Six-month functional outcomes also worse

Positive daily fluid balance—more daily fluid input than output—was associated with higher mortality in intensive care unit (ICU) patients with traumatic brain injury (TBI), a prospective study of observational cohorts found.

At the patient level, a mean daily positive fluid balance was associated with both higher ICU mortality (OR 1.10, 95% CI 1.07-1.12 per 0.1 L increase) and worse functional 6-month outcome (OR 1.04, 95% CI 1.03-1.04 per 1-point decrease of the Glasgow Outcome Scale Extended [GOSE] per 0.1 L increase, reported Mathieu van der Jagt, MD, of the University Medical Center in Rotterdam, the Netherlands and co-authors in Lancet Neurology.

Daily fluid balance is the difference between daily inputs (beverages, food, and metabolic processes) and outputs (insensible, respiratory, urinary, and stool loss).

“Our findings, in combination with previous evidence, argue for a more rigorous policy of normovolemia, carefully avoiding both hypervolemia and hypovolemic as indicated by mean neutral fluid balances, given the harm associated with both mean negative and positive fluid balances,” van der Jagt and colleagues wrote.

“However, further research is needed to establish how to implement this knowledge while still respecting individualized approaches (e.g., based on hemodynamic monitoring) and taking into account cerebral perfusion pressure.”

The researchers conducted patient- and center-level analysis on data from two observational cohorts, CENTER-TBI in Europe (data collected from December 2014 to December 2017) and OzENTER-TBI (included patients admitted February 2015 to March 2017) in Australia. Patients included in the study were 16 or older; they were admitted to the ICU within 24 hours of injury with a clinical diagnosis of TBI and had a CT scan of the brain (n=2,125; median age 50, 74% male).

Injury severity overall was mild in 34%, moderate in 17%, and severe in 50%. Hypertonic saline or mannitol was used in 27%. Cranial and extracranial surgery were performed in 42% and 31%, respectively, and 3% overall had renal replacement therapy.

Sensitivity analyses that adjusted for cerebral perfusion pressure and serum sodium did not change the study’s conclusions.

In an accompanying editorial, Patrick Kochanek, MD, of University of Pittsburgh in Pennsylvania, and Ruchira Jha, MD, of Barrow Neurological Institute in Phoenix, Arizona, noted that the present results are “long overdue, and although the findings are based on observational studies, they suggest the need to meticulously titrate fluid use in patients with severe traumatic brain injury, while avoiding dehydration.”

“Perfect euvolemia seems an impractical goal, yet it might warrant closer daily attention than previously appreciated,” they added.

Salient negative effects of hypovolemia and dehydration—for example, reduced cerebral perfusion pressure with associated vasodilation and increased intracranial pressure, or renal failure—suggest maintaining a euvolemic hyperosmolar state in severe TBI, Kochanek and Jha explained. Yet, avoiding hypovolemia may lead to hypervolemia.

“Over the past decade, the deleterious consequences of fluid overload have been highlighted in studies of acute lung injury and sepsis, including worsened lung function and prolonged duration of mechanical ventilation and ICU stay,” the editorialists wrote.

In the present study, center-level analysis showed that higher positive fluid balance also was associated with higher ICU mortality (OR 1.17, 95% CI 1.05-1.29) and worse functional outcome (OR 1.07, 95% CI 1.02-1.13).

Unlike fluid balance results, fluid input findings differed for patient-level versus center-level analysis. Higher mean daily fluid input in patient-level analysis was, like fluid balance, associated with higher ICU mortality (OR 1.05, 95% CI 1.03 to 1.06 per 0.1 L increase) and worse functional outcome (OR 1.04, 95% CI 1.03 to 1.04 per 1-point decrease of the GOSE per 0.1 L increase). However, center-level analysis showed no association between higher fluid input for mortality or functional outcome.

Center-level analysis showed that across 55 hospitals, the mean daily fluid input ranged from 1.58 L to 4.23 L (median 2.91 L) and median average daily fluid balance from -0.85 to +1.13 L (median +0.37 L).

“We found substantial differences in fluid management policies between centers across Europe and Australia,” Van der Jagt and colleagues noted. “The substantial variation observed in our study is in line with earlier studies showing between-center differences in intensive care management of patients with traumatic brain injury.”

“Guidelines could help reduce treatment variation in clinical practice, and awareness of guidelines has increased,” the authors continued. “The variation that we observed might be due to the fact that the Brain Trauma Foundation’s guidelines do not include recommendations about fluid management.”

A 2018 consensus statement on fluid therapy in patients in neurocritical care recommended a goal of normovolemia and avoiding restrictive fluid policies (negative fluid balances), but did not include a statement on risks of positive fluid balance, they added.

Limitations of the analysis include the possibility of confounding in observational data.

“These findings suggest that positive fluid balance might be an underappreciated factor contributing to adverse outcomes,” Van der Jagt and co-authors wrote. “This finding is clinically relevant since a positive fluid balance could be readily modifiable by less liberal fluid administration.”

  1. Positive daily fluid balance was associated with higher mortality in intensive care unit patients with traumatic brain injury (TBI), a prospective study of observational cohorts found.

  2. In combination with other evidence, the findings support a policy of normovolemia, avoiding both hypervolemia and hypovolemic as indicated by mean neutral fluid balances.

Paul Smyth, MD, Contributing Writer, BreakingMED™

This research was funded by the European Commission 7th Framework program, Australian National Health and Medical Research Council, and Transport Accident Commission Victoria, Australia. Additional funding was obtained from Hannelore Kohl Stiftung, OneMind, Integra LifeSciences, and Neurotrauma Sciences.

van der Jagt declared no competing interests.

Kochanek declared no competing interests. Jha is a paid consultant and on the scientific advisory board for Biogen.

Cat ID: 474

Topic ID: 82,474,501,728,791,572,730,474,192

Author