Diffusion tensor imaging in extremely low birth weight infants managed with hypercapnic vs. normocapnic ventilation.

Abstract

Permissive hypercapnia is a ventilatory strategy used to prevent lung injury in ventilated extremely low birth weight (ELBW, birth weight ≤1,000 g) infants. However, there is retrospective evidence showing that high CO2 is associated with brain injury. The objective of this study was to compare brain white matter development at term-equivalent age in ELBW infants randomized to hypercapnic vs. normocapnic ventilation during the first week of life and in healthy non-ventilated term newborns. Twenty-two ELBW infants from a randomized controlled trial were included in this study; 11 received hypercapnic (transcutaneous PCO2 [tcPCO2] 50-60 mmHg) ventilation and 11 normocapnic (tcPCO2 35-45 mmHg) ventilation during the first week of life while still intubated. In addition, ten term healthy newborns served as controls. Magnetic resonance imaging (MRI) with diffusion tensor imaging (DTI) was performed at term-equivalent age for the ELBW infants and at approximately 2 weeks of age for the control infants. White matter injury on conventional MRI was graded in the ELBW and control infants using a scoring system adopted from literature. Tract-based spatial statistics (TBSS) was used to evaluate for differences in DTI measured fractional anisotropy (FA, spatially normalized to a customized template) among the ELBW and term control infants. Conventional MRI white matter scores were not different (7.3 ± 1.7 vs. 6.9 ± 1.4, P = 0.65) between the hypercapnic and normocapnic ELBW infants. TBSS analysis did not show significant differences (P < 0.05, corrected) between the two ELBW infant groups, although before multiple comparisons correction, hypercapnic infants had many regions with lower FA and no regions with higher FA (P < 0.05, uncorrected) compared to normocapnic infants. When compared to the control infants, normocapnic ELBW infants had a few small regions with significantly lower FA, while hypercapnic ELBW infants had more widespread regions with significantly lower FA (P < 0.05, fully corrected for multiple comparisons). Normocapnic ventilation vs. permissive hypercapnia may be associated with improved white matter development at term-equivalent age in ELBW infants. This effect, however, was small and was not apparent on conventional MRI. Further research is needed using larger sample sizes to assess if permissive hypercapnic ventilation in ELBW infants is associated with worse white matter development.