Burden of Helicobacter pylori Infections and Associated Risk Factors among Women of Child Bearing Age in Addis Ababa, Ethiopia

Background Early detection and treatment of Helicobacter pylori (H. pylori) infection in women of child bearing ages may reduce the risk of maternal health disorder. This study was conducted to determine the burden of H. pylori infections and associated risk factors among women of child bearing ages in Kolfe Keranio Subcity Woreda 9 Health Centers, Addis Ababa, Ethiopia. Methods Facility based cross sectional study design was conducted from April to October 2015. The study recruited 195 pregnant and 137 nonpregnant women with age range of 16-40 years. Sociodemographic data of study participants were collected by structured questionnaire. Venous blood was analyzed to determine hemoglobin, H. pylori stool antigen test kit was used to assess H. pylori infection, and fresh fecal (stool) was used to examine intestinal parasites among study subjects. Data was entered and analyzed using SPSS version 19. Bivariate and multivariate logistic regression model using odds ratio (OR) at 95% confidence interval (CI) were calculated. P-value less than 0.05 was taken as statistically significant. Results The overall burden of H. pylori infection among study participants was 29% (96/332). H. pylori infection was statistically significantly associated with pregnancy status (AOR: 1.825, CI (1.42-2.15), P=0.020), history of hyperemesis gravidarum (AOR=7.028, C.I (2.47-19.99), P=0.018), and low hemoglobin value (AOR=0.177, CI (0.083–0.379), p=0.003). There was no statistically significant association between H. pylori infection and sociodemographic characteristics and some expected risk factors like smoking, Khat chewing, alcohol drinking habit, and presence of intestinal parasites. Conclusion In this study, H. pylori infection was still a public health problem in the study area. H. pylori infected women also had high rate of anemia compared to women who had not H. pylori infected. Hence clinician and other responsible bodies should give a special attention for women who had been infected with H. pylori. Further large case control studies are warranted to understand more the role of H. pylori, HG, and other associated risk factors.


Introduction
. . Background. Maternal mortality remains a major challenge to health systems worldwide and improving maternal health has been on the global health agenda for many years [1]. Ethiopia has one of the highest rates of maternal mortality in Africa [2]. H. pylori is an inflammatory gastric bacterial infection especially prevalent in developing countries. It is the most common bacterial infection worldwide, infecting almost half of people in developed countries and 80% of people in developing countries [3]. Study conducted in Ethiopia showed that there are gaps on screening and treatment of infectious diseases including H. pylori that could contribute to high maternal morbidity [3]. H. pylori infection has a role in the pathogenesis of various pregnancy related disorders through different mechanisms, mainly focused on iron deficiency anemia, thrombocytopenia, fetal malformations and fetal growth restriction [4]. Chronic infection with H. pylori 2 International Journal of Chronic Diseases in pregnant women may also cause hyperemesis gravidarum (HG). This particular complication affects 0.3 to 2.0% of all pregnancies. Previous studies showed that HG has a strong correlation with H. pylori infection [5,6].
Epidemiological studies suggested that anemia presumably due to iron deficiency is correlated to H. pylori infection. It is related to a high level risk of low birth weight, prematurity, and perinatal mortality. It is associated with anemia caused by iron deficiency regardless of the presence or absence of peptic ulcer disease. H. pylori is a cause of gastrointestinal blood loss, reduction in iron absorption, and increase of iron absorption by the bacteria [7,8].
Moderate to severe maternal anemia has been associated with an increased risk of poor reproductive outcomes, including low birth weight and preterm birth deliveries [9]. Early detection and treatment of multiple causes of severe anemia may reduce the risk of prenatal and maternal mortality [10].
The burden of H. pylori infection among various adults and pediatrics groups in Ethiopia is available and the burden could vary in the range of 80-93%. However, none of them address disease burden among women of child bearing ages [11]. Diagnosis of H. pylori infection includes invasive techniques (requiring endoscopy) such as rapid Urease test, culture, and histology, and noninvasive methods such as serology, urea breath test (UBT), and stool antigen test. Invasive methods due to ethical issues and UBT because of the use of radioactive materials are prohibited in pregnancy [9,11]. Serologic testing is the most common noninvasive diagnostic method for H. pylori and is relatively inexpensive and convenient; H. pylori tested by stool antigen test (HPSA) which is an enzymatic immunoassay to detect bacterial antigen of actual ongoing infection in stool is a reliable noninvasive marker in the primary diagnosis and in the monitoring of posttreatment outcome [9,11]. Maternal health is complex and it requires different studies to identify the problem in each level of maternity health services. Hence this study aimed to determine the burden of H. pylori infection and associated factors among women of childbearing ages in the study area.

Methods
. . Study Setting and Context. Facility based cross-sectional study design was conducted among women of childbearing ages from April to October 2015. The study was conducted in Kolfe Keranio subcity, Woreda 9 Health Center, Addis Ababa, Ethiopia. The health centers were delivering service for about 150-200 out patients per day. On average 40-50 pregnant women visited the antenatal care clinics on daily bases. The health facility was well established and trained with different professional staff. The laboratory section was also fairly well equipped with laboratory equipment.
. . Study Population. Three hundred thirty-two women of child bearing age, who were attended antenatal care clinics and outpatient services in the health centers, were used as study populations.
. . Inclusion and Exclusion Criteria. All volunteer child bearing ages of pregnant and nonpregnant (16-45 years) women were included. However, women on triple therapy for H. pylori infection in the past two weeks of data collection period and presentation with severe medical conditions that necessitated urgent care were excluded from the study.
. . Sample Size and Sampling Procedures. The sample size was determined using single population proportion formula, considering the assumption that prevalence of H. pylori infection was taken as 24.1% previous study conducted in Nigeria [12]. By considering 95% confidence interval (CI) and 5% of marginal level, the total sample size was calculated 332. We factored a 10% nonresponse rate in the sample size calculation. Simple random sampling was used to select the study participants. Each day, before the provision of health education, the names of every pregnant woman attending the antenatal clinic were taken. Fifty percent of the women were then randomly selected until the required sample size was attained. The probability sampling was employed to avoid selection bias.
. . Data Collection and Laboratory Procedure. A convenience sample of participants was recruited during the study period. For nonpregnant study participants the most common reasons for presentation themselves to outpatients' clinic were generalized medical healthcare seeking and in the same time, for pregnant women study participants the most common reasons for their presentation to the antenatal care clinic in the health center were for their prenatal and antennal care follow-up during pregnancy. Using a standard questionnaire, interviews with adult participants were conducted and information was collected on sociodemographic data and some clinical and behavioral variables.
Stool (fecal) Specimen Collection and Laboratory procedure: H. pylori infection was examined by using H. pylori stool antigen test kits (CTK Biotech HpSA kit, San Diego, CA 92121 Inc., USA). A random stool (fecal) specimen in a clean, dry receptacle was collected. Then the stool collection device with the specimen's ID number (patient ID sticker) was labeled. The stool collection device was opened by unscrewing the top and used the collection stick to randomly pierce in 2-5 different sites, twisting the collection stick into the fecal specimens to help collection if necessary. All inner grooves of the collection stick were filled with fecal specimen. However, excess fecal specimen on the outside of grooves was scraped off. The collection stick and tighten securely to close the stool collection device was replace. According to the manufacturer leaflet provided with the H. pylori test kits the relative sensitivity, relative specificity, and overall agreement were 96.7%, 93.8%, and 94.9%, respectively. Its analytical sensitivity was 100% positive detection rate at 1 ng/mL of pylori lysate antigen in fecal specimens. All reagents were ready to use as supplied. Unopened test devices were properly stored at 2-30 ∘ C. It was ensured that the test device is brought to room temperature before opening, when stored at 2-8 ∘ C. The test device was stabled through the expiration date printed on the sealed pouch. The kits were not stored in the freezer or exposed to temperatures above 30 ∘ C. Lastly the stool collection device vigorously was shacked and ready for H. pylori test. An H. pylori stool antigen (HPSA) test was International Journal of Chronic Diseases 3 performed according to the manufacturer's recommendation.
The tests were done with fresh stool samples. The results were evaluated within 10 minutes and tests with any change of color of the test-line were interpreted as positive.
Venous blood samples were collected following standard operating procedures (SOP) by well-trained laboratory personnel and were analyzed using cell dyne methods (Hemocount 30 TS hematology analyzer machine, Germany) to determine the hemoglobin level of study subjects and then hemoglobin value <12mg/dl for nonpregnant and <11.0 mg/dl for pregnant women was considered as anemic according to hematology analyzer machine user manual. The manufacturer supplied controls were run every morning to ensure that the analyzer was operating within 2.0 standard deviations. The analyzer automatically sampled blood, processed, analyzed, and printed out the hemoglobin concentration levels.
Intestinal parasites identification was done by direct smear microscopy and formol-ether stool concentration technique. The stool samples were processing based on standard operation procedures described in Monica Cheesbrough [13].
. . Data Management and Quality Assurances. Before data collection, adequate training was given for data collectors and supervisors. Data collectors were instructed to check the completeness of each data before submission. Quality of data collection process and all laboratory procedures were supervised and monitored by the principal investigator.
. . Statistical Analysis. The collected data were checked for completeness and consistency and summarized using Microsoft office Excel. The data were coded and analyzed using SPSS version 19.0 (SPSS Inc. Chicago, USA) software. Descriptive statistics was employed for the analysis of demographic data. Association between the burden H. pylori infection and risk factors was assessed by Chi-square (X 2 ) tests. We used odds ratios as a measure of association, with a 95% confidence interval. Variables with p-values < 0.2 at bivariable analysis and these with biological plausibility with respect to H. pylori infection were put into backward stepwise multivariable logistic regression to determine predictors for H. pylori infection in study subjects. P-value less than 0.05 was taken as statistically significant.

Results
. . Study Subjects Characteristics. From the total 332 study participants that were involved in this study, 195 (58.7%) were pregnant and the remaining were nonpregnant women. The mean (±SD) age of the respondents was 27.3 ± 4.7 years. Majority (82.5%) of the study subjects were married. Sixtytwo (18.7%) of study subjects were illiterate. Most (70.2%) of childbearing women lived with a family number of four or more. Among the pregnant women, 44.6% and 28.7% were become pregnant for the first and second time respectively. Sixty percent and 31.7% pregnant women were within a gestational period of 1-12 weeks and 13-24 weeks, respectively. One hundred ninety (57.2%) of study participants had gastrointestinal illness. From the total of pregnant women, There was no statistically significance association between H. pylori infection and sociodemographic characteristics (age group, marital status, educational background, occupational status, and number of people in the household), and some expected behavioral and clinical risk factors like habits of drinking alcohol, cigarette smoking, Khat chewing, drinking tea and coffee, water source for drinking purpose, gestational period (week), gravidity (number of pregnancy), parity (number of children), and presence of intestinal parasites, whereas categorical Chi-square (X 2 ) analysis showed that there was statistically significant association between H. pylori infection and pregnancy status (X 2 = 68.61, P=0.001), history of hyperemesis gravidarum in pregnant women (X 2 =5.259, P=0.001), history of gastrointestinal illness (X 2 =7.3235, p=0.007), and anemia status (X 2 =68.61, P=0.001) of the study subjects [ Table 2].
In this study the overall burden of H. pylori infection among study subjects was 29%. Burden of H. pylori infection among pregnant and nonpregnant women is 21.9% and 33.8%, respectively [ Figure 1].
Results of multiple logistic regression analysis showed that pregnant mothers were 1. 8

Discussion
In this study burden of H. pylori infection among pregnant women is lower than study reported 45.2% in Uganda and 52.4% in Belgium, Brussels [14,15]. However, this study finding is higher than studies conducted in France and Zanzibar that reported a prevalence of 21.5% and 17.5%, respectively [16,17]. The variation might be due to the difference in study settings, study population, and the laboratory method. In the study of Uganda and Belgium, Brussels, serological antibody test was used as compared to the current study. In the current study burden of H. pylori infection among pregnant women was similar to study reported 33.3% in US-Mexico and 24.1% in Nigeria [12,18].
In the current study, H. pylori infection among in the age group of 31-35 years was higher (36.8%) compared to in the age group of 21-25 years (21.4%). This might be due to H. pylori acquired during young age but it is asymptomatic, which lead to develop gastritis, peptic ulcers, and gastric carcinoma, usually during late adulthood [11,16].   The results of this study also showed that H. pylori infection was significantly high in the pregnant population with hyperemesis gravidarum accounting 29.6%, and this finding agrees with study conducted in Jeddah, Saudi Arabia, However, in case of Saudi Arabia there was a case control study with hyperemesis gravidarum and without hyperemesis gravidarum [19]. But this study showed higher result than the study done in Izmir, Turkey, which showed that women who were H. pylori infected with hyperemesis gravidarum were 22.2% [20]. The possible explanation for this result is that an association of H. pylori and hyperemesis gravidarum could be that an increased accumulation of fluid and a displacement of intracellular and extracellular volume occur as a result of increase in steroid hormones, and this condition results in a change of pH which could lead to the manifestation of a latent H. pylori infection in the gastrointestinal tract. The increased level of steroid hormones and human chorionic gonadotropins (HCG) during pregnancy also leads to changes in PH and motility of GI tract; this change favors activities of H. pylori infection [21].
In this study, 34.7% of H. pylori infected study participant women had history of gastrointestinal illness. This finding is relatively lower than study done in Chile, which reported 68.6% of study participant women had H. pylori infection 8 International Journal of Chronic Diseases among complaining dyspepsia [22]. This indicates that one of the major consequences of H. pylori infection is its effect on acid production in the stomach. The bacteria affect the stomach cells that control stomach acid secretion. This can lead to overproduction of hydrochloric acid, paving the way for ulceration, then the acid producing cells themselves are affected, and less acid is secreted, causing low stomach acid. The consequence of low stomach acid is low B12, since it is difficult to assimilate the nutrient from animal protein if you do not have adequate stomach acid, leading to the fact that gastritis or stomach inflammation always accompanies infection [23,24].
In these study findings, high prevalence of H. pylori infection was seen in anemic (72.1%) study women and H. pylori infection has been found to be associated with anemia. Our findings are similar to other studies in Baghdad [10], Jeddah, Saudi Arabia [23], Zanzibar [17], and India that showed that women who had been infected with H. pylori, indicating that they had high prevalence of anemia [17,24]. The probable finding of this results could be due to some possible mechanism by which H. pylori affects iron metabolism by decreased absorption resulting from chronic gastritis [23], decreased gastric juice ascorbic acid concentration, increased hepcidin production associated with H. pylori gastritis, uptake of iron by H. pylori for growth, and decreased availability of iron by sequestration of iron in lactoferrin in the gastric mucosa and bacterium host competition for dietary iron supply [14]. Another explanation most commonly offered for this relationship could be also based upon the development of H. pylori associated chronic gastritis with resultant achlorhydria and reduced ascorbic acid secretion leading to reduced intestinal iron absorption. Besides an association between anemia [23] and H. pylori includes occult blood loss from erosive gastritis [23] and sequestration and utilization of iron by the organism [25].
In this study another point observed is that pregnant women were significantly associated with H. pylori infection. These findings agree with that of Cardaropoli and Ehab reports [4,26]. This probably proposed that pregnant women are among the most vulnerable groups for H. pylori infection and a reduction of gastric acid production during early pregnancy as a result of increased accumulation of woman's body fluid, steroid hormone changes, and immunologic tolerance which could lead to the activation of latent H. pylori infection, which can exacerbate dyspepsia, nausea, and vomiting symptoms, because of underlying undiagnosed peptic ulcer disease, which in turn may affect maternal gastric absorption. Mild to moderate dyspepsia is commonly associated with nausea and vomiting and complicates about 50% of all pregnancies and it diminishes women's life quality and social functions during early pregnancy [27]. However, in most women, these symptoms resolve by fluid and vitamin supplementation as well as dietary modification.

Conclusion
In this study H. pylori infection was still a problem among women child bearing age in study area. H. pylori infection was statistically associated with pregnancy status, low hemoglobin value, and presence of hyperemesis gravidarum in pregnant women. As H. pylori infection can lead to serious gastrointestinal problem throughout the individual's life, we would like to emphasize how important it is for H. pylori screening and eradication treatment to be conducted in the country. Therefore clinician and other responsible bodies should give a special attention for women who have been infected with H. pylori; however further large scale case control studies are warranted among study participants to understand between H. pylori infection and related risk factors.

Data Availability
The data used to support the ïňĄndings of this study cannot be shared in a publicly available data repository system, because there is no such a data repository system in the country. However, the data are available from the authors upon request.

Ethical Approval
The study was approved by Institutional Review Board (IRB) of Addis Ababa University, School of Medical Laboratory Science. In addition, official letter permission was obtained from the study site. Moreover, permission was obtained from respective health facilities.

Consent
Written consent was obtained from all study subjects for those who could write and read; for the illiterate ones they are verbally informed by data collector, and then informed consent was obtained. The authors kept the privacy and confidentiality of the study participants throughout the study process.

Conflicts of Interest
The authors declare that they have no conflicts of interest.