Search

Objective: Claudication is an early-stage subtype of peripheral artery disease that may result in surgical intervention. The purpose of this study was to compare the types of elective bypass in a population of Black and White patients with claudication and to identify actionable areas that may explain the higher limb amputation rates observed in Black patients. Methods: We identified Black and White patients undergoing elective infrainguinal bypass for claudication using data from the Vascular Quality Initiative registry (2011-2018). Distal bypass target was classified as 1) at or above the popliteal artery (above-knee) or 2) below the popliteal, including all tibial, peroneal, dorsalis pedis and tarsal/plantar arteries (below-knee). Graft conduit was categorized as vein or prosthetic. We calculated one-year risk of major lower limb amputation with 95% confidence intervals (CI) for all combinations of target and conduit, stratified by race. We then forced an intervention on the data to estimate the post-intervention incidence of amputation and the proportion of racial disparity eliminated and 95% CI if everyone received the gold standard treatment (above-knee, vein). Results: We identified 8,401 infrainguinal bypass revascularizations for claudication (12% Black/88% White); 82% were performed above-knee and 44% of those used a vein conduit. The remaining 18% of bypasses were to below-knee arteries, of which 78% used a vein conduit. The remaining 1,534 (18.3%) bypasses were to below-knee arteries, including 1,197 (78.4%) using vein and 330 (21.6%) using prosthetic material. One-year incidence of major limb amputation was 1.6% (1.3%,1.9%) among all grafts, 1.2% (1.0%,1.5%) in above-knee grafts, and 3.3% (2.4%,4.2%) in below-knee grafts. Black patients had higher incidence of amputation across bypass target/conduit groups compared to White patients; the difference was greatest for below-knee revascularizations (8.9%, 95% CI: 4.8%,12.3% v 2.1%, 95% CI: 0.9%,2.9%). Pre-intervention amputation incidence was 1.3% (1.0%,1.5%) in White patients and 3.6% (1.9%,5.2%) in Black patients. Overall amputation incidence decreased significantly after our gold standard intervention was applied for both White (1.0%; 0.7%,1.2%) and Black (2.0%; 0.7%,3.2%) patients. The risk difference (i.e. the disparity gap) between Black and White patients was lessened post-intervention by 56.8% (31.9%,116.3%). Conclusions: We observed risk of amputation higher than would be expected based on other studies of the natural history of claudication without bypass surgery; future work should identify the mechanisms that alter the natural history of claudication via open bypass surgery. Interventions below-knee or using a prosthetic graft were particularly harmful and their harm was concentrated in Black patients.

Objective: Peripheral artery disease (PAD) is a common atherosclerotic disorder that reduces blood flow to the lower extremities. There are significant racial disparities in the outcomes of patients with PAD. Ideally, understanding the result of inadequate implementation of proven interventions on health disparities might provoke a policy and culture shift in the provision of care for minoritized persons. We demonstrate a method that leverages knowledge of interventions that we know should be implemented to optimize care for people with PAD, to identify the extent to which disparities might be mitigated by more universal uptake of those interventions. Methods: We compare racial disparities as they exist currently with disparities that would exist under complete uptake of interventions to improve PAD and vascular care. This counterfactual world is modeled by upweighting people who received the proposed interventions to represent the study population at the start of follow-up. It relies on the assumption that we have measured (and adjusted for) all confounders of the intervention (a hypothesized mediator of the observed health disparities) and the outcome. A unique feature of the application of this method to this problem is our ability to demonstrate how, even when guideline-based care appears evenly distributed by race/ethnicity, increasing access to such care may still decrease health disparities. This method was previously proposed for estimating potential reductions in racial disparities for people with HIV. Results: Across 100,579 infrainguinal revascularizations among Black (n=15,442), Hispanic (n=5,506) and White (n=67,651) patients treated for symptomatic PAD, Black (56.5%) and Hispanic patients (57.6%) were slightly less likely than White patients (59.8%) to receive Class I recommended aspirin and statin therapy prior to their procedures. One-year risk of amputation was 2.7% (95% CI: 2.6%,2.8%) in White patients, 5.8% (5.4%,6.2%) in Black patients and 5.6% (5.0%,6.2%) in Hispanic patients. Had all patients received the appropriate anti-ischemic medications, amputation risk would have been lower across all three race-ethnicity groups. However, specifically, the risk difference (RD) for Black vs White patients would have reduced by 8% (-3%,20%) and the RD for Hispanic vs White patients would have reduced by 18% (-1%,36%). Conclusions: Not only is provision of aspirin and statin for infrainguinal revascularization good for all PAD patients, it is also a social justice imperative.

Excess adiposity, which affects 30% of the world’s population, is associated with risk of coronary heart disease (CHD), yet the potential reductions in CHD burden attainable through shifts in the population distributions of adiposity are unclear. Risk of CHD conveyed by excess adiposity is mostly mediated by the associated metabolic dysregulation, with manifestations such as hypertension and diabetes. Considering these metabolic pathways, we estimated the effect of hypothetical population reductions in general adiposity [body mass index (BMI)] or visceral adiposity [indexed by waist circumference (WC)], each consistent with lifestyle modification, on the risk of incident CHD in a US-based biracial population. The study population included 13,610 ARIC study participants aged 45-64 years, after excluding those with CHD (667) or chronic conditions associated with weight change (969) at baseline. Our hypothetical intervention reduced general adiposity (BMI) or waist circumference (WC) by 5% relative to the temporal trend observed under no intervention; the intervention was applied only among those with BMI > 24 kg/m2 (or WC>88 cm). For example, an individual who increased from a BMI of 25.2 to 27 over the study period under no intervention would increase from 24 to 25.7 following the intervention. Incident CHD was ascertained from 1987 to 2001. CHD risk differences were estimated comparing the intervention to no intervention. Over the follow-up time, 736 (BMI analysis) and 712 (WC analysis) incident CHD events occurred. For the BMI analysis, the median BMI (kg/m2) at the end of follow-up was 28.2 under no intervention and 25.6 under the hypothetical intervention. The cumulative 12-year incidence of CHD and 95% CI under no intervention was 6.3% (5.9, 6. 8%) and the risk difference following the hypothetical BMI change was -0.6% (-1.0, -0.1%). For the WC analysis, the median WC (cm) at the end of follow-up was 100.6 under no intervention and 98.3 under the hypothetical intervention. The cumulative 12-year incidence of CHD and 95% CI was 6.2% (5.8, 6. 7%) under no intervention and the risk difference following the hypothetical WC change was -1.0% (-1.4, -0.5%). Hence, 9% and 16% of CHD events occurring in this study population over 12 years could have been prevented by an annual 5% shift in BMI and WC, respectively. We estimated that meaningful reductions in CHD risk could result from modest reductions in adiposity that are consistent with what might be observed under public health efforts aimed at lifestyle modification. CHD risk reduction was larger under a hypothetical WC reduction compared to a BMI reduction, consistent with metabolic dysregulation associated with visceral adiposity. Public health messages and clinical recommendations that promote modification in WC, perhaps with reference to clothing size, would offer more concrete targets than similar messages for composite measures such as BMI.