Ventricular shunting for hydrocephalus in children: patients, procedures, surgeons and institutions in English Canada, 1989-2001.

INTRODUCTION AND
PURPOSE: Ventricular shunting remains the principle and most generally applicable method to treat hydrocephalus in children. This paper describes the demographics of this treatment in English Canada during the period of 1989 to March 2001.
METHODS: Hospital discharge records were obtained for patients less than 18 years who had a shunt inserted or revised. A database was constructed relating patients and procedures to hospital discharges based on scrambled patient identifiers, year of birth, sex, postal code and diagnoses. OBSERVATIONS: 5,947 patients underwent ventricular shunting procedures for hydrocephalus in this period. 261 surgeons working in 73 institutions provided 12,106 interventions (Shunt insertions: ventriculoperitoneal--5009, ventriculoatrial--119, ventriculopleural--28. Revisions: 6,950). Infection was deemed to have occurred in 1,059 procedures. Over the study period, the median number of procedures performed per surgeon per year was 2, with 75 % of surgeons performing 5 or fewer procedures in children per year. Although many surgeons operated on children throughout the thirteen years of the study, many did not acquire substantive cumulative experience. Overall infection rate was 8.6 %. Surgeon infection rates were greater than or equal to 20 % during the first four years of practice and thereafter they fell to and remained in the 10 % range. The mean shunt survival at 12 months of individual surgeons varied between 50 - 60 %, regardless of the number of years of experience of the surgeon; however, performance variability as measured by the standard deviation of 12 month survival rates for all surgeons, adjusted for years of experience, ranged widely until the fifth year of practice. The average number of procedures per year for treating hospitals was 2 with 75 %, providing 12 or fewer services annually. Over the entire study, 50 % of institutions provided 10 or fewer procedures. The mean institutional infection rate was 11.4 % (SD 23, median--6.0).
CONCLUSIONS: Quality monitoring of infection rate and duration of shunt function remains critical as many surgeons and hospitals provide care to children with hydrocephalus infrequently. Variability in infection rates and shunt survival at 12 months are a function of surgeon experience, measured by years in practice. Variability in outcome decreases with increasing surgeon experience.