Primary therapy for the infant with NEC is medical. Surgery is indicated only for complications: perforation, bowel necrosis, overwhelming sepsis. Nevertheless, the decision to operate or continue observing can be difficult. In general, it is better to err on the side of performing an operation, rather than waiting until it is too late to make a difference. Though some have attempted to use ultrasound or paracentesis to improve the accuracy of the decision-making process, ultimately the decision is based on the good judgment that comes with experience.
Drainage has fallen into disfavor of late, but it is still occasionally useful, specifically for the extremely critically ill micropremie with an obvious perforation who cannot be transported safely, needs more time for resuscitation, or has other anomalies that need to be sorted out. In these cases, peritoneal drainage can buy some time. But drainage is rarely definitive and one can justify performing a laparotomy within 48–72 h whether the patient has improved (able to safely undergo a more formal surgical procedure) or deteriorated (requires an escalation in therapy). It is useful to make small bilateral lower quadrant incisions, irrigate with warm normal saline solution, and place three ¼-in. Penrose drains: one that passes between the two incisions, one that reaches into the pelvis from the right-sided incision, and one that passes into the left upper quadrant from the left side. In this way, the abdomen is widely drained, injury to the liver is avoided, and the drains can be sutured to themselves rather than the patient.
We have all seen infants months after a laparotomy for NEC with a huge and misshapen scar that crosses the entire upper abdomen and is usually cross-hatched with suture scars. Some surgeons question why such an incision is ever necessary. A small right lower quadrant incision that rarely needs to cross the midline, unless there is bowel to be dealt with near the ligament of Treitz, is all that is necessary. The entire abdomen in a small premature is available through this incision and the ultimate appearance is far superior. The stoma and mucous fistula can be brought out in the corners of the wound and the incision can be closed at the level of Scarpa’s fascia and steri-strips or cyanoacrylate glue. There is also a tendency to wait an excessively long time to close the stoma in these infants, which puts them at risk for parenteral nutrition-associated liver failure. Closing the stoma usually allows them to tolerate enteral nutrition sooner. If the patient is stable, one should consider closing the stoma within 2 weeks of resection. There is even a trend towards performing a primary anastomosis at the original operation and avoiding a stoma altogether. At any rate, continuity should be re-established no more than 4–6 weeks after resection.
The most difficult situations arise when the there is loss of nearly the entire small intestine. Despite recent advances in the field of intestinal transplantation, it is still very unlikely that an infant with essentially no intestine can be made to survive long enough to undergo a successful transplant. Nevertheless, it will likely soon be a viable option and every case should be assessed individually and with intimate involvement of the parents. It is also important to remember that the length of the remaining intestine does not tell the whole story: some of it was likely ischemic and though now viable, it might be dysmotile or otherwise dysfunctional.