Effects of Carvedilol Therapy on QT-Interval Dispersion in Congestive Heart Failure

Abstract

In cases of chronic congestive heart failure, QT-interval dispersion is a strong predictor of death. Carvedilol therapy appears to decrease QT-interval dispersion. We investigated whether carvedilol reduces QT-interval dispersion in congestive heart failure and whether this pharmaceutical agent has additional effects on elderly patients.

Seventy-seven ambulatory patients who had chronic congestive heart failure were evaluated for hypertension, diabetes mellitus, smoking, alcohol abuse, concomitant medications, and QT-interval dispersion. Carvedilol therapy was then initiated. Six months later, we re-evaluated the same variables, as well as morbidity and mortality rates, and number of hospitalizations. The patients were divided into 2 groups: Group I, aged <65 years (n=42); and Group II, aged ≥65 years (n=35). Statistics were analyzed with the Student's t-test, χ² test, and Cox regression model.

At 6 months, both groups showed significantly decreased QT-interval dispersion values compared with baseline values (76.9 ± 29.3 vs 104.3 ± 41.5 ms, respectively; P ≤0.0001). An elevated QT-interval dispersion value at baseline increased morbidity P = 0.041) but not hospitalization (P >0.05). Group II had a smaller reduction in QT-interval dispersion than did Group I (24.41 ± 29.36 and 30.98 ± 32.70 ms, respectively), but this difference was not significant.

We conclude that, in ambulatory patients with chronic congestive heart failure, long-term carvedilol therapy significantly decreases QT-interval dispersion, and this effect is uniformly distributed between patients aged <65 years and those aged ≥65 years. (Tex Heart Inst J 2003;30:176–9)

Congestive heart failure (CHF) is increasingly common and entails a high mortality rate: it causes more than 43,000 deaths annually in the United States. As many as half of these deaths are sudden. ¹ A large percentage of sudden deaths, especially in patients with mild-to-moderate symptoms, are due to malignant ventricular arrhythmias (ventricular tachycardia and ventricular fibrillation). Several clinical studies have suggested that, in the standard electrocardiogram (ECG), interlead variability of the QT interval, defined as QT-interval dispersion (QTD), reflects inhomogeneity in ventricular repolarization. ^2,3 An important condition underlying QTD is patchy myocardial fibrosis, resulting from myocardial ischemia, ventricular dilatation, and neurohormonal activation. ^4,5 Increased dispersion of the ventricular recovery time appears to increase the risk of ventricular arrhythmias and sudden death. ² This phenomenon has been observed not only in patients with chronic heart failure but also in those with mitral valve prolapse and familial long-QT syndrome. ^6,7 This effect has also been shown to be a marker for increased cardiac death in patients with peripheral arterial disease and myocardial infarction (MI). ^8–10

With regard to age groups, a cohort study aimed at evaluating chronic diseases and their risk factors in elderly patients showed that, compared with patients in the lowest tertile for QTD, those in the highest tertile (>60 ms) had a 2-fold risk for cardiac death (hazard ratio, 2.5; 95% confidence interval [CI], 1.6–4.0) and sudden cardiac death (hazard ratio, 1.9; 95% CI, 1.0–3.7), and a 40% increased risk for mortality of all causes (hazard ratio, 1.4; 95% CI, 1.2–1.8). Adjustment for potential confounding variables (history of myocardial infarction, hypertension, and overall QTc) did not change the estimated risk. These results showed that increased QTc dispersion is both a strong and an independent risk factor for cardiac death in the elderly. ¹⁰

Consequently, cardiologists have become increasingly interested in QTD in patients who have CHF, because such dispersion could theoretically be a surrogate marker of prognostic significance, as well as a tool for monitoring the effectiveness of antiarrhythmic therapies. ¹¹ Several trials have shown that β-adrenergic blockade reduces the mortality rates in patients with mild-to-moderate CHF. One of the most widely studied drugs that has proved beneficial in this setting has been carvedilol: a nonselective, 3rd-generation β-blocker. This agent has been shown to reduce the rate of sudden death, possibly by decreasing QTD. To date, no studies have investigated the relationship between CHF, QTD, carvedilol therapy, and advanced age. Therefore, given the evidence that QTD is an important marker of cardiac death, both in the elderly and in patients with CHF, we undertook a prospective study to determine 1) the effects of carvedilol therapy on QTD in patients with CHF and 2) specifically whether elderly patients with CHF (aged ≥65 years) would receive any additional observable benefit from carvedilol therapy in comparison with younger patients with CHF (aged <65 years).

Methods

Study Design and Entry Criteria

This prospective study was performed from July 1997 through February 2001 at our tertiary care university hospital. The series included 77 ambulatory patients (53 men and 24 women), aged 22 to 79 years (mean, 60.2 ±12.6 years). All the patients had symptoms and objective evidence of chronic heart failure, including a left ventricular ejection fraction of <0.45 and pul monary venous congestion or edema. To avoid any confounding variables, we excluded patients who had experienced a recent MI (within 6 months).

Each patient underwent a 12-lead ECG and a transthoracic echocardiogram (ATL Apogee CX 200, Advanced Technology Laboratories; Bothell, Wash), according to the recommendations of the American Society of Echocardiography. ¹² Carvedilol was started at 3.125 mg twice a day and was titrated up to 50 mg twice a day as tolerated. After treatment had begun, we recorded hospitalizations and morbidity, defined as a new supraventricular arrhythmia (such as new atrial fibrillation), acute MI, “flash” or worsening pulmonary edema, and worsening New York Heart Association (NYHA) functional status.

To investigate the effects of carvedilol in elderly patients, we created 2 categories of patients, divided according to age. Group I consisted of 42 patients aged <65 years, and Group II consisted of 35 patients aged ≥65 years. This investigation conformed to the principles outlined in the Declaration of Helsinki.

Measurement of QT Intervals

At the time of enrollment (before carvedilol treatment) and 6 months after the initiation of therapy, patients underwent a 12-lead ECG that was recorded at a paper speed of 25 mm/s by means of a 1-channel recorder (CardioMax FX-11, Fukuda Denshi Company, Ltd.; Tokyo, Japan). We evaluated QTD in as many of the 12 leads as possible (minimum, 9 leads) by means of a digitizing tablet (Digitizer KD 3200, Graphtec Corporation; Yokohama, Japan) that was interfaced to a personal computer. The tablet's cursor was equipped with a magnifying glass. All measurements were made by a single operator who was blinded to the patients' demographic information and diagnoses. Each measurement was stated as the mean value of 3 consecutive QT intervals. The QTD was defined as the difference between the maximum and minimum QT intervals.

Statistical Analysis

The data were analyzed by an independent statistical center, which did not participate in the study design, recruitment, or data collection. The Cox proportional hazards regression model and the χ² test were used to determine which variables were significantly associated with QTD reduction, morbidity, and mortality. The Student's t-test was used to compare the reduction in QTD before and after carvedilol therapy in the overall population and in the 2 groups categorized by age. Data were presented as the mean ± standard deviation. Results were considered significant if they yielded a P value of <0.05.

Results

The most common cause of CHF was ischemic heart disease, which was present in 42 patients (54.5%). On enrollment in the study, most patients were mildly to moderately symptomatic: 42 patients (54.5%) were in NYHA functional class II; 32 (41.5%) were in class III, and 3 (4.0%) were in class IV. All 77 patients were in sinus rhythm and were receiving additional CHF therapy, including angiotensin-converting enzyme (ACE) diuretics (74/77; 96.1%), ACE inhibitors (73/77; 94.8%), digitalis (37/77; 48.0%), nitrates (19/77; 24.7%), and amiodarone (16/77; 20.8%). Concomitant conditions included a history of hypertension (37/77; 48%), diabetes mellitus (22/77; 28.6%), and hypercholesterolemia (26/77; 33.8%). Thirty-five patients (45.5%) were smokers, and 13 patients (16.9%) admitted to alcohol abuse (>2 drinks per day).

All the patients reached a mean carvedilol dosage of 16.7 ± 7.8 mg twice a day. During the follow-up period (mean, 21 ± 13.7 mo), 9 deaths (11.7%) occurred, most of which were cardiovascular in origin (sudden death, 6; progressive heart failure, 2; and bacterial pneumonia, 1). There was no significant difference between groups I and II in the maximal dosage of carvedilol achieved (15.6 ± 7.8 vs 17.6 ± 7.8 mg twice a day, respectively), in adjuvant CHF drug treatment, or in baseline QTD values (P>0.05). After 6 months of carvedilol therapy, both groups had significantly reduced QTD values (76.9 ± 29.3 ms) compared with baseline values (104.3 ± 41.5 ms) (P<0.0001). Elevated baseline QTD values were equated with increased morbidity (P= 0.041) but not with increased frequency of hospitalization (P>0.05). The QTD reduction was more pronounced in Group I than in Group II (30.98 ± 32.70 vs 24.41 ± 29.36 ms, respectively), but this difference did not achieve statistical significance (P>0.05).

Discussion

Since its initial description in 1990 by Day and co-authors, ² QTD has been recognized as a significant prognostic marker in several disease settings, ^6–9 especially those involving ischemic heart disease. ¹³ Spargias and associates ¹⁴ demonstrated the importance of QTD after an MI with heart failure in selected patients; according to a multivariate analysis, the QTD value was an independent predictor of mortality of all causes. However, using this variable, Zabel and coworkers ¹⁵ failed to predict subsequent risk in a large series of infarct survivors. There is also evidence that, after an anterior MI, a high QTD value constitutes indirect evidence of residual viability in ischemic or hibernated areas of myocardium. ¹⁶ Of note, in the elderly, increased QTD values are also associated with higher mortality rates, as shown in the Rotterdam cohort. ¹⁰ Nonetheless, in patients with chronic CHF, the prognostic usefulness of QTD is highly controversial. Initial studies showed an increased incidence of ventricular arrhythmias and sudden cardiac death in CHF patients who had an increased QTD value, ^6,17,18 but those results could not be reproduced in larger series. ^19,20

Despite new therapeutic advances in the treatment of patients who have chronic CHF, morbidity and mortality rates remain high: half of the patients succumb to sudden death. In several studies, β-adrenergic blockade therapy has proved beneficial in improving cardiac function and reducing the all-cause and cardiac mortality and morbidity rates. ^21–24 This regimen may prevent sudden death by altering ventricular remodeling, decreasing catecholamine effects that could generate arrhythmias, and reducing myocardial ischemia. ²⁵ The literature contains few descriptions of the effects of carvedilol on QTD in patients who have chronic CHF. ^26,27

After 6 months of carvedilol therapy, our patients had a significant reduction in QTD. This finding corroborates previously published data and reveals that, in reducing QTD, the effects of carvedilol are uniformly distributed between groups of patients aged <65 years and ≥65 years. Although our study did not provide long-term mortality data and was not designed to compare morbidity parameters, it clearly showed that a significant difference exists. This finding indicates that there is an independent correlation between higher morbidity rates and higher baseline QTD values.

Study Limitations

In our study, the sample size was adequate for assessing the effects of carvedilol on QTD reduction; for intergroup comparisons of morbidity and mortality rates, however, statistical analysis was limited by the small sample size. Moreover, although we took great care to use consistent criteria in defining the QT interval (which was measured by a single experienced observer who was blinded to the patient data), the study was limited by manual measurements of the QTD ^28,29 and its lack of reproducibility. Nevertheless, errors between the observed and automatic measurements are reportedly less important in QTD than in QTc dispersion. ³⁰

Conclusions

We found that long-term carvedilol therapy resulted in a significant decrease in QTD values in ambulatory patients with CHF. This effect was uniformly distributed between patients aged <65 years and those aged ≥65 years. Further studies involving larger sample sizes are needed to define the implications of these findings in clinical practice and to determine the method's prognostic value in patients who have chronic chronic heart failure.