Quality improvement project to reduce intraventricular haemorrhage in very preterm infants failed due to increased life-sustaining intensive care at low gestational age.

Abstract

In October 2016, a single centre quality improvement programme 'IBIS' (from the German "Intraventrikuläre Blutungs-Inzidenz Senken"), an intraventricular haemorrhage (IVH) prevention bundle had the Specific Measurable Achievable Reasonable Time-bound (SMART) Aim to reduce IVH incidence from 29.6% to 18.5% within 2 years for very preterm (gestational age (GA) between 23 0/7 and 29 6/7). Monocentric retrospective cohort study of a failed quality improvement project using prospectively collected data including 1276 live-born very preterm infants between 2010 and 2023. Primary outcome measures were IVH, severe IVH and death or severe IVH. Statistical analysis included propensity score matching. IBIS, an ongoing IVH prevention bundle based on a systematic literature search starting in October 2016. Comparing pre-IBIS patients (2010-September 2016) to IBIS patients (October 2016-2023) revealed no reduction in IVH (25.1% vs 25.1%) or severe IVH (9.6% vs 9.1%). Instead, mortality (24.8% vs 13%) and delivery room mortality (9% vs 5%) were almost halved and fewer infants died after primary palliative care (8.5% vs 4.7%) or after redirection from intensive to palliative care (15.1% vs 7.9%). Longitudinal analysis revealed no trends for IVH or severe IVH over the entire period. Limiting the analysis to the IBIS period reveals a significant trend for fewer IVH (p=0.001). Propensity score matching revealed significant reduction for severe IVH (OR 0.62, 95% CI 0.39 to 0.99) and death or severe IVH (OR 0.42, 95% CI: 0.3 to 0.59) but not for IVH. The IBIS SMART Aim to reduce IVH failed. Simultaneously, survival was significantly increased by the increased provision of life-sustaining intensive care at lower GAs, which possibly led to an increased risk for acquiring IVH. Our results highlight the need for a better understanding of the effects of extending perinatal interventional activity to lower GAs on adverse outcome monitoring.