Exploring the impact of spontaneous intestinal perforation and necrotizing enterocolitis treatment modality on shunt outcomes in premature infants.

OBJECTIVE

Spontaneous intestinal perforation (SIP), necrotizing enterocolitis (NEC), and posthemorrhagic hydrocephalus (PHH) are common complications in premature infants. SIP and NEC may require surgery, including peritoneal drainage (PD) and/or laparotomy. PHH typically requires shunting, generally using either a ventriculoperitoneal (VPS) or ventriculoatrial (VAS) shunt. The authors hypothesized that prior abdominal instrumentation with possible adhesion formation in patients undergoing laparotomy may affect VPS placement and longevity. This study examined the impact of SIP or NEC treatment on shunt outcomes in 43 premature infants.

METHODS

A retrospective review of infants treated for SIP or NEC and PHH at the authors' institution from 2007 to 2024 was conducted. SIP/NEC treatment modality and shunt outcomes were extracted from the electronic medical record. Categorical variables were analyzed using the chi-square or Fisher's exact test and nonparametric variables were analyzed using the Mann-Whitney U-test.

RESULTS

Forty-three eligible patients were identified (23.3% female, 76.7% Black), the majority of whom were diagnosed with SIP (72.1%). Most patients were treated with laparotomy (65.1%), with only 34.9% receiving definitive PD. A VPS was the predominant shunt type (81.4%), with a VAS being less common (18.6%). Among patients with VPS placement, 51.4% required distal (peritoneal catheter) revision. When comparing shunt outcomes between infants treated with laparotomy versus PD, primary shunt type selection was significantly different, with all cases of initial VAS placement occurring in the laparotomy group (p = 0.04). Patients who underwent laparotomy were also significantly older at initial shunt placement (median 126.5 vs 94 days, p < 0.01) and had a longer period between SIP/NEC treatment and shunt placement (median 100 vs 77 days, p = 0.04). There was no significant difference in the rate of distal VPS revision (p = 0.63), number of VPS revisions required (p = 0.38), or rates of conversion from a VPS to a VAS (p = 0.14) between the laparotomy and PD cohorts. However, infants who underwent definitive PD were significantly more likely to have a VPS maintained, even if revision was required, without conversion to a VAS (80.0% vs 39.3%, p = 0.01).

CONCLUSIONS

In this retrospective review of 43 premature infants with SIP or NEC and PHH, definitive PD was associated with higher rates of initial VPS placement and long-term maintenance compared with laparotomy. Although limited by a small sample size, these findings suggest that initial SIP/NEC treatment type may influence VPS longevity. Recognizing the role of prior abdominal surgery may inform neurosurgical decision-making and minimize shunt revisions.