Measles Vaccination and Inflammatory Bowel Disease

A case-control study of measles vaccination and inflammatory bowel disease. Feeney M, Clegg A, Winwood P, Snook J. Lancet 1997;350:764-66. Summary: There are two schools of thought about the pathogenesis of inflammatory bowel disease (IBD). The first emphasizes the role of the immune system in disease development, whereas the other focuses on a disease-transferring agent. The latter viewpoint implicates such micro-organisms as the measles virus in the pathogenesis of Crohn's disease. We discuss here an article presenting a study on the influence of measles vaccination on the later development of IBD. Eighty-three patients with Crohn's disease and 57 patients with ulcerative colitis, all born after 1968, were identified from hospital records. For each of those patients, two controls were chosen from the same general practitioner list. The patients and controls were matched for sex and year of birth. The vaccination history was monitored by means of the general practitioner's vaccination records or other medical records. The vaccination records were inadequate in 15% of the IBD cases, leaving 140 patients. In the control group, 16% of the patients did not have an adequate vaccination record. These control subjects were substituted with a matched reserve. In the group of IBD patients, 56.4% had been vaccinated, whereas in the control group, 57.1% had been vaccinated. Thus, there was no evidence of a correlation between IBD and measles vaccination with live attenuated virus. Similarly, no evidence of an association was found between IBD and vaccination for pertussis, diphtheria, tetanus, or polio. Comments: A few years ago, the public, particularly in the United Kingdom, was alarmed by a report stating that live attenuated measles vaccine rendered an odds ratio for developing ulcerative colitis or Crohn's disease of 2.53 and 3.01, respectively (Lancet 1995;345:1071-74) . This article reported a randomized trial by the Medical Research Council (MRC) on measles vaccination. The prevalence of IBD in a cohort of 3967 recipients of live measles vaccine was compared with the prevalence in a control cohort from the National Child Development Study. This control cohort consisted of all subjects born in 1 week in 1958 and was used because the unvaccinated subjects from the MRC trial had been lost in the follow-up. Additionally, the partners of the participants of the MRC trial were used as a control group, although their vaccination history was not known. The study design and the interpretation of the data were heavily criticized because both control groups did not account for a good match with the MRC trial group (Lancet 1995;345:1062-63;Lancet, 1995:345:1362; Lancet 1995;345:1363) . Furthermore, comparing people from the vaccinated group with their partners, the increased risk of IBD was not significant. However, the investigators in the MRC study provided a large amount of epidemiologic (Lancet 1994;344:508-10;Lancet 1996;348:515-17) and histopathologic(J Med Virol 1993;39:345-53) evidence in favor of a measles hypothesis, although not all findings could be confirmed by others(Lancet 1995;345:199; Gut 1996:38:211-15) . A subsequent case-control study by the investigators in the MRC study remained inconclusive because of inadequate documentation of the vaccination history(Eur J Gastroenterol Hepatol 1995;7:385-90) . The present article is therefore of great importance in this discussion because a case-control study is the most appropriate way of defining the risk of vaccination. Of note, the study shows that vaccination with live attenuated virus does not increase the risk of developing IBD in later life. In contrast to the MRC trial, the control group was matched for both sex and year of birth. Furthermore, the authors reasonably account for vaccination history of patients and control subjects. In the MRC study, the Schwartz strain of live attenuated virus vaccination was used, whereas the current study contains 11 different although closely related vaccines, some of which are identical to the Schwartz strain. In theory, this could mean that the increased risk for Crohn's disease is related to the vaccine strain used. In practice, this hypothesis is difficult to test considering the omissions in the immunization records. The study does not exclude a role for wild-type measles virus infection in the pathogenesis of CD that was suggested in an epidemiologic study reporting development of IBD in later life in three out of four cases of exposure to measles virus in utero (Lancet 1996;348:515-17) . When in utero or perinatal measles infection plays a role in the pathogenesis of CD, a decrease of the disease incidence is expected in the coming years, because most of the future mothers will have been vaccinated. Indirectly, the study provides an argument against an association of measles infection and CD, in that the incidence of CD during the past decades has increased, whereas vaccination has reduced the number of measles infections in this same period (Immunisation against infectious disease. Salisbury DM, ed. London 1996:125-27) . It is now commonly accepted that IBD is likely to be triggered by some environmental factor that causes a chronic inflammatory response in the genetically susceptible person (Nat Med 1995;12:1241-43) . Therefore, research should be focused on other pathogens, perhaps on pathogens that share characteristics with the measles virus. Genetic and immunological studies will provide important data on disease susceptibility and may thus identify the pathogen(s) that triggers IBD. For now, it seems that measles virus vaccination does not increase the risk of IBD in later life, and therefore it should not be an argument that determines vaccination policy. Pieter C. F. Stokkers Merlijn van den Berg Edmond Rings Division of Pediatric Gastroenterology and Nutrition; University of Amsterdam; Academic Medical Center, G2-205; Amsterdam, The Netherlands