Author + information
- Oussama Wazni, MD∗ ( and )
- Bryan Baranowski, MD
- ↵∗Reprint requests and correspondence:
Dr. Oussama Wazni, Out Patient Department, Section of Cardiac Electrophysilogy, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, Ohio 44195.
Major advances in the treatment of cardiovascular disease and better implementation of aggressive risk factor reduction strategies have resulted in significant declines in overall cardiovascular mortality over the past 3 decades (1). Despite these advances, the rate of sudden cardiac death (SCD) has remained largely unchanged over the same period (2). SCD continues to account for >50% of all cardiovascular deaths and up to 20% of all deaths (3). Furthermore, survival following an out-of-hospital cardiac arrest remains dismal despite improvements in resuscitative care (4). As such, substantial reductions in the incidence of SCD will require significant improvements in primary prevention strategies.
The current approach to prevention of SCD involves the placement of an implantable cardiac-defibrillator (ICD) in high-risk individuals. For such an approach to be effective, it needs to be applied to populations at elevated risk for SCD. However, identifying those at elevated risk has proven to be problematic (3). The best-known predictors of SCD are the degree of left ventricular systolic dysfunction and the severity of heart failure symptoms (5). In such patients, the use of ICD therapy has demonstrated clear survival benefits (6,7). Unfortunately, the use of these criteria to reduce SCD incidence in the general population has significant limitations. In fact, most SCD victims in the community do not have a pre-existing history of depressed ejection fraction or a clinical history of heart failure (8–10). As such, our current strategy fails to identify and affect the vast majority of individuals who suffer a SCD event.
Significant efforts have been made to improve risk prediction models for SCD in the general population by incorporating additional clinical characteristics (11–13). Coronary artery disease (CAD) is the most common substrate underlying SCD, noted in up to 75% of cases (3). As such, it is not surprising that CAD and risk factors for CAD (hypertension, hypercholesterolemia, diabetes, kidney disease, obesity, and smoking) are also predictive of SCD (3). Unfortunately, these risk factors are also strongly associated with competing causes of cardiac death. As such, their ability to improve patient selection for device-based therapy is limited. ICD therapy specifically targets SCD events. The utilization of screening markers that are also associated with competing modes of death significantly limits the effectiveness of defibrillation therapy, which only targets SCD.
To improve the performance of prediction models and to identify patients who would receive the maximum benefit from ICD therapy, we need to identify markers that specifically predict SCD. The identification of markers specifically associated with SCD may also further improve our understanding of the complex and poorly understood pathophysiology of this disease process.
In this issue of the Journal, Hussein et al. (14) report their findings of a significant, graded association between cardiac troponin T levels, measured by a highly set sensitive assay (hsTnT), and the risk of SCD in a large community-based population. Most of the participants in the study were older with an average age of 72.8 years at the time of enrollment (range 65 to 100). They were followed for a median of 13.1 years. The association between hsTnT levels and the risk of SCD was first tested using hsTnT as a continuous variable. Subsequently, 3 groups (low-, intermediate-, and high-risk) were identified using the Cox proportional hazard ratios of SCD for each decile of detectable hsTnT compared with those for participants with undetectable levels. Overall, higher levels of hsTnT were associated with an increased risk for SCD (unadjusted hazard ratio [HR] for +1 log hsTnT: 2.04; 95% confidence interval [CI]: 1.78 to 2.34). The incidence of SCD was 3.4%, 6.6%, and 8.7% for the low-, intermediate-, and high-risk groups (fully adjusted HR: 1 vs. 1.55 vs. 1.89, respectively; p = 0.005).
The association between hsTnT and SCD was attenuated after adjusting for other clinical risk factors as well as low ejection fraction, heart failure, and myocardial infarction (HR: 1.26; 95% CI: 1.05 to 1.62). This finding suggests that hsTnT may serve as a marker for these other conditions and thus co-segregate them. However, the fact that the association persisted after making these adjustments implies that hsTnT may also either directly influence the risk of SCD or serve as a marker for SCD that is more strongly related to the underlying pathophysiology of SCD than this group of covariates is. Establishing with better clarity, the cause and effect relationship between hsTnT and the pathophysiology of SCD will require further investigation. The data presented by Hussein et al. (14) is an important first step in establishing this connection.
The hypothesis that hsTnT would carry a specific association with an elevated risk of SCD appears very plausible. As the investigators point out, subclinical cardiac myocyte injury, as reflected by low levels of circulating hsTnT, may provide the anatomical substrate for scar-related electrical re-entry, which would support and propagate lethal arrhythmias. Low levels of circulating hsTnT may also be an indicator of ongoing subclinical myocardial inflammation that may result in cell membrane instability. This instability could theoretically result in increased automaticity, triggered activity or result in a dispersion of phase 3 refractoriness that could pre-dispose the patient to the sudden development of malignant arrhythmias. The intriguing prospect of using biomarkers is their presumptive ability to detect processes that may pre-dispose the patient to a risk for SCD prior to the onset of clinically overt cardiovascular disease. Myocardial inflammation, cardiac remodeling, and scar formation could theoretically be detected and addressed at very early stages of the disease process when the opportunity to intervene would have the greatest impact on survival.
Unfortunately, despite the association, the data presented do not provide any information to determine whether hsTnT would help improve SCD prediction models and the investigators acknowledge that this was not the aim of the study. To establish its usefulness in predicting SCD, a statistically significant improvement in discrimination (C-index) after adding hsTnT to current prediction models would need to be demonstrated. Attempts to do so with other clinical markers associated with SCD have been limited to mostly nonsignificant improvements in the risk prediction model (11). This is likely due in part to the fact that these markers are not specific to SCD and are associated with an increased risk for cardiovascular death as well as all-cause mortality. It appears as if hsTnT may be vulnerable to this limitation. As presented in Hussein et al.'s paper, hsTnT was also associated with the risk of all-cause mortality and nonsudden cardiovascular mortality. Despite the fact that the association between hsTnT and SCD appeared to be stronger with a higher HR, a definitive conclusion regarding the strength of the association could not be made as the CI overlapped.
Even if hsTnT is unlikely to improve risk prediction models for SCD, this biomarker may be useful as a screening tool for the general population, where access to traditional screening with echocardiography is often difficult and cost-prohibitive. It is possible that TnT could play a role in identifying patients who require further evaluation and risk stratification. The likely problem with this strategy, however, is that the highly sensitive assay used to assess for TnT may be too sensitive. In the report presented by Hussein et al. (14), 67% of patients (2,989 of 4,431) had baseline detectable levels of hsTnT. Furthermore, only 46% of the population had a level <5.0 pg/ml and fell into a low-risk category. This suggests that if applied as a solitary screening tool to a large, older, general population, follow-up assessment would be necessary in over 50% of those screened. It is possible that these findings are skewed by the fact that the marker was assessed in an older population that carried a more significant burden of cardiovascular risk factors. Its utility as a screening tool in a younger healthier patient population may be more plausible but requires further investigation. Due to the high prevalence of elevated hsTnT, this biomarker may be more useful for its negative predictive value. Of the 2,039 subjects with hsTnT ≤5.0 pg/ml, only 69 patients had a SCD event. Using this cutoff, the negative predictive value of a low level of hsTnT is 96.6%.
In summary, Hussein et al. (14) present novel work that demonstrates an association between hsTnT and SCD. The clinical applications of such a finding have yet to be established and will require further investigation. Although hsTnT was also associated with the risk of all-cause mortality and nonsudden cardiovascular mortality, it appears to potentially have a stronger association with SCD. This finding is significant as it suggests that hsTnT levels may be directly related to the underlying pathophysiology of SCD. A better understanding of the pathophysiology of SCD is critical as we continue to look for novel therapeutic prevention approaches.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Zheng Z.J.,
- Croft J.B.,
- Giles W.H.,
- Mensah G.A.
- Deo R.,
- Albert C.M.
- Epstein A.E.,
- Dimarco J.P.,
- Ellenbogen K.A.,
- et al.
- Stecker E.C.,
- Vickers C.,
- Waltz J.,
- et al.
- Fishman G.I.,
- Chugh S.S.,
- Dimarco J.P.,
- et al.
- Buxton A.E.,
- Lee K.L.,
- Hafley G.E.,
- et al.,
- for the MUSTT Investigators
- Hussein A.A.,
- Gottdiener J.S.,
- Bartz T.M.,
- et al.