Background: The epidemiology of rubella in Costa Rica changed during recent decades, shifting the susceptible groups to the reproductive age. This study estimates the burden of congenital rubella syndrome (CRS) from 1996 to 2001 in this country.
Methods: Three methods to calculate CRS incidence were used. A retrospective search ("Observed cases") was conducted using hospital discharge records of children born from 1996 to 2001 with selected codes of ICD9 and ICD10 consistent with CRS and children <3 months of age with a positive serologic test for rubella IgM antibody at the National Children's Hospital (NCH). Cases were classified as either suspected, compatible or confirmed CRS and congenital rubella infection. "Expected" incidence of CRS was calculated using reported cases of rubella (women 15-45 years of age) and fertility rates, assuming CRS probability of 0.9 during the first trimester of pregnancy and 0.5 of asymptomatic rubella cases. "Estimated" CRS cases were calculated using incidence rates reported from modeling analysis during epidemic and endemic years.
Results: Of the 577 discharge charts reviewed and the 66 children reported as rubella IgM(+), 40 compatible CRS cases, 45 confirmed, and 4 with congenital rubella infection cases were identified. The range of annual incidence rate of CRS (per 1000 live births) was as follows: "Observed" = 0.00-0.33, "Expected" = 0.00-0.35 and "Estimated" = 0.5-1.5. Compared with the estimated number of CRS cases, only 27.2% of CRS cases were detected from the retrospective search and 10.1% would be expected when calculated using rubella reported cases.
Conclusions: The under-detection of CRS cases using rubella reported cases in women of reproductive age and retrospective search of CRS reinforces the importance of suspecting CRS in the presence of a single compatible manifestation. Laboratory confirmation is indispensable to implement CRS elimination strategies and should be done in every suspected case.