Incidence and Risk Factors of Limb Ischaemia in Adult Patients Receiving Veno-Arterial Extracorporeal Membrane Oxygenation: A Systematic Review and Meta-Analysis.

The included studies were observational cohort and case-control designs, so they can describe associations but cannot prove that ECMO or any risk factor caused limb ischaemia. There were no non-ECMO control groups in the included studies, so the incidence may reflect critical illness, immobility, vasopressors or coagulopathy as well as ECMO. Heterogeneity was high (I2 = 91.3%) because limb ischaemia definitions, detection methods, cannulation strategies and patient populations varied across studies. Only English-language studies were included, and studies with a Newcastle-Ottawa Scale score below 5 were excluded. Risk-factor estimates were unstable: PAD was driven by a single influential study, unsuccessful percutaneous cannulation came from two studies, and height came from three studies. The funnel plot was asymmetric, and trim-and-fill adjustment gave a pooled incidence of 9.5% (95% CI: 4.2% to 15.1%). Most studies used peripheral cannulation, so results may not apply to central or alternative strategies. ECMO duration analyses were limited by inconsistent reporting units.

The authors report no funding and no conflicts of interest.

Do not treat pre-existing PAD as a proven, stable risk factor for every ECMO patient. The pooled odds ratio was 6.12, but sensitivity analysis showed the result was driven by a single influential study and was not robust.