Accuracy of intracranial pressure monitoring: systematic review and meta-analysis.

INTRODUCTION Intracranial pressure (ICP) measurement is used to tailor interventions and to assist in formulating the prognosis for traumatic brain injury patients. Accurate data are therefore essential. The aim of this study was to verify the accuracy of ICP monitoring systems on the basis of a literature review. METHODS A PubMed search was conducted from 1982 to 2014, plus additional references from the selected papers. Accuracy was defined as the degree of correspondence between the pressure read by the catheter and a reference "real" ICP measurement. Studies comparing simultaneous readings from at least two catheters were included. Drift was defined as the loss of accuracy over the monitoring period. Meta-analyses of data from the studies were used to estimate the overall mean difference between simultaneous ICP measurements and their variability. Individual studies were weighted using both a fixed and a random effects model. RESULTS Of 163 articles screened, 83 compared two intracranial catheters: 64 reported accuracy and 37 drift (some reported both). Of these, 10 and 17, respectively, fulfilled the inclusion criteria for accuracy and zero drift analysis. The combined mean differences between probes were 1.5 mmHg (95 % confidence interval (CI) 0.7-2.3) with the random effects model and 1.6 mmHg (95 % CI 1.3-1.9) with the fixed effects model. The reported mean drift over a long observation period was 0.75 mmHg. No relation was found with the duration of monitoring or differences between various probes. CONCLUSIONS This study confirms that the average error between ICP measures is clinically negligible. The random effects model, however, indicates that a high percentage of readings may vary over a wide range, with clinical implications both for future comparison studies and for daily care.