Elevated troponin I levels on admission predict long-term mortality in patients with acute cerebral infarction following thrombolysis

Cardiac diseases are frequently accompanied by elevated cardiac troponin (cTn) concentration. Elevated concentrations of cTn, including cTnI and cTnT, have been observed in many non-cardiac diseases, such as acute cerebral infarction (ACI) [1], subarachnoid hemorrhage [2], chronic kidney disease [3], severe infection [4], pulmonary embolism [5], cancer [6], and other diseases [7]. Elevated cTn has a prognostic value for short- and long-term mortalities in acute ischemic diseases or subarachnoid hemorrhage cases [2, 8, 9]. Considering the close relationship between acute cardiac disease and ischemic stroke, the current American Heart Association/American Stroke Association guidelines recommend the assessment of blood cTn concentration in patients diagnosed with ACI [10]. Elevated cTnI concentrations are observed in patients with ACI [1], indicating ascending risks of short-term mortality [1, 9]. When referring to patients with ACI treated with thrombolytic therapy, there are relatively few studies concerning the association between cTnI concentration and long-term mortality [11].

The exact mechanism of cTnI elevation in ACI remains controversial [1, 8, 9]. Some authors have proposed that it occurs owing to concomitant cardiac disease [1, 8, 9] or that cTnI in ACI is a biomarker of unstable atherosclerotic plaques in the cardiac coronary blood circulation [12]. Other studies have suggested that cTnI elevation following cardiac injury may be caused by an abnormal sympathetic nervous system condition, in which cortical areas controlling autonomic functions are destroyed, which is followed by ACI (e.g., insular cortex) [13].

Patients and relatives can determine the potential outcomes when prognostic information for ACI is provided. At least 30% of all patients with ACI who were treated with recombinant tissue-type plasminogen activator (rt-PA) thrombolytic therapy by intravenous infusion were more likely to have limited or no disability based on assessment using the National Institutes of Health Stroke Scale (NIHSS) at 3 months [14]. The mortality rate at 3 months was 17% in the rt-PA group [14]. In patients without intravenous thrombolysis, the 5-year mortality rate after the first ischemic stroke was approximately 41% [15]. However, few studies have reported on the long-term mortality (> 5 years) owing to troponin elevation in patients with ACI following thrombolytic therapy. Therefore, the main purpose of this study was to describe the patient characteristics, long-term mortality, and frequency and determinants of elevated plasma cTnI levels in patients with ACI following thrombolysis.

In 2000, James first reported the relationship between mortality and cTnI elevation in patients with ACI [18]. Herein, we observed cTnI elevation in patients with ACI treated with rt-PA therapy. Despite adjusting for four possible confounders, cTnI was associated with a 1.8-fold increased risk of 5-year mortality. The cTnI level was also associated with the 3-month mortality rate. The ascending risk was maintained during the entire 5-year follow-up period. The ascending risk of mortality in patients with ACI treated with intravenous thrombolysis at admission is consistent with a previous study’s results on 3-month mortality [11]. The proportions of our patients with cTnI elevation and ACI treated with rt-PA therapy (33%) were similar to those reported previously (36%) [11]. We extended the observation time to approximately 5.7 years, whereas other studies assessed short-term or in-hospital outcomes [1, 9].

Elevation of cTnI level in the context of ACI in patients treated with rt-PA therapy was associated with older age and higher occurrence of CHF, CAD, AF, and RI. These findings are consistent with the following previously-reported data [1, 9]: (a) cTnI elevation has been reported in both acute and chronic patients with CHF, which may be attributed to the vulnerability of myocardial cells, chronic subclinical ischemia, or left ventricular insufficiency [8]; (b) RI was reported to be the cause of cTnI elevation [3]; (c) overall, 12.5% of the patients who first developed AF showed elevated cTnI concentration [19], which was explained by the fact that the heart experienced ischemia, with consequent cardiomyocyte necrosis. Both CHF and AF are strong predictors of mortality. Another study reported mild cTnI elevation in 830 stable patients with vascular disease, which included 491 and 341 patients with CAD and stroke, respectively [8]. We found that cTnI concentration > 0.03 μg/L independently predicted the all-cause 5-year mortality (HR, 1.76). Cardiovascular disease is considered the leading cause of long-term mortality in patients with stroke [20]. However, in our study, CAD was not significantly associated with mortality in the univariate or multivariate Cox regression models.

We observed a positive association between elevated blood cTnI concentrations and stroke severity (scored using the NIHSS) (p < 0.05), consistent with the result of a previous study [1]. Furthermore, the 3-month stroke severity (assessed using the mRS score) also significantly influenced mortality (odds ratio, 1.26; p < 0.001), consistent with the result of a previous study [11]. Jan et al. reported that elevated cTn levels were significantly associated with right insular cortex involvement [13]. A previous work reported that impairment of the insular lobe (right or left) increases with elevated cTn concentration [9]. However, Barber et al. did not identify a significant association between cTnI elevation and insular cortex stroke [21]. Similarly, we cannot make a conclusion concerning the association between elevated cTnI levels and the corresponding part of the affected brain artery territory (right or left insular cortex). Jan et al. reported that elevated cTn levels are more often observed in patients with acute cardioembolic stroke [13]. However, we did not find that the etiological classification of ACI differed between the normal and elevated cTnI levels.

Faiz et al. observed no significant correlation between rt-PA therapy and long-term mortality (558 ± 254 days) [22]. However, rt-PA therapy can improve the 3-month prognosis of ACI patients [14, 23]. The 3-month mortality rate in the present study was approximately 13%, slightly higher than that reported in a previous work (5%) [11], but similar to that reported in another study (17%) [14]. This contradiction is possibly attributed to our small sample size. The 5-year mortality rate in our patients with ACI was approximately 31%, which was significantly lower than that observed in patients with acute stroke with similar follow-up periods (40–60% fatality rate within 5 years after ACI) [24, 25]. This may be attributed to thrombolysis therapy and different patient characteristics. In our study, patients who died had a higher likelihood of undergoing ET (8/44, 18% vs. 12/101, 11%). However, we could not determine whether ET resulted in more deaths. ET was associated with considerable functional improvement in several randomized trials [25]. The higher mortality rate may be attributed to the obstruction of large intracranial vessels in patients with ACI, which would indicate that massive cerebral infarction is the main cause of death and can lead to increased mortality [27]. Cao et al. reported a 1-month mortality rate of approximately 36.2% in patients with troponin T high-sensitivity elevation who underwent ET [28], which was slightly higher than that observed in our study. However, ET did not increase the mortality rate, suggesting that the severity of disease based on the NIHSS score is more relevant than that in patients with ACI [20].

In this study, we present data that show a significantly higher WBC count in patients with cTnI elevation on admission, a finding consistent with that of a previous study [24]. These observations may result from stress that the patients feel when ACI occurs, thus resulting in increased WBC counts, simultaneous hyperexcitation of the sympathetic system, and release of several catecholamines, which may cause excessive contraction of cardiac muscle cells and, ultimately, increased cTn levels, particularly in patients with cardiac disease. This has also been demonstrated in stress-induced ischemia tests [29].

Many studies have reported the relationship between elevated cTnI concentration and the ischemic territory of cerebral infarction [30, 31], especially in the right part of the insular lobe. The insular lobe is part of the center of the autonomic nervous system, which is associated with the cerebral control of the sympathetic and parasympathetic outflow to the heart. An acute ischemic lesion in the insula results in the imbalanced control of the heart, which ultimately increases cTn concentrations through myocardial cell necrosis [32]. However, our study did not find a significant association between these factors.

The exact mechanism between elevated TnI and ACI is not clear. We found that the TnI level was elevated in ultra-early stage of disease. With retesting, estimation of TnI in many patients may increase or decrease, but none of them had sustained increase. One explanation of the condition is the cardiac stress reaction in patients with excessive release of catecholamine, resulting in the elevation of TnI [21]. Another explanation is that patients with elevated TnI level were accompanied with CAD or the presence of culprit plaques in the coronary arteries. Nevertheless, the troponin elevation in an acute ischemic stroke study reported that patients with elevated Tn in ACI underwent coronary angiography, suggesting that the coronary disease-causing plaque and stenosis in such patients were less frequent than those in patients with acute coronary syndrome without ST elevation [33]. Future studies are needed to record TnI concentration dynamically and cardiac examination (catecholamine, electrocardiograph, and coronary angiography).

This study has some limitations. First, all patients received intravenous rt-PA thrombolytic therapy. This is a unique population of patients with ACI. Second, a large number of patients were excluded because of missing cTnI values on admission or the loss of mRS data during follow-up, resulting in a small sample size. Third, we did not confirm heart disease by coronary CTA or coronary angiography, although we conducted electrocardiography and cardiac ultrasonography. In particular, we confirmed heart disease via clinical examinations. Nevertheless, the cardiologists did not establish definite diagnoses of acute myocardial infarction. Fourth, we did not measure the cTnI concentration continuously, although we determined the index of patients with cTnI concentrations > 0.1 μg/L until the levels were normalized. Unfortunately, we did not test all patients continuously. Fifth, our results were insufficient to demonstrate differences in the ACI subtypes. Scheitz et al. reported an evident contrast in the ratio of cardioembolic stroke between a cTnI elevated and a normal group [9]. Sixth, although new test systems, such as hypersensitive assays of troponin, are frequently used for detecting cardiovascular disease in patients with ACI, we used the conventional method to test cTnI. Future studies may use these newly developed troponin immunoassays to identify more cases of cardiovascular diseases. Moreover, we only detected the single biomarker, cTnI, other indexes, such as N-terminal prohormone of brain natriuretic peptide, copeptin, matrix metalloproteinase-9, and S100 β, are all independently associated with various clinical outcomes [34].

Thrombolysis-treated patients with ACI who have elevated cTnI levels more often had serious strokes compared to those who had no cTnI elevation. These patients also had more concomitant conditions compared to those with normal cTnI levels. Furthermore, we confirmed the ascending risk of 5-year mortality in these patients with elevated cTnI levels on admission. Our data indicated the possibility that cTnI evaluation in these patients may be predictive of long-term mortality. More studies are needed to further clarify the mechanisms underlying elevated troponin levels in patients with ACI.