Search

Objectives: To test the hypothesis that the lack of protective effect of antenatal steroids (ANS) on bronchopulmonary dysplasia (BPD) could derive from overadjustment during analysis, caused by controlling for factors that are intermediate in the causal pathway between treatment and outcome.Methods. A cohort of infants 23 to 32 weeks of gestation, admitted to 10 tertiary-level neonatal units in Lombardy (Northern Italy) in 1999–2002, was prospectively studied; 1118 neonates out of 1314 survived to 36 weeks; 15.9% developed BPD (oxygen requirement at 36 weeks); 82% were treated with ANS.Results. In univariate analysis, ANS were not significantly protective against BPD; some intermediate factors (mechanical ventilation, greater severity of illness as measured by CRIB score, patent ductus arteriosus) were positively associated with (i.e. were risk factors for) BPD (OR=11.0, 1.46, 4.42 respectively, all P<0.001), and negatively associated with (i.e. prevented by) ANS (OR=0.58, 0.92, and 0.58 respectively, all P<0.001). In multiple logistic regression models without the above-mentioned intermediate risk factors, ANS-treated infants had a lower risk of BPD (OR 0.59; 95% CI 0.36, 0.95, P=0.03); other factors significantly associated to BPD were male sex (OR=2.07), late-onset sepsis (OR= 4.31), and birthweight (OR=0.62 for 100 g increase), all P<0.001. When also intermediate risk factors for BPD, which are prevented by ANS, were added to the model, ANS effect disappeared, thus demonstrating the existence of overadjustment.Conclusions. These results support the hypothesis that incorrect methods of analysis can obscure ANS protection against BPD. Using proper adjustment, ANS appeared to decrease the risk of BPD in this cohort.

Background: Bronchopulmonary dysplasia (BPD) remains a significant complication of prematurity, and wide variations of BPD rates in different neonatal units have been reported. These differences have been ascribed to different practices, especially in mechanical ventilation.Since mechanical ventilation is often started in the delivery room (DR), we wanted to investigate if different approaches to DR intubation affected the development of BPD.Methods: We examined a cohort of 1338 newborns, with birthweight < 1500g (mean 1086 g) and gestational age < 32 weeks (mean 28.7 weeks), admitted to 12 tertiary-level neonatal units in Lombardy (Northern Italy) participating in a local network in 1999–2002 and surviving to 36 weeks; 226 (16,1%) developed BPD defined as oxygen need at 36 weeks.Results: BPD rates significantly differed between hospitals (from 8.5% to 27.7%), as well as rates of intubation in the DR (from 28.8% to 73.6%), and rates of mechanical ventilation (from 32.2% to 84%). Centers with high intubation rates had higher ventilation (R2 = 0.7, linear regression analysis) and higher BPD (R2 = 0.35) rates. In univariate analysis, intubation in the DR was associated with increased risk of BPD (odds ratio 5.9), while surfactant administration did not affect BPD.Differences in individual infants characteristics (GA, BW, antenatal steroids, CRIB score) did not explain the differences in intubation rates across centers. Differences between centers were highly significant (P<0.001, logistic regression)Conclusions: These results support the hypothesis that differences in initial management of VLBWI between hospitals partly explain differences in BPD, and that these differences are not explained by case-mix, but probably by different policies of treatment. Moreover, this enlights the problem of a wide variation in practices, that deserves further investigation. In these situations, potentially better practices, including a more selective approach to the endotracheal intubation in the DR, could achieve better clinical results.