Prospective study of methodological issues in intracranial pressure monitoring in patients with hydrocephalus

OBJECT: Continuous intracranial pressure (ICP) monitoring is performed in selected patients with hydrocephalus to determine whether shunt placement is required. The mean ICP is usually calculated from end-hour readings manually recorded by nurses. The aim of this study was to evaluate the accuracy of manual recordings by comparing nurses' end-hour ICP readings with those of an online computerized ICP monitoring system that records one ICP value per second. METHODS: Continuous ICP monitoring was performed using a fiberoptic extradural sensor in 115 patients with hydrocephalus of different origins. A notebook computer was connected to an ICP monitor and was programmed to register one ICP value per second. In all patients, mean ICP values were calculated from data recorded manually by nurses at the end of every hour and from data recorded by the computer within the preselected time period. The two methods were compared using correlation analysis and the Bland and Altman method. The median number of ICP values noted manually by the nurses in each patient was 17 (interquartile range 15-18 readings), and that recorded by the software was 61,200 (interquartile range 54,000-64,800 readings). The correlation coefficient of the mean ICP values recorded by both methods was r = 0.99 (p < 0.001). The Bland and Altman analysis revealed a mean difference of 0.3 +/- 1.26 mm Hg between the two methods and that they were equally valid with all mean ICP values. CONCLUSIONS: The recording of end-hour ICP values by nurses is an accurate method of calculating the mean ICP after prolonged ICP monitoring in patients with hydrocephalus.