Selection of patients for prophylactic treatment with implantable cardioverter-defibrillator

Discovery of the clinical utility of implantable cardioverter-defibrillator (ICD) to convert ventricular fibrillation and fast ventricular tachycardia to sinus rhythm has improved the prognosis of selected high-risk patient groups. Initially, ICD therapy was used in patients who had experienced a life-threatening arrhythmia event. After completion of seminal prophylactic ICD trials, this therapy has been expanded to patients considered to be at high risk of sudden unexpected cardiac death (SCD) but who have not experienced a life-threatening arrhythmia event (1,2). However, identification of the patients assumed to be at high risk for the future occurrence of ventricular tachyarrhythmia that could be treated by ICD is a challenge and has been under continuing debate among the clinicians and scientists (3,4).

Despite the multiplicity of factors known to be related to an increased risk of SCD in observational studies, the only clinical practice currently used to predict such an event is the measurement of left ventricular ejection fraction (LVEF). Use of this measurement for stratifying risk is based on the results of randomized ICD trials, which have shown a mortality benefit of this therapy for patients with depressed LVEF (< 35%) (1,2). However, measurement of LVEF falls far short of goals in optimal risk stratification of SCD, such as identification of those patients who will experience SCD due to a reversible arrhythmia within a specified time period (i.e. 3–5 years) and exclusion of those who will not experience SCD. The majority of patients who will experience SCD do not have a low LVEF, and many patients with a low LVEF may be at low risk of SCD (5). Therefore, clinical scientists have sought other methods to predict and prevent these events to find more suitable candidates for prophylactic ICD therapy.

Most promising tools in the prediction of SCD in observational follow-up studies have been the measurements of heart rate variability/turbulence, T-wave alternans, and certain markers from the standard 12-lead ECG (5). Despite the promising results, none of these measurements is widely used in clinical practice for selection of patients to ICD therapy, mainly due to the lack of the randomized ICD trials using these measurements as pre-defined inclusion criteria.

In this issue of the journal, Amlie and co-workers describe the results of a prospective, multicentre, observational study assessing the utility of a novel semi-invasive method, Wedensky Modulation (WM), in predicting the appropriate device therapy of the recipients of the ICD (6). This method uses a short low-amplitude electrical impulse, which is synchronized to the QRS complex between the precordial and dorsal thoracic patches. Changes in the following QRS-T are registered and analyzed with the wavelet decomposition leading to a 3-dimensional surface pattern called the wavelet envelope. The authors conclude that potentially life-threatening ventricular arrhythmia events could be predicted by the WM test and that in combination with other tests the WM test might be clinically useful.

Although the results are of potential interest, several caveats should be recognized in the generalizability of the results of the study by Amlie et al. The electrophysiological basis for the presence of abnormal WM test has not been clearly defined. Changes in the morphology of the QRS complex after the subthreshold stimulation could well be explained by subtle abnormalities in ventricular conduction that are not clearly visible in standard 12-lead ECG. Unfortunately, the duration of QRS complex was not measured in all patients from the 12-lead ECG. Some other ECG characteristic, such as fractionated QRS complex that can easily be measured from the standard ECG, can also reveal subtle ventricular conduction abnormalities (7). Comparison of the results of the WM test with these measures would have been helpful in attempting to show that WM test provides more useful information for risk stratification than the standard 12-lead ECG. Similarly, changes in T-wave morphology after subthreshold stimulation could potentially disclose heterogeneity of ventricular repolarization. Again, a comparison with other repolarization abnormalities obtained from the standard ECG, such as QT interval dispersion, and duration of the end of the T-wave or abnormalities in T-wave morphology form vectorcardiograms would have been helpful (8). Thus, the results of this study do not provide convincing evidence on the superiority of this method as compared to standard 12-lead ECG in the prediction of arrhythmia events.

The ideal test for selecting the patients for ICD therapy would be a method or a combination of methods that could identify the patients with a LVEF < 35% who are at a very low risk for ventricular tachyarrhythmia events, or those with a LVEF > 35% who are at a high risk. Recent data show that inappropriate shocks are relatively common in ICD patients and that the shocks are associated with an increased mortality (9). Thus, unnecessary implantation of prophylactic ICD to low-risk patients should be avoided. In this study using the WM method, 22% of the patients with a negative WMI-R or WMI-T test result experienced a ventricular tachyarrhythmia event. The negative predictive accuracy of the WMI test remained at a low level, suggesting that this test is not useful for this purpose. This is opposite to the results obtained using the T-wave alternans test as a risk marker, which has yielded high negative predictive accuracy results (5).

Despite the above-mentioned caveats, any efforts to improve the risk stratification of the patients for SCD are welcomed. As the authors state, the combination of the WM test could give some important clinical information. To achieve this goal, a large amount of further studies will be needed, such as comparison of the predictive accuracy of WM test with other non-invasive tests, and the additional benefit obtained by combination of the WM test results with other test results. Finally, a randomized ICD trial is needed to convince the medical community on the clinical utility of any non-invasive test or their combination beyond the measurement of LVEF in selecting the patients for prophylactic ICD therapy.

Declaration of interest: The author reports no declarations of interest. The author alone is responsible for the content and writing of the paper.

This work was supported in part by grants from Sigrid Juselius Foundation, Helsinki, Finland, and the Foundation for Cardiovascular Research, Helsinki Finland.