Human cytomegalovirus (CMV) is a member of the family Herpes viridae and belongs to the subfamily beta herpes viridae. CMV has worldwide distribution, infects humans of all ages and all socioeconomic groups, and with no seasonal or epidemic patterns of transmission [1]. It is the most common cause of congenital infection with birth prevalence of about 0.5% (range 0.2-2.5 percent), and a common cause of deafness and intellectual impairment worldwide [2-4].

In utero transmission of CMV can occur following primary maternal infection during pregnancy but can also occur in women with natural immunity, either because of the reactivation of latent virus or by reinfected with a different strain [5]. Postnatal, CMV is also transmitted from mother to child through breastfeeding and close contact [6]. The transmission risk is the proportion of mothers undergoing a primary infection in a given trimester and/or the preconception period who transmitted CMV to the fetus [7].

Risk of congenital infection is much higher during primary infection of the mother with transmission rate of 30%-40% compared with 0.15%-2.2% during reactivations and reinfection [2,8]. Seroprevalence studies among pregnant women worldwide have shown seropositivity rates for previous infection ranging from 50% in highly developed countries to 100% in developing countries [9,10]. Cytomegalovirus is one of the most common causes of congenital infections; however, in Ethiopia seropositivity rate of cytomegalovirus among pregnant women is yet undetermined; that might more worsens the outcome of the disease.

The aim of the study was to determine the seroprevalence of CMV in pregnant women who attended the Department of Obstetrics and Gynecology of St. Paul’s Hospital Millennium Medical College. The recognition of CMV among pregnant women in the country helps to develop effective preventive protocols. Therefore, this study brought an important input on CMV burden data to design appropriate control measures for Ethiopian mothers and children. This study also, highlights improving awareness to clinicians and creating a public concern for the communities.

A cross sectional study was conducted in St. Paul’s Hospital Millennium Medical College (SPHMMC), Addis Ababa, Ethiopia. It is an urban setting and tertiary hospital for Ethiopian and teaching hospital for national and international students. The hospital provides out and in patient services with 370 beds. Accordingly, patients being seen at SPHMMC come from all over the Northern, Western, Southern and Eastern parts of Ethiopia. However, most pregnant women seeking consultation at SPHMMC are predominantly from Northern Oromia regional state and Addis Ababa city administration. The ranges of monthly ANC service users at SPHMMC, at the time of the study, were 500–750.

All pregnant women who were more than the age of 17, and who were volunteer to give consent to participate in the study were eligible to be included. Pregnant women who were less than 17 were excluded.

Sample size calculation and sampling technique

The sample size for the study was calculated using the formula (n = (zα/2)2 p (1-p)/d2) for estimating a single population proportion at 95% confidence interval (CI) (Zα/2 = 1.96), 5% margin of error, and 10% non-respondents rate based on IgM seroprevalence of CMV from a study in Sudanees pregnant women 6.0% [10]. Therefore, the minimum sample size for seroprevalence of CMV survey was 97. However, a total of 200 pregnant women were consecutively selected to increase the finding or detection rates of cytomegalovirus infection rate in the study settings.

Data collection

A total of 26 standard structured questionnaires were designed to collect information regarding socio-demographics and risk related data. The questionnaire was first developed in English and translated into Amharic (the local language), and then pre-tested in non-selected health institutions to assess the content validity, appropriateness, and question comprehensibility. The questionnaire was revised accordingly. Three data collectors from the institution in the study area were selected. Training was given to the data collectors for two day on how to conduct the interview, content of the questionnaire, data quality, and ways to approach respondents. The first author checked the questionnaires for completeness every day. Incomplete questionnaires were excluded. Five percent of the interviewed participants were randomly selected and re-interviewed by the first author.

Venous blood samples were collected from 200 pregnant women attending antenatal clinic between June 1 and July 30, 2014. Samples were collected under aseptic conditions, and transported using ice-box to Ethiopian public health institute (EPHI) and the experiment was performed in EPHI.

CMV-specific immunoglobulin (Ig) IgG and IgM were analyzed by using the ELISA test kits (Diagnostic Automation, Inc., USA) according to manufacturer’s guideline. Briefly, purified CMV antigen is coated on the surface of micro wells. Patient serum was then added to wells. If the antigen is present then it will bind to the CMV IgM/ IgG specific antibody. All unbound materials are washed away and an enzyme conjugate is added to the well. The conjugate, then binds to the antibody-antigen complex. Excess enzyme conjugate is washed off and TMB Chromogenic Substrate is added. Intensity of the color generated by the bound conjugate is proportional to the amount of IgG/IgM specific antibody present in the sample. Results are then read by a microwell reader compared in a parallel manner with calibrator and controls. Based on the findings the results were communicated to clinicians.

Interpretation

Interpretation was made by using the manufacturer guideline i.e CMV G Index of 0.90 or less is seronegative for IgG antibody to CMV (<1.1 IU/ml) was considered negative and CMV G Index of 1.00 or greater, or IU value greater than 1.2 was positive. It indicates prior exposure to the CMV virus. (>1.2 IU/ml). CMV M Index of less than 0.90 is negative for IgM antibody to CMV and positive was considered when CMV M Index of 1.0 or greater.

Data analysis

The data was entered (with double entry) and cleaned with Epidata version 3.1, and analyzed by using SPSS version 20. Statistical significance was considered when P-value<0.05. CMV-IgG and CMVIgM prevalence was determined by dividing the number of CMV infected individuals to the total number of individuals screened for CMV infection. Frequency distribution tables were used to quantify participant’s age range, gestation, occupation, parity and risk factors of CMV positivity rate. Chi-square test and Fisher’s exact test was used to see the association of risk factors to CMV incidence rate. Multivariate logistic regression analysis was used to quantify the effect of different clinical and obstetrical risk factors on CMV seroprevalence. 95% confidence intervals were calculated for odds ratio. Values were considered statistically significant when P-value<0.05.

Socio-demographic and Obstetrical characteristics

A total of 200 pregnant women were enrolled in the study. The mean (SD) of maternal age, parity and gestational age were 26.7 (4.7) years, 1.75 (1.5), 7.3 (2.2) months, respectively. Almost all, 197 (98.5%) of participants were married; and 64 (32.0%) had completed a secondary education. More than half (66.5%) of participants were housewives; 14 (7.0%) were government employees and the least 5 (2.5%) were students. Regarding gestational age distribution, nearly two third of the mothers 139 (69.5%) were in third trimester pregnancy and followed by 22.0% second trimester pregnancy (Table 1).

Table 1: Socio-demographic variables of 200 pregnant women at St. Paul’s Hospital Millennium Medical College, Addis Ababa, Ethiopia, 2014.

The mean CMV positivity age was 26.1 with standard deviations’ of 4.9. The frequency of CMV infection was highest among pregnant women in the 3rd trimester (69.5%), than those with second and first trimester pregnancy 21.5 and 9.0% respectively (Table 2).

Table 2: Distributions of CMV with Obstetrical, socio-demographical and clinical characteristic of the pregnant women in St. Paul’s Hospital Millennium Medical College, Addis Ababa, Ethiopia 2014.

Out of the total 200 pregnant women 177 (88.5%; 95% CI: 84.0- 94.0) were positive for anti-CMV-IgG antibodies and 31 (15.5%; 95% CI: 10.5-21.0) were positive for CMV-IgM. Twenty two (11.0%) individuals were sero-negative for CMV. These were categorized into 4 types of responses. The first category were immune to CMV {IgG (+) plus IgM (-)}. This constituted 73.5% of the women. The second group was those with primary infection {IgG (+) plus IgM (+)} and this consisted of 15.0% respondents. The third group also had twenty two women who were seronegative {IgG (-) plus IgM (-)}. The last category of women was those with {IgG (-) plus IgM (+)}, and therefore, one individual was included in this category and had a recent primary infection (Table 3).

Table 3: Seroprevalence of CMV-specific IgG and IgM antibodies among pregnant women (n=200) in St.Paul’s Hospital millennium Medical College, Addis Ababa, Ethiopia 2014.

Risk factors

Among the participants: 8.5% were between age groups of ≤ 20, 38.5% were between 21-25, 36.5% were between 26-31 and 16.5% were above 32.The infection prevalence among age groups ≤ 20, 21-25, 26-31 and ≥ 32 was 88.2% (CI: 0.1-3.2), 83.1% (CI: 0.3-10.3), 93.1% (CI: 0.2- 8.8), and 90.1% respectively. However, based on the logistic regression analysis showed no statistically significant variables were found. There was no also significant association detected between sero-positivity rate of cytomegalovirus and educational status, occupation, gestational age, and parity.

None of the pregnant women had history of jaundice; whereas 24.0% had history of abortion and 17.0% had at least one frequency of abortion. Among those who had history of abortion 93.7% (CI: 0.6- 38.6) were sero-positive. From the participants 7.0% had a history of dead child. Among mothers who had history of child lose 85.7% (CI: 0.3-6.3) were positive for the infection. Pregnant women who had no toddler in the house hold were 33.5%. Among those mothers who had no children in the household 85.1% were seropositive. However, there was no a single independent factor for the infection of cytomegalovirus (Table 4).

Table 4: Association of CMV with Obstetrical, socio-demographical and clinical characteristic of the pregnant women in SPHMMC, Addis Ababa, Ethiopia 2014.

This is the first study to examine the rates of CMV infection among pregnant women in Ethiopia. The overall incidence rate of CMV IgG among pregnant women in our study was 88.5%, which is similar with other studies in different African countries: 77.3% in Kenya [11], 97.5% in Sudan, [12], 96% in Egypt [13], 92% in Nigeria [10], and 87% in Gambia [14], and even in one of European country [15]. However, the results of this study were higher than those reported in developed countries. A relatively low seroprevalence, 40%-60%, is reported, 46.8% from France [16], 56.3% in Finland [17], and 60% in the United States [9]. It seems that the prevalence of CMV infection observed in this study was similar to that reported in other developing communities but higher than in the developed one. This may be attributed to the low socioeconomic status and poor hygienic practices which might play significant roles in increasing the rate of CMV infection.

In the present study seroprevalence of CMV IgM was 15.5% among the pregnant women, which reflected an active recent infection and reactivation of the virus. This finding was to be 4-7 times higher than in other different developing countries; 4% in Nigeria [10], 8.1% in Kenya [11], 6% in Sudan [12], 2.5% in Iran [18], 1.7% in Korea [19] and/or even developed world studies from Belgium, Brazil, Taiwan, Cuba and Finland [20-24]. This might be due to low socio-economic status, the number of toddlers, and poor hygienic practice. Poor hygienic practice, in particular is a key source for the prevalent of infection. However, similar rate of active infection has been reported from Poland [25], 15.9% in Kashmir valley [26], and from India [27]. Luckily, one woman (0.5%) out of the 31 IgM-positive Ethiopian women tested was IgG negative, indicating probably a recent primary infection.

In the current study there was no a single independent factor, which signifies cytomegalovirus infection rate among pregnant women (P ≥ 0.05). These results suggest that all maternal age groups have equal chance of being infected by this virus. This is in agreement with other studies [10,12]. However, our finding is in contrast with other studies, marital status, illiterate women and women with high parity were at higher risk for CMV infection [11,18,28-30]. This might be due to the difference in socio-demographic characteristics, various cultures and behaviors among these settings might have influence and determine epidemiology of CMV. Low socioeconomic status has also been found as a strong risk factor for acquisition CMV infection [29].

In the current study 22 (11.0%) of the participants were observed to be susceptible to primary CMV infection during their pregnancy. These women are at risk for congenital infection resulting from a primary infection due to maternal primary CMV infection, which leads to fetal infection in approximately 40% of cases [31]. Since the incidence rate of primary infection among pregnant women is high, they are a critical group because the risk of congenital infection/following/after/resulting from a primary infection is much higher during primary infection in the pregnant women [2,31]. Therefore it would be beneficial to properly inform this category of women on the need of further investigations to detect prenatal infection and planning of appropriate intervention such as use of hyperimmune globulin or consented termination of pregnancy as an option.

Possible preventive measures including improved hygiene behavior of seronegative pregnant women should be emplaced and routine maternal screening for primary infection using IgG avidity has to be made. Once diagnosed with the cytomegalovirus infection, treatment with hyperimmune globulin and safe administration of oral ganciclovir to mothers of CMV-infected fetuses, with no teratogenic side effects when given in the early stages of pregnancy should be supported [32,33].

This study brought a new epidemiology of etiological agent in Ethiopia. It showed the high prevalence of cytomegalovirus infection in the setting. The study also showed the high prevalence of primary (recent and/or re-activation) infection and indicates the high risk of feta anomalies. Since this is the first study in the setting, the significance of information that contained in the study is not undermined by clinicians, researchers, and policy makers.

The following Limitations should be taken into consideration; failure to use CMV IgG avidity test, which helps to identify between primary infection and reactivation. We were unable to address all the potential risk factors to CMV positivity like HIV/ADIS sero-status and income of the participants. The other limitation is the lack of long term follow-up examination of pregnant women and their offspring born to CMV IgM-positive mothers in order to document seroconverting and to detect fetal anomalies.

In general, this study showed that there is high seroprevalence rate of CMV infection among pregnant women at our center; and is likely to be a reflection of the overall high prevalence among adult Ethiopians. These data should therefore, help to create awareness for clinicians in Ethiopia, that rapid and accurate diagnosis of CMV infection in pregnant women is critical to prevent major associated complications. Clinicians should take CMV into consideration during management of congenital viral related diseases. Future studies, including large scale surveillance throughout Ethiopia are needed before national screening and universal prevention measures are considered.

The authors are very grateful to St. Paul’s Hospital Millennium Medical College, Dr. Zerihun Abebe, Dr. Lia Tadesse, and finance offices for their kind cooperation in addition to providing the fund. We are also grateful to Dr. Jason Bell, for his kindness and unreserved co-operation starting from the project development and facilitating the reagent purchasing. Finally, we would like to say thank you for all the patients for their willingness to participate in this study. Finally, we would not ignore the co-operation of Ethiopian public health institute for their collaboration.

YM, Principal investigator of the study, study design, data collection, laboratory work, and data analysis; BN & DB, Co-investigator, study design and participate in clinical data collection; FC, laboratory work; AD, study design and support during lab work; SS, study design; all authors contributed to the write up. All authors commented and approved the final manuscript.

Ethical approval for this study was obtained from the Institutional review board (IRB) of St. Paul’s Hospital Millennium Medical College Research Ethics Committee (ref: 2014/P.M23/119) with 16/05/2014 date of approval. Women gave written informed consent before taking part.

St. Paul’s Hospital Millennium Medical College, under the Research directorate, fully funded the project. The program was focused on five streams. Among them, maternal and child stream was one of the area to be funded and this paper was selected and sponsored by the organization.