Derivation and validation of the Patient risk, Limb severity and ANatomical complexity of disease (PLAN) score

Abstract

Background: In patients with chronic limb-threatening ischemia (CLTI), accurate assessment of the patient's prognosis is crucial to guide decision-making and therapy. The Global Vascular Guidelines on CLTI suggest the multifactorial acronym PLAN, consisting of estimations of Patient risk, Limb severity and Anatomical complexity of disease. After a recent meta-analysis, we constructed and validated a risk score consisting of the GermanVasc CLTI, Wound Ischemia and foot Infection (WIfI), and Global Limb Anatomic Staging System (GLASS) scores. Methods: A retrospective analysis of patients suspected of having CLTI who were evaluated in our multidisciplinary team between 2018 and 2019 was performed. A random limb per patient was chosen for analysis. A Cox proportional hazards model was created using previously published hazard ratios for the GermanVasc score, and remodeled to include—and fit a model on—the WIfI and GLASS score. Bootstrap analysis and corrected c-statistics were performed. Secondary analysis was performed on limbs not included in the model creation, providing Kaplan-Meier amputation-free survival curves and comparing receiver operating characteristics plots between the PLAN model and the individual component scores. A scatterplot of PLAN risk points vs 1-year amputation or death probability was created. Results: A total of 394 patients were included, with a median follow-up of 22.7 months. Complementing the GermanVasc score with 0, 6, 7, or 8 points for WIfI stages 1 to 4, respectively, and 3, 6, 7, or 12 points for GLASS stages 0 to 3 respectively, with 0 points if no imaging was available, showed a good discrimination (optimism corrected-c = 0.71 [0.67-0.77]). Overall 1-year amputation-free survival was 83.2% (79.3%-87.1%), ranging from 94.0% (89.5%-98.8%) in the low-risk group (0-8 points), to 55.9% (45.8%-68.1%) in the high-risk group (22-61 points). PLAN risk points had a significantly higher area under the receiver operating characteristic curve than WIfI (P < .01), GLASS (P < .01), and GermanVasc (P < .05). Conclusions: A single score integrating the components of the Global Vascular Guidelines' PLAN concept for revascularization, combining GermanVasc CLTI score, WIfI, and GLASS, resulted in a significant improvement of risk assessment in patients with CLTI.