Malformation risks higher among infants exposed in utero to TMP-SMX vs beta-lactams

First-trimester exposure to trimethoprim-sulfamethoxazole (TMP-SMX) poses a much higher risk of severe cardiac malformation and cleft lip and palate, among others, when compared with exposure to beta-lactam antibiotics during the same period, according to a study.

Researchers used data from the Merative MarketScan Commercial Database (2006-2022) and looked at pregnant women aged 15–49 years who were treated for urinary tract infection (UTI) and linked liveborn infants.

First-trimester antibiotic prescription fill of nitrofurantoin, TMP-SMX, fluoroquinolones (ciprofloxacin, levofloxacin, ofloxacin), and beta-lactams was evaluated in relation to congenital malformations (any and by organ system).

The analysis included 71,604 eligible pregnancies (median maternal age 30 years), of which 42,402 (59.2 percent) were exposed to nitrofurantoin, 3,494 (4.9 percent) to TMP-SMX, 3,663 (5.1 percent) to fluoroquinolone, and 22,045 to beta-lactams. The median gestational age was 62, 26, 18, and 63 days, respectively.

The absolute risk of any malformation was 19.8 per 1,000 infants with beta-lactams, 21.2 per 1,000 infants with nitrofurantoin, 23.5 per 1,000 infants with fluoroquinolones, and 26.9 per 1,000 infants with TMP-SMX. With beta-lactams as the comparator, the risk of any congenital malformation was elevated with TMP-SMX (risk ratio [RR], 1.35, 95 percent confidence interval [CI], 1.04–1.75) but similar with nitrofurantoin (RR, 1.12, 95 percent CI, 1.00–1.26) and fluoroquinolones (RR, 1.18, 95 percent CI, 0.87–1.60).

Specifically, TMP-SMX was associated with increased risks of severe cardiac malformations (RR, 2.09, 95 percent CI, 1.09–3.99), other cardiac malformations (RR, 1.52, 95 percent CI, 1.02–2.25), and cleft lip and palate (RR, 3.23, 95 percent CI, 1.44–7.22) compared with beta-lactams. However, the corresponding risk difference estimates for these specific malformations included the null.

Consistent results were obtained across sensitivity analyses.