BACKGROUND: To evaluate the effects of extending the practice of bedside burr hole for intracranial pressure (ICP) monitoring to the entire ICU team. DESIGN: a 10-year observational cohort study. SETTING: A general-neurologic 12-bed ICU of a University Hospital with 10 full-time specialists in anaesthesia-intensive care. PATIENTS: patients admitted for acute neurologic lesions requiring ICP monitoring. MEASUREMENTS: ICP monitoring was performed by means of intraparenchymal fiberoptic cathe-ters. Rates of successfully performed procedures and complications were compared between 2 study periods: January 1990 - August 1997, in which the procedure was performed by 2 experienced physicians (group 1), and September 1997-July 2000 in which the procedure was extended to the entire team (group 2). RESULTS: Two hundred and seventy patients had 293 catheters positioned (group 1=180; 2=113). The procedure was successfully performed in all cases. Occurrence of complications was similar in the 2 groups: minor surgical wound infections (3.3%; 2.6%); meningitis (0.5%; 0); scalp (3.9%; 2.6%) and dural (5.0%; 6.2%) bleeding; intracranial haematoma (1.1%; 1.8%). Two of these latter (1 for each group) required surgical evacuation. Twelve anaesthetist-intensive care physicians were trained, and they were able to perform burr hole for ICP monitoring without help after 2-3 assisted procedures. CONCLUSIONS: Extending the practice of ICP monitoring to the entire ICU team is safe and feasible. To decide whether or not to implement this technique, one should consider the high costs on one side and prompt availability of ICP monitoring on the other. Haemorrhagic and infectious complications are comparable to those of neurosurgical series.