Nonspecific elevations in troponins
Alcalai R, Planer D, Culhaoqlu A, et al. Acute coronary syndrome vs nonspecific troponin elevation: clinical predictors and survival analysis. Arch Intern Med. 2007 Feb 12;167(3):276-281.
In 2000, the American College of Cardiology (ACC) and the European Society of Cardiology (ESC) jointly produced a recommendation for a new definition of myocardial infarction. This proposal based the diagnosis primarily on the elevation of biomarkers specific to cardiac tissue, troponin T and troponin I. Since that time, as use of these blood tests has escalated, it is apparent that elevations in these biomarkers do not always translate into thrombotic coronary artery occlusion. Instead, we have seen that they are positive in a variety of clinical settings. These include sepsis, renal failure, pulmonary embolism, and atrial fibrillation. This investigation attempts to characterize the differences among patients presenting with acute coronary syndrome (ACS) and nonthrombotic troponin elevation (NTTE), to report on outcomes for each, and to note the positive predictive values (PPV) for elevated troponins across clinical settings.
Two hospitals in Israel collected data on all adult patients who experienced an elevation in troponin T (defined as at least 0.1 ng/mL) at any time during their hospital stay. Six hundred and fifteen patients were evaluated by age, sex, cardiovascular risk factors, history of ischemic heart disease, left ventricular function (LVF) by echocardiogram, serum creatine phosphokinase (CPK), and creatinine levels, as well as by which hospital service each had been admitted under. The highest troponin T value was used in the analysis, along with the creatinine level taken on the same day. Two physicians, one a specialist in internal medicine and the other a specialist in cardiology, independently determined the principal diagnosis in accordance with the ACC/ESC guidelines for thrombotic ACS and used other diagnostic studies for alternative diagnosis for conditions known to cause NTTE.
Patients were followed up for causes of mortality for up to two-and-a-half years. Kappa (k) was calculated for physician agreement regarding the principal diagnosis. To assess independent odds ratios and their 95% confidence intervals (CIs) of predictor variables for ACS, an unconditional multiple logistic regression analysis was used. The PPV for troponin T in the diagnosis of ACS was calculated. In-house mortality rates were measured. Long-term risk of death was assessed using Cox proportional hazard models.
The diagnosis of ACS was made in only 53% (326) of the patients. Forty-one percent (254) had NTTE, and the diagnosis was not determined in 6% (35). The diagnoses comprising NTTE included—in order from most to least common—cardiac non-ischemic conditions, sepsis, pulmonary diseases, and neurologic diseases. Using the multivariate analysis, the diagnostic predictors for ACS were history of hypertension or ischemic heart disease, age between 40 and 70 years, higher troponin levels (greater than 1.0 ng/mL), and normal renal function. Extreme age and admission to a surgical team were negative predictors for ACS. Gender, presence of diabetes, and LVF did not appear to make a difference.
The PPV of an elevated troponin T for ACS among all patients was only 56% (95% CI, 52%-60%). It became lower (27%) in those older than 70 with abnormal renal function and higher (90%) in those with a troponin T greater than 1.0 ng/mL and normal renal function. In-house mortality for all patients was 8%; for those with ACS, it was 3%, while for those with NTTE, it was—at 21%—almost eight times higher than the ACS group (P<0.001). Patients were followed up for mortality for a median of 22 months. The long-term mortality was also significantly better (P<0.001) for those with a diagnosis of ACS than for those with NTTE.