Prognostic value of extravascular lung water index in critically ill patients: a systematic review of the literature

Abstract

Background: The prognostic value of extravascular lung water (EVLW) has been widely investigated; however, a wide range of its predictive accuracy has been reported.

Study design: A meta-analysis of diagnostic test studies was conducted.

Setting and population: Various patient populations in the intensive care unit were included, such as burned patients and patients with acute lung injury/acute respiratory distress syndrome and sepsis.

Selection criteria: A computerized search of PubMed, Current Contents, CINAHL, and EMBASE from inception until March 1, 2011, was performed to identify potentially relevant articles. The inclusion criteria were studies investigating the prognostic value of EVLW in critically ill patients. There was no language restriction in the searching.

Index tests: The EVLW index (EVLWI) was used.

Reference tests: The outcome was mortality (including in-hospital mortality, intensive care unit mortality, and 28-day mortality).

Results: We analyzed data from 11 studies and 9 countries involving 670 patients. Overall, the EVLWI was significantly higher in nonsurvivors than in survivors, with a mean difference of 5.06 mL/kg (95% confidence interval, -7.53 to -2.58). The heterogeneity was significant with I(2) = 90%. The pooled statistics of diagnostic accuracy together with relevant 95% confidence interval were as follows: sensitivity, 0.81 (0.72-0.88); specificity, 0.66 (0.55-0.76); diagnostic odds ratio, 8.84 (3.83-20.4), positive likelihood ratio, 2.44 (1.69-3.52); negative likelihood ratio, 0.28 (0.16-0.46).

Limitations: The sample sizes of included studies were small.

Conclusion: The EVLWI appears to be a good predictor of mortality in critically ill patients.