P2522A new scoring system to stratify post-surgical valvular patients during cardiovascular rehabilitation: derivation and validation study

Abstract

Abstract Background Guidelines underline the importance of Cardiovascular Rehabilitation (CR) in post-surgical valvular patients both for the functional recovery and the monitoring of complications. However, there are no established indicators to better categorise their risk and to identify the real probability of recovery. Purpose The aim of this study is to propose and validate a scoring system to appropriately stratify post-surgical valvular patients in order to individualise CR programmes. Methods A retrospective study was conducted on 1480 post-surgical valvular patients hospitalized in our CR Unit (902 M – 578 F; median age of 64 years, IQR 53–73). 485 patients underwent single heart valve repair, 408 single heart valve replacement, 237 single heart valve surgery and additional interventions, 249 multiple valve interventions and 101 multiple heart valves and additional interventions. Subjects were randomised in two groups for data analysis: a Derivation (D; n=1000) and a Validation (V; n=480) group. Initially, in group D we assessed the predictive value of anamnestic, clinical and laboratory variables for major complications and functional recovery. We created two scoring systems for these outcomes and, subsequently, we validated them on group V. Finally, we interlaced them in an operative algorithm. Results Chronic kidney disease (OR 2.588; 95% CI 1.232–5.436; p=0.012), sternal surgical re-synthesis (OR 7.757; 95% CI 2.042–29.471; p=0.003), post-surgical transfusions (OR 2.419; 95% CI 1.407–4.161; p=0.001) and Troponin T peak >1400 μg/L (OR 2.441; 95% CI 1.418–4.200; p=0.001) were independent predictors for the occurrence of major complications in group D. Age (OR 0.958; 95% CI 0.9339–0.977; p<0.001), post- surgical transfusions (OR 1.981; 95% CI 1.160–3.380; p<0.001) and METS at admission (OR 0.032; 95% CI 0.017–0.061; p<0.001) were independent predictors of a higher functional recovery in group D. When the two scoring systems were validated on group V, we obtained a z score of 0.07 (p=0.941) for the major complications risk score and a z score of 1.23 (p=0.219) for the functional recovery stratification system, respectively, indicating a very reliable model. We proceeded to build an operative algorithm to stratify patients and propose personalised CR strategies. Conclusions We identified predictors to stratify the risk of complications and to define the probability of recovery in post-surgical valvular patients undergoing CR. The proposed final operative algorithm may be a unique tool to support the cardiologist to tailor rehabilitation programmes. This may lead to better outcomes and reduction of healthcare expenditure with optimisation in the use of available resources. Acknowledgement/Funding None