AFJOG

ORIGINAL RESEARCH Gestational age was confirmed either by the earliest available ultrasound or the earliest estimation based on menstrual dates or palpation. A detailed history was taken, and a focused clinical examination performed on all the participants. All participants were tested for SARS-CoV-2 using polymerase chain reaction (PCR). This was initially performed specifically for the study until it became standard of care with universal testing on all patients on admission. 12–14 SARS-Cov-2 immunoglobulin G (IgG) testing on blood samples was performed to determine SARS-CoV-2 exposure in asymptomatic women once the test became available from August 2021. 15 The latter was therefore only done on a subset of pregnant women. Other causes of PTL and PPROM were investigated as per unit protocol and included the following: white cell count (WCC) and C reactive protein (CRP), urine microscopy, culture, and sensitivity (MCS); and vaginal swabs for MCS which included testing for Gardnerella and Group B Streptococcus. The primary outcome measured was COVID-19 prevalence in PTL and PPROM as compared to the prevalence in patients at a similar gestation not in labour. Secondary outcomes were other causative factors of PTL and PPROM in our setting. Ethical Considerations Ethical and health authority regulatory permissions were obtained (Biomedical Research Ethics Council BREC/00001918/2020). Written informed consent was obtained from all participants. All residual clinical data was de-identified and captured with a unique study number. Data was collected in a password protected Excel spreadsheet on a dedicated computer behind firewall-protected servers. Sample size A minimum sample size of 100 women with a minimum of 50 cases and 50 controls were estimated using PASS 2021 software (Power Analysis and Sample Size Software NCSS, LLC. Kaysville, Utah, USA, ncss.com/software/pass.) based on a presumed prevalence of COVID-19 exposure in the case group of 39.5% and 15% in the control group from published studies done globally. 5–9 This sample size was required to achieve 80% power to detect an odds ratio of 3.7. The test statistic used was the two-sided Z test with pooled variance. The significance level of the test was targeted at 0.050. Statistical Analysis Data were analysed using IBM SPSS version 28. Frequency tables were used to describe categorical variables, while summary statistics such as mean and standard deviation, or median and interquartile range were used depending on whether the variables were normally distributed or not. Comparisons between groups were achieved for continuous normally distributed variables using t-tests, for numeric variables that were not normally distributed or counts using Mann-Whitney tests, and for categorical variables using Fisher’s exact 2-sided or Chi square tests as appropriate. RESULTS One hundred and six participants were recruited, of which 56 were in the case group and 50 in the control group. Table 1 shows the demographic characteristics. The mean age was 28 years (range 21-35 years). The median gestational age was 33 weeks (range 30-35 weeks) in both groups and 23 patients (21.7%) were nulliparous. Table 1. Demographic characteristics of studygroups Variable Cases Controls Total p-value Age Mean 27 30 28 0.054 16-19 6 (10.7%) 7 (14%) 13 (12.3%) 0.081 20-24 17 (30.4%) 6 (12%) 23 (21.7%) 25-34 25 (44.6%) 23 (46%) 48 (45.3%) 35-44 8 (14.3%) 14 (28%) 22 (20.7%) Variable Cases Controls Total p-value BMI* <25 15 (26.8%) 16 (32%) 31 (29.2%) 0.695 25-29.9 24 (42.9%) 16 (32%) 40 (37.7%) >=30 17 (30.3%) 18 (36%) 35 (33.1%) Gravidity Primigravid (1) 13 (23.2%) 8 (16%) 21 (19.8%) 0.530 Multigravid (2-4) 37 (66%) 38 (76%) 75 (70.1%) Grandmultigrav- id (5 and >) 6 (10.8%) 4 (8%) 10 (9.1%) Parity Nulliparous 14 (25%) 11 (22%) 25 (23.6%) 0.285 Parous (P1-P3) 42 (75%) 37 (74%) 79 (74.5%) Multiparous (P4 and >) 0 2 (4%) 2 (1.9%) Gestational age (weeks) Median 33 33 33 0.868 28-33 31 (55.4%) 29 (58%) 60 (56.6%) 0.777 34-37 25 (44.6%) 21 (42%) 46 (43.4%) *BMI- Body Mass Index Early PTL (28-33 weeks) occurred in 31 (55.4%) women and in 25 (23.6%) women PTL occurred between 34-37 weeks. Thirty- one women (29.2%) had a BMI <25kg/m 2 , 40 (37.7%) a BMI of 25-29.9kg/m 2 and 35 (33.1%) a BMI ≥30 kg/m 2 . Three women (2.8%) reported smoking and alcohol consumption respectively, and one woman (0.9%) reported recreational drug use. Recent coitus within the preceding five days was reported by 13 women (23.2%) in the case group compared to fivewomen (10%) in the control group, (p=0.070). Twenty-six women (46.4%) in the case group had human immunodeficiency virus (HIV) infection compared to 20 women in the control group (40%) (p = 0.505). Among HIV-infected participants, 40 women (87%) were virally suppressed (HIV viral load <50 copies/ml) and 6 women (13%) were not. A history of TB infection within the preceding two years was present in nine women (16.1%) in the case group and in one woman (2%) in the control group (p=0.018). Six of the 106 women (5.7%) had positive syphilis serology. A history of PTL in a previous pregnancy was present in 13 (23.2%) of the case group eight women (16%) of the control group, (p=0.465). Thirteen women (28.3%) in the control group had a history of caesarean delivery in a previous pregnancy compared to five women (8.9%) in the study group, (p=0.006). A history of gynecological complications such as ectopic pregnancies, terminations of pregnancies, and sexually transmitted in infections were comparable between the groups. Four women (7.1%) in the case group had a history of three previousmiscarriagesasopposedto thecontrol group inwhichno patients reported three previous miscarriages. Fifty-one women (48.9%) used Depo Provera as their method of contraception, while 37 (34.9%) did not use any form of contraception prior to the index pregnancy. The prevalence of COVID-19 infection was 14.6% compared to 8% in the control group (p = 0.503). Two participants, 1 (3.3%) in the case and 1 in the control group (2%), had a positive SARS-CoV-2 PCR test (p=0.68). Five women (23.8%) from the case and 4(12.5%) from the control group were SARS- CoV-2 IgG positive (p=0.456). Urine microscopy and culture (MC&S) was positive in 5 (8.9%) of the case group compared to 1 (0.5%) in the control African Journal of Obstetrics and Gynaecology | Volume 1 | Issue 2 | 2023 | 21