Risk factors for periventricular-intraventricular haemorrhage severity in preterm infants: a propensity score-matched analysis.

Background: Most previous studies comparing etiological studies in infants with and without periventricular-intraventricular haemorrhage (PV-IVH) concluded that younger gestational age (GA) was associated with a higher prevalence rate of PV-IVH. However, only a few studies have examined the risk factors associated with the severity of PV-IVH after removing the influence of GA. Therefore, we investigated the risk factors apart from GA for PV-IVH severity in preterm infants less than 28 weeks.

Methods: This was a retrospective case-control study of preterm infants born in West China Second Hospital with PV-IVH between 2009 and 2020. PV-IVH was defined using cranial ultrasound screening. Preterm infants were divided into no PV-IVH and PV-IVH groups, and preterm infants with PV-IVH were divided into mild and severe PV-IVH groups. Groups were matched in a 1:1 ratio using propensity score calculated from GA. Variables were collected from infant-mother pairs. A stepwise forward multivariate logistic regression model was adopted to select factors that affected PV-IVH in preterm infants.

Results: A total of 429 preterm infants were included. The total incidence of PV-IVH in preterm infants was 55.6%, and the incidence of mild and severe PV-IVH was 28.7% and 26.9%, respectively. We matched 162 infants with no PV-IVH with 162 infants with PV-IVH. The results suggested that electrolyte disorder (OR 2.79, 95% CI: 1.34-5.77), early-onset sepsis (OR 1.76, 95% CI: 1.01-3.08), thrombocytopenia (OR 2.87, 95% CI: 1.10-7.48), invasive mechanical ventilation (OR 4.21, 95% CI: 1.86-9.55), and male sex (OR 2.16, 95% CI: 1.29-3.60) were independently associated with PV-IVH. Then, we matched 87 infants with mild PV-IVH with 87 infants with severe PV-IVH. The results suggested that electrolyte disorder (OR 2.88, 95% CI: 1.29-6.45), thrombocytopenia (OR 5.73, 95% CI: 1.91-17.14), and invasive mechanical ventilation (OR 10.54, 95% CI: 1.16-95.85) were independently associated with severity of PV-IVH.

Conclusions: Regardless of GA, electrolyte disorder, early-onset sepsis, thrombocytopenia, invasive mechanical ventilation, and male sex contributed to PV-IVH in preterm infants, and electrolyte disorder, thrombocytopenia, and invasive mechanical ventilation contributed to severe PV-IVH. These risk factors may combine to predict the incidence of PV-IVH in preterm infants.