In this study, we determined the seroprevalence of HCMV infection in adult populations of Pakistan. Our results reveal a high seroprevalence of HCMV infection, with an average of 93.2 % of individuals aged ≥ 18 years, showing the presence of HCMV-specific IgG antibodies indicative of persistent infection. This is consistent with studies reported from Asia, South America and the Caribbean [10, 12, 24] but higher than that reported from North America, Europe, Australia and Africa [7, 8, 15, 16, 25]. The IgM seroprevalence was 4.3 %, which is similar to a previously reported estimate [26] but higher than 0.9–3 % IgM seroprevalence in several other countries [11, 12, 27–30]. However, much greater variations in serum IgM seroprevalence (0–19.5 %) have been reported from developing countries [9, 31]. 95.3 % of individuals who were IgM seropositive were also seropositive for IgG. In the absence of either a confirmed sero-conversion or IgG avidity testing it was not possible to define what proportion of these cases represented primary infection or non-primary reactivation of IgM. Interestingly, a recent analysis of HCMV IgM seroprevalence in women of reproductive age in USA also found that 97.5 % of IgM seropositive women were CMV IgG seropositive [32]. IgG avidity testing in countries with a high prevalence of HCMV infection, such as Korea and Turkey; have shown that none of the women with an IgM seropositive and IgG seropositive sero-status had evidence of a primary infection [33, 34]. As such, we would suggest that the great majority of IgM seropositive cases within our cohort represent viral recurrent reinfection or reactivation rather than primary infection.

Despite the general very high seroprevalence of HCMV infection, 6.6 % of the population remained seronegative. It is likely that good hygiene, hand washing and limited sharing of edibles and used utensils serve to limit infection rates in some populations [35]. The low percentage of individuals with a HCMV-IgM positive and HCMV-IgG negative profile (0.2 %) indicates that the great majority of infections occur during childhood although our data also reveal a significant increase in seroprevalence of men between the ages of 25 and 34 years.

Significant differences apparent in relation to the seroprevalence of HCMV infection in different locations are likely to be attributed to the marked differences in socioeconomic status between the government and private hospital patient populations. In particular, a very high proportion of the patients visiting JPMC belong to the low socioeconomic strata of the society in whom HCMV-IgG seroprevalence reached 95 %.

When we compared the HCMV-IgG seropositivity between the different age groups, we observed a rise in HCMV-IgG seropositivity from age group ≤ 25 years to age group 35–45 years. However, this increase is not as high as has been observed in previous studies and reflects the high baseline prevalence of infection by the time of early adulthood [16, 36]. Nevertheless, our findings concur with other studies from areas of high seroprevalence where a positive HCMV-IgG response was observed in more than 90 % of the individuals in the age group 18 to 35 years [9, 12].

Gender differences in HCMV seroprevalence have been noted previously [7, 15, 16, 36]. In our study, younger adult males are shown to be much more vulnerable to a primary HCMV infection than females and seroprevalence within men increased by 12 % between the age groups ≤ 25 years and 35–45 years. Women can acquire primary infection through their own infected children’s urine and saliva [37]. Therefore those women who remain HCMV seronegative as they enter early adulthood are more likely to be at risk of acquiring HCMV infection from their children than men at this stage. Importantly, nearly all women have had a primary HCMV infection as they reach childbearing age in contrast to developed countries where many women remain seronegative at reproductive age [15, 16, 25]. This is important in relation to the potential susceptibility to primary infection during pregnancy as this is associated with risk of congenital transmission. The seroprevalence of HCMV infection remained high in women of all ages and was still found in 96 % of those aged over 45 years. Interestingly, the seroprevalence of HCMV was slightly reduced in men aged > 45 years. Reasons for this age-related reduction are unclear. It is very unlikely that seroprevalence of HCMV was lower during early adulthood in the current group of individuals aged > 45 years. As such, it is possible that HCMV infection has been associated with excess mortality within males, such that the proportion of HCMV seronegative individuals may have increased slightly over time. Indeed, infection has regularly been correlated with a significant increase in mortality rates in older individuals and it is possible that this effect is more pronounced within male subjects in developing countries.

Previous studies examining the relation between HCMV seroprevalence and socioeconomic status in developing countries have examined small sample size compared to studies undertaken in developed countries [7, 18, 38]. By targeting a relatively large population, we were able to demonstrate an inverse relationship between socioeconomic status and HCMV-IgG seroprevalence, which is consistent with several previous reports [15, 17, 18, 36, 39]. However, in contrast to a previous study [7], household size does not seem to contribute to the disparity in persistent HCMV infection within our population. It has been suggested that people at the lower ends of the education and income spectrum are likely to spend longer time with HCMV infection [17], primarily due to more crowded living conditions [40]. However the definitive reason for the increased seroprevalence of HCMV within poorer communities remains unclear.

Univariate analysis showed a significant association between ‘married’ marital status and HCMV IgG seroprevalence, which is reported to reflect increased viral exposure via vertical and horizontal modes of transmission [37, 41]. The prevalence of HCMV infection did not appear to be influenced by ethnicity in our population.

A unique feature of our study was an epidemiological assessment of sociodemographic factors in relation to IgM sero-status as this has not been widely studied previously. An important finding was that only two individuals were IgM seropositive but IgG seronegative, a pattern which implies recent primary infection. However, false positive HCMV IgM results due to interfering infections in IgG seronegative individuals may exist. As such, as discussed above, IgM sero-status is likely to reflect a humoral response to viral reinfection or reactivation [28, 42]. Indeed, HCMV infection is unusual in that IgM seropositivity can occur in individuals with persistent infection, whereas an IgM antibody response is generally observed only in the setting of acute infection for other pathogens. A striking observation was the increased seroprevalence of HCMV-IgM in association with decreasing household size. No association was seen between IgM and increasing crowding index and so increased frequency of HCMV transmission due to close contact with infected individuals is unlikely to be an explanation. Indeed our data may suggest an opposite explanation, in that smaller families are less likely, at any given time, to house an individual with primary infection. As such, it may be that the regular environmental exposure to HCMV in larger families serves to boost established HCMV-specific immunity and control endogenous viral replication. In this regard, individuals who do not get such regular environmental challenge may undergo more frequent episodes of endogenous viral replication, which would stimulate recurrent IgM immune responses. Interestingly, a recent study of IgM seroprevalence in US women observed a non-significant increase in prevalence within single women compared to married women (6.8 % vs 3.8 %) which could be compatible with this observation [32]. Further epidemiological and virological studies will be required to investigate this observation further.

The two major strengths of this study included a large sample size (N = 1000) and first estimate for HCMV seroprevalence in diverse subgroups of population in a developing country. Our comprehensive analysis was based on a questionnaire which gathered data about living conditions and provided an insight into the reasons that could account for such a high seroprevalence of this virus in a developing country.

Congenital infection with HCMV is a significant cause of morbidity and mortality and several candidate HCMV vaccines are under investigation. However, the development of a vaccine programme must take into account the potential risk of increasing the proportion of women who remain HCMV seronegative as they enter pregnancy. Our findings reveal that up to 6 % of women of childbearing age remain HCMV seronegative in Pakistan, and the risk of primary infection within this group may be significant given our observation that around 0.5 % of men of a similar age may be undergoing primary infection each year, an event followed by prolonged viral secretion, which could serve to drive infection in the women who remain uninfected at this age.

Our study serves as a preliminary work towards improved understanding about the status of HCMV infection in Pakistan; mainly so because such a large number of participants with sociodemographic data has not been studied for the assessment of this particular virus.