(A-171) Comparative Risk of Preterm Birth Among Pregnant Individuals Treated for Urinary Tract Infection with Different Antibiotics

Background: Guidelines recommend screening and treatment of urinary tract infections (UTIs), including asymptomatic bacteriuria, during pregnancy due to the risk of serious maternal and infant consequences such as preterm birth. However, evidence is scarce about the comparative effectiveness of different antibiotics to clear the infection and ultimately prevent preterm birth. Furthermore, it is unclear whether timing of UTI during pregnancy may modify the risk of preterm birth.

Objective: To evaluate the risk of preterm birth by antibiotic among pregnant individuals with UTI, overall and by gestational age at UTI.

Methods: We conducted an active comparator, new-user cohort study using the Merative MarketScan Commercial Database (2006-2022). We included singleton pregnancies diagnosed with UTI in the outpatient setting and dispensed an oral antibiotic at ≥6 weeks gestation. We required evidence of pregnancy-related care by the antibiotic dispensing (index) date. We excluded those with recent UTI, pyelonephritis, kidney stones, or hospitalization; autoimmune conditions; or fetal chromosomal abnormalities. Preterm birth was defined as live birth from 20 to <37 weeks gestation. We used log-binomial regression to estimate risk differences (RD) and risk ratios (RRs) with 95% confidence intervals (CIs). We used standardized mortality ratio weights to account for potential confounders and inverse probability weights to account for competing risks (i.e., spontaneous abortion, termination, stillbirth). We performed analyses overall and by trimester of index UTI. Sensitivity analyses restricted the cohort to the ICD-10 era.



Results: We identified 150,218 eligible pregnancies treated for UTI (66% nitrofurantoin, 34% β-lactam). Median age of the birthing parent was 29 years (IQR, 24-33). Median gestational age at index differed by antibiotic (nitrofurantoin, 90 days; β-lactams, 85 days). Among live births (93.3% of sample), 12.7% were preterm. Compared to nitrofurantoin, the risk of preterm birth was similar for β-lactams (weighted RD per 100, 0.01; 95% CI: -0.4, 0.4). In subgroup analyses, estimates did not differ by trimester. Weighted RD estimates per 100 births were -0.3 (95% CI: -0.8, 0.2) during the first trimester; 0.4 (95% CI: -0.4, 1.1) during the second trimester; and 0.3 (95% CI: -0.7, 1.4) during the third trimester for β-lactams vs nitrofurantoin. Results were consistent on the relative scale as well as in sensitivity analyses.

Conclusions: Among pregnant individuals treated for UTI, the risk of preterm birth was similar for nitrofurantoin and β-lactams–both commonly used antibiotics. Risk did not vary by trimester of UTI.