Microbial resistance to antibiotics has always been the elephant in the room that no one wants to confront. Hospitals use the mantra of antimicrobial stewardship by attempting to target antibiotic use, but this endeavour is less in vogue for outpatient infections. We have seen major resistance to antimicrobials occurring with malaria, tuberculosis and gonorrhoea, but to date we have managed to have other antimicrobials to substitute. The same holds for antibiotics to treat urinary tract infections (UTI). However, this situation is not expected to last forever. Exposure to antibiotics is prevalent, as we are even exposed to the residues from antibiotics given to animals.¹ Antibiotic use in pregnancy potentially adds risk to the fetus with changes in maternal gastrointestinal and vaginal microbes.²

The frequency of UTIs has produced a situation where antimicrobial resistance has already appeared, with Escherichia coli resistance to fluoroquinolones and penicillins. This situation has brought forth strategies to limit inappropriate antibiotic use and reduce the risk of antimicrobial resistance for UTIs among women.^{3, 4} In a report in this issue,⁵ the presence of extended-spectrum β-lactamase (ESBL) was examined among a select group of very high-risk pregnant women in Korea. A higher rate of ESBL was found among E. coli isolates in 2015–20 than in 2010–15. The presence of ESBL in E. coli was related to previous antibiotic use, as one might expect. The concern with ESBL is that the antimicrobial resistance carries over to many of the usual antibiotics used to treat UTIs. An additional concern, found in the study, was that those women with ESBL resistance delivered infants with a higher rate of early-onset neonatal sepsis. This indicates a potentially increased invasive potential for ESBL isolates as well as a deleterious effect of antibiotic use in pregnancy.

This increased trend of ESBL resistant isolates among Gram-negative bacteria needs to be studied in Europe and the USA, as antibiotic use in Asian countries is often unregulated, leading to both higher antibiotic use and higher antibiotic resistance. Additionally, the select group of high-risk pregnancies that made up the study population in this report was unusual. The indication of admission to this high-risk unit had a rate of premature rupture of membranes of 39% and a preterm labour of 36%, in addition to an incompetent cervix, or short cervical length. Besides this, the study population had incidences of twins of 18%, triplets of over 1.4%, incompetent cervix of over 17%, and cervical cerclage of about 10%. This indicates that the study may not be fully reproducible in other populations. Still, the trend identified here bears watching in other Obstetrics and Gynaecology units, not only for infection of the mother, but for early-onset neonatal sepsis. The issue of early-onset neonatal sepsis is especially concerning because E. coli, either with or without ESBL, may eventually replace Group B Streptococcus as a leading cause of early-onset neonatal sepsis; indeed, this has already occurred in some neonatal units.