Associations of Ureaplasma status with early white blood cell count and interstitial radiographic changes in extremely preterm infants by gestational age

Ureaplasma species (spp.) have been associated with changes in white blood cell (WBC) count and radiographic abnormalities in preterm infants, but whether these associations differ according to gestational age (GA) remains unclear. This study aimed to evaluate whether GA modifies the associations of Ureaplasma status with early WBC count and interstitial radiographic changes in extremely preterm infants (EPIs). A retrospective cohort study was performed among infants born at < 28 weeks’ GA between September 2018 and August 2025. Infants who underwent nasopharyngeal swab collection for Ureaplasma screening within 12 h after birth were included. Ureaplasma spp. were detected using simultaneous amplification and testing (SAT). WBC count was obtained from complete blood counts within 24 h after birth, and chest X-ray was obtained on postnatal day 7. The primary outcome was WBC count; the secondary outcome was interstitial radiographic changes. Interaction analyses were performed using linear regression for WBC count and Firth’s penalized logistic regression for radiographic interstitial changes because of sparse data. A total of 251 EPIs were included. Within each GA subgroup, Ureaplasma- positive infants had significantly higher WBC counts and a higher frequency of interstitial radiographic changes than Ureaplasma- negative infants (all P < 0.05). Among Ureaplasma- positive infants, those born at < 25 weeks’ gestation had higher WBC count (22.0 × 10 9 /L vs. 16.5 × 10 9 /L), and a higher rate of interstitial radiographic changes (59.3% vs. 19.4%) than those born at ≥ 25 weeks’ gestation (all P < 0.05). In the multivariable linear regression model, Ureaplasma status was independently associated with a higher early neonatal WBC count at the mean GA (β = 11.06, 95% CI 8.16 to 13.97, P < 0.001). A significant interaction between Ureaplasma status and GA was observed ( β = −2.45, 95% CI − 4.68 to − 0.22, P = 0.031). For interstitial radiographic changes, the interaction estimate was imprecise and did not provide reliable evidence of GA-dependent effect modification. In exploratory stratified analyses, subgroup-specific estimates were also imprecise, so no firm conclusions could be drawn for associations within each gestational-age subgroup. Among EPIs, the association between Ureaplasma status and early postnatal WBC count was significantly modified by GA, with stronger associations observed at lower GA. Although evidence for GA-dependent effect modification on interstitial radiographic changes was inconclusive because of sparse data, exploratory analyses suggested a possible positive association between Ureaplasma status and interstitial radiographic changes, with imprecise estimates.