Background Cesarean section (CS) is often complicated by surgical site infection (SSI) that may happen to a woman within 30 days after the operation. This study was conducted to estimate the prevalence of SSI and identify the factors associated with SSI. Methods A hospital-based analytic cross-sectional study was conducted based on the review of medical records of 1069 women who underwent CS in two public hospitals in Harar city. The post-CS SSI is defined when it occurred within 30 days after the CS procedure. Factors associated with SSI were identified using a multivariable binary logistic regression analysis. The analysis outputs are presented using an adjusted odds ratio (aOR) with a corresponding 95% confidence interval (CI). All statistical tests are defined as statistically significant at P-values<0.05. Results The prevalence of SSI was 12.3% (95% confidence interval (CI): 10.4, 14.4). Emergency- CS was conducted for 75.9% (95% CI: 73.2, 78.3) of the women and 13.2% (95% CI: 11.3, 15.4) had at least one co-morbid condition. On presentation, 21.7% (95% CI: 19.3, 24.3) of women had rupture of membrane (ROM). Factors significantly and positively associated with post-CS SSI include general anesthesia (aOR = 2.0, 95%CI: 1.10, 2.90), ROM (aOR = 2.27, 95%CI: 1.02, 3.52), hospital stay for over 7 days after operation (aOR = 3.57, 95%CI: 1.91, 5.21), and blood transfusion (aOR = 4.2, 95%CI: 2.35, 6.08). Conclusion The prevalence of post-CS SSI was relatively high in the study settings. Screening for preoperative anemia and appropriate correction before surgery, selection of the type of anesthesia, close follow-up to avoid unnecessary prolonged hospitalization, and careful assessment of membrane status should be considered to avoid preventable SSI and maternal morbidity. Copyright:
A fifteen months data, from October 11, 2018 to December 31, 2019, was extracted from patient records during March 1 to March 15, 2020 from two public hospitals in Harar city, namely Jugol General Hospital and Hiwot Fana Specialized University Hospital (HFSUH), Eastern Ethiopia. Harar city is located at 526 km towards the east of Addis Ababa, the Ethiopian capital. According to the 2007 census conducted by the Central Statistical Agency, the total population of the region is estimated to be 183, 415 [15]. Ethiopia’s health service is structured into a three tier system: primary healthcare unit (health post (serve 3000–5000 people), health center (serve 15000–40,000 people), and primary hospital (serve 60,000–100,000 people)), secondary level healthcare (general hospital: serve 1.0–1.5 million people), and tertiary level healthcare (specialized hospital: 3.0–5.0 million people). When this study was conducted, there were two public hospitals (general and specialized hospitals), two private general hospitals, one police-, and one non-governmental fistula hospital in the region. In addition to these hospitals in the city, there were 29 private clinics, 26 health posts, eight health centers, and one regional laboratory. In the two hospitals, in terms of human resource to provide CS services, there were 7 gynecology and obstetrics senior specialists, 33 residents, and 3 integrated emergency surgical officers (only in Jugol General Hospital). Hospital-based analytic cross-sectional audit of patient records (from October 11, 2018 to December 31, 2019) was conducted to estimate the magnitude of post-CS SSI and identify the associated factors among women who underwent CS in two public hospitals. The source population for this study was all women who underwent CS at the public hospitals in Harar city. Those women for whom CS procedure was conducted in the specified period are considered as our study population. We reviewed all the medical records of women who underwent CS in the specified period. Medical records with incomplete values for the outcome and important predictor variables were excluded. We conducted a census of all cesarean sections conducted in the two hospitals during the period of October 11, 2018 to December 31, 2019. Over this period, we found 1069 complete medical records documented with CS birth (806 records from HFSUH and 263 from Jugal General Hospital). We identified records of eligible women by reviewing hospital registry books in the operation room and at labor and obstetric wards. Data collection checklists are developed after reviewing variables in patient’s medical records. Five diploma midwives working at the study hospitals reviewed patient medical records and collected data using the data collection checklists. Two senior midwives, one at each hospital, supervised data collection process, and the principal investigator closely oversaw the overall data collection activity. The data collected included variables on characteristics related to demography, obstetrics, operation, comorbidity, and a post-CS SSI status. Age and residence are the only demographic characteristics for which values are collected. Data on obstetrics characteristics included parity (which refers to the number of pregnancies carried to fetal viability), gestational age (measured in weeks from last menstrual period to the date of CS), presence of labor before operation, duration of labor, membrane status before operation, duration of membrane rapture before operation, and chorioamnionitis (an inflammation of the fetal membranes, amnion and chorion, mostly due to bacterial infection). Regarding comorbidity data on cardiac disease, diabetes mellitus, hypertension, Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome (HIV/AIDS), and anemia are collected. Variables collected to characterize the CS operation were the professional level of a physician conducting the operation, operation type (elective versus emergency: a CS is emergency operation when it was done due to unexpected or acute obstetric emergencies, when the mother’s or fetus’s life or well-being is in direct jeopardy or risk [16]), duration of the operation, type of anesthesia (regional versus general), prophylactic antibiotic, duration of the operation, pre-operation hematocrit count, blood transfusion, type of abdominal incision (lower transverse versus vertical), and post-operation hospital stay in days), and a post-CS SSI status which was defined according to the Center for Disease Control and Prevention’s (CDC) standard [5]. Post-CS SSI was the outcome variable and measured using five items with a yes/no response. The items used include: fever≥38°C, purulent discharge from surgical site, at least one sign of inflammation (pain/tenderness, localized swelling, redness or heat), abscess, and diagnosed wound infection by a physician/surgeon. A post-CS SSI was defined if there was at least one ‘yes’ response to any of the five items within 30 days after operation [5]. The data were entered into Epi-data version 3.1 and exported to Stata version 16 software for analysis. Descriptive data analysis was made using frequency and percentages. Difference of proportion between two samples was conducted to see if there is a significant difference in the proportion of post-CS SSI between population groups. Bivariable and multivariable binary logistic regression analyses were conducted to identify factors associated with post-CS SSI. Variables that showed statistically significant association (p<0.25) in the bivariable binary logistic regression are entered in the multivariable model. Model fitness of the final model was checked using the Hosmer-Lemeshow test in Stata (post-estimation command: estat gof) and it demonstrated a good fit with Pearson chi2 (133) = 149.7 and P-value = 0.153. Adjusted odds ratio (aOR) along with 95% CI was estimated to identify factors associated with post-CS SSI. Multicollinearity was checked using variance inflation factor (VIF) (higher VIF suggests possible existence of collinearity), and we removed a variable, ‘presence of anemia’, which showed collinearity with another variable, ‘co-morbid conditions’. Statistical estimates were considered as significant at P-value < 0.05. This study obtained ethical approval from the Haramaya University, College of Health and Medical Sciences, Institutional Health Research Ethics Review Committee (IHRERC) with a reference number of IHRERC/016/2020. Formal letters of permission were written from the College to administrators of both Jugol General Hospital and HFSUH. Data collection was started after obtaining informed, written, voluntary and signed consent from the medical directors of the two hospitals involved. The ethics review committee waived obtaining informed consent from patients to access data. We collected the data anonymously without patient identifying information to maintain confidentiality of patient information.