Necrotizing enterocolitis with portal venous gas in a 780 g infant: drain, laparotomy, or wait
A 26-week, 780 g boy born by emergency cesarean for preeclampsia is now 11 days old. He has been fed with breast milk and has had two episodes of bloody stool. Today his abdomen is distended and shiny, and radiographs show pneumatosis intestinalis in the right lower quadrant with portal venous gas. Platelets are 58,000, lactate 3.8 mmol/L, and he requires dopamine at 8 mcg/kg/min to keep mean arterial pressure at 32 mmHg. He is intubated on high-frequency ventilation with an FiO2 of 0.45. Abdominal examination is soft enough to seem unremarkable to the resident, who suggests another abdominal film in six hours and continuing antibiotics.
1. How do you stage this infant's disease and what is your operative decision? Justify your choice among laparotomy, bedside peritoneal drain, and continued medical therapy.
Read the worked answer
Pneumatosis plus portal venous gas, thrombocytopenia at 58,000, lactate 3.8, and vasopressor dependence place this infant in Bell stage IIIA at minimum. Portal venous gas in a deteriorating preterm infant is a strong predictor of full-thickness necrosis, and the 'soft abdomen' on palpation is unreliable in a 780 g baby on high-frequency ventilation. Waiting six hours for another film is the wrong decision.
The genuine question is laparotomy versus primary peritoneal drainage. Randomized trials in extremely low birth weight infants with perforated or surgically indicated necrotizing enterocolitis found no significant difference in death or neurodevelopmental impairment between the two strategies, so neither is categorically superior. The practical framing is that a drain is a resuscitative bridge, not a definitive operation.
Given 780 g, HFOV at 0.45, and pressor dependence, I would place a bedside peritoneal drain under local anesthesia in the NICU while neonatology continues resuscitation, with a planned reassessment at 12 to 24 hours. If the drain returns feculent material, or the lactate and pressor requirement do not improve, I would proceed to laparotomy. The decision to operate is made now; only the route is staged.
2. At laparotomy you find 25 cm of frankly necrotic ileum with a single perforation; the adjacent bowel is dusky but not clearly dead. How do you decide between resection with anastomosis and resection with a stoma?
Read the worked answer
First, resect only what is unambiguously dead. Dusky but viable-looking segments are better managed with a planned second look at 24 to 48 hours than with wide margins that risk short bowel syndrome; the length you leave behind governs everything that follows.
Second, reconstruct based on three variables: residual bowel length, stability, and contamination. In a 780 g infant who has already lost 25 cm of ileum, a proximal stoma invites high-output losses, poor growth, and TPN-associated cholestasis, so the remaining length is precious. If the bowel ends are healthy and well perfused, the infant is not coagulopathic or acidotic at the end of the case, and the field is not grossly contaminated, primary anastomosis is reasonable and avoids a second operation on a stoma.
If the ends are questionable, if there is residual ischemia, or if the infant is unstable, I would exteriorize as an end stoma with a mucus fistula brought out adjacently so that continuity is straightforward later. But whichever route I choose, I would measure and document the residual length from the ligament of Treitz, because that number dictates nutritional strategy and prognosis.
3. Outline the postoperative plan and the specific findings that would send this infant back to the operating room.
Read the worked answer
Postoperatively: NICU care with lung-protective ventilation, pressor support as needed, broad-spectrum antibiotics, and total parenteral nutrition through a dedicated central line. No enteral feeding until the ileus resolves and the gastric and stoma output are non-bloody; then trophic breast milk feeds advanced slowly. Monitor platelets, lactate, acid-base status, and abdominal examination serially, and keep the abdominal wall visible rather than under drapes.
Triggers to return to the operating room include persistent or worsening metabolic acidosis, new or increasing pneumoperitoneum, discoloration or induration of the abdominal wall, feculent drain or wound output, an infant who simply is not improving within 24 to 48 hours of the index operation, and any planned second look for bowel of uncertain viability. A normal-appearing abdomen is not reassuring in the face of a rising lactate.
After discharge, the issues that matter are anastomotic or stoma strictures - which present as obstruction weeks to months later and need a contrast study - TPN-related cholestasis, growth, and neurodevelopmental follow-up. Roughly a quarter of surgical NEC survivors develop a stricture, so a low threshold for imaging is appropriate.