Prognostic implications of troponin I elevation in emergency department patients with tachyarrhythmia

Background Tachyarrhythmias are very common in emergency medicine, and little is known about the long‐term prognostic implications of troponin I levels in these patients. Hypothesis This study aimed to investigate the correlation of cardiac troponin I (cTnI) levels and long‐term prognosis in patients admitted to the emergency department (ED) with a primary diagnosis of tachyarrhythmia. Methods A retrospective cohort study was conducted between January 2012 and December 2013, enrolling patients admitted to the ED with a primary diagnosis of tachyarrhythmia and having documented cTnI measurements. Clinical characteristics and 5‐year all‐cause mortality were analyzed. Results Of a total of 222 subjects with a primary diagnosis of tachyarrhythmia, 73 patients had elevated levels of cTnI (32.9%). Patients with elevated cTnI levels were older and presented significantly more cardiovascular risk factors. At the 5‐year follow‐up, mortality was higher among patients with elevated cTnI levels (log‐rank test P < 0.001). In the multivariable Cox regression analysis, elevated cTnI was an independent predictor of all‐cause death (hazard ratio, 1.95, 95% confidence interval: 1.08‐3.50, P = 0.026), in addition to age and prior heart failure. Conclusion Patients admitted to the ED with a primary diagnosis of tachyarrhythmia and high cTnI levels have higher long‐term mortality rates than patients with low cTnI levels. cTnI is thus a biomarker with predictive capacity for mortality in late follow‐up, conferring utility in the risk stratification of this population.


| INTRODUCTION
Cardiac troponin I (cTnI) is a marker of myocardial injury used in establishing a diagnosis of myocardial infarction (MI). This marker is used in addition to ischemic signs and symptoms, with chest pain being one of the most common symptom. 1 However, certain populations, including elderly, diabetic and female patients, may have atypical presentation, leading to worse prognosis. In these populations, the suspicion of acute coronary syndrome (ACS) leads to troponin testing. [2][3][4][5] Consequently, it is common for patients admitted to the emergency department (ED) to undergo cTnI determination, despite not presenting cardiac-related symptoms, such as chest pain.
Likewise, cardiac arrhythmias are very common in the ED and are a reason for cTnI determination as they may occur in the case of MI and may be associated with chest pain, syncope or palpitations. Therefore, it is common to request cTnI to rule out ACS in patients with a primary diagnosis of tachyarrhythmia in the ED. Elevated cTnI levels have been described in patients with cardiac arrhythmias not related to ACS, [6][7][8][9] but the long-term prognostic value of this elevation is not yet known.

| Troponin I
All analyses of cTnI levels were performed using the same immunoassay techniques (cTnI-Ultra Siemens, Advia Centaur) in the same laboratory. The reference limit for the positivity of the cTnI test was >39 ng/L, corresponding to the 99th percentile of a reference control group, with analytical imprecision expressed by a coefficient of variation below 10%. The analytic performance of this assay has been previously validated. 10 cTnI levels were determined at the time of admission and 6 hours later. We recorded the number of cTnI determinations for each patient, the maximum value and the dynamic rise/fall patterns (changes >20% in cTnI levels). Patients with evidence of a dynamic pattern were classified as acute myocardial injury or as type 2 MI, following Saaby's criteria 11 ; those without evidence of a dynamic pattern were considered to have a chronic myocardial injury.

| Events
The primary outcome of the study was all-cause mortality at the 5-year follow-up. The secondary outcomes were readmission rates for HF and for MI and the combination of all-cause mortality and readmission rates for MI or HF at the 5-year follow-up. The follow-up events were obtained from patients' electronic clinical records and death registers.

| Statistical analysis
Continuous variables are reported as median and interquartile range (IQR). Categorical variables are presented as counts and percentages.
The baseline characteristics of patients were compared using the Kruskal-Wallis test for continuous variables and Pearson's χ 2 test for categorical variables. Proportions were compared using Fisher's exact test. Survival analysis was performed with the Kaplan-Meier method and compared using the log-rank test. The associations of quantitative and qualitative variables with survival were analyzed using the univariate and multivariate Cox proportional risk models. In the adjusted model for the multivariate mortality analysis, all the variables that were significant in the univariate analysis were introduced. Backward stepwise selection was used with the input P-value <0.05 to determine the predictors of mortality. To assess the discrimination benefit of adding cTnI (in a continuous manner) to a clinical model, Harrell's C statistics were used; continuous net reclassification improvement (NRI) was used for reclassification prediction. The results are presented as a hazard ratio (HR) with 95% confidence intervals (CIs). Differences were considered statistically significant at P < 0.05. STATA 13.0 (College Station, Texas) was used for all analyses.

| Baseline characteristics
A total of 1021 patients were admitted to the ED with a primary diagnosis of tachyarrhythmia between January 2012 and December 2013; of these patients, 222 patients had their cTnI levels tested (21.7%) ( Figure 1). The median age was 69 years, and half were male. Of these patients, 73 (32.9%) had elevated cTnI levels. Table 1 shows the demographic, clinical and analytical variables according to the presence or absence of elevated cTnI levels. Patients with elevated cTnI levels were older, with a higher prevalence of hypertension, prior history of MI and chronic kidney disease. There were no differences in sex or in comorbidities, such as smoking, COPD, or cerebral or peripheral vascular disease, and the most frequent symptom was chest pain in both groups. At the time of admission, patients with elevated cTnI levels had higher HR (147 vs 133 bpm, P < 0.001), a tendency to lower SaO2, a lower eGFR (64.2 vs 84.1 mL/min/1.73 m 2 , P < 0.001), as well as lower values of hemoglobin (12.4 vs13.9 g/dL, P = 0.050) and higher glycemia (143 vs 122 mg/dL, P = 0.043). Of all the patients admitted, 28% of those with elevated cTnI levels needed hospital admission compared to 6% without elevated cTnI levels (P < 0.001).
Hospital mortality was similar in both groups (P = 0.55). There were differences in the need for hospital admission depending on the type

| DISCUSSION
This study shows that one-third of patients with a principal diagnosis of tachyarrhythmia admitted to the ED had elevated cTnI levels. This group of patients was identified as a high-risk population due to older age, worse cardiovascular risk profile, and higher long-term mortality.
Therefore, cTnI emerges as a biomarker with predictive capacity for mortality in long-term follow-up, conferring utility in the stratification of risk in this population. In the ED, cTnI testing is a common tool used for the diagnosis of MI. However, elevated cTnI levels are frequently encountered in various clinical situations, such as pulmonary thromboembolism, HF, and sepsis and is associated with poor prognosis. [12][13][14] In previous studies, it has been shown that a high proportion of patients admitted to the ED present cTnI elevation in clinical contexts other than acute atherothrombotic plaque disruption (type 1 MI) and that this elevation is associated with a worse prognosis in followups. 15,16 In addition, it has been observed that the prognosis was similar in patients who fulfill diagnostic criteria for type 2 MI and for non-ischemic myocardial injury, being worse compared to patients with type 1 MI. 17 Moreover, patients with cTnI elevation and without chest pain have been found to present higher mortality rates. 18 In addition, when comparing different laboratory troponin testing in patients in the ED without ACS, cTnT, and cTnI elevations have both important prognostic information regarding all-cause mortality. 19 In conclusion, these data indicate that cTnI elevation is a reflection of acute or chronic myocardial injury that is not always attributable to ACS and should be considered an entity in itself. 20  This finding is consistent with that of Schueler et al study. 25  HF patients are predisposed to develop arrhythmias. It has been described that highly complex, interactive, and dynamic changes in mechanical, structural neurohormonal properties might predispose the failing heart to tachyarrhythmias. In our study, there were no differences in the history of HF between the cTnI elevation groups or readmission for HF in the long term, but prior HF was an independent mortality risk factor in the long term. This result may reflect a relationship between tachyarrhythmias and structural changes describing an underlying structural heart disease, which might explain cTnI elevation.
Therefore, the mechanisms justifying elevated cTnI concentrations in tachyarrhythmia may be multiple and influenced by other factors that are not yet fully clear. It is possible that shortening of the diastolic period during tachyarrhythmia leads to the appearance of subendocardial ischemia, which explains the elevation of cardiac biomarkers. 27 Another potential pathophysiological mechanism was described by Goette et al 28   study, several markers of oxidative stress mediated by angiotensin II were increased in pigs subjected to rapid atrial pacing. In other experimental studies, it has been observed that myocardial overdistension represents another tachycardia-dependent mechanism that would justify the troponin elevation, since there is a direct association between the parallel increase in proBNP and troponins. 29 It is also possible that changes in the permeability of the cardiac myocyte membrane secondary to myocardial injury may be sufficient to generate the release of cardiac troponin to the cytoplasm, without structural damage being seen. 30 Thus, overall, it seems very likely that the cause of cTnI elevation in tachyarrhythmia is multifactorial and is more frequent in the elderly population and in patients with multiple cardiovascular risk factors and comorbidities.
Because in our study, the adverse prognosis of patients with ele-

| Limitations
This study has certain limitations. First, cTnI testing in the ED was left to the decision of the treating physician, who may not have measured cTnI in patients with a lower risk perception, such as younger adults with fewer cardiovascular risk factors. This decision could represent a selection bias. Despite this limitation, the study reflects the real-life clinical scenario in the ED. Second, as it was a retrospective study, the evaluation of coronary heart disease was not performed in all patients with elevated cTnI; hence, we cannot draw conclusions on this aspect.
However, in the short-and long-term follow-up, we did not observe readmission for MI or HF. This result strongly suggests that elevated cTnI levels do not correlate closely with the burden of coronary atherosclerotic disease. Third, the study did not aim to identify the mechanism by which elevated cTnI levels predict cardiovascular risk in the context of tachyarrhythmia.

| CONCLUSIONS
This study shows that cTnI elevation in patients with a primary diagnosis of tachyarrhythmia admitted to the ED is frequent and is associated with an increased risk of long-term mortality. These patients presented a worse cardiovascular risk profile and more comorbidities.
Our data prompt the need for more research to understand the mechanisms of elevated cTnI levels in patients with tachyarrhythmia and the association with cardiovascular risk.