Mortality associated with laparotomy-confirmed neonatal spontaneous intestinal perforation: A prospective 5-year multicenter analysis

Background Spontaneous intestinal perforation (SIP) has been recognized as a distinct disease entity. This study sought to quantify mortality associated with laparotomy-confirmed SIP and to compare it to mortality of laparotomy-confirmed necrotizing enterocolitis (NEC). Methods Data were prospectively collected on 177,618 very-low-birth-weight (VLBW, 401-1500 g) neonates born between January 2006 and December 2010 admitted to US hospitals participating in the Vermont Oxford Network (VON). SIP was defined at laparotomy as a focal perforation of the intestine without features suggestive of NEC or other intestinal abnormalities. The primary outcome was in-hospital mortality. Results At laparotomy, 2036 (1.1%) neonates were diagnosed with SIP and 4076 (2.3%) with NEC. Neonates with laparotomy-confirmed SIP had higher mortality (19%) than infants without NEC or SIP (5%, P = 0.003). However, laparotomy-confirmed SIP patients had significantly lower mortality than those with confirmed NEC (38%, P < 0.0001). Mortality in both NEC and SIP groups decreased with increasing birth weight and mortality was significantly higher for NEC than SIP in each birth weight category. Indomethacin and steroid exposure were more frequent in the SIP cohort than the other two groups (P < 0.001). Conclusions In VLBW infants, the presence of laparotomy-confirmed SIP increases mortality significantly. However, laparotomy-confirmed NEC mortality was double that of SIP. This relationship is evident regardless of birth weight. The variant mortality of laparotomy-confirmed SIP versus laparotomy-confirmed NEC highlights the importance of differentiating between these two diseases both for clinical and research purposes.