For infants with surgical necrotizing enterocolitis, does primary anastomosis or stoma formation provide shorter parenteral nutrition?

The STAT Trial was a multicenter, multinational randomized controlled trial conducted at 12 centers across Europe, Canada, South America, Singapore, and New Zealand from April 2010 to January 2019 [ 1 ]. The study was approved by the Institute of Child Health/Great Ormond Street Hospital Research Ethics Committee/National Research Ethics Service (09/H0713/58), and each institution obtained individual ethical approval. The trial was registered with the International Standard Randomized Controlled Trial Register (ISRCTN): ISRCTN01700960. Pediatric surgeons obtained written consent before surgery. However, enrollment and randomization occurred during laparotomy after each infant was assessed for findings of necrotizing enterocolitis (NEC) and the extent of NEC. Infants who were determined to benefit from either intervention were then included in the study. Randomization occurred intraoperatively after inspecting the bowel for exclusion criteria below. A weighted minimization method was then used to evaluate how assigning an infant to each intervention could affect the balance of the following factors: 1. Weight at enrollment (2000 g) Weight at enrollment (2000 g) 2. Mechanical ventilation Mechanical ventilation 3. Inotropic support Inotropic support 4. Focal vs. multifocal disease Focal vs. multifocal disease 5. Area of intestinal involvement Area of intestinal involvement This approach maintains randomness while reducing confounding by assigning each new participant to keep the trial arms as balanced as possible. Blinding was not possible as randomization occurred during laparotomy when disease findings and extent of disease were assessed. Patients were evaluated in the surgery clinic at 3 months and 1 year. Seventy-nine infants were enrolled following parental consent. Each site received individual ethics approval. Suspected necrotizing enterocolitis (NEC) requiring laparotomy due to perforation or failed medical management. Exclusions included: no NEC no NEC focal intestinal perforation focal intestinal perforation extensive NEC requiring resection, likely to result in short bowel syndrome extensive NEC requiring resection, likely to result in short bowel syndrome NEC isolated in the colon with high bleeding risk NEC isolated in the colon with high bleeding risk intraoperative instability intraoperative instability Consent declined Consent declined Following resection of necrotic bowel, infants underwent either primary anastomosis or formation of a stoma. The primary outcome was defined as the duration of parenteral nutrition in days. Secondary outcomes included mortality, need for additional surgery, and intestinal complications (e.g., anastomotic leak, stoma-related complications, obstruction, recurrent NEC). A sample size of 66 provided 80% power to detect a 14-day difference in time to full enteral feeds, with 13 more added for anticipated 20% mortality. An interim analysis was performed after recruiting 40 participants, halfway to the goal enrollment. Gestational age, birth weight, and sex were similar between groups, with no difference in bowel involvement or extent of NEC. Median duration of parenteral nutrition was significantly shorter in the primary anastomosis group (30 days; range 4–105) compared to the stoma group (51 days; range 3–310) ( p < 0.05). Mortality rates were similar between groups. Intestinal complications were more frequent in the stoma group (46%) versus the anastomosis group (16%) ( p < 0.05). At laparotomy for NEC, when there is no disease distal to the resected intestine, primary anastomosis enhances the recovery from NEC, reduces the risk of multiple intestinal complications and does not increase adverse outcomes. NEC remains one of the most serious gastrointestinal emergencies in the neonatal intensive care unit, primarily affecting premature infants. Risk factors include prematurity, low birth weight, antibiotic exposure, and formula feeding, while protective factors include breastfeeding, feeding protocols, and probiotics [ 2 , 3 , 4 , 5 ]. Infants requiring surgery for NEC face the highest risks of poor outcomes, including death, short bowel syndrome, and long-term nutritional dependence [ 6 ]. The optimal surgical approach, whether to perform primary anastomosis or stoma formation, has remained a debated issue [ 7 ]. This trial is the first randomized controlled study to provide evidence on this topic, addressing a significant gap in the literature. The multinational nature of the STAT Trial, covering 12 centers, supports generalizability. However, 73% of participants came from just three countries (Canada, the UK, and Serbia). There may be potential differences in concurrent practices between locations that could modify the effect of the surgical approach. The absence of a center in the primary analysis may have weakened the ability to account for practice variation or demographics in determining the effect of surgical approach. Similarly, the lack of detail on feeding strategies post-operatively limits the interpretation of the primary outcome (duration of parenteral nutrition). Feeding protocols across institutions often differ in duration of small volume feeds before advancing, speed of advancing to full volume feeds, and timing of re-introduction of feeds after surgical NEC [ 8 , 9 ]. These protocols directly impact the duration of parenteral nutrition, the study’s primary outcome. In addition, while the sample size was calculated to detect a difference in days to full feeds, the analysis used days on parenteral nutrition as the primary outcome. This is a subtle distinction. Infants may achieve full feeds but still require parenteral nutrition in the setting of poor weight gain from presumed malabsorption. In this case, the time to full feeds is shorter than the timeframe of requiring parenteral nutrition. Furthermore, while randomization was designed to balance illness severity, surgeon discretion in selecting cases for randomization may have introduced selection bias, limiting the generalizability. Demographic characteristics of the infants who were not included in the study are not provided. It is unclear how this group differs from those included in the study. The study was not powered for subgroup analysis, such as examining outcomes in extremely low birth weight or critically ill infants who are at higher risk of mortality after surgical NEC [ 10 ]. Despite these limitations, the STAT Trial provides compelling evidence that primary anastomosis may lead to quicker recovery in terms of parenteral nutrition needs, without an increase in complications or mortality.