By Samuel Nadler, MD, PhD
Critical Care, Pulmonary Medicine, The Polyclinic Madison Center, Seattle; Clinical Instructor, University of Washington, Seattle
Dr. Nadler reports no financial relationships relevant to this field of study.
SYNOPSIS: The use of balanced salt solutions rather than isotonic saline or colloids may improve in-hospital mortality in patients admitted with septic shock.
SOURCE: Raghunathan K, et al. Association between initial fluid choice and subsequent in-hospital mortality during the resuscitation of adults with septic shock. Anesthesiology 2015;123:1385-1393.
Appropriate fluid resuscitation is a foundation of appropriate sepsis care. Which fluid type is best remains unclear. There are studies comparing crystalloids and colloids, starch solutions, albumin, etc. In practice, however, combinations of these options are often employed. Previously, Raghunathan et al demonstrated a superiority of balanced salt solutions when compared with isotonic saline.1 This study examined the effects of combinations of crystalloid and colloids on in-hospital mortality from sepsis.
This retrospective cohort study between January 2006 and December 2010 evaluated 60,734 patients admitted with sepsis who had received at least 2 L of volume resuscitation, required vasopressors, had not undergone any major surgical procedure, and remained in the ICU for at least 2 days. Four categories of exposures were defined: isotonic saline alone (Sal); isotonic saline and balanced salt solutions (Sal + Bal); isotonic saline and colloids (Sal + Col); and all three (Sal + Bal + Col). The primary outcome was in-hospital mortality with secondary analyses looking at length of stay and cost per day among survivors. As there were significant differences in baseline characteristics in each group, risk adjustments for 27 known comorbidities were used, including inverse probability weighting, propensity score matching, and hierarchical logistic regression methods.
In the study cohort, most patients (n = 44,347) received Sal, while 3651 patients received Sal + Bal, 11,038 received Sal + Col, and 1698 received Sal + Bal + Col. Using various risk adjustment methods, patients in the Sal + Bal cohort had the lowest absolute mortality (17.64-18.83%) as compared with Sal (20.19-21.35%), Sal + Col (24.16-29.94%), or Sal + Bal + Col (19.23-25.15%). In pairwise comparisons, Sal + Bal was associated with the lowest mortality whether Col were used (relative risk [RR], 0.84; 95% confidence interval [CI], 0.76-0.92; P < 0.001) or not (RR, 0.76; 95% CI, 0.70-0.89; P < 0.001). Conversely, administration of Col was not associated with an increased risk when Bal were used but did have an increased mortality rate when in combination with Sal (RR, 1.14; 95% CI, 1.08-1.19; P < 0.001). This effect persisted when the Col were restricted to albumin and hetastarch was excluded in the analysis. Additional sensitivity analysis demonstrated that the difference in mortality with Col administration could be due to an unidentified confounder but that the difference in mortality between Sal and Sal + Bal was robust. Secondary outcomes such as hospital length of stay and costs per day were comparable in the Sal vs Sal + Bal group, but were higher in the cohorts receiving colloids (Sal + Col and Sal + Bal + Col).
COMMENTARY
This study extends the conclusions of a previous analysis of similar data with respect to the choice of fluid for sepsis resuscitation. The baseline differences encountered included a lower rate of congestive heart failure in the Sal + Bal and Sal + Bal + Col cohorts and a higher rate of liver disease in the Sal + Col cohort. With inverse probability weighting and propensity score matching adjustments, these differences can be eliminated but other confounders may still bias results. When evaluating the outcomes in these cohorts at day 2, there are a few striking results. The Col-containing groups had the highest rates of mechanical ventilation, vasopressor use, bicarbonate infusions, total parenteral nutrition, and diuretic needs. Also telling is that the rate of tracheostomy at day 2 in the Sal + Bal+ Col group was 7.07% as compared with 3.12%, 4.74%, and 3.86% in the Sal, Sal + Bal, and Sal + Col groups, respectively, implying that clinicians saw this cohort as having the greatest likelihood of prolonged mechanical ventilation early in the course of the hospital stay.
While this is a retrospective analysis, some confidence can be gained that Bal seem to improve patient outcomes, regardless of Col co-administration. Looking at mortality at the hospital level, with an increasing proportion of balanced salt solution use, there is decreasing mortality. In general, one would not expect clinicians to adjust the proportion of Bal vs Sal based on perceptions of severity of illness. But this may belie that hospital-level interventions and protocols may have an effect on sepsis mortality.
Overall, this study adds to the evidence that Bal improve patient outcomes with sepsis. There has yet to be shown a significant risk or cost associated with their use, and lactated ringers is as prevalent in most ICUs as Sal. It may be time to start using Bal routinely as part of sepsis resuscitation.
REFERENCE
-
Raghunathan K, et al. Association between the choice of IV crystalloid and in-hospital mortality among critically ill adults with sepsis. Crit Care Med 2014;42:1585-1591.
