Extravascular lung water levels are associated with mortality: a systematic review and meta-analysis

Table 2 The results of the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) assessment of the evidence certainty on the association between the extravascular lung water and mortality

Outcome	Relative effect (95% CI)	No. of patients		No. of participants (studies)	Downgrade factors					Certainty of the evidence (GRADE) 18
Outcome	Relative effect (95% CI)	Survivors	Non-survivors	No. of participants (studies)	Risk of bias	Inconsistency	Imprecision	Indirectness	Publication bias	Certainty of the evidence (GRADE) 18
Mortality	OR 1.69 (1.22–2.34)	464/905 (51.3%)	441/905 (48.7%)	905 Adult patients (11 studies)	Serious^a	Serious^b	Not serious	Serious^c	Not serious	⊕○○○ VERY LOW
Mortality	WMD − 4.97 mL/kg (− 6.54; − 3.41)	680/1246 (54.6%)	566/1246 (45.4%)	1246 Adult patients (17 studies)	Serious^a	Serious^b	Not serious	Serious^c	Not serious	⊕○○○ VERY LOW

CI confidence interval, OR odds ratio, WMD weighted mean difference
^aDowngraded by one level for the risk of bias because ten of 18 studies were evaluated as high risk of bias according to the QUIPS tool and four of 11 studies did not report adjusted OR. Nevertheless, no differences regarding the relative effects were observed between high and moderate and low risk of bias in subgroup analysis
^bDowngraded by one level for inconsistency: substantial heterogeneity is seen between studies (I² > 75%)
^cDowngraded by one level for indirectness because different cut-off values for elevated EVLW definition (> 7 mL/kg/m² in two studies, > 10 mL/kg/m² in eight studies, not available in eight studies)

ISSN: 1364-8535