Overuse of Cardiac Troponin Among Hospitalized Patients: A Cohort Study of Biomarker “Superusers”

Cardiac troponin assays are an essential test for evaluating patients with suspected acute coronary syndrome and provide necessary information to make a diagnosis of acute myocardial infarction (AMI) [1]. Serial measurements allow clinicians to determine the pattern of troponin, which can help distinguish between AMI and troponin elevation secondary to other conditions [2]. Concerns about the overuse of troponin assays have been previously raised in the literature [3]. In a prior investigation at our institution evaluating troponin use, we observed an exceptionally high number of assays had been performed on some patients during a single hospital stay [3]. That observation led us to conduct a descriptive study of the patients who had the highest number of troponin assays ordered for them, patients that we refer to as "superusers". To better understand the patterns of care associated with these encounters, we reviewed the medical records of a sample of patients on whom a large number of troponin assays had been performed during a single hospitalization.

This was a single-center study conducted at an academic tertiary care hospital. Data were retrieved for all patients hospitalized between 2012 and 2015 on whom at least one troponin assay had been performed. All tests assayed for troponin T (Roche Diagnostics, Indianapolis, IN, USA), with a reference range of < 0.03 ng/mL. All patients who had ≥ 20 troponin T assays ordered within a single encounter were identified as “superusers” for this investigation. The cutpoint of 20 was chosen based on the frequency distribution of assays per patient in order to balance comprehensiveness with feasibility of detailed chart reviews. Data were extracted from a review of medical records, including clinical history, demographics, comorbidities, and outcomes.

This investigation was approved by the University of Florida Institutional Review Board-01 (protocol number 201501010) and conformed with the Helsinki Declaration of 1964, as revised in 2013. The need to obtain informed consent was waived.

Of the 43,063 inpatient encounters featuring at least one troponin assay, 0.12% (n = 51) were identified as superusers. The mean (standard deviation) age of the patients was 59.9 (20.0) years, and the mean length of stay was 30.8 (18.8) days. Of these patients defined as superusers, all had at least one elevated troponin value, and 34 were men (67%). The baseline characteristics of superusers are described in Table 1.

Of these 51 encounters, almost all patients (n = 47) had cardiology consultations; in approximately one-half of these patients (n = 25) the result of the cardiology consult was a recommendation for trending troponins (Table 2). In none of the superusers did repeated measures of troponin significantly alter the management plans. Only 29 patients were diagnosed with a myocardial infarction (MI); the remainder were diagnosed with acute myocardial injury. In 12 of these encounters, these patients were documented as not being candidates for invasive angiography due to either comorbidities or personal preferences.

While following the trend of troponins is a common response to suspected cardiac events, clinicians should only do so if they anticipate the results might alter how they manage the patient. In our sample of troponin “superusers,” none of the patients experienced a change in their care due to the results of the assays. While formal criteria for the appropriate use of troponin assays have not been pursued by any professional cardiology societies, the lack of impact on care plans of the results from these assays suggests that for the population we studied, ordering a large number of troponin assays could be considered low value care. Furthermore, a substantial portion of the “superusers” were never diagnosed with MI, suggesting that any elevation in their troponin level was chronic. For patients who are not candidates for invasive procedures (whether due to significant comorbidities or the patient’s own documented preferences), trending troponins that are already known to be elevated will likely not change the course of care.

Troponin elevation has been shown in numerous settings to be a poor prognostic indicator, but that risk can be ascertained without repeated measurements of troponin. Facilities may need to reconsider how troponin assays can be ordered when newer high-sensitivity troponin assays become available. These new assays detect troponin in the majority of a healthy population with higher sensitivity for detecting MI, possibly at the cost of lower specificity [4]. Accordingly, concerns exist regarding whether this will lead to an increase in cardiovascular resource utilization.