Advanced

Will this hemodynamically unstable patient respond to a bolus of intravenous fluids?

Bentzer, Peter LU ; Griesdale, Donald E.; Boyd, John; MacLean, Kelly; Sirounis, Demetrios and Ayas, Najib T. (2016) In JAMA: the journal of the American Medical Association 316(12). p.1298-1309
Abstract

IMPORTANCE Fluid overload occurring as a consequence of overly aggressive fluid resuscitation may adversely affect outcome in hemodynamically unstable critically ill patients. Therefore, following the initial fluid resuscitation, it is important to identify which patients will benefit from further fluid administration. OBJECTIVE To identify predictors of fluid responsiveness in hemodynamically unstable patients with signs of inadequate organ perfusion. DATA SOURCES AND STUDY SELECTION Search of MEDLINE and EMBASE (1966 to June 2016) and reference lists from retrieved articles, previous reviews, and physical examination textbooks for studies that evaluated the diagnostic accuracy of tests to predict fluid responsiveness in... (More)

IMPORTANCE Fluid overload occurring as a consequence of overly aggressive fluid resuscitation may adversely affect outcome in hemodynamically unstable critically ill patients. Therefore, following the initial fluid resuscitation, it is important to identify which patients will benefit from further fluid administration. OBJECTIVE To identify predictors of fluid responsiveness in hemodynamically unstable patients with signs of inadequate organ perfusion. DATA SOURCES AND STUDY SELECTION Search of MEDLINE and EMBASE (1966 to June 2016) and reference lists from retrieved articles, previous reviews, and physical examination textbooks for studies that evaluated the diagnostic accuracy of tests to predict fluid responsiveness in hemodynamically unstable adult patients who were defined as having refractory hypotension, signs of organ hypoperfusion, or both. Fluid responsiveness was defined as an increase in cardiac output following intravenous fluid administration. DATA EXTRACTION Two authors independently abstracted data (sensitivity, specificity, and likelihood ratios [LRs]) and assessed methodological quality. A bivariate mixed-effects binary regression model was used to pool the sensitivities, specificities, and LRs across studies. RESULTS A total of 50 studies (N = 2260 patients) were analyzed. In all studies, indices were measured before assessment of fluid responsiveness. The mean prevalence of fluid responsiveness was 50% (95%CI, 42%-56%). Findings on physical examination were not predictive of fluid responsiveness with LRs and 95%CIs for each finding crossing 1.0. A low central venous pressure (CVP) (mean threshold 8mmHg) was associated with fluid responsiveness (positive LR, 2.6 [95%CI, 1.4-4.6]; pooled specificity, 76%), but a CVP greater than the threshold made fluid responsiveness less likely (negative LR, 0.50 [95%CI, 0.39-0.65]; pooled sensitivity, 62%). Respiratory variation in vena cava diameter measured by ultrasound (distensibility index >15%) predicted fluid responsiveness in a subgroup of patients without spontaneous respiratory efforts (positive LR, 5.3 [95%CI, 1.1-27]; pooled specificity, 85%). Patients with less vena cava distensibility were not as likely to be fluid responsive (negative LR, 0.27 [95%CI, 0.08-0.87]; pooled sensitivity, 77%). Augmentation of cardiac output or related parameters following passive leg raising predicted fluid responsiveness (positive LR, 11 [95%CI, 7.6-17]; pooled specificity, 92%). Conversely, the lack of an increase in cardiac output with passive leg raising identified patients unlikely to be fluid responsive (negative LR, 0.13 [95%CI, 0.07-0.22]; pooled sensitivity, 88%). CONCLUSIONS AND RELEVANCE Passive leg raising followed by measurement of cardiac output or related parameters may be the most useful test for predicting fluid responsiveness in hemodynamically unstable adults. The usefulness of respiratory variation in the vena cava requires confirmatory studies.

(Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
JAMA: the journal of the American Medical Association
volume
316
issue
12
pages
12 pages
publisher
American Medical Association
external identifiers
  • scopus:84995752823
  • wos:000383959800021
ISSN
0098-7484
DOI
10.1001/jama.2016.12310
language
English
LU publication?
yes
id
b7680a49-fc7c-41d5-ac5c-e98fa025286c
date added to LUP
2016-12-01 10:21:53
date last changed
2017-09-18 11:31:50
@article{b7680a49-fc7c-41d5-ac5c-e98fa025286c,
  abstract     = {<p>IMPORTANCE Fluid overload occurring as a consequence of overly aggressive fluid resuscitation may adversely affect outcome in hemodynamically unstable critically ill patients. Therefore, following the initial fluid resuscitation, it is important to identify which patients will benefit from further fluid administration. OBJECTIVE To identify predictors of fluid responsiveness in hemodynamically unstable patients with signs of inadequate organ perfusion. DATA SOURCES AND STUDY SELECTION Search of MEDLINE and EMBASE (1966 to June 2016) and reference lists from retrieved articles, previous reviews, and physical examination textbooks for studies that evaluated the diagnostic accuracy of tests to predict fluid responsiveness in hemodynamically unstable adult patients who were defined as having refractory hypotension, signs of organ hypoperfusion, or both. Fluid responsiveness was defined as an increase in cardiac output following intravenous fluid administration. DATA EXTRACTION Two authors independently abstracted data (sensitivity, specificity, and likelihood ratios [LRs]) and assessed methodological quality. A bivariate mixed-effects binary regression model was used to pool the sensitivities, specificities, and LRs across studies. RESULTS A total of 50 studies (N = 2260 patients) were analyzed. In all studies, indices were measured before assessment of fluid responsiveness. The mean prevalence of fluid responsiveness was 50% (95%CI, 42%-56%). Findings on physical examination were not predictive of fluid responsiveness with LRs and 95%CIs for each finding crossing 1.0. A low central venous pressure (CVP) (mean threshold 8mmHg) was associated with fluid responsiveness (positive LR, 2.6 [95%CI, 1.4-4.6]; pooled specificity, 76%), but a CVP greater than the threshold made fluid responsiveness less likely (negative LR, 0.50 [95%CI, 0.39-0.65]; pooled sensitivity, 62%). Respiratory variation in vena cava diameter measured by ultrasound (distensibility index &gt;15%) predicted fluid responsiveness in a subgroup of patients without spontaneous respiratory efforts (positive LR, 5.3 [95%CI, 1.1-27]; pooled specificity, 85%). Patients with less vena cava distensibility were not as likely to be fluid responsive (negative LR, 0.27 [95%CI, 0.08-0.87]; pooled sensitivity, 77%). Augmentation of cardiac output or related parameters following passive leg raising predicted fluid responsiveness (positive LR, 11 [95%CI, 7.6-17]; pooled specificity, 92%). Conversely, the lack of an increase in cardiac output with passive leg raising identified patients unlikely to be fluid responsive (negative LR, 0.13 [95%CI, 0.07-0.22]; pooled sensitivity, 88%). CONCLUSIONS AND RELEVANCE Passive leg raising followed by measurement of cardiac output or related parameters may be the most useful test for predicting fluid responsiveness in hemodynamically unstable adults. The usefulness of respiratory variation in the vena cava requires confirmatory studies.</p>},
  author       = {Bentzer, Peter and Griesdale, Donald E. and Boyd, John and MacLean, Kelly and Sirounis, Demetrios and Ayas, Najib T.},
  issn         = {0098-7484},
  language     = {eng},
  month        = {09},
  number       = {12},
  pages        = {1298--1309},
  publisher    = {American Medical Association},
  series       = {JAMA: the journal of the American Medical Association},
  title        = {Will this hemodynamically unstable patient respond to a bolus of intravenous fluids?},
  url          = {http://dx.doi.org/10.1001/jama.2016.12310},
  volume       = {316},
  year         = {2016},
}