed a fall in titre. Among the 9 cases having initial titres of 1: 160 or greater, none showed a rise in titre; in 4 cases the titre fell and in 5 others there was no change. It may be expected that carriers detected before the first bleeding will already have raised titres in their sera as compared with those detected after the first bleeding. The latter were presumably nega- tive contacts during their first bleeding. It is in the case of this second group that a more pronounced TABLE 4 DISTRIBUTION OF VIBRIOCIDAL TITRES OF PAIRED SERA FROM CASES, CARRIERS AND NEGATIVE CONTACTS Cases Carriers Negativecontacts Titre 1st 2nd 1st 2nd 1st 2nd Sample Sample Sample Sample Sample Sample 0 1 5 19 18 29 28 20 1 - - - I I 40 - - 1 1 7 3 80 - - 2 3 9 3 160 2 1 4 5 13 11 320 1 3 6 2 7 15 640 5 1 9 9 12 10 1 280 1 1 4 6 9 10 2 560 - - 4 7 9 9 5120 - - 5 3 3 4 10240 - - 2 2 - 4 20 480 - - 3 2 1 2 40960 - - - 1 - - Total 11 11 59 [ 59 100 [ 100 TABLE 5 DISTRIBUTION OF AGGLUTININ TITRES OF PAIRED SERA FROM CASES, CARRIERS AND NEGATIVE CONTACTS Case Carriers Negativecontacts Titre 1st 2nd 1st 2nd 1st 2nd Sample Sample Sample Sample Sample Sample 0 6 3 16 4 31 23 20 3 1 14 9 15 17 40 1 3 6 9 16 23 80 - - 6 9 15 5 160 - 4 7 5 7 11 320 - - 5 8 6 6 640 - - 2 6 5 8 1 280 1 - 2 3 3 4 2560 - - - 5 2 1 5120 - - 1 1 - I 10240 - - - - - - 20 480 _- - - - - 40960 _- - - - - Total f 11 ] 11 59 59 100 100 TABLE 6 PERCENTAGE DISTRIBUTION OF THE VIBRIOCIDAL RESPONSES OF PAIRED SERA FROM CASES, TWO CATEGORIES OF CARRIERS AND NEGATIVE CONTACTS HAVING LOW OR HIGH INITIAL TITRES a Carriers Vibriocidal Cases Detected Detected 'Negative response before after contacts 1st bleeding 1st bleeding Persons with Initial titre of < 1 : 80 i 50 (1) 46 (6) 44 (4) 46 (6) 56 (5) 8(1) - Persons with initial titre of > 1 :160 > 4-fold rise - 37 (7) 28 (5) 24 (13) None 56 (5) 16 (3) 33 (6) 43 (23) > 4-fold fall 44 (4) 47 (9) 39 (7) 33 (18) a The figures in parentheses indicate number of sera examined. 392 SEROLOGICAL STUDIES ON CHOLERA PATIENTS AND HOUSEHOLD CONTACTS IN CALCUTTA FIG. 1 INITIAL VIBRIOCIDAL ANTIBODY PATTERNS OF BOTH CARRIERS AND NEGATIVE CONTACTS 100 CARRIERS 90 80 70 60. IINON-IWIUNE IL 39X IWIUNE a5 611 a!40 30 fl 20 - 10 f l _ 10 c 20 20 40 80 160 320 640 1280 2560 5120 1024020480 Vibriocidal titre rise of titre would be expected in their second sam- ples. But from the data available in Table 6 it ap- pears that there is no significant difference between the two categories of carriers, nor between carriers and negative contacts. Persons having low initial titres (< 1: 80) show a greater tendency towards rise of titre than those with high initial titres (> 1: 160) who in contrast exhibited a higher percentage of titre reductions. The over-all picture in the table suggests that the initial serological patterns in a highly endemic area such as Calcutta are essentially the same in negative contacts and carriers. Mosley et al. (1968) observed that Vibrio cholerae infection, whether symptomatic or asymptomatic, predomi- nantly occurred in individuals with low titres. Fig. 1 shows the initial vibriocidal antibody patterns of both the carriers and the negative con- tacts. It can be seen that there is bardly any difference in the initial antibody levels of the two groups. It is difficult to explain why only the carriers showed subsequent infection when both groups had an almost similar number of susceptible non-immunes. Table 7 shows the agglutination response of paired sera similar to those presented in Table 6. Of the 11 cases, 10 had low initial agglutinin titres and 40% of this group showed a 4-fold or greater rise in titre. Both categories of carriers in the lower initial titre group showed considerable rise in titre, but it is interesting to note that in the group with higher NEGATIVE CONTACTS IMMUNE 58% Vibriocidal titre initial titres (> 1: 80) the carriers detected before their first bleeding showed fewer rises (7 Y.) of titre in their second sample than did those detected after their first bleeding (56°/). Negative contacts also TABLE 7 PERCENTAGE DISTRIBUTION OF THE AGGLUTININ RESPONSE OF PAIRED SERA FROM CASES, TWO CATEGORIES OF CARRIERS AND NEGATIVE CONTACTS HAVING LOW OR HIGH INITIAL TITRES a Carriers Agglutination Cases Detected Detected Negative response before after contacts 1st bleeding 1st bleeding Persons with initial titre of C 1: 40 > 4-fold rise 40 (4) 56 (10) 44 (8) 34 (21) None 60 (6) 44 (8) 56 (10) 61 (38) 4-fold fall - - - 5 (3) Persons with initial titre of > 1 : 80 > 4-fold rise - 7 (1) 56 (5) 13 (5) None 86 (12) 33 (3) 47 (18) > 4-fold fall 100 (1) 7 (1) 11 (1) 40 (15) a The figures in parentheses indicate number of sera examined. 393 JOINT ICMR-GWB-WHO CHOLERA STUDY GROUP FIG. 2 VIBRIOCIDAL ANTIBODY PATTERN BY AGE-GROUP (A) CARRIERS & NEGATIVE CONTACTS (1st SAMPLE) c u8 bU-jzu 64U-Z bbU '5 'ZO" 80 WHO OC 7-- Vibriocidal titre (B) CARRIERS (2nd SAMPLE) 160-320 640-2 560 ' 5 120 Viri;rlliro (C) NEGATIVE CONTACTS (2nd SAMPLE) - <10 years of age __-- 10-25 years of age >25 years of age exhibited relatively fewer rises and more falls of titre compared with carriers, both in the low and high initial titre groups. As the level of vibriocidins in contacts may be related to the probability of contact with the vibrios, an attempt was made to show in Table 8 the titres in each category when grouped according to the number of established carriers within the household. In the first group of households with no carriers, it can be seen that at the time of the first bleeding, 81 % of the household contacts had no significant antibody titres. As the number of carriers within the house- hold increased, a definite shift towards a significant antibody level can be observed. In households having 1-4 carriers, the percentage of contacts with insignificant antibody levels was reduced to 48% and in households having 5 or more carriers, this figure was further reduced to 29%. The carriers in these households also showed a similar tendency to raised antibody levels in their sera with a rise in the number of carriers per household. These findings also indicate that there is not much difference in titres between carriers and negative contacts in a highly endemic situation, where there is the possibility that individuals of the latter group may become infected any time prior to the second bleeding. There is ample evidence of a rise in titre in individuals along with a rise in the number of infected persons in the household. Moreover, in the group of houses in which no carrier could be detected, none showed a titre higher than 1: 2560, whereas in the other groups of houses with carriers there were many individuals with titres beyond that level. This may point towards frequent circulation of vibrios among the persons in houses with multiple infections. The vibriocidal antibody pattern according to age-group is shown in Fig. 2, the higher of the titres against either Ogawa or Inaba having been used. The initial titres of carriers and negative contacts of the three age-groups (<10 years, 10-25 years and >25 years) are shown together in the first part (A) of the figure. During their initial bleeding these carriers were actually negative contacts as they were- detected after their first bleedings. The titres in the second samples from carriers and negative contacts, by age-group, are shown in the second and third parts (B and C) of the figure, respectively. It appears from the figure (part A) that 42% of the children below 10 years of age had low titres (<1: 80) in the first sample but that higher titres were also present in the group. In the second part of the figure (B), however, the carriers in this age-group (<10 years) showed a peak (64%) in the 1: 640-1 : 2560 titre region, clearly indicating that this age-group re- sponded better to the antigenic stimulus. The third part of the figure (C) shows that among the negative contacts below 10 years of age the pattern is almost similar to that of their initial titre except in the 70 - . 60K .- 50K -'- 40X v) -E 30 -4=, 20L XL 100 _ V1U11U%,1UQ1 LILIV. VIUIIV%.IUUI LILIIU 394 SEROLOGICAL STUDIES ON CHOLERA PATIENTS AND HOUSEHOLD CONTACTS IN CALCTUTTA TABLE 8 PERCENTAGE OF INDIVIDUALS SHOWING DIFFERENT VIBRIOCIDIN TITRES AS INFLUENCED BY NUMBER OF CARRIERS IN THEIR RESPECTIVE HOUSEHOLDS Number Percentage of individuals having of carriers Subject Serum vibriocidin titres a per category collection - household >80 160-320 640-2560 >5120 1st - - - - Cases 2nd Ist None Carriers 2nd - - - - Negative 1st 81(17) 10 (2) 10 (2) - contacts 2nd 33 (7) 43 (9) 24 (5) - 1st - 33 (2) 67 (4) - Cases 2nd 17 (1) 50 (3) 33 (2) - 1st 43 (6) 14 (2) 43 (6) - 1-4 Carriers 2nd 14 (2) 14 (2) 43 (6) 29 (4) Negative 1st 48 (15) 23 (7) 23 (7) 6 (2) contacts 2nd 42 (13) 23 (7) 19 (6) 16 (5) 1st 40 (2) 20 (1) 40 (2) - Cases 2nd 80 (4) 20 (1) - - 1st 36 (16) 18 (8) 24 (11) 22 (10) >5 Carriers 2nd 44 (20) 9 (4) 38 (17) 9 (4) Negative 1st 29 (14) 25 (12) 42 (20) 4 (2) contacts 2nd 29 (14) 23 (11) 38 (18) 10 (5) a The higher of thetitres against either Ogawa or Inaba has been used and the values are shown as reciprocals. The figures in parentheses indicate the number of sera examined. 1: 640-1: 2560 titre region where a slight increase is apparent: this probably indicates that they were not altogether free from infection. Antibody response in individuals over 25 years of age is not well marked compared with that of the individuals between 10-25 years of age. Fig. 3 gives the vibriocidal response of carriers and negative contacts in their second samples by age- groups depending on their low (<1: 80) or high (>1: 160) initial antibody titre. It is clear from the figure that there is a difference in the pattern of vibriocidal response amongst those having low or high initial titres. None of those having a low initial titre showed a rise of titre to 1: 5120 or higher, as did those with high initial titre. Carriers and negative contacts who had high initial titres showed an almost similar picture and thereby indicated that the negative contacts were also infected at some stage: this infection was missed or it took place before the period of follow-up. The distribution of vibriocidin titres by sex is presented in Table 9. No significant influence of sex on the vibriocidin titres of carriers and negative contacts was observed. 4 395 JOINT ICMR-GWB-WHO CHOLERA STUDY GROUP FIG. 3 VIBRIOCIDAL RESPONSE BY AGE-GROUP ACCORDING TO LEVEL OF INITIAL TITRE CARRIERS INITIAL TITRE '1:80 ' 5 120 ' 80 160-320 640-2 560 >5 120 NEGATIVE CONTACTS F INITIAL TITRE >1:160 10 0 ' 80 160-320 640-2 Vibriocidal titre 60-320 640-2 560 >5 120 Vibriocidal titre WO 0C7 b <10 years of age ---- 10-25 years of age -- >25 years of age The temporal aspects of the vibriocidin titres are shown in Table 10 for the entire study population including those from whom only a single serum sample was taken. Similar tabulations for the carriers and negative contacts did not demonstrate appreciable differences from the total population and are not, therefore, shown here. Inspection of this table reveals that there is a definite seasonal variation in the antibody titres of the population. In general, an increase in titre coincides with the peak of the cholera season. Titres were generally low in April, began to increase in early May and reached peak in mid- to late-June. Although the population sample was small during July, there appears to have been a definite fall in titre level coincident with the decline of clinical disease in the population. DISCUSSION Recent reports from East Pakistan (Mosley et al., 1968, 1969) indicate that 77% of individuals with inapparent infection showed a 4-fold or greater rise in vibriocidal titres in their second blood samples 80 70 60 50C400 W=, 40 XL 30 20 10 a) c0 396 SEROLOGICAL STUDIES ON CHOLERA PATIENTS AND HOUSEHOLD CONTACTS IN CALCUTITA TABLE 9 PERCENTAGE DISTRIBUTION OF VIBRIOCIDIN TITRES ACCORDING TO SEX _rumVibriocidin titre aCategory Sex collection 080 16320 640-2560 >5120 1st 40 (2) 20 (1) 40 (2) - Male 2nd 60 (3) 40 (2) - - Cases _ 1st - 33 (2) 67 (4) - Female 2nd 33 (2) 33 (2) 33 (2) - 1st 38 (12) 16 (5) 25 (8) 22 (7) Male 2nd 41 (13) 13 (4) 38 (12) 9 (3) Carriers 1st 37 (10) 19 (5) 33 (9) 11 (3) Female 2nd 33 (9) 11 (3) 37 (10) 19 (5) 1st 54 (26) 17 (8) 25 (12) 4 (2) Male 2nd 40 (19) 21 (10) 33 (16) 6 (3) Negative contacts 1 st 38 20) 23 (12) 35 (18) 4 (2) Female 2nd 31 (16) 31 (16) 27 (14) 12 (6) a The higher of thetitres against either Ogawa or Inaba has been used and the values are shown as reciprocals. The figures In parentheses indicate the number of sera examined. TABLE 10 PERCENTAGE DISTRIBUTION OF RELATIVE VIBRIOCIDIN TITRES IN THE TOTAL STUDY POPULATION BY DATE OF COLLECTION Sera collected ________ - Vibriocidin titres a between < 80 160-320 640-2560 > 5120 3-17 April 83 (25) 7 (2) 10 (3) - 18 April-2 May 40 (37) 28 (26) 24 (22) 8 (7) 3-17 May 30 (26) 20 (17) 36 (31) 15 (13) 18 May-I June 38 (33) 16 (14) 40 (35) 6 (5) 2-16 June 33 (17) 26 (13) 33 (17) 8 (4) 17 June-1 July 18 (8) 13 (6) 47 (21) 22 (10) 2-14July 67(18) 26 (7) 4 (1) 4 (1) a The higher of the titres against either Ogawa or Inaba hasbeen used and the values are shown as reciprocals. The figures in parentheses indicate the number of sera examined. obtained after an interval of 10 days. They also found that initial vibriocidal antibody titres in blood increased with increasing age, and that the infection, symptomatic or asymptomatic, occurred in individu- als with low initial titres. The present serological study in the highly endemic- area of Calcutta, has, however, indicated a different picture. It has been found that only 44%-46% of carriers with low initial vibriocidin titres (.1: 80) and 28 %-37 % of those with high initial titres (>1: 160) showed a 4-fold or greater rise of vibrio- cidin titre in their second serum specimens taken after an interval of 13-26 days. Negative contactsS showed rises in antibody titre almost similar to those of the carriers (Table 6). This study has also shown that the vibriocidal antibody patterns of both carriers and negative contacts are almost the same (Fig. 1). Although 61 % of the carriers had high titres (>1: 160) during their initial bleedings they were later found to be infected with V. cholerae, but none developed clinical symptoms of cholera. It appears from the above findings that an individual with high vibriocidin titre might become a carrier but 397 JOINT ICMR-GWB-WHO CHOLERA STUDY GROUP may not suffer from an attack of clinical cholera. In this connection it is interesting to recall the observa- tion of McCormack et al. (1969) that 13 of their 58 admitted cholera patients had vibriocidal titres above 1: 40. Vibriocidin titres were found to increase with age (Fig. 2, part A) the rise being greater at the titre level of 1: 640-1: 2560. A slightly greater effect was seen in the 10-25-years age-group than in persons over 25 years of age. Although 46% of the children below 10 years of age showed low titres (.1: 80) in their initial blood samples, 30% of them had titres above 1: 640. This may have been due to the high degree of infection in Calcutta and the frequent circulation of vibrios amongst children. It may be mentioned here that during the cholera carrier studies in Calcutta in 1968, two newborn (twin) babies were found infected with V. cholerae within 1 month of their discharge from the maternity hospital; their parents were established carriers, the father being a long-term carrier. This shows that children among the shanty dwellers in Calcutta are exposed to cholera infection very early in life, obviously due to the presence of heavy infection in their households. That the infection is circulating freely in the communities is further illustrated by the fact that with increase in the number of carriers within the households, a larger number of negative contacts developed significant antibody titres in their second blood samples (Table 8). This also suggests that the negative contacts had at some stage been infected with V. cholerae, which could not be isolated. It is rather interesting to note that the agglutination titres by the microtitration method on the other hand showed a more significant difference between the carriers and the negative contacts, the former show- ing better response than the latter. It was also observed that 10 out of 11 cases (Table 5) had low initial titres (<1 40) on admission which may have contributed towards their becoming cholera patients, if it is presumed that agglutinating antibodies have any protective action against cholera. No similar indications were, however, obtained from vibriocidal antibody titrations, although these are usually believed to be more sensitive and specific (Ganguly et al., 1969). In any case the nature of protective antibodies against cholera is still not clear (WHO Scientific Group on Cholera Immunology, 1969). The data available from the present study indicate that probably one cannot depend solely on sero- logical findings in a highly endemic situation like that in Calcutta, for retrospective diagnosis of cholera infection. The serological picture is almost similar in carriers and negative contacts, and this may suggest that bacteriological techniques available at present are not adequately sensitive to detect very low-grade infections. Mosley et al. (1968) also suggested that some isolations might be missed. Another possibility is that the negative contacts may have been infected with vibrios within the 10-days period after the first sample was taken. Annual anticholera inoculation may have also accounted for the higher titres amongst the negative contacts, but authentic information regarding this could not be obtained. On the basis of the above data it may be said that a more intensive study of the serological behaviour of a cross-section of the community at different times would probably throw more light on the problem of immunity against cholera. ACKNOWLEDGEMENTS The authors are grateful to Colonel B. L. Taneja, formerly Director-General, Indian Council of Medical Research, and to Dr C. L. Mukerjee, formerly Director of Health Services, Government of West Bengal, for their help. The authors had useful discussions with Dr Y. Watanabe of the Department of Microbiology, University of Texas Medical Branch, Galveston, Texas, USA, for which they are grateful. Thanks of the authors are also due to Mr S. N. Sikdar for statistical analysis, Mr S. Paul for technical assistance and Mr Biswanath De for drawing the original graphs. Finally, the authors wish to express their thanks to Dr A. Mondal and Dr P. M. Manji of the Infectious Diseases Hospital, Calcutta, and Dr G. C. Das of the Calcutta Corporation for their excellent co-operation during the studies. 398 SEROLOGICAL STUDIES ON CHOLERA PATIENTS AND HOUSEHOLD CONTACTS IN CALCUTTA 399 RE'SUMt ENQUETE SEROLOGIQUE DANS DES FAMILLES AlTEINTES PAR L'INFECTION CHOLtRIQUE A CALCUTTA EN 1968 Des investigations serologiques ont ete men6es dans des familles oiu etait survenu un cas de chol6ra avere. Elles ont port6 sur les malades eux-m8mes et sur les contacts familiaux qu'ils aient ete, A 1'examen bacterio- logique, reconnus porteurs de germes ou exempts de toute infection a Vibrio cholerae. Au total, 170 paires de serums preleves a intervalle de 13-26 jours ont ete exa- min6es, une microtechnique etant utilisee pour la re- cherche des agglutinines et des anticorps vibriocides. On a releve une plus forte proportion de hausses des titres d'anticorps chez les sujets faiblement positifs (titres <1: 80) que parmi ceux dont le serum reagissait fortement (titres >160) lors du premier examen. Chez 44%-46% des porteurs de germes A faibles titres d'anti- corps vibriocides et chez 28%-37% des porteurs pre- sentant des titres eleves, 1'examen du second echantillon de serum a montr6 une el6vation de quatre fois des titres. Une reponse quasi similaire a et6 observee chez les contacts bacteriologiquement n6gatifs. On a constate que lorsqu'un contact presentait a l'origine des titres elev6s, il pouvait par la suite devenir porteur de germes du cholera sans etre atteint d'une forme clinique de l'affection. L'enquete a egalement r6vele que l'augmen- tation du nombre des porteurs au sein des familles avait pour cons6quence d'accroitre la proportion des con- tacts bact6riologiquement n6gatifs dont le second 6chan- tillon de s6rum renfermait des anticorps a des titres eleves. Dans les premiers echantillons de serum, les titres d'anticorps vibriocides augmentaient en fonction de l'age des sujets; cependant, apr6s 25 ans, les titres 6leves etaient moins nombreux que dans le groupe d'age 10-25 ans. Des titres de 1: 640 ou plus ont ete trouv6s chez 30% des enfants ages de moins de 10 ans. Les auteurs estiment qu'il n'est pas indique de fonder un diagnostic r6trospectif de l'infection a V. cholerae dans une r6gion de forte end6micite sur les seules donnees serologiques. REFERENCES Benenson, A. S., Saad, A. & Paul, M. (1968a) Bull. Wld Hlth Org., 38, 267-276 Benenson, A. S., Saad, A. & Mosley, W. H. (1968b) Bull. Wld Hlth Org., 38, 277-285 Deb, B. C., Sinha, R. & Shrivastava, D. L. (1969) Indian J. med. Res., 57, 167-174 Ganguly, R., Deb, B. C. & Shrivastava, D. L. (1969) Indian J. med. Res., 57, 815-822 Joint ICMR-GWB-WHO Cholera Study Group, Cal- cutta, India (1970a) Bull. Wld Hlth Org., 43, 379- 387 Joint ICMR-GWB-WHO Cholera Study Group, Cal- cutta, India (1970b) Bull. Wld Hlth Org., 43, 407- 412 McCormack, W. M., Rahman, A. S. M. M., Chowdhury, A. K. M. A., Mosley, W. H. & Phillips, R. A. (1969) Bull. Wld Hlth Org., 40, 199-204 Mosley, W. H., Ahmad, S., Benenson, A. S. & Ahmed, A. (1968) Bull. Wld Hlth Org., 38, 777-785 Mosley, W. H., McCormack, W. M., Ahmed, A., Chowdhury, A. K. M. A. & Barui, R. K. (1969) Bull. Wld Hlth Org., 40, 187-197 Sinha, R., Deb, B. C., De, S. P., Abou-Gareeb, A. H, & Shrivastava, D. L. (1967) Bull. Wld Hlth Org., 37. 89-100 Sinha, R., Deb, B. C., De, S. P., Sircar, B. K., Abou- Gareeb, A. H. & Shrivastava, D. L. (1968) Indian J. med. Res., 56, 964-978