Chest pain in patients with cancer: prevalence of myocardial infarction and diagnostic performance of high-sensitivity cardiac troponin T/I

Abstract

Abstract Background Among patients with active or past cancer presenting with acute chest discomfort to the emergency department (ED), multiple uncertainties remain regarding the prevalence of acute myocardial infarction (AMI), outcomes, and the diagnostic utility of recommended diagnostic tools. Purpose We aimed to evaluate 1) the prevalence of AMI among consecutive patients with a history of cancer presenting with acute chest pain to the ED; 2) time to discharge from the ED and hospitalization rate; 3) the diagnostic accuracy of 34 predefined chest pain characteristics (CPCs) and ECG signatures; 4) the diagnostic accuracy of hs-cTnT and hs-cTnI concentrations; and 5) the performance of the European Society of Cardiology (ESC) 0/1h-hs-cTnT/I algorithms. Methods Patients presenting with chest pain to the ED were prospectively enrolled in an international multicenter diagnostic study. Cancer status was prospectively assessed. Final diagnosis was centrally adjudicated by two independent cardiologists, applying the 4th universal definition of AMI. Findings were externally validated in an independent, multicenter cohort. Results Among 8267 patients, 711 (8.6%) had cancer. Cancer patients had a higher burden of cardiovascular risk factors and preexisting cardiac disease, and more often a final diagnosis of AMI (184 [26%] patients with cancer vs 1562 [21%, P<0.001] without cancer). This difference was largely due to a significantly higher proportion of NSTEMIs in cancer patients (23% vs 17%; p<0.001). Total length of stay in the ED was significantly longer in patients with cancer versus those without (5.2h versus 4.3h, p<0.001). The hospitalization rate was higher in cancer patients (49.8%) than in non-cancer patients (34.3%, P<0.001), driven by non-AMI diagnoses (36.3% vs 21.3%, P<0.001). All-cause mortality at 5-years was 34.4% in patients with cancer versus 8.9% in patients without cancer (P<0.001). In cancer patients, high-sensitivity cardiac troponin (hs-cTn) T, but not hs-cTnI, concentrations had lower diagnostic accuracy for NSTEMI versus those without cancer (hs-cTnT area under the receiver-operating characteristics curve 0.89 [95%CI 0.86-0.92] vs 0.94 [95%CI 0.93-0.94], p<0.001). In cancer patients, the ESC 0/1h-hs-cTnT/I-algorithms maintained a very high safety but had lower efficacy with significantly more patients remaining in the observe zone (with hs-cTnT: 39.0% vs 20.0%, P<0.001). Similar findings were obtained in the external validation cohort. Conclusions Patients with current or past cancer have a substantially higher prevalence of AMI as the cause of acute chest pain. Time to discharge from the ED and the rate of hospitalization are increased, mainly due to non-AMI causes. The diagnostic performance of hs-cTnT/I and the efficacy of the ESC 0/1-h-hs-cTnT/I-algorithms is reduced.