Dose–response relationship of mannitol and intracranial pressure: a metaanalysis

Abstract

Object Brain edema can increase intracranial pressure (ICP), potentially leading to ischemia, herniation, and death. Edema and elevated ICP are often treated with osmotic agents to remove water from brain tissue. Mannitol is the osmotic diuretic most commonly used in the intensive care unit; however, despite its clinical importance, treatment protocols vary from center to center, and the dose–response relationship is not understood. The goal of this metaanalysis was to aggregate and analyze data from studies in which authors have described the dose–response relationship between mannitol and ICP. Methods The authors identified 18 studies that quantitatively characterized the dose–response relationship of mannitol and ICP. We also examined study designs and mannitol administration protocols. Results Meta-regression found a weak linear relationship between change in ICP (ΔICP) and dose (ΔICP = 6.6 × dose − 1.1; p = 0.27, R2 = 0.05). The lack of statistical significance could reflect the variation in protocols among studies and the variation in patients both within and among studies. However, the authors found a highly significant difference (p < 0.001) in decrease in ICP when the initial ICP was higher or lower than 30 mm Hg. Nonlinear regression suggested that ICP decrease is greatest shortly after mannitol is given (R2 = 0.63). Finally, the authors found that recent studies tend to include fewer patients and set a lower ICP threshold for mannitol administration but report more parameters of interest; the duration of mannitol's effect was the most frequently unreported parameter. Conclusions Despite its clinical importance, the determination of the mannitol dose–response curve continues to be challenging for many reasons. This metaanalysis highlights the need for a consensus of methods and results required to determine this important relationship.