Study area and study design
This study was conducted in Addis Ababa, the capital city of Ethiopia from March to May, 2019. Health facility based cross-sectional study design was employed to recruit pregnant women by simple random sampling method from antenatal care centers of two government hospitals namely: Tikur Anbessa Speciallized Hospital (TASH) and Zewditu Memorial Hospital (ZMH) and two private health care facilities: Ananiya, and Hemon mother and child health care specialty centers located in Addis Ababa.
Sample size determination and recruitment of study participants
The sample size was calculated using single population proportion formula based on the previous report of 21.2% of ASBU in the northern Ethiopia [22]. With 95% confidence interval, 10% non-response rate and 5% marginal error, the calculated sample size was 283.
First, total sample size calculated was allocated into the study hospitals proportional to the number of pregnant women attending selected antenatal care centers. Once clear explanation about the study was given to each pregnant women coming to antenatal care center for routine antenatal care service, they were requested to be involved in the study. Pregnant women excluded from the study were those <18 years of age, those with symptomatic urinary tract infection, who took two or more glasses of fluid one hour before clinic attendance, who were treated with antimicrobials during the last one week and women with current symptoms of sexually transmitted infections. Among those who volunteered to participate and fulfill the inclusion criteria, half of them were recruited randomly. Two rolled paper coded as number one, and zero were offered to participants to pick one and those who picked a rolled paper labeled with number one were recruited.
Data and sample collection
Information about age, residence, religion, marital status, educational status, monthly income, frequency of sexual intercourse per week, frequency of vaginal douching per day, history of urinary tract infection in this pregnancy; past history for chronic diseases (diabetes mellitus, hypertension), history of sexually transmitted diseases, stage of pregnancy and parity were collected from each woman through interview during sample collection. Current status of the pregnant women for HIV/AIDS and Hepatitis B surface antigen was reviewed from the recent information sheets of pregnant women’s chart. After instructing the women to take perineal care, a clean-catch midstream urine samples were collected into sterile containers and transported to Microbiology Laboratory of Aklilu lemma Institute of Pathobiology in ice box within 3-4 hours of collection.
Urine sample processing and pathogen isolation
A loopful (10 micro liter) of well mixed urine sample was inoculated on to plate count agar and grown for 24h at 37°C. Number of colonies grown on each plate was counted to determine colony forming units per milliliter (CFU/ml) of the urine sample. A urine sample containing ≥ 105 cfu/ml was considered as having ASBU [9] . For determination of the type of organisms involved, colonies from plate count agar plates were picked and inoculated to various selective media such as Eoseine Methylene Blue (EMB) agar to differentiate Escherchia coli, Klebsiella species, and Citrobacter species; Cystiene-lactose-electrolyte-deficient (CLED) medium to differentiate proteus species; biosculine agar and enterococcus agar to identify Enterococcus species after incubating for 24 hrs at 37°C. In addition, urine samples from women with ASBU were also directly plated to selective media. Further, biochemical tests like triple sugar iron agar, lysine iron agar, citrate, urease, indole, potassium hydroxide (KOH), and catalase tests were employed to further identify type of bacterial species involved[23-25].
Investigation of antimicrobial susceptibility of bacterial isolates
The antimicrobial susceptibility tests of the isolates were performed according to the Clinical Laboratory Standards Institute (CLSI) guideline[26] using Kirby-Bauer disk diffusion method on Muller-Hinton agar plates (Oxoid, CM0337 Basingstoke, England). The following antimicrobial (Sensi-Discs, Becton, Dickinson and Company, Loveton, USA) and disc potencies were used: amoxicillin + clavulanic acid (Amc) (20/10 µg), ampicillin (Amp) (10 µg), cephalothin (CF) (30 µg), ceftriaxone (Cro) (30 µg), ciprofloxacin (Cip) (5 µg), gentamicin(Gm) (10 µg), streptomycin(S) (10 µg), sulfisoxazole (G), sulfamethoxazole+ trimethoprim (Sxt) (23.75/1.25 µg), tetracycline (Te) (30 µg), chloramphenicol (C), (30 µg), vancomycin (Va) (30 µg), kanamycin (K), (30 µg) erthromycin (Ery) (15 µg), Oxacillin, (ox), (1µg), clindamycin (DA) (2 µg), penicillin (P) (10 µg). Escherichia coli ATCC 25922 and Staphylococcus aureus ATCC25923 were used as a quality control organisms for gram negative and gram positive organisms, respectively. The interpretation of the categories of susceptible and resistant was based on the CLSI guidelines [26]. Isolates were regarded as multi-drug resistant (MDR) when they were resistant to at least two or more antimicrobials belonging to different classes [27].
Data analysis
Descriptive statistics: mean, standard deviation, frequency and percentage were used to summarize different variables. Logistic regression was used to model association of pre-defined independent variables with the binary outcome. P-value less than 0.05 was as indicator for statistical significance.