A Risk Score for Predicting Long-Term Mortality Following Off-Pump Coronary Artery Bypass Grafting

Tomasz Kamil Urbanowicz; Michał Michalak; Aleksandra Gąsecka; Anna Olasińska-Wiśniewska; Bartłomiej Perek; Michał Rodzki; Michał Bociański; Marek Jemielity

doi:10.3390/jcm10143032

A Risk Score for Predicting Long-Term Mortality Following Off-Pump Coronary Artery Bypass Grafting

J Clin Med. 2021 Jul 7;10(14):3032. doi: 10.3390/jcm10143032.

Authors

Tomasz Kamil Urbanowicz¹, Michał Michalak², Aleksandra Gąsecka³, Anna Olasińska-Wiśniewska¹, Bartłomiej Perek¹, Michał Rodzki¹, Michał Bociański¹, Marek Jemielity¹

Affiliations

¹ Department of Cardiac Surgery and Transplantology, Poznan University of Medical Sciences, 61-701 Poznan, Poland.
² Department of Computer Science and Statistics, Poznan University of Medical Sciences, 61-701 Poznan, Poland.
³ 1st Chair and Department of Cardiology, Medical University of Warsaw, 02-091 Warsaw, Poland.

Abstract

Background: Off-pump coronary artery bypass grafting (OPCAB) comprises 15-30% of all bypass grafting surgeries. The currently available perioperative scores such as Euroscore and STS score do not specifically predict long-term mortality after off-pump procedures. The neutrophil-to-lymphocyte ratio (NLR) is one of the new, easily accessible markers of inflammation with proven predictive value in cardiovascular diseases. We aimed to develop the first risk score for long-term mortality after OPCAB and to determine if the perioperative value of NLR predicts long-term mortality in OPCAB patients.

Methods: In total, 440 consecutive patients with multivessel stable coronary artery disease undergoing OPCAB were recruited. Differential leukocyte counts were obtained by a routine hematology analyzer. Data regarding mortality during a median follow-up time of 5.3 years were obtained from the Polish National Health Service database. An independent population of 242 patients served as a validation cohort.

Results: All-cause mortality was influenced by different clinical risk factors. In multivariate regression analysis, chronic obstructive pulmonary disease, stroke history, post-operative NLR and LVEF were independent predictors of mortality. Combing all independent predictors predicted long-term all-cause mortality with 68.5% sensitivity and 71.5% specificity (AUC = 0.704, p < 0.001). After weighing these variables according to their estimates in a multivariate regression model, we developed a score to predict mortality in patients undergoing OPCAB (PREDICT-OPCAB Score, ranging from 0 to 10). Patients with a high score were at higher risk of mortality within the median 5.3 years of follow-up (score 0-3: 8.3%; 4-6: 27.0%; 7-10: 40.0%; p < 0.001 for score 0-3 vs. 4-6 and 7-10). This association was confirmed in the validation cohort.

Conclusions: We developed and validated the first simplified risk score to predict mortality following OPCAB based on easily accessible clinical factors. This risk score can be used when obtaining a patient's informed consent and as an aid in determining treatment.

Keywords: coronary artery bypass grafting; long-term mortality; multivessel coronary artery disease; neutrophil-to-lymphocyte ratio; off-pump; prediction; risk stratification.