Open Access

Ulrich H. Thome, Jens Dreyhaupt, Orsolya Genzel-Boroviczeny, Bettina Bohnhorst, Manuel Schmid, Hans Fuchs, Oliver Rohde, Stefan Avenarius, Hans-Georg Topf, Andrea Zimmermann, Dirk Faas, Katharina Timme, Barbara Kleinlein, Horst Buxmann, Wilfried Schenk, Hugo Segerer, Norbert Teig, Benjamin Ackermann, Roland Hentschel, Matthias Heckmann, Rolf Schlösser, Jochen Peters, Rainer Rossi, Wolfgang Rascher, Ralf Böttger, Jürgen Seidenberg, Gesine Hansen, Harald Bode, Maria Zernickel, Rainer Muche, Helmut D. Hummler

• Background: Levels or fluctuations in the partial pressure of CO₂ (PCO₂) may affect outcomes for extremely low birth weight infants. Objectives: In an exploratory analysis of a randomized trial, we hypothesized that the PCO₂ values achieved could be related to significant outcomes. Methods: On each treatment day, infants were divided into 4 groups: relative hypocapnia, normocapnia, hypercapnia, or fluctuating PCO₂. Ultimate assignment to a group for the purpose of this analysis was made according to the group in which an infant spent the most days. Statistical analyses were performed with analysis of variance (ANOVA), the Kruskal-Wallis test, the χ² test, and the Fisher exact test as well as by multiple logistic regression. Results: Of the 359 infants, 57 were classified as hypocapnic, 230 as normocapnic, 70 as hypercapnic, and 2 as fluctuating PCO₂. Hypercapnic infants had a higher average product of mean airway pressure and fraction of inspired oxygen (MAP × FiO₂). For this group, mortality was higher, as was the likelihood of having moderate/severe bronchopulmonary dysplasia (BPD), necrotizing enterocolitis (NEC), and poorer neurodevelopment. Multiple logistic regression analyses showed an increased risk for BPD or death associated with birth weight (p < 0.001) and MAP × FiO₂ (p < 0.01). The incidence of adverse neurodevelopment was associated with birth weight (p < 0.001) and intraventricular hemorrhage (IVH; p < 0.01). Conclusions: Birth weight and respiratory morbidity, as measured by MAP × FiO₂, were the most predictive of death or BPD and NEC, whereas poor neurodevelopmental outcome was associated with low birth weight and IVH. Univariate models also identified PCO₂. Thus, hypercapnia seems to reflect greater disease severity, a likely contributor to differences in outcomes.