Association of timing of surgery and outcomes in preterm infants with surgical necrotizing enterocolitis and intestinal perforation

Abstract

Background: To investigate the association between the timing of surgery from the day of NEC/SIP diagnosis and clinical outcomes in preterm infants. Study Design: A retrospective cohort study comparing clinical outcomes of infants undergoing laparotomy at three clinically relevant time points (less and more than 48 hours, 96 hours, and 168 hours [7 days]) following NEC/SIP diagnosis. Results: Infants with NEC/SIP (N = 97) receiving surgical invention >96 hours (34/97) had significantly lower gestational age (25.5 weeks [24.0; 26.9] vs 27.0 [25.0; 31.3]; p = 0.006), had lower birth weight (687 grams [600; 902] vs 940 [710; 1495]; p<0.001), had pneumoperitoneum less often on the abdominal x-ray (29.4% vs 57.1%, p = 0.017), had hemorrhagic (p = 0.04) and reparative (p = 0.003) lesions more often on intestinal histopathology, had PDA diagnosed more often (76.5% vs 50.8%, p = 0.02), required assisted ventilation more frequently (p = 0.013), and received parenteral nutrition for longer duration (112 days [76.5; 145] vs 65.0 [23.0; 112], p = 0.004) following surgery compared to the infants receiving surgical intervention before 96 hours (n = 63/97). In NEC-only sub-cohort, infants receiving laparotomy >48 hours (n = 29/75) had lower median gestational age, lower birth weight, less pneumoperitoneum, and higher acute kidney injury than those receiving surgery <48 hours. On logistic regression, the odds of death were not significantly different (OR 0.65 [0.28, 1.54], p = 0.32) for infants receiving laparotomy ≤48 hours following NEC/SIP compared to subjects undergoing surgery >48 hours. The odds of intestinal failure (>60 days of parenteral nutrition) were 4.5 times (CI 1.56, 14.3), p = 0.005) higher for those having surgery >96 hours from NEC/SIP diagnosis. Conclusion: There was no significant difference in death among infants receiving surgery within 48 hours following surgical NEC/SIP diagnosis compared to those receiving surgery at ≥ 48 hours of diagnosis. However, infants receiving surgery >96 hours were more likely to receive parenteral nutrition for longer time. A prospective study is needed to understand the continuous relationship between time to surgery and outcomes.

Keywords: laparotomy; necrotizing enterocolitis; neonate; outcomes; prematurity.

The time to surgery from NEC onset variable ranges from widely from 1 to 3456 hours. If we were to subset it by those under 200 hours, we can see:

The distribution is quite even. Those under 48 hours binary exposure represent comparing the bottom 50^th percentile to the upper 50^th percentile, those under 96 hours binary exposure represent comparing the bottom 60~ percentile to the upper 40^th, those under 168 represent comparing the bottom 75~ percentile to the upper 25^th. Any cutoff above 168 hours we result in too small sample size to detect a meaningful difference.

Figure 1:

The patient flow chart showing infants with surgical NEC included in the study.

Figure 2:

Association between time to surgery (hours) from NEC diagnosis and probability of in-hospital death for those operated within 7 days of diagnosis. In continuous modeling of the association between time to surgery and outcomes in the initial seven days (<168 hours), the adjusted OR of mortality is 1.36 [0.79–2.25], p=0.264 per standard deviation of time to surgery adjusted for birth weight and Penrose drain placement.