For 50 years the workhorse of the treatment of hydrocephalus has been the cerebrospinal fluid (CSF) shunt, a valve-regulated drain that diverts CSF from the central nervous system to another site where it can be reabsorbed into the vascular circulation at low pressure—most commonly the peritoneal cavity. The problems with CSF shunting are the infectious and mechanical complications, the management of which requires hospital admissions, surgical procedures, and the consumption of other resources. In the early 1990s, improvements in surgical instrumentation drove a resurgence of interest in an alternative treatment: endoscopic third ventriculostomy (ETV). ETV entails creation of a hole in the floor of the third ventricle to allow escape of CSF out of the ventricular cavities into the subarachnoid spaces where, presumably, it can be reabsorbed in the normal fashion. Early proponents conceived of ETV as a cure for hydrocephalus, but subsequent experience has demonstrated both early and late relapses and, like CSF shunting, a small rate of disabling and life-threatening complications.1,2 Few methodologically serious comparisons of ETV and CSF shunting have appeared.1
The authors from Ann Arbor, Salt Lake City, and Vancouver, BC, conducted a case-control study comparing 28 children who underwent ETV with matched controls treated with shunts identified from an institutional database. The outcome was “days free of treatment for hydrocephalus,” calculated as the total days of follow-up, less time spent hospitalized for hydrocephalus-related treatment, and less a 14- day addition for each hospital admission to account for the impact of the prehospitalization illness and postdischarge recovery time. Controls were matched to cases on the basis of age, etiology of hydrocephalus, and number of previous surgical procedures. Costs were estimated from the perspective of the third-party payor and were discounted at an annual rate of 5% beginning from the time of the index procedure. For each matched pair, the accounting was terminated at the conclusion of the shorter follow-up.
ETV yielded more days free of treatment and consumed fewer resources than CSF shunting, but the differences were small and did not attain significance. The initial costs of ETV are greater than CSF shunting, but in this population the cost streams crossed at 36 months due to the more frequent, late complications of shunting. The analysis was robust with respect to variations in cost assumptions, but variation of the ETV success rate from the observed 55% to a hypothetical but attainable 75%3,4 caused the cost streams to cross much earlier, at 12 months instead of 36. In a quality of life analysis, a single instance of postoperative hemiparesis and a late postoperative death erased the superior effectiveness of ETV. The authors reiterate a concern expressed repeatedly in the recent neurosurgical literature that patients and their families be counseled that even a successful ETV is not a cure.2 Indefinite follow-up is appropriate for prompt detection of late deterioration.