Gastro-intestinal tract perforation in neonates

Abstract

Background: Gastro-intestinal tract (GIT) perforation in neonates is a serious problem associated with high mortality due to resulting sepsis. Co-morbid factors, eg. prematurity, respiratory problems, low birth weight, and nutritional factors, negatively affect the outcome. Objectives: To review the management outcome of gastro-intestinal tract perforation in neonates in KwaZulu-Natal and identify factors that require attention for better survival of neonates with GIT perforation. Design: Retrospective study of consecutive complete data sets of patients presenting with a diagnosis of GIT perforation. Setting: Department of Paediatric Surgery, Nelson R. Mandela School of Medicine, University of Natal, Durban, South Africa. Subjects: Fifty four neonates treated for gastro-intestinal tract perforation between January 1998 and January 2003. Main outcome measures: Morbidity as determined by complications and mortality. Results: More males (69%) were affected than females (31%). The median birth weight was 2.3 kg and median age at presentation was four days. Eighty nine percent were referred from peripheral hospitals. Abdominal distension was the leading symptom and sign (74%). Co-morbid factors were present in 89%, with prematurity as the leading factor (52%). Necrotising enterocolitis (NEC) was the main cause of perforation (33%) and the terminal ileum was the most common site. Most (56%) were treated by excision and primary repair of perforations. Sepsis was the leading complication (44%) and major cause of death (72%). Mortality was highest (56%) in perforations due to other primary pathology followed by NEC (53%). Overall mortality was 46%. Conclusion: It is essential to prevent secondary perforations by early recognition and management of primary pathology. Management of pneumoperitoneum in neonates with respiratory difficulties should be included in resuscitation before transfer. Rectal temperature monitoring and herbal enemas should be strongly discouraged.