Prognostic factors for lower limb revascularization outcomes

Abstract

Aim: The aim of this thesis was to investigate the applicability of preoperative magnetic resonance angiography (MRA) examinations regarding assessment of infrapopliteal lesion severity; to assess the agreement, reproducibility and predictive value of established anatomical classifications and a new Infrapopliteal Total Atherosclerotic Burden (I-TAB); to investigate the prevalence of 2-4 point major adverse limb event (2-4P MALE) in a nationwide patient population with peripheral arterial disease (PAD) who underwent lower limb revascularization; and to identify predictors of post-operative MALE. Methods: Patients with chronic limb-threatening ischemia (CLTI) who underwent infrapopliteal revascularization at one (Papers I-II) to three (Paper III) vascular centres between 2008 and 2016 were included. Preoperative MRA (paper I-III) and perioperative digital subtraction angiographies (DSA, paper I-II) were assessed by up to four different observers. Intra- and interobserver agreement were analyzed using Krippendorff’s . Receiver operating characteristics (ROC) curves and the associated area under the curve (AUCroc) were used to analyze the predicted value of anatomical classifications, including I-TAB, regarding limb salvage. Population-based health data registries were used to study the subsequent MALE incidences in the entire Swedish revascularized PAD cohort between 2008-2016. Kaplan-Meier estimates were used for time-to-event analysis for intermittent claudication (IC) and CLTI separately. Stepwise Cox proportional hazard models were used for predictor candidate analysis. Results: Krippendorff’s  showed poor agreement for the Trans-Atlantic Steering Committee (TASC II) grading especially regarding the choice of target artery (0.12-0.19 for MRA and 0.39-0.48 for DSA in 68 patients with examinations available for assessment). A tentative agreement was observed for individual infrapopliteal arteries (0.64-0.79 for MRA and 0.66-0.84 for DSA). Krippendorff’s -values for tentative agreement was also observed for I-TAB (MRA 0.76 (95% CI 0.70-0.81) and DSA 0.79 (95% CI 0.74-0.84)). In paper III, 240 MRA examinations were assessed and graded according to the global anatomic severity score (GLASS), TASC II and I-TAB. None of these could predict the risk for major limb amputation after revascularization (AUCroc was 0.61, 0.68 and 0.63 respectively). The MALE cumulative incidence was high for both IC and CLTI regardless of applied MALE definition. In IC, the one-year event probability for 2-4P MALE were 3.3 (95% CI 2.9-3.6), 8.3 (95% CI 7.8-8.9) and 21.4 (95% CI 20.6-22.2), respectively. For CLTI, the one-year event probability was 21.1 (95% CI 20.4-21.7), 38.3 (95% CI 37.6-39.1) and 46.9 (95% CI 46.1-47.7) for 2-4P MALE, respectively. Chronic kidney disease correlated with higher 2-4P MALE. Previous lower limb revascularization was associated with a higher subsequent risk of MALE in IC patients. In CLTI patients, previous lower limb amputation was associated with subsequent MALE. Conclusions: The interobserver agreement of infrapopliteal arterial lesion assessment from preoperative MRA and DSA examinations is poor for TASC II which appears to be on account of the choice of target artery. The proposed I-TAB score is more reproducible and may be a better tool for clinical decisions and scientific evaluations than TASC II. There is no clear association between the preoperative infrapopliteal GLASS or TASC II grade, or the proposed I-TAB, and subsequent lower limb amputation following revascularization. A reliable prediction of subsequent amputation risk could thus not be established from preoperative MRA examinations. We observed very high 2-4P MALE incidences in a population-based revascularized lower limb PAD cohort. Previous lower limb revascularization was associated with subsequent MALE in patients with IC and previous lower limb amputation was associated with subsequent MALE in patients with CLTI. Chronic kidney disease correlated with higher event rates across all studied MALE definitions.