Date of Award

January 2011

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)



First Advisor

Dennis D. Spencer

Subject Area(s)

Neurosciences, African studies


The treatment of neurosurgical disease in the developing world presents challenges on numerous levels, not the least of which are the litany of logistical and infrastructural barriers which stand in the way of delivering care. The scarcity of neurosurgeons combined with limited mobility in developing countries requires a reconsideration of approaches to treatment; this is particularly true of ventriculoperitoneal (VP) shunt insertion for the treatment of hydrocephalus. While VP shunt implantation is the mainstay of treatment in the United States, it is prone to failure requiring rapid access to neurosurgical care making shunt dependency a dangerous proposition in this setting. Endoscopic third ventriculostomy (ETV) offers an alternative to VP shunt dependency and has been shown to be effective in treating hydrocephalus at intervals of up to 14 months.

The purpose of this study is to compare the five-year survival for children treated with endoscopic third ventriculostomy (ETV) to those treated with ventriculoperitoneal (VP) shunt implantation for myelomeningocele-associated hydrocephalus. Because of the well known dangers of shunt dependency, it is hypothesized that patients treated with ETV will have a survival advantage when compared with patients treated with VPS at a follow-up interval of five years.

In order to address this hypothesis, a retrospective observational study of children treated with ETV or VP shunt implantation for myelomeningocele-related hydrocephalus at the CURE Children's Hospital of Uganda was carried out, including a control group consisting of myelomeningocele patients who had not developed hydrocephalus.

Survival status was determined for 128 of 131 study participants (98%). 47/128 (37%) of patients had expired at five years post-treatment, and 55/128 (42%) patients had expired at a mean follow-up interval of 84.4 months. Only two cases of death were attributable to the development of hydrocephalus, none as a result of treatment failure. Kaplan-Meier survival analysis found no statistically significant relationship between survival and method of treatment for hydrocephalus (p=0.45), sex of patient (p=0.53), HIV status (p=0.69), age at repair ( p=0.34), or myelomeningocele level (p=0.12). Survival analysis performed for districts with community based rehabilitation (CBR) programs and districts without CBR programs revealed a significant interaction (p=0.001).

The uniformly high mortality across all groups suggests that the chief causes of long-term mortality are both powerful and independent of hydrocephalus. The only correlation with survival identified in this study, the presence of a CBR program providing in-home rehabilitation, fulfills both of these criteria: myelomeningocele patients require long-term rehabilitation regardless of the development of hydrocephalus and these programs exert a powerful influence on survival. This substantial difference in long-term survival highlights the fact that children in communities without CBR programs are not receiving life-saving supportive care, in part due to a lack of parental understanding of the need for longitudinal care. Established cultural beliefs about myelomeningocele, hydrocephalus, and disability in general also hampered efforts to improve survival. Understanding the practical barriers to the delivery of care in a developing country as well as the cultural mores through which diseases are understood are critical to effectively treating disease across cultures and continents.