Sociodemographic factors associated with IgG and IgM seroprevalence for human cytomegalovirus infection in adult populations of Pakistan: a seroprevalence survey

Saira Ibrahim, Anwar A. Siddiqui, Amna R. Siddiqui, Waquaruddin Ahmed, Paul A. H. Moss, El-Nasir M. A. Lalani

The seroprevalence of human cytomegalovirus (HCMV) infection ranges from 30 to 90 % in developed countries. Reliable estimates of HCMV seroprevalence are not available for Pakistan. This study determined the seroprevalence and sociodemographic factors associated with HCMV infection in adult populations of Karachi, Pakistan. A seroprevalence survey was conducted on 1000 adults, including residents of two semi-urban communities, and visitors to a government and a private hospital. Questionnaire-based interviews were conducted. Sera were analysed for HCMV-specific IgG and IgM. Chi-square or Fisher's exact test was used for comparing sociodemographic variables against seropositivity of HCMV-IgG or IgM. Multiple logistic regression modeling was performed for IgG seroprevalence and adjusted odds ratios were computed. The seroprevalence of HCMV-IgG and IgM was 93.2 and 4.3 % respectively. 95.3 % of individuals who were IgM seropositive were also seropositive for IgG. Around 6 % (15/250) of women of childbearing age remained uninfected and were therefore susceptible to primary infection. HCMV-IgG seroprevalence was associated with being female (p = 0.001), increasing age (p = 0.002) and crowding index (p = 0.003) and also with lower levels of both education (p < 0.001) and income (p = 0.008). Seroprevalence also differed significantly by marital status (p = 0.008) and sampling location (p < 0.001). A logistic regression model for HCMV-IgG seroprevalence showed associations with being female (OR = 1.89; 95 % CI: 1.10-3.25), increasing age (OR = 3.95; 95 % CI: 1.79-8.71) and decreasing income (OR = 0.72; 95 % CI: 0.54-0.96). A strong association was observed between increased seroprevalence of HCMV-IgM and decreasing household size (p = 0.008). Seroprevalence of HCMV is very high in Pakistan, although 6 % of women of childbearing age remain at risk of primary infection. The IgM seropositivity observed in some individuals living in small household size (1-3 individuals) with persistent HCMV infection could have resulted from a recurrent HCMV infection. Future longitudinal research in pregnant women and neonates is required to study the trends in HCMV seroprevalence over time in Pakistan for the development of a potential HCMV prevention and vaccination programme.