Aim: Safe limits of arterial partial pressure of carbon dioxide (PaCO2) and acidosis in premature infants are not well defined. Both respiratory and systemic illness along with center-specific ventilation strategies contribute to PaCO2 fluctuations and acid-base imbalances during the critical time period of first 72 h of life. This study evaluated the association between early blood gas parameters and intraventricular hemorrhage (IVH) in preterm infants.
Methods: This retrospective observational study included neonates with a gestational age (GA) of ≤29 wks, who had at least 7 blood gas analysis done within the first 72 h of life. By adjusting for known variables that predispose to IVH, multivariable logistic regression analysis was used to study the association of PaCO2 and acid-base measures with the risk of IVH.
Results: Between 2013-2016, among 272 neonates who met inclusion criteria and were assessed for IVH on cranial ultrasound within first week of life, 101 neonates [mean GA of 25 ± 1.5 wks] had IVH and 171 neonates [mean GA of 25 ± 1.6 wks] had normal scans. After adjustment for confounding variables, higher values of maximum lactate (OR = 1.18, 95% CI = 1.1-1.3, p < .0001) and maximum base deficit (OR = 1.19, 95% CI = 1.1-1.2, p < .0001) within 72 h of life increased the likelihood of any grade of IVH. However, time-weighted average PaCO2, maximum and minimum PaCO2 had no statistically significant effect on the risk of IVH. The relationship remained unchanged even when moderate-severe IVH was considered as the primary outcome.
Conclusion: Severe metabolic acidosis rather than hypo/hypercapnia during the first 72 h of life was associated with higher odds of IVH in infants born at ≤29 wks of gestation. Future studies determining levels of PaCO2 that is safe for premature brain would need to control for the metabolic component of acidosis.
Keywords: Acute brain injury; blood gas; carbon dioxide; extreme prematurity; intraventricular hemorrhage; metabolic acidosis.