Obesity paradox in peripheral artery disease

Background & aims: Previous studies have suggested an obesity survival paradox in patients with peripheral artery disease (PAD). We investigated the influence of obesity and underweight on adverse in-hospital outcomes in PAD.

Methods: Patients diagnosed with PAD based on ICD-code I70.2 of the German nationwide database were stratified for obesity, underweight and a reference group with normal-weight/over-weight and compared regarding adverse in-hospital outcomes.

Results: Between 01/2005-12/2015, 5,611,484 inpatients (64.8% males) were diagnosed with PAD; of those, 8.9% were coded with obesity and 0.3% with underweight. Obese patients were younger (70 (IQR 63/76) vs. 73 (66/80) years, P < 0.001), more frequently female (36.7% vs. 35.1%, P < 0.001), had less cancer (4.9% vs. 7.9%, P < 0.001) and had less treatment with major amputation (2.6% vs. 3.2%, P < 0.001) compared to the reference group. Overall, 277 876 (5.0%) patients died in-hospital. Obese patients showed lower mortality rate (3.2% vs. 5.1%, P < 0.001) compared to the reference group and reduced risk of in-hospital mortality (OR, 0.617 [95%CI 0.607-0.627], P < 0.001). This "obesity paradox" was demonstrated in obesity classes I (OR, 0.475 [95%CI 0.461-0.490], P < 0.001), II (OR, 0.580 [95%CI 0.557-0.605], P < 0.001), and III (OR, 0.895 [95%CI 0.857-0.934], P < 0.001) and was independent of age, sex and comorbidities. Underweight patients revealed higher in-hospital mortality (6.0% vs. 5.1%, P < 0.001) compared to the reference group (OR, 1.179 [95%CI 1.106-1.257], P < 0.001) and showed higher prevalence of cancer (22.0% vs. 7.9%, P < 0.001).

Conclusions: Coding for obesity is associated with lower in-hospital mortality in PAD patients relative to those with normal-weight/over-weight. This obesity survival paradox was independent of age, gender and comorbidities and observed for all obesity classes.