Varicella is a common infection of childhood, regardless of sex or race. Approximately 90% of the infections occur in children between the ages of 1 and 14 years]. Seroepidemiologic studies in developed countries have shown that by the age of 20 years, over 90% of the population has been exposed to the infection and are protected against a new infection. However, recent studies have shown an increase in hospitalization due to varicella zoster virus (VZV) infection in young adults living under crowded conditions. Incidence of VZV primary infection seems to be decreasing in individuals younger than 15 years; hence, increasing in young adults. In fact, the incidence has increased two times in patients 15 to 19 years, and five times in patients over 19 years, as has been reported. These studies indicate that the proportion of adults susceptible to VZV infection is increasing. The infection in young adults is often more severe, with pneumonia as a common complication and with a mortality 25 times higher than that observed in children. Furthermore, prime-infection during pregnancy is associated with high maternal morbidity and mortality and represents a risk for fetal infection.

Recent reports suggest that the age of VZV primary infection is rising in the U.S. and in the UK, resulting in an increase of symptomatic cases in adults.

The aim of this work was to study the seroprevalence to VZV infection in children and young adults in Mexico, in order to determine the size of the population susceptible to symptomatic VZV infection. The influence of socioeconomic and demographic variables on seroprevalence was also studied.

Materials and Methods

Population

During 1987–1988 the Health Secretariat of Mexico performed a National Serologic Survey to create a National Serum Bank. The collection of the serum samples was done according to a probabilistic master-sampling frame based on the General Population Census data. The survey included all 32 states of Mexico, and was designed to include individuals 1- to 90-years-old from all socioeconomic levels and from all geographic zones of the country. A total of 32,200 housings were visited and more than 70,000 serum samples were collected. Each individual completed a questionnaire detailing personal, socioeconomic and other demographic data.

To define socioeconomic level, an index was constructed which included number of persons per room, type of material of the floor of the house, availability of municipal water, waste disposal, and years of education of the head of the family. Urban population was defined as an area with more than 2,500 inhabitants; a rural population was defined as an area with less than 2,500 inhabitants.

A subsample from this survey representing individuals 1- to 29-years-old from all regions of the country was selected; sample size was calculated with an expected seroprevalence of 80% and considering the socioeconomic and demographic regionalization of the country proposed by Kunz et al.This regionalization divides the country into eight regions and is an indicator of the level of development. The total number of serum samples to study was 3,737.

Determination of VZV infection

Total IgG antibodies against VZV were determined with a commercial immunoassay kit (VARELISA, Merck, Germany). Positive and negative reference serum (included in the kit) were included in each plate. Cut-off value was defined as: mean of the O.D. of the negative reference samples + (0.1 × the mean of the O.D. of the positive reference samples). Patients with O.D. values below the cut-off point were considered seronegative, hence susceptible to symptomatic VZV infection. Specimens with O.D. values close to the cut-off point were tested again.

Statistics

Seroprevalence to infection and its association with sociodemographic characteristics were described as percentages with 95% confidence intervals. Chi square test was used to compare VZV seroprevalence and the sociodemographic variables among the regions. Seroprevalence in the different age groups was analyzed with the chi square test for trend, and odds ratio with 95% confidence intervals.

Results

VZV serology was positive in 3,273 of the 3,737 individuals studied (87.6%); and 464 were seronegative (12.4%) or susceptible to infection.describes seroprevalence according to age; the population susceptible to infection decreased from 18% in the 1- to 14-year-old group to 5.2% in the 25- to 29-year-old group. In the group of patients 15- to 19-years-old, 17% were susceptible to infection. The tendency of a higher susceptibility in lower age groups was significant (p <0.0001) and indicate that age is a risk factor for infection. The role of socioeconomic and demographic variables as risk factors for susceptibility to VZV infection was analyzed with a multivariate model and is presented in. Low socioeconomic levels, living in urban or in metropolitan areas, overcrowding, and gender were not significantly associated with a risk for susceptibility to VZV infection. However, low educational level was found as a risk factor for susceptibility to infection.

Discussion

We found that in Mexico about 18% of the 1- to 19-year-old population studied is susceptible to VZV infection, and decrease with age to 5.2% in individuals 25- to 29-years-old. Thus, in Mexico adolescents and young adults have a susceptibility to VZV infection higher than expected. We further studied socioeconomic and demographic variables as risk factors for infection and found that only a low level of education was a risk factor for susceptibility to infection. We speculate that the higher susceptibility to infection in individuals with lower education is probably related to the fact that they attend school less frequently and have a lower exposure to other infected children.

Susceptibility to VZV infection in adolescents and young adults is increasing in various countries and has already influenced vaccination practices. The age-specific incidence of VZV infection during l991 in the UK and from 1972–1978 in the U.S. are shown in and compared with the seroprevalence found in our study from l987–1988. Interestingly, the infection during the first 5 years of life is higher in Mexico than in the U.S. and in the UK; however, in Mexico almost no new cases occur after the age of five, whereas in the U.S. and the UK, infections increase continuously. As a result of this contrasting epidemiology, 17% of the adolescent group (15- to 19-years-old) were susceptible to VZV infection in Mexico; whereas <10% of the adolescents in the U.S. and in the UK were susceptible. Even by the age of 29 years, 5.2% of the young adults in Mexico were still susceptible to infection. We have no clear explanation for these differences, but we speculate that in Mexico most infections occur early in life (during the first 5 years of age) and in contrast to U.S. and UK, susceptibility to primary infection decreases with age. The contrasting change in seroprevalence after the age of 20 years in our country might be due to a cohort phenomenon; circulation of varicella virus might have decreased in our population, thus favoring the increase of seronegative or susceptible individuals. We should note the differences in the three studies being compared, UK data are from serosurvey and U.S. data from interview survey.