Coronary artery disease, left ventricular function and cardiac biomarkers determine all-cause mortality in cancer patients-a large monocenter cohort study.

Cancer patients are at risk of suffering from cardiovascular diseases (CVD). Nevertheless, the impact of cardiovascular comorbidity on all-cause mortality (ACM) in large clinical cohorts is not well investigated. In this retrospective cohort study, we collected data from 40,329 patients who were subjected to cardiac catherization from 01/2006 to 12/2017 at University Hospital Heidelberg. The study population included 3666 patients with a diagnosis of cancer prior to catherization and 3666 propensity-score matched non-cancer patients according to age, gender, diabetes and hypertension. 5-year ACM in cancer patients was higher with a reduced left ventricular function (LVEF < 50%; 68.0% vs 50.9%) or cardiac biomarker elevation (high-sensitivity cardiac troponin T (hs-cTnT; 64.6% vs 44.6%) and N-terminal brain natriuretic peptide (NT-proBNP; 62.9% vs 41.4%) compared to cancer patients without cardiac risk. Compared to non-cancer patients, NT-proBNP was found to be significantly higher (median NT-proBNP cancer: 881 ng/L, IQR [254; 3983 ng/L] vs non-cancer: 668 ng/L, IQR [179; 2704 ng/L]; p < 0.001, Wilcoxon-rank sum test) and turned out to predict ACM more accurately than hs-cTnT (NT-proBNP: AUC: 0.74; hs-cTnT: AUC: 0.63; p < 0.001, DeLong's test) in cancer patients. Risk factors for atherosclerosis, such as diabetes and age (> 65 years) were significant predictors for increased ACM in cancer patients in a multivariate analysis (OR diabetes: 1.96 (1.39-2.75); p < 0.001; OR age > 65 years: 2.95 (1.68-5.4); p < 0.001, logistic regression). Our data support the notion, that overall outcome in cancer patients who underwent cardiac catherization depends on cardiovascular comorbidities. Therefore, particularly cancer patients may benefit from standardized cardiac care.