Send to

Choose Destination
See comment in PubMed Commons below
Crit Care Med. 2004 Apr;32(4):986-91.

Osmole gap in neurologic-neurosurgical intensive care unit: Its normal value, calculation, and relationship with mannitol serum concentrations.

Author information

Department of Neurology, Washington University School of Medicine, St. Louis, MO, USA.



To determine a) if the admission osmole gap, the difference between osmolality and osmolarity, is the same in the neurologic-neurosurgical intensive care unit (NNICU) population as in healthy controls; b) which of 11 osmole gap formulas, or osmolality, correlates best with mannitol serum concentrations; c) whether osmole gap correction for plasma water content improves this correlation; and d) whether the osmole gap can predict mannitol serum concentrations.


Prospectively collected data.


NNICU of a tertiary teaching hospital.


Ten NNICU patients on mannitol and eight not on mannitol, and 95 healthy controls.




We compared the admission osmole gap between all 18 NNICU patients and healthy controls and the correlation between osmole gap or osmolality and mannitol serum concentrations in ten NNICU patients while receiving mannitol. The osmole gap was calculated using 11 osmolarity formulas (six corrected for plasma water content). Student's t-test was used to compare the mean osmole gap between control and patient groups.We found that the mean osmole gap in healthy subjects and NNICU patients was not different. There were no statistically significant differences between any of the 11 osmole gap formulas and the correlation of osmole gap with serum mannitol concentrations; the highest R =.80, with formula 4, 1.86 (sodium + potassium) + (blood urea nitrogen/2.8) + (glucose/18) + 10, requires the least laboratory measurements. Osmolality had the lowest correlation with mannitol concentration (R =.60), significantly lower than any of the osmole gap calculations. Plasma water content correction did not improve this correlation. The osmole gap-mannitol serum concentrations relationship is 1 to 0.81, not accurate enough to predict specific mannitol serum concentrations.


The osmole gap correlates better with mannitol serum concentrations than osmolality, and although it cannot predict a specific mannitol serum concentration, a normal osmole gap concentration, as we find at trough times, indicates sufficient clearance for a new mannitol dose.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Lippincott Williams & Wilkins
    Loading ...
    Support Center