Acute Limb Ischemia in Patients on Veno-Arterial Extracorporeal Membrane Oxygenation (VA-ECMO) Support: A Ten-Year Single-Center Experience.

BACKGROUND

Veno-arterial extracorporeal membrane oxygenation (VA-ECMO) has been increasingly utilized as a life-saving modality in patients with cardiopulmonary compromise. Acute limb ischemia (ALI) has been reported when the femoral artery was accessed, and it was associated with higher mortality in patients on extracorporeal membrane oxygenation (ECMO). However, prior studies were limited by small sample size and the lack of long-term follow-up. We aimed to investigate the predictors of ALI in ECMO and the effect of ALI on long-term mortality.

METHODS

Retrospective institutional chart review was performed. Patients who underwent VA-ECMO (Jan/2008-Jan/2018) were identified. Primary outcomes were ALI and 4-year mortality. Logistic regression analysis was used to investigate the predictors of ALI. Survival analysis methods were used to examine 4-year mortality.

RESULTS

A total of 377 patients were included. Vascular complications took place in 149 (39.5%) patients. The majority was ALI (107, 75.4%). Eleven patients had limb loss. ALI patients were younger (mean age: 50.2 ± 15.9 vs. 54.1 ± 15.5; P = 0.03), were more likely to have history of peripheral arterial disease (PAD) (7.6% vs. 3.0%; P = 0.047), were less likely to be on aspirin (40.2% vs. 54.4%; P = 0.013), and were more likely to be on vasopressors (97.1% vs. 88.0%; P = 0.007). Arterial cannula size was not associated with ALI so as the concurrent use of ventricular offloading devices (intra-aortic balloon pump or Impella (Abiomed Inc, Dancers, MA)). The use of distal perfusion catheter (DPC) was not protective against ALI in this cohort. However, DPC was associated with less likelihood to require vascular intervention (20.1% vs. 32.0%; P = 0.009). In adjusted analysis, aspirin use was protective against ALI [adjusted odds ratios (aOR) 95% CI: 0.52(0.30 - 0.90); P = 0.018]. The only predictor of ALI was the use of vasopressors [aOR (95% CI): 6.8 (1.5 - 30.4); P = 0.012]. For those who were successfully decannulated, 4-year survival was 65.1% in patients without ALI versus 46.8% in ALI (P = 0.044). After adjusting for potential risk factors, 4-year mortality hazard was significantly higher in patients with ALI [HR: 1.80(1.04 - 3.12); P = 0.035].

CONCLUSIONS

Patients requiring ECMO are critically ill. The development of ALI is detrimental to this population. This effect extends beyond the acute period. ALI increased 4-year mortality risk by 80%. The use of DPC did not protect against ALI, but it was associated with less vascular interventions.