(692) Serum Osmolarity Versus Osmolal Gap Monitoring for Mannitol-Associated Acute Kidney Injury

Introduction: Mannitol is used to manage elevated intracranial pressure (ICP) with varying monitoring strategies to mitigate risk of acute kidney injury (AKI). The 2020 Guidelines for Acute Treatment of Cerebral Edema in Neurocritical Care Patients (pts) recommend osmolar gap (OG) over serum osmolarity (sOSM) thresholds ( < 320 mOsm) to monitor for AKI with mannitol, based on very low-quality evidence. This study compared AKI incidence between these monitoring strategies.

Methods: This single-center, retrospective case-control study included adults admitted to an ICU from December 2022 and June 2024 who received > 2g/kg of mannitol within a 48-hour (hr) period. Pts with incomplete lab data or those with kidney disease on renal replacement therapy at admission were excluded. The primary outcome was incidence of AKI within 24 hrs of each discrete mannitol administration (admin), comparing those with sOSM < 320 to ≥ 320 and OG < 20. Secondary outcomes included OG in pts who developed AKI, ICP 1 and 4 hrs post-mannitol, incidence of refractory ICPs requiring deep sedation, neuromuscular blocker (NMB) or pentobarbital use, ICU length of stay (LOS), and hospital mortality.

Results: The study included 23 pts with 125 discrete mannitol admins, 47 with sOSM < 320 and 79 with sOSM ≥ 320 and OG < 20. The population was 50% male, 56% white with no baseline renal dysfunction. Pts were admitted to neurosciences (52%) and trauma (48%) ICUs. There were no AKI instances in either group. Median OG in the sOSM ≥ 320 group was 5.7. There were 9 mannitol admins with sOSM < 320 with ICP monitoring; 5 had ICP > 20 for 2 hrs pre-dose. There were 78 mannitol admins with sOSM ≥ 320 and OG < 20 with ICP monitoring; 53 had ICP > 20 for 2 hrs pre-dose. Median ICP pre-mannitol was 27 in both groups. Median ICP was 13 (sOSM < 320) and 19 (sOSM ≥ 320) 1 hr and 22 and 21 4 hrs post-mannitol, respectively. Secondary outcomes were not compared between groups due to overlapping pts and small sample size. Many pts required deep sedation (65%), NMB intermittent (48%) or continuous dosing (26%), and pentobarbital (17%). Median ICU LOS was 7 days, with 48% hospital mortality.

Conclusions: No difference in AKI incidence was found between the two strategies, however results may be limited by the small sample size and lack of AKI incidence overall.