Abstract:

Objective: Infant hydrocephalus is estimated to affect more than 100,000 new infants per year in sub-Saharan Africa (SSA). Bugando medical centre (BMC), a government funded and patient cost-shared referral center, serves over 13 million people in the lake and western regions of Tanzania. The goal of this study were to characterize the infant population affected by hydrocephalus who were presented to Bugando medical centre and treated with a ventriculo-peritoneal shunt (VPS), and to determine the early complication rate of the VPS insertion surgical procedure and to assess its potential risk factors.

Methods: Data was prospectively collected from every patient aged less than 1 year old who were diagnosed with hydrocephalus and admitted to Bugando medical centre for primary VPS, over a period of 7 months. Demographics, maternal history, pre-operative studies, surgical procedure and up to first follow-up surgical complication were analyzed and the risk factors associated with surgical complications were determined.

Results: 125 infants were included for analysis. 75% were younger than 6 months of age, and 56% were males. Only 7% mothers had a gestational ultrasound, 98% of them did not receive pre-conception folic acid, and 25% delivery at home. Congenital hydrocephalus accounted for the majority (56%) of the hydrocephalus etiologies followed by post-infectious (22.4%) and myelomeningocele-associated (16%). The mean head circumference on admission was 51.4 cm +/- 6.3 cm. The median age at shunt surgery was 137 days, and over a 1/3 of patients were operated without a radiologic exam. The majority of shunts were placed on a right parieto-occipital location. 13 patients had a previous intra-ventricular endoscopic procedure. Overall, at least one surgical complication was found in 33.6% of patients up to the first follow-up assessment (median follow-up time of 70 days); VPS infection was the most common complication. The post-operative mortality was 9%. The risk factors associated with early surgical complications were tumor-related etiology, larger head circumference and longer post-operative hospital stays.

Conclusion: In a region where most infant hydrocephalus cases were congenital in nature, most patients presented to the hospital in a late stage, with no pre-natal diagnosis and with large head circumferences. The pre-operative investigations were not uniformly performed and the surgical complications, led by VPS infection, where unacceptably high. Younger patient age, previous endoscopic procedure, the surgeon performing the procedure and anatomical location of the VPS had no association with surgical complications.

These results were discrepant from those previously published in SSA by mission hospitals that were sub-specialized in pediatric neurosurgery, but were similar to the remaining studies from East Africa. In order to improve maternal and neonatal care in the lake region of Tanzania, the development of a strict peri-operative protocol for VPS insertion and the increase in the number of endoscopic procedures are imperative.