Predictors of gangrenous necrotizing enterocolitis and extent of disease. Early laparotomy versus peritoneal drainage

Evaluation of known predictors of gangrene in neonates with necrotizing enterocolitis [NEC] and identification of those suggestive of severe disease, requiring expeditious laparotomy rather than primary peritoneal drainage as a definitive treatment. This is a retrospective review of data collected from the medical records of newborns with confirmed NEC, treated at the Maternity and Children's Hospital, Dammam, Kingdom of Saudi Arabia, from May 1993 to May 2004. Fifty-five cases were selected for the study, 23 had successful medical management and 32 underwent laparotomy. Of this group, 15 had peritoneal drainage prior to laparotomy. Nine known clinical, radiological and laboratory features suspicious of bowel perforation or gangrene were evaluated. The operated group was classified according to the extent of disease into isolated, multifocal or pan intestinal and the distribution of these 9 criteria was calculated for each of the 3 groups. Comparison was then carried out between the group with isolated NEC and those with extensive disease. Isolated NEC was present in 8 [25%], multifocal NEC in 19 [59%] and pan intestinal NEC in 5 [16%] of the operated cases. Pneumoperitoneum and palpable abdominal mass were the most specific and predictive signs of perforated or gangrenous bowel in NEC. Severe pneumatosis intestinalis and gasless abdomen were also highly specific and predictive of the same but had a low prevalence. Abdominal wall erythema, persistent metabolic acidosis, portal vein air, gasless abdomen and severe pneumatosis intestinalis were found to be associated with severe or extensive gangrene. Palpable abdominal mass and fixed dilated loops were increased in cases of isolated NEC. Portal vein air was associated with the highest mortality. Pneumoperitoneum, though the only absolute evidence of bowel perforation, cannot predict the extent of disease. Peritoneal drainage is a useful stabilizing procedure but the presence of any of the above mentioned criteria which are associated with severe disease necessitate a quick decision in favor of laparotomy. The absence of these signs, however, cannot rule out extensive or progressive NEC and failure to improve after peritoneal drainage also requires an emergency laparotomy, regardless of birth weight or gestational age

Outcome in definite and advanced neonatal necrotizing enterocolitis

To evaluate and compare the outcome of newborns with definite [Bells stage II] and advanced [Bells stage III] necrotizing enterocolitis [NEC] and to assess the role of primary peritoneal drainage. This study was conducted in the Department of Pediatric Surgery, Maternity and Children's Hospital, Dammam, Kingdom of Saudi Arabia. Medical records of all cases diagnosed with NEC or suspected NEC between May 1993 and June 2003 were reviewed retrospectively. A total of 67 cases meeting the criteria for Modified Bells stage II and III disease were selected for the study. Twenty-five [37%] cases were treated medically and 42 [63%] needed surgical intervention, namely primary peritoneal drainage [PPD] with or without salvage laparotomy [SL] [n=25], or primary laparotomy [PL] [n=17]. Data regarding patient demographics, neonatal history, clinical presentation, laboratory and radiological features, operative findings, complications and mortality were collected and compared between the medical and surgical group and between the 2 surgical groups. The overall mortality was 37%, 8% in the medical group versus 55% in the surgically treated group. The PPD group had the highest mortality [72%] versus 29% in the PL group. In the PPD group, 14 [56%] needed SL and only 3 [12%] survived without laparotomy. The mean gestational age and birthweight were 32.1 weeks and 1713 gms in the PPD group as compared to 35.7 weeks and 2484 gms in the PL group. The PPD group were more critically sick than the PL group. The average length of time from onset to laparotomy was longer, 6.6 days in the PPD group versus 2.2 days in the PL group. In the 31 cases undergoing laparotomy, the terminal ileum was involved most frequently followed by the cecum and right colon. The PPD + SL group had a higher mortality in isolated, multifocal, and pan involvement of bowel when compared to the PL group. All babies with only isolated involvement in the PL group survived. The surgical mortality of NEC in our hospital is very high due to advanced disease. The PPD group in our study fared badly, as it comprised a poor risk group with ongoing sepsis, with the bias being in favor of PPD as these babies were unfit to undergo major surgery. Early SL after PPD in those showing signs of persistent disease may improve the outcome in this group