An analysis of early volume resuscitation and the association with prolonged mechanical ventilation

Background Previous studies have found that intravenous fluid administration within the first 24 h may be associated with prolonged mechanical ventilation (PMV). We examined the association between initial 24 h fluids and PMV in combat casualties. Methods This is a secondary analysis of a previously described dataset from the Department of Defense Trauma Registry (DODTR). We included casualties with at least 24 h on the ventilator and no significant traumatic brain injury. The definition of PMV and associations were constructed using univariable and multivariable logistic regression models. Results We identified 1508 casualties available for analysis for this study - 1275 in the non-PMV cohort (<9 days on ventilator vs. 233 in the PMV cohort (>= 9 days on ventilator). Explosives comprised the most common mechanism of injury for both groups (72% vs. 75%) followed by firearms (21% vs. 16%). The composite injury severity score (ISS) was lower in the non-PMV cohort (18 vs. 30, p < .001). There were lower volumes of all resuscitation fluid within the first 24 h in the non-PMV cohort. When adjusting for composite ISS and mechanism of injury in a multivariable logistic regression model with PMV as the outcome, crystalloid volume (unit odds ratio [UOR] 1.07) and colloid volume (UOR 1.03) were both associated with PMV. Conclusions We found that volume of resuscitation fluids were substantially higher in the PMV cohort. Our findings suggest the need for caution with the routine use of crystalloid and colloid in the first 24 h of resuscitation.