Invasive mechanical ventilation as a risk factor for acute kidney injury in the critically ill: a systematic review and meta-analysis

Abstract

Introduction: Mechanical ventilation (MV) is commonly regarded as a risk factor for acute kidney injury (AKI) in the critically ill. We investigated the strength of this association and whether settings of tidal volume (Vt) and positive end-expiratory pressure (PEEP) affect the risk for AKI.

Methods: We performed a systematic review and meta-analysis using studies found by searching MEDLINE, EMBASE, and references in relevant reviews and articles. We included studies reporting on a relation between the use of invasive MV and subsequent onset of AKI, or comparing higher with lower Vt or PEEP and subsequent onset of AKI. All studies clearly stating that MV was initiated after onset of AKI were excluded. We extracted the proportion with and without MV and AKI. We included 31 studies on invasive MV.

Results: The pooled odds ratio (OR) for the overall effect of MV on AKI was 3.16 (95% CI 2.32 to 4.28, P<0.001). Nearly all subgroups showed that MV increases the risk for AKI. The pooled OR for studies with a multivariate analysis including MV as a risk factor for AKI was 3.58 (95% CI 1.85 to 6.92; P<0.001). Different settings of Vt and PEEP showed no effect.

Conclusions: Invasive MV is associated with a threefold increase in the odds of developing AKI and various Vt or PEEP settings do not modify this risk. The latter argues in favour of a haemodynamic origin of AKI during MV.

Figure 1

Flow chart of search results and study selection.

Figure 2

Forest plot of the studies comparing the patients with and without mechanical ventilation on the onset of acute kidney injury categorised according to inclusion diagnosis. MV+, with mechanical ventilation; MV-, without mechanical ventilation; IV, inverse variance; CI, confidence interval. ^a Only OR reported.

Figure 3

Forest plot of the studies reporting on mechanical ventilation as a risk factor for acute kidney injury in multivariate analysis. MV+, with mechanical ventilation; MV-, without mechanical ventilation; IV, inverse variance; CI, confidence interval; PEEP, positive end-expiratory pressure.

Figure 4

Forest plot of the studies comparing higher with lower tidal volume on the occurrence of acute kidney injury. IV, inverse variance; CI, confidence interval; Vt, tidal volume. ^aAcute kidney injury pre-randomisation omitted; ^bdata extracted from bar graph.

Figure 5

Forest plot of the studies comparing higher with lower positive end-expiratory pressure on the occurrence of acute kidney injury. IV, inverse variance, CI, confidence interval; PEEP, positive end-expiratory pressure. ^aAcute kidney injury pre-randomisation omitted; ^bdata extracted from bar graph; ^cdialysis rates exclude patients receiving dialysis at the time of enrollment.