Neonatal surgery in Africa: a systematic review and meta-analysis of challenges of management and outcome.

Background: Advances in diagnostic techniques and perioperative care have greatly improved the outcome of neonatal surgery. Despite this, disparity still exists in the outcome of neonatal surgery between high-income countries and low-income and middle-income countries. This study reviews publications on neonatal surgery in Africa over 20 years with a focus on challenges of management, trends in outcome, and potential interventions to improve outcome.

Methods: We did a literature review by searching PubMed and African Index Medicus for original articles published in any language between January, 1995, and September, 2014, with the search terms "neonatal surgery" and "Africa", further supplemented by "(surgery OR anaesthesia) AND (neonatal OR newborn) AND (developing countries OR Africa)". A data extraction sheet was used to collect information, including type of study, demographics, number of cases, outcome, challenges, and suggestions to improve outcome. For the meta-analysis, data were analysed by χ(2) test or Student's t-test as appropriate. In all, the significance level was set to p<0·05.

Findings: We identified 859 published papers, of which 51 studies from 11 countries met the inclusion criteria. The 16 studies in the first 10 years (before 2005; group A) were compared with the 35 in the last 10 years (2005-14; group B). Nigeria (n=32; 62·7%), South Africa (n=7; 13·7%), Tanzania (n=2; 3·9%), and Tunisia (n=2; 3·9%) were the predominant source of the publications, of which were retrospective in 38 (74·5%) studies and prospective in 13 (25·5%) studies. The mean sample size of the studies was 97·8 (range 5-640). Overall, 4989 neonates were studied, with median age of 6 days (range 1-30). Common neonatal conditions reported were intestinal atresia in 28 (54·9%) studies, abdominal wall defects in 27 (52·9%), anorectal malformations in 24 (47·1%), and Hirschsprung's disease, necrotising enterocolitis, and volvulus neonatorum in 23 (45·1%) each. Mortality was lowest (<3%) in spina bifida and facial cleft procedures, and highest (>50%) in emergency neonatal surgeries involving bowel perforation, bowel resection, congenital diaphragmatic hernia, oesophageal atresia, and ruptured omphalocele or gastroschisis. Overall average mortality rate was higher in group A than in group B (36·9% vs 29·1%; p<0·001), but mortality did not vary between the groups for similar neonatal conditions. The major documented challenges were delayed presentation and inadequate facilities in 39 (76·5%) studies, dearth of trained support personnel in 32 (62·7%), and absence of neonatal intensive care in 29 (56·9%). The challenges varied from country to country but did not differ in the two groups.

Interpretation: Improvement has been achieved in outcomes of neonatal surgery in Africa in the past two decades, although several of the studies reviewed are retrospective and poorly designed. Cost-effective adaptations for neonatal intensive care, improved health-care funding, coordinated neonatal surgical care via regional centres, and collaboration with international partners are potential interventions that could help to address the challenges and further improve outcome.

Funding: None.