Epidemiology of ventriculostomy in the United States from 1997 to 2001.

Ventriculostomy is a common practice in neurosurgery, but the annual trend of this procedure in the United States has not been reported in the literature. This study evaluates the annual trend during a recent 5-year period. Between 1997 and 2001, a retrospective review was undertaken concerning all patients in the Nationwide Inpatient Sample (NIS) who had undergone ventriculostomy. The population sample represented approximately a 20% stratified sample of nonfederal hospitals in the United States. The annual number of patients who underwent ventriculostomy during the study period ranged from 20,586 to 25,634. Most patients were male (53.4%), with a mean age of 44.8 years, were commercially insured (46.0%) and had a median annual income above $25,000 (84.4%). Most frequent ICD-9-CM diagnoses were subarachnoid haemorrhage, intracerebral haemorrhage and obstructive hydrocephalus, respectively. The majority of ventriculostomies were performed in large, private, not-for-profit, metropolitan, teaching institutions. Mean length of hospital stay was 19.2 days. Regarding discharge status for patients who had undergone ventriculostomy, approximately one-quarter died in the hospital, one-third were discharged home and one-third were transferred to another institution. No demographic variables changed during the study with the exception of location of ventriculostomy in a teaching hospital, which increased from 64.4% in 1997 to 77.4% in 2001. Patient and hospital demographic characteristics were consistent during the study period. By extrapolation of the data, the prevalence of ventriculostomy in the United States averaged 24,380 per year. This study is the first to comprehensively document data concerning the epidemiology of this common procedure.