Bethesda, MD - The microvolt T-wave alternans (TWA) test continues to live up to billing as an independent risk discriminator in patients with ischemic cardiomyopathy and a low LVEF in a study appearing in the May 2, 2006 issue of Journal of the American College of Cardiology [1]. Compared with an absence of TWA, a positive test or one with "indeterminate" results pointed to more than twice the risk of all-cause and arrhythmic death over an average of 18 months.

"Importantly, we also demonstrated that the increased risk associated with a nonnegative microvolt TWA result is mediated primarily through higher rates of arrhythmic mortality," write the authors, Dr Theodore Chow (Ohio Heart and Vascular Center, Cincinnati) and colleagues.

The study, the group writes, is the largest yet to explore the TWA test's prowess as a risk stratifier in patients with ischemic cardiomyopathy and LV dysfunction, the largest population included in the January 2005 introduction of Medicare coverage for primary-prevention implantable cardioverter defibrillators (ICDs). Reimbursement for the TWA test itself, which screens for subtle beat-to-beat variations in the shape of the electrocardiographic T wave during exercise stress, began in March 2006. Its primary use is to identify patients who qualify for prophylactic-ICD reimbursement based on LVEF but who might safely avoid receiving the device.

Of the group's 768 prospectively enrolled patients with ischemic cardiomyopathy and an LVEF <35%, who were required to be in sinus rhythm, 67% tested "nonnegative" for microvolt TWA, the traditional way of combining positive results with indeterminate ones. In addition to showing significant mortality increases in that two thirds of the population, the study also suggested that:

• The increased all-cause mortality was "equally robust" among patients with an LVEF <30%, a subgroup that seemed to account for most of the elevated risk associated with a nonnegative TWA test.
• The test significantly predicted mortality in the overall population and in the subgroup that didn't receive an ICD, but it wasn't predictive among those who were implanted with a device.
• Among patients with ICDs, however, a nonnegative TWA finding was a significant predictor of either all-cause mortality or device-delivered shock.
• The test's power as a risk stratifier was independent of other predictors, such as LVEF, QRS-interval length, and certain demographic features, cardiovascular comorbidities, and medical therapies.

All-cause mortality was higher among patients who were classified as TWA positive compared with those who tested negative (HR 2.08, 95% CI 1.18-3.66; p=0.01). But there was no such significant finding for arrhythmic mortality.

In another subset analysis, TWA-indeterminate patients showed higher mortality risks compared with those who were TWA negative for both all-cause death (HR 2.78, 95% CI 1.55-4.99; p=0.0006) and arrhythmic death (HR 3.62, 95% CI 1.44-9.13; p=0.006).

The novel observation that all-cause and cause-specific mortalities among TWA-indeterminate and TWA-positive patients are statistically similar shows—apparently for the first time, according to the authors—that the traditional study practice of combining them into one nonnegative group is "statistically appropriate."