INFLUENCE OF MODERATE MALNUTRITI ON ON MORBIDITY AND ANTIBODY RESPONSE FOLLOWING VACCINATION WITH LIVE, ATTENUATED MEASLES VIRUS VACCINE’

D. N. McM~rray,~ S. A. Loomis,3 L. J. Casazza,4 and H. Rey5

Following vaccination against measles, a group of moderately malnourished Colombian children produced as much antibody and had the same degree of post-vaccination morbidity as did their well-nourished cohorts. It a@ears on the basis of these findings that live, attenuated measles vaccine is both safe and effective in moderately malnourished populations.

Introduction tions for vaccination and other disease management programs in developing parts The influence of malnutrition upon sus- of the world, since more than half the chil- ceptibility to infection has been well-docu- dren in these areas are suffering from some mented in man (I, 2). Hence increased degree of moderate malnutrition (II). morbidity and mortality observed in mal- Measles infection is known to cause nourished populations can be explained, at severe disease and death in badly malnour- least in part, by the detrimental effects of ished populations (12). Yet it has been malnutrition on various aspects of the shown repeatedly that vaccination with humoral (3, 4), cell-mediated (5, 6), and live, attenuated measles virus stimulates secretory (7, 8) immune responses. These antibodies in malnourished populations effects have been most clearly demonstrated (13, 14), and there is experimental evidence in children suffering from the more severe suggesting that post-vaccination morbidity forms of protein and energy malnutrition, in such populations is no more severe than although recent studies indicate that sub- in normal populations (14, 15). Due to the clinical malnutrition may also significantly severity of the natural infection in mal- impair a child’s ability to respond immuno- nourished children, however, physicians in logically to infectious agents (9, 20). The many parts of the world are reluctant to latter findings have far-reaching implica- vaccinate with the live, attenuated virus for fear of causing severe illness in vac-

lWork supported by the Tulane University-Uni- cinees. versidad de1 Valle International Center for Medical Many early vaccine studies failed to ade- Research (ICMR) and by a grant (AI-10050) from the quately document the vaccinee’s nutri- National Institute of Allergy and Infectious Diseases, tional status, and lack of long-term follow- National Institutes of Health, United States Public Health Service. Article also appearing in Spanish in up left some uncertainty about the ability the ISoletin de la Oficina Sanitaria Panamericana, of malnourished children to maintain an 1979. adequate anti-measles antibody titer during ZAssistant Professor, Departament of Medical Mi- crobiology and Immunology. Texas A and M Univer- a time when their nutritional status tiay be sity, College Station, Texas, U.S.A. deteriorating. SOffice of International Health, United States In an attempt to clarify these points, Department of Health, Education, and Welfare, Rockville, Maryland, U.S.A. humoral antibody response and post-vac- 4Department of Population and Nutrition Projects, cination morbidity were studied in groups World Bank, Washington, D.C.. U.S.A. 5Associate Professor, Department of Pediatrics, of normal and moderately malnourished Universidad de1 Valle, Cali,Colombia Colombian children subsequent to adminis-

52 McMurray et al. l MALNUTRITION AND MEASLES VACCINES 53 tration of live, attenuated measles virus employed. A close approximation has been vaccine. The children were followed for demonstrated between virus neutralization more than a year to determine the influence titers and HI titers in human and animal of chronic malnutrition upon antibody sera containing moderate to high levels of titers. antibody (18, 19). For 21 consecutive days following vaccine Test Subjects and Methods administration the children were visited at home by a nurse who noted the presence of A total of 74 children from a poor urban conjunctivitis, diarrhea, skin eruptions, or population in Cali, Colombia, were in- fever. Symptoms appearing before 5 days cluded in this study. All came from a post-vaccination were not considered re- homogeneous cultural and socioeconomic lated to the vaccination. Children showing background, and all lived in the same part these early symptoms were excluded from of the city. Thirty-nine children were the morbidity analysis. classified as normal and 35 as suffering For purposes of statistical analysis, chil- from some degree of malnutrition (26 from dren with grade II and grade III malnutri- grade I, 7 from grade II, and 2 from grade tion were grouped together. The analysis of III) on the basis of failure to gain weight variance was used to test for a significant appropriately for their age (16). All the influence of nutritional status on either children were 10 months of age when vac- antibody response or post-vaccination mor- cinated. None had suffered a clinical bidity. measles infection prior to vaccination and none was seriously ill at the time. Results Each child received a single dose of live, attentuated measles virus vaccine (Attenu- Before vaccination approximately 17 per vax, Merck, Sharp and Dohme) intra- cent of the children had detectable levels of muscularly in the midlateral muscles of the anti-measles antibodies yielding titers of at thigh. Blood samples were taken imme- least 1:s. The existence of this pre-vaccina- diately before vaccination and 2, 8, and 14 tion antibody was not related to the nutri- months afterwards. Sera were assayed for tional status of the children, nor did it sig- measles antibody using a micro-modif?‘ca- nificantly alter their subsequent response tion of the hemagglutination-inhibition to the vaccine. (HI) technique (17). Commercial reagents As the figures in Table 1 indicate, chil- (Flow Laboratories, Inc., Maryland) were dren’s nutritional status did not influence

Table 1. Measles antibody titers yicldcd by sera of vaccinated normal and malnourished childmm.

Log2 HI titers (mean f SEMa) Time Normal Grade I Grades II and III F-ratio P post-vaccination (n = 39) malnutrition malnutrition (n = 26) (n = 9)

0 (pre-vaccination) 0.49 * 0.12 0.58 f 0.15 0.67 f 0.29 .I03 N.S.b 2 man ths 6.54 -I 0.10 6.87 zt 0.30 6.67 f 0.31 .076 N.S. 8 man ths 6.21 + 0.19 6.39 f 0.22 6.00 + 0.45 .422 N.S. 14 months 6.05 zt 0.15 6.00 + 0.21 5.71 + 0.22 .563 N.S.

%EM = Standard error of the mean. bN.S. = Not significant (p > 0.1). 54 PAHO BULLETIN . vol. 13, no. 1, 1979 their antibody response to the vaccination. terval did not change the results signifi- Children in all three groups responded cantly , and no consistent reduction in very well to the attenuated measles virus, antibody titers was observed for those chil- and their serum titers showed only a very dren who became progressively more mal- gradual reduction over the subsequent 14 nourished. months. It was found that about 20 per cent of The nutritional grouping of children at the children exhibited one or more clinical the 0, 2, 8, and 14-month intervals shown signs during the first 4 days post-vaccina- in Table 1 is based upon each child’s nutri- tion. For the remainder of the children, tional status at the time of vaccination. the incidence of these sequelae was not in- During the 14-month period 20 children fluenced significantly by nutritional status. became more malnourished; specifically, Table 2 shows the proportion of daily visits 13 children normal at vaccination later fell to children in each nutritional group that into grade I, 5 grade I children descended detected the indicated sign. In general, the into grade II, and two grade II children incidence of post-vaccination morbidity entered grade III. Nevertheless, grouping was quite low. When present, fever never the children according to their actual nu- exceeded 40°C (104°F) in any child. The tritional status at each post-vaccination in- mean of the highest temperatures for chil-

Table 2. Incidence of post-vaccination sequelae in normal and malnourished children vaccinated against measles.

Average percentage of days with sign for children in each group (mean rt SEMa) Sign Normal Grade I Grades II and III F-ratio P (n = 30) malnutrition malnutrition (n = 23) (n = 6)

Conjunctivitis 13.9 f 3.8 8.1 f 2.3 10.2 f 5.2 0.716 N.Lb Diarrhea 14.4 f 3.2 12.6 f 4.3 0 - - Skin eruption 7.7 f 2.7 13.6 f 3.2 6.4 f 3.5 1.226 N.S. * Fever 13.9 f 3.2 13.4 f 4.2 16.7 z!z 5.8 0.115 N.S.

aSEM = Standard error of the mean. bN.S. = Not significant (p > 0.1).

Table 3. First appearance of sequelae following measles vaccination of normal and malnourished children.

First day of sign (mean + SEMa) Sign Normal Grade I Grades II and III F-ratio P (n = 30) malnutrition malnutrition (n = 23) (n = 6)

Conjunctivitis 14.9 f 1.3 17.5 f 1.4 14.0 * 3.3 1.110 N.S.b Diarrhea 13.3 f 1.3 16.2 f 1.5 - - - Skin eruption 18.7 ziz 1.0 15.6 + 1.3 15.4 f 3.9 1.840 N.S. Fever 11.0 * 1.6 15.2 i 1.8 8.8’f 0.4 2.810 N.S.

aSEM = Standard error of the mean. bN.S. = Not significant (p > 0.1). McMurray et al. l MALNUTRITION AND MEASLES VACCINES 55 dren with fever in each nutritional group in cell-mediated immune mechanisms was as follows: normal, 37.9%; grade I, rather than deficient antibody production 37.9%; grades II and III, 38.O’C. (21). Reduced cellular immunity has been The rate of onset of clinical sequelae in demonstrated in marginally malnourished the three groups of children was also exam- children (9), and we have observed both ined. The results, summarized in Table 3, reduced skin test responsiveness and in vitro indicate that nutritional status did not sig- lymphocyte transformation in our children nificantly influence the interval between with grade II and III malnutrition (un- vaccination and first appearance of the published results). However, these circum- signs. Fever appeared somewhat earlier in stances did not seem to exacerbate the in- grade II and III children, but the differ- fection produced by the vaccine strain. ences involved were not statistically sig- The low levels of antibody detected in nificant. some of the children before vaccination were probably the result of exposure to na- Discussion tural infection at an earlier age. Some of this antibody could have been residual IgG We have demonstrated that moderately of maternal origin, although this is usually malnourished children do not suffer in- metabolized by the child within the first 6 creased morbidity and do respond as suc- months of life. The presence or absence of cessfully as their normal cohort in produc- pre-vaccination antibody did not influence ing antibody following vaccination with the subsequent response to vaccination. live, attenuated measles virus vaccine. The maintenance of high HI titers Previous investigators have examined the during the 14 months following vaccination response to measles vaccination of popula- is similar to the patterns obtained in studies tions known to suffer from malnutrition, of normal children (22, 23). The literature but few specific data on the nutritional reports no longitudinal study of measles status of the vaccinees were reported (13, titers in normal and malnourished vac- 14). Hence little information is available cinees. comparing the post-vaccination morbidity The observation that even a progressive and antibody responses of normal and mal- decline in nutritional status (observed in 20 nourished children from a homogeneous of the study children) failed to adversely population. affect their antibody levels illustrates the Our results support the earlier observa- efficacy of measles vaccination in such a tions, which suggest that the antibody re- population. It should be noted, however, sponse to live, attenuated measles virus is that the rontinued high titers in our popula- normal in malnourished children (20). tion could be related to the booster effect of The apparent lack of nutritional status in- exposure to wild-type virus in the environ- fluence on post-vaccination morbidity that ment, as well as to residual antibody pro- we observed in our Colombian children duced against the vaccine strain. should encourage the use of live, attenuated Antibody titers have been shown to wane measles vaccine in other moderately mal- somewhat more quickly in institutionalized, nourished populations-populations that and presumably isolated, populations (23). include over half of all the children in It has also been demonstrated that a decline many developing countries (II). in the post-vaccination titer to very low It has been postulated that increased mor- levels does not necessarily mean loss of pro- bidity and mortality following natural in- tection, since a rapid anamnestic response fection with measles in malnourished in such individuals may adequately cope populations may be related to deficiencies with a natural infection (22). 56 PAHO BULLETIN l vol. 13, no. 1, 1979

Chandra (8) has recently demonstrated cination with living agents such as polio, that severely malnourished children have BCG, cowpox, and measles can precipitate a significantly less IgA and specific anti- series of metabolic events resulting in nega- measles antibody in nasopharyngeal secre- tive nitrogen balance and anorexia (24, 25). tions following vaccination with the live, Kielmann has demonstrated significant re- attenuated vaccine. We found reduced ductions in weight-for-age in vaccinees less levels of IgA in the tears and saliva of our than 6 months of age (26). Although we moderately malnourished children (IO), were unable to demonstrate increased mor- although we did not measure specific anti- bidity in the malnourished children who measles activity in these secretions. The received live measles vaccine, it is pos- apparent sensitivity of the secretory immune sible that the deteriorating nutritional system to even moderate malnutrition may status noted in 20 of the children during the be a very significant factor in the response subsequent months was related, at least in to vaccination and the susceptibility to part, to the subclinical measles virus infec- . natural infection with measles, since secre- tion. However, given the high risk of tory IgA represents the first line of defense natural exposure in this population and the against inhaled virus landing on mucosal excellent antibody response in both normal surfaces. and malnourished children, vaccination There is a well-documented synergism with measles vaccine as early as 10 months between infection and nutritional status. of age is warranted. Even the mild infections produced by vac-

ACKNOWLEDGMENTS

The authors express their appreciation to and subsequent field work; to Ms. M. Ms. A. de Aly for expert technical assist- McMurray for help with study coordination ante; to Ms. 0. Vanegas, Ms. B. Mosquera, and data control; and to Ms. G. Wilkerson and Ms. M. C. de Rodriguez for vaccination for data analysis.

SUMMARY

Seventy-four Colombian children from a poor tional status had no influence on post-vaccina- urban population were given one dose of a live, tion morbidity. In addition, all the children attenuated measles virus vaccine at the age of 10 responded serologically to the same degree, months. Nearly half of these children (35) were regardless of nutritional status, mean hemag- suffering from some degree of moderate malnu- glutination-inhibition titers being only slightly trition, as indicated by lack of the weight ex- reduced in all nutritional groups 14 months pected for their age. after vaccination. On the basis of these findings, In terms of the incidence and onset dates of the use of live, attenuated measles vaccine in fever, conjunctivitis, diarrhea, and skin erup- moderately malnourished children appears to be tions in this population, the children’s nutri- safe and effective. McMurray et al. l MALNUTRITION AND MEASLES VACCINES 57

REFERENCES

(I) Mata, L. J. Malnutrition-infection inter- rian children. Am J Dis Child 103-402, 1962. actions in the tropics. A7nJ Trap Med Hyg 24: (15) Hendrickse, R. G., D. Montefiore, P. M. 564, 1975. Sherman, and H. M. Van der Wall. Studies ori (2) Scrimshaw, N. S., C. E. Taylor, and J. E. measles vaccination in Nigerian children. Br Gordon. Interactions of Nutrition and Znfec- Med J 1:470, 1964. tion. WHO Monograph Series, World Health (16) Gomez, F., R. Ramos-Galvan, J. Cravio- Organization, Geneva, 1968, 329 pp. to, and S. Grenk. Malnutrition in infancy and (3) Sirisinha, S., R. Edelman, R. Suskind, C. childhood with special reference to Kwashior- Charupatana, and R. E. Olson. Complement kor. In: Advancesin Pediatrics. Yearbook Publi- and CS-proactivator levels in children with cations, Inc., New York, 1955, p. 131. protein-calorie malnutrition and effect of die- (17) Rosen, L. Hemagglutination and hemag- tary treatment. Lancef 1:1016, 1973. glutination-inhibition with measles virus. Viro- (4) Mata, L. J., and W. P. Faulk. The im- logy 13:139, 1961. mune response of malnourished subjects with (18) Cutchins, E. C. A comparison of the special reference to measles. Arch Latinoam hemagglutination-inhibition,neutralizationand Nutr 233345, 1973. complement fixation tests in the assay of anti- (5) Edelman, R., R. Suskind, R. E. Olson, body to measles. J Zmmunol i&788, 1962. and S. Sirisinha. Mechanisms of defective (19) Fulginiti, V. A., and C. H. Kempe. A delayed cutaneous hypersensitivity in children comparison of measles neutralizing and hemag- with protein-caloriemalnutrition. Lancet 1:506, glutination-inhibition antibody titres in indiv- 1973. ual sera. Am J Epidemiol 82:135, 1965. (6) Smythe, P. M., M. Schonland, G. G. (20) Ifekwunigwe, A.E.C. Mass immunization Brereton-Stiles, H. M. Coovadia, H. J. Grace, programme against measles, smallpox, and W.E.K. Loening, A. Mafoyane, M. A. Parent, tuberculosis in a war situation. In: Symposia of and G. H. Vos. Thymolymphatic deficiency and the Swedish Nutrition Foundation, Vol. IX. depression of cell-mediated immunity in protein- Almquist and Wiksell International, Stockholm, calorie malnutrition. Lancet 2:939, 1971. 1971, pp. 150-152. (7) Sirisinha, S., R. Suskind, R. Edelman, C. (21) Smithwich, E. M., and S. Berkovich. Asvapaka, and R. E. Olson. Secretory and serum The effect of measles virus on the in vitro lym- IgA in children with protein-calorie malnutri- phocyte response to tuberculin. In: R. T. Smith tion. Pediatrics 55:166, 1975. and R. A. Good (eds.). Cellular Recognition. (8) Chandra, R. K. Reduced secretory anti- New York, Appleton-Century-Crofts, 1969, pp. body response to live attenuated measles and 131-137. 7 poliovirus vaccines in malnourished children. (22) Reynolds, D. W., and A. Start. Immuni- Br Med J 2:583, 1975. ty to measles in children vaccinated before and (9) Ziegler, H. D., and P. B. Ziegler. Depres- after 1 year of age. Am J Dis Child 124:848, sion of tuberculin reaction in mild and moderate 1972. protein-calorie malnourished children following (23) Ueda, S., Y. Okuno, Y. Sakamoto, N. BCG vaccination. Johns Hopkins Med J 137;59, Sangkawibha, P. Tuchinda, S. Bukkavesa, P. 1975. !,- Ochatononda, Y. Yamada, K. Suzuki, Y. (10) McMurray, D.N., H. Rey, L. J. Casazza, Tanami, F. Kusano, Y. Hayakawa, and T. and R. R. Watson. Effect of moderate malnutri- Kurose. Studies on further attenuated live tion on concentrations of immunoglobulins and measles vaccine: VIII. Estimated duration of enzymes in tears and saliva of young Colombian immunity after vaccination without natural in- children. Am J Clin Nutr 30:1944, 1977. fection. Biken J 17:11, 1974. (II) Goldsmith, G. A. Current status of mal- (24) Brown, R. E. Interaction of nutrition nutrition in the tropics. Am J Trap Med Hyg 23: and infection in clinical practice. Pediutr Clin 756, 1974. North Am 24:241, 1977. (12) Gordon, J. E., A.A.J. J ansen, and W. (25) Beisel, W. R. The influence of infection Ascoli. Measles in rural Guatemala. J Pediutr or injury on nutritional requirements during 66:779, 1965. adolescence. In: J. I. McKigney and H. N. (13) Brown, P., M. Basnight, and D. C. Munro (eds.). Nutrient Requirements in AdoEes- Gajdusek. Response to live attenuated measles cence. M.I.T. Press, Cambridge, Mass., 1976, p. vaccine in susceptible island populations in 259. Micronesia. Am J tifiidemiol 82:115, 1965. (26) Kielmann, A. A. Weight fluctuations (14) Katz, S. L., D. C. Morley, and S. after immunization in a rural preschool child Krugman. Attenuated measles vaccine in Nige- community. Am J Clin Nuts 30:592, 1977.