SURGERY FOR NEC IN HUMAN INFANTS-- A RANDOMIZED TRIAL

Yale University, New Haven CT

Investigators

Linked publications & trials

Abstract

Necrotizing enterocolitis (NEC) is a severe inflammatory disease of the intestine that occurs in premature and low birth weight infants. In its most severe form, NEC results in perforation of the bowel mandating surgical intervention. It is the most common surgical disease of the newborn and results in over 2500 deaths per year in this country. The mortality rate for perforated NEC remains largely unchanged over the past 10 years. Traditional surgical therapy includes laparotomy, bowel resection, and creation of intestinal stomas (LAP). Simple primary peritoneal drainage (PPD) of the abdominal cavity is a novel, minimally invasive alternative. Experience with both of these treatments is broad but their outcomes have never been compared in a meaningful fashion. Our primary aim is to perform a clinical trial to determine whether peritoneal drainage improves survival over laparotomy and bowel resection in babies less than 1500 grams. Our secondary aims are to determine whether PPD affects gastrointestinal morbidity, length of hospital stay, and cost of care. The unique resources of UCSF Stanford Health Care will allow us to enroll a population of sufficient sample size yet maintain the rigid standardization of operative technique and patient care necessary for a successful surgical trial. We will enroll infants less than 1500 grams with perforated NEC and randomize their operative treatment to PPD or LAP. Postoperative care will follow an adopted clinical pathway. Statistical analysis will use survival as the primary outcome. Inference will be based on a "cure rate" model" and a log rank statistic. The design is sequential allowing for an "early" look at the data after 37 patients. If evidence is overwhelmingly in favor of one group we will stop. Otherwise we will continue until 88 patients are accrued. Secondary variables will be analyzed by Cox regression with adaptively chosen covariates from a stated list involving sample reuse methods to make inferences regarding parameters. As with many problems in Pediatric Surgery, single institutions do not have sufficient experience with NEC to complete meaningful studies. We believe that we have the necessary personnel and resources in place to complete the first randomized controlled trial for neonatal surgery in the United States.