Serological evaluation of the role of cytomegalovirus in the pathogenesis of IDDM: a prospective study

Abstract

SummaryTo study the possible temporal association between primary cytomegalovirus infection and the appearance of islet cell autoantibodies or the development of insulin-dependent diabetes mellitus (IDDM) cytomegalovirus antibodies were analysed from follow-up sera of 46 initially non-diabetic siblings of diabetic children who either manifested clinical IDDM (22 siblings) or turned islet cell antibody positive (24 siblings) during the prospective observation (mean follow-up time 2.9 years). Secondly, cytomegalovirus antibodies were analysed during pregnancy in 96 mothers whose child presented with IDDM before the age of 7 years and in 96 control mothers who gave birth to a non-diabetic child. Thirdly, a case-control series including 90 newly-diagnosed young children with IDDM and their 90 control subjects was analysed. No seroconversions were found in cytomegalovirus antibodies during the follow-up of the 46 siblings indicating no temporal association with islet cell antibody seroconversion or manifestation of clinical diabetes. During the followup 17 (37 %) siblings were constantly seronegative and 29 (63 %) seropositive for cytomegalovirus IgG and there was no difference between islet cell antibody positive and negative siblings. Cytomegalovirus IgG and IgM were not different in pregnant mothers who gave birth to a subsequently diabetic child compared to control mothers, or in newly-diagnosed diabetic children compared to control children. Cytomegalovirus IgA was higher in newly-diagnosed diabetic children than in control children (p < 0.005). This difference disappeared when only cytomegalovirus IgG positive individuals were analysed. No correlation was found between islet cell antibodies and cytomegalovirus antibodies in newly-diagnosed diabetic patients. The results do not support the hypothesis that primary cytomegalovirus infections could initiate the cascade leading to autoimmune destruction of the beta cells. [Diabetologia (1995) 38: 705-710]

Viruses and Diabetes

“…Beside several case reports of recent‐onset IDDM with a preceding or concurrent Coxsackie B infection,33 epidemiological studies34 have shown a high frequency of IgM antibodies against Coxsackie B in children newly diagnosed with IDDM 32,35. There are, however, also contradictory results, since some reports have been unable to find any differences in antibody prevalence and titer against Coxsackie virus36 or even decreased levels of antibodies 37.…”

“…Single‐case reports of IDDM after cytomegalovirus (CMV) infection have been reported in children68 and adults 69‐71. More indirect evidence came from studies showing that 35% of newly diagnosed IDDM patients showed serological evidence of recent CMV infection and association with islet cell autoantibodies 70,72‐74. On the other hand a prospective study from Sweden screening 16,474 newborns for congenital CMV infection was unable to detect any correlation of CMV infection with diabetes 75.…”

Maternal virus infections in pregnancy and type 1 diabetes in their offspring: Systematic review and meta‐analysis of observational studies

“…Ten publications were included—1 letter and 9 articles (Figure ). Of these, 6 were nested case‐control studies using samples collected as part of prospective birth cohorts . One study contained 2 case groups with substantial patient overlap; the case group with more detailed virus detection data was included in meta‐analysis.…”

“…Three studies utilised radioimmunoassay (RIA) and 9 employed enzyme linked immunosorbent assay (ELISA) for virus detection . One study tested IgG avidity for CMV and another used plaque neutralisation assay (PNA) for EV . Five additionally performed nucleic acid testing to detect EV RNA .…”

Viruses in type 1 diabetes

“…In addition to EVs, several other viruses have been analyzed in the DIPP cohort including adeno‐, influenza A , respiratory syncytial, rota‐, noro‐, parecho‐ and rhinoviruses ( , , and unpublished observations). Our previous studies in other Finnish cohorts screened also adeno‐, mumps, cytomegalo‐, Epstein‐Barr, and parainfluenzaviruses . None of these viruses have shown similar risk association with T1D as EVs.…”