Enhanced risk stratification utilizing highly sensitive cardiac troponin T in cancer patients with atrial fibrillation presenting to an emergency department

Background Atrial fibrillation (AF) and cancer are frequent comorbidities. However, data for risk stratification in cancer patients presenting with AF in the setting of an emergency department (ED) is still lacking. This study thought to evaluate the role of highly sensitive cardiac troponin T (hs-cTnT) in cancer patients presenting with AF in an ED setting. Methods This study used data from a single-centre retrospective observational Heidelberg Registry of Atrial Fibrillation (HERA-FIB) stratified by presence of cancer and hs-cTnT. Patients were consecutively included within HERA-FIB when presenting with the diagnosis of atrial fibrillation to the emergency department (ED) of our Hospital between June 2009 and March 2020. Diagnosis for the history of cancer and a composite endpoint (EP) consisting of all-cause mortality, stroke, major bleeding or myocardial infarction were retrospectively accessed using a sequential follow-up. Patients with history of cancer and at least one available hs-cTnT value were included within this study. For comparison of continuous parameters between 3 groups Kruskal Wallis test and Dunn Post-hoc analysis was used. For comparison of categorical variables Chi² test and p for trend was utilized. Results 1,544 (15.1%) of AF patients within HERA-FIB presented with the diagnosis of cancer. Patients with the known cancer were stratified in hs-cTnT categories < 5 ng/L; 5-14 ng/L > 14 ng/L. Here, 48 (3.1%) patients showed an hs-cTnT < 5 ng/L, 425 (27.5%) showed an hs-cTnT between 5-14 ng/L and the majority of patients (1,071; 69.14%) presented with an hsc-TnT >14 ng/Regarding the composite endpoint, 3 (6.2)% in patients with hs-cTnT <5 ng/L, 77 (18.1%) of the patients with hs-cTnT 5-14 ng/L and 496 (46.3%) of the patients with hscTnT > 14 ng/L. The HR for the composite EP was 3.12 (95%CI 2.10-4.96) for patients with hs-cTnT 5-14 ng/L compared to patients with an hs-cTnT < 5 ng/L and 10.00 (95%CI 6.54-15.30) for patients with an hs-cTnT > 14 ng/L. In cancer patients, the Area under the curve (AUC) for hs-cTnT in prediction of the composite EP was 0.709 (95%CI 0.685-0.731), p<0.0001. In a Cox proportional hazard regression model adjusted for age, sex, c-reactive protein C, creatinine, arterial hypertension, presence of coronary artery disease and frist diagnosed AF hs-cTnT > 14 remained an independent risk factor for the composite endopoint aHR: 6.67 (2.13-21.04), p=0.0012. Conclusions Hs-cTnT could be utilized for risk stratification in AF patients presenting with known of cancer diagnosis to an ED, adding valuable information for this vulnerable patient subset.