Sriram Padmanabhan, MD, Helme Silvet, MD, Jatin Amin, MD, and Ramdas G. Pai, MD Loma Linda, Calif

Transcription

1 Congestive Heart Failure Prognostic value of QT interval and QT dispersion in patients with left ventricular systolic dysfunction: Results from a cohort of 2265 patients with an ejection fraction of <40% Sriram Padmanabhan, MD, Helme Silvet, MD, Jatin Amin, MD, and Ramdas G. Pai, MD Loma Linda, Calif Background Increased QT interval and QT dispersion have been associated with higher mortality in populationbased studies and in patients with myocardial infarction. However, the prognostic significance of these measurements in patients with left ventricular (LV) systolic dysfunction is not clear. Methods and Results Rate corrected QT interval (QTc) and QT dispersion (QTd) were measured by means of an automated method from digitized echocardiograms in 2265 patients with an LV ejection fraction 40% and were related to survival. Increased QTc was strongly related to mortality in the whole group and in subsets on the basis of age and the level of LV systolic dysfunction. There was a graded increase in mortality rate with an increase in QTc. The effect of QTc on mortality was incremental to the effects of age and ejection fraction. QT interval was measurable in 6 leads in 1193 patients in whom QTd was computed. QTd higher than the mean value of 35 ms was associated with an increase in all cause mortality (P.04). Its mortality impact was most pronounced in the older patients, patients with more severe LV dysfunction, and patients with increased QTc. Conclusions Both QTc prolongation and increased QTd are associated with higher mortality rate in patients with moderate and severe LV dysfunction. (Am Heart J 2003;145:132-8.) Prolongation of QT interval is associated with an increased mortality rate in population-based studies, and a variety of cardiac disorders, including heart failure. 1-7 However, the impact of increased QT dispersion (QTd) on prognosis is uncertain. Although QTd has been associated with increased mortality rate in small series of patients with heart failure, some of the larger prospective studies have not confirmed these findings To our knowledge, there are no studies addressing their prognostic importance as a function of the level of left ventricular (LV) systolic dysfunction. This study examines the prognostic implications of QT and QTd prolongation in a large cohort of patients with moderate and severe LV dysfunction. From the Department of Cardiology, Loma Linda University and VA Medical Center, Loma Linda, Calif. Submitted September 6, 2001; accepted March 13, Reprint requests: Ramdas G. Pai, MD, FRCP(E), FACC, Professor of Medicine and Director of Research/Fellowship Program, Division of Cardiology, Loma Linda University Medical Center, Anderson St, Loma Linda, CA ramdaspai@yahoo.com Copyright 2003, Mosby, Inc. All rights reserved /2003/$ doi: /mhj Methods Study population This retrospective cohort study included patients at Loma Linda Veterans Affairs (VA) Medical Center. All patients undergoing echocardiographic examinations between July 1990 and June 1998 were enrolled if they had simultaneous electrocardiograms (ECGs). For patients with multiple studies, the first examination was included. Of the 8931 patients meeting these criteria, 2265 patients had an estimated left ventricular ejection fraction (LVEF) 40%, and they formed the study cohort. The institutional review board approved the use of the echocardiographic database for all retrospective studies. Echocardographic examination All patients underwent standard 2-dimensional and Doppler scanning echocardiographic examinations. LVEF was assessed visually by a level-3 trained echocardiographer and entered into the database at the time of the examination. Anatomic measurements were performed according to the recommendations of the American Society of Echocardiography. 12 A visual estimate of ejection fraction (EF) by an experienced interpreter has been shown to correlate very closely with that obtained by means of radionuclide ventriculography and to be equivalent or superior to other quantitative echocardiographic methods. 13,14

2 American Heart Journal Volume 145, Number 1 Padmanabhan et al 133 Figure 1 Effect of rate corrected QTc on mortality rate in patients with moderate and severe LV systolic dysfunction. ECG analysis Digitally stored ECG data on the MUSE system (Marquette Medical Systems, Milwaukee, Wis) was used for analysis. The QT interval was computed for all the patients by use of QT Guard software (Marquette Medical systems, Milwaukee, Wis). The QT Guard software determines T-wave compliability by means of principal component analysis. T-wave offset, T peak, and T end were computed by means of the least squares fit algorithm 15 and were threshold-based measurements. All filters for analysis of QTd were at default settings as determined by means of Marquette Medical Systems in its analysis. 15 QTd was measured as the difference between the longest and shortest QT intervals in the 12-lead ECG with at least 6 measurable leads. Thus, QTd was obtained in 1193 patients. The resolution of QT Guard is 4 ms, and the filter range is 0.05 to 100 Hz. Follow-up and mortality data Once the population was characterized, follow-up data of these patients were gathered by use of the VA Hospital medical records, VA mortality database, prescription refills, and clinic visits. Different causes of mortality were not identified. Alive status was confirmed by means of prescription refills or clinic visits. In a minority of patients, direct telephone contact by a member of the cardiology team was used as a means of determining the alive status. Only 76 patients (3%) were lost to follow-up. Database management and statistical analysis The clinical, electrocardiographic, echocardiographic, and follow-up databases were merged by use of the Microsoft Access software program (Microsoft Corp, Seattle, Wash). The effect of binary variables on survival was analyzed by use of the Kaplan-Meier method, and the curves were compared by use of the log-rank statistic. StatView software version 5.01 (SAS Insitute, Cary, NC) was used for statistical analysis. The proportional hazards model was used as a means of looking at the effect of continuous variables on survival and correcting for the effect of other confounding variables. A P value.05 was considered to be significant.

3 134 Padmanabhan et al American Heart Journal January 2003 Figure 2 Effect of QTc on mortality rate: QTc prolongation was predictive of an increased mortality rate in subsets of patients on the basis of age and rhythm. Results Baseline patient characteristics Almost all the patients were male (97%). The mean patient age was years, the mean EF 29% 9%, the mean rate corrected QT interval (QTc) was ms (range ms), and the mean QTd was ms (4-168 ms). A total of 1830 patients (81%) were white, 222 patients (10%) were African American, and 157 patients (7%) were Hispanic. In a mean follow-up period of 855 days, there were 712 deaths. Effect of rate corrected QT interval on mortality Patients with a longer QTc by use of a threshold of 440 or 450 or 460 ms had an increased mortality rate compared with patients with a shorter QTc (P.0004, P.0001, and P.0003, respectively). As shown in Figure 1, patients with a QTc 450 ms (close to the mean) had a 5-year mortality rate of 75%, compared with 52% for the remaining patients (P.0001). Mortality impact of QT prolongation was seen in all subsets of patients, including those with an LVEF 30% (P.01) or those with an LVEF 30% (P 0.001), those in sinus rhythm (P.0001) or atrial fibrillation (P.05), those aged 65 years (P.013), or those aged 65 years (P.002) (Figures 1 and 2). Effect of QT dispersion on mortality Patients with a QTd greater than the mean value of 35 ms had a higher mortality rate. Patients with a QTd

4 American Heart Journal Volume 145, Number 1 Padmanabhan et al 135 of 35 ms had a 5-year mortality rate of 58%, compared with 45% for the remaining patients (P.04) (Figure 3). The mortality rate impact of increased QTd was more striking in patients with a LVEF 30% (85% vs 45%, P.03) and patients with a QTc 450 ms (90% vs 50%, P.03) (Figures 4 and 5). Figure 3 Prognostic implications of rate corrected QT interval and QT dispersion in multivariate analysis The effect of QTc on mortality rate persisted after correcting for age and EF by use of the proportional hazards model (P.0001 for age and EF and P.004 for QTc). However, QTd was not related to mortality rate after correcting for these variables. Discussion The results of this study show that both computermeasured QTc and QTd are predictive of mortality rate in patients with moderate and severe LV systolic dysfunction, which is defined as an LVEF 40%. This LVEF cutoff level was used because mortality rate increases exponentially when the EF drops lower than this level. Effect of QTd on all-cause mortality rate in patients with moderate and severe LV systolic dysfunction. Prognostic implications of rate corrected QT interval Despite the absence of evidence from the Framingham study, increased QTc has generally been associated with an increased mortality rate in large population-based studies and smaller studies in patients with cardiac disease. 1-7 Data from multiple population-based studies that enrolled mostly healthy subjects indicate that QTc prolongation increases mortality rate. Data from the Strong Heart Study on 1839 subjects showed that QTc and QTd, measured by means of the same method used in our study, are independent predictors of mortality rate. 1 The Rotterdam study, which enrolled 5241 subjects, showed QTc prolongation to be an independent predictor of the cardiac and all-cause mortality rate. 7 A Danish study that enrolled 3455 subjects aged 30 to 60 years reported both QTc 430 ms and QTd 80 ms to be independent predictors of mortality rate. 3 However, in a population-based study from Finland enrolling 10,000 subjects, QT prolongation was associated with a higher mortality rate only in men with heart disease. 5 There are very little data about the prognostic implications of QT prolongation in patients with heart failure or LV dysfunction. The largest study addressing this issue, the United Kingdom Heart Failure Evaluation and Assessment Risk Trial (UK-HEART), enrolled 554 ambulatory patients with congestive heart failure. 6 Both QTc and QTd were found to be predictors of mortality rate only in univariate, but not in multivariate, analysis. Our study, the largest in this subject area, strongly supports the prognostic implications of QTc prolongation in patients with LV dysfunction. Prognostic implications of QT dispersion The largest congestive heart failure (CHF) study addressing the prognostic implications of QTd is the Diamond-CHF study. 8 Of the 1518 patients enrolled, QTd was measurable in 703, and QTd was not predictive of mortality rate. Pinsky et al 9 found QTd 140 ms to be a predictor of mortality rate in patients awaiting cardiac transplantation. Fei et al, 10 in a study involving 60 patients with CHF caused by idiopathic dilated cardiomyopathy, found QTd to be a predictor of ventricular tachycardia on the Holter monitor, but not death. Of 205 patients with CHF, Galinier et al 11 found a QTd 80 ms to be predictive of sudden death in nonischemic, but not in ischemic, cardiomyopathy. Thus, the value of QTd for prognostication in patients with CHF has not been clear, because small studies do not have enough power to show its effect on mortality rate. Our study, involving a very large group of patients with LV dysfunction, does support the prognostic value of automated QTd measurement in patients with LV dysfunction. However, the prognostic value of QTd seems to be less than that of QTc. The impact of an increase in QTd was greater in patients with greater degrees of LV dysfunction and patients with prolonged QTc, the groups with a poorer prognosis.

5 136 Padmanabhan et al American Heart Journal January 2003 Figure 4 Effect of QTd on mortality rate stratified by means of ejection fraction and age. Manual versus automated QTd measurement Automated QTd measurement by means of the QT- Guard program is reproducible and has been well validated. 1,15-17 The reproducibility of this method has been studied by Xue and Reddy. 15 They performed QTd measurements serially on ECGs at baseline and after 30 minutes, 1 day, 1 week, and 30 days, and the mean difference was 7.6 ms. Okin et al 1 have used this software to analyze the ECGs from the Strong Heart Study (n 1839) and found it to be a reproducible measurement, with a strong relation to mortality rate. Thus, we feel this tool is valid, and this feeling is also supported by our data. Study limitations This was a retrospective study, and detailed clinical and pharmacologic data were not available. Almost all patients were men, and thus the data may not be applicable to women. However, it is the largest available study that has evaluated the value of QTc and QTd in patients with LV dysfunction. Significance and conclusions This study established both prolongation of QTc and increase in QTd to be predictors of mortality rate in patients with LV dysfunction. Increased QTc was a stronger predictor of mortality compared to QTd. QTd

6 American Heart Journal Volume 145, Number 1 Padmanabhan et al 137 Figure 5 Effect of QTd on mortality rate stratified by QTc and rhythm. may have an important prognostic value in patients with greater degrees of LV dysfunction and in patients with LV dysfunction who have QTc prolongation. References 1. Okin PM, Devereux RB, Howard BV, et al. Assessment of QT interval and QT dispersion for prediction of all-cause and cardiovascular mortality in American Indians: the Strong Heart Study. Circulation 2000;101: de Bruyne MC, Hoes AW, Kors JA, et al. Prolonged QT interval predicts cardiac and all-cause mortality in the elderly: the Rotterdam Study. Eur Heart J 1999;20: Elming H, Holm E, Jun L, et al. The prognostic value of the QT interval and QT interval dispersion in all-cause and cardiac mortality and morbidity in a population of Danish citizens. Eur Heart J 1998;19: Brooksby P, Batin PD, Nolan J, et al. The relationship between QT intervals and mortality in ambulant patients with chronic heart failure: the United Kingdom Heart Failure Evaluation and Assessment of Risk Trial (UK-HEART). Eur Heart J 1999;20: Karjalainen J, Reunanen A, Ristola P, et al. QT interval as a cardiac risk factor in a middle aged population. Heart 1997;77: Goldberg RJ, Bengtson J, Chen ZY, et al. Duration of the QT interval and total and cardiovascular mortality in healthy persons (the Framingham Heart Study experience). Am J Cardiol 1991;67: de Bruyne MC, Hoes AW, Kors JA, et al. QTc dispersion predicts cardiac mortality in the elderly: the Rotterdam Study. Circulation 1998;97: Brendorp B, Elming H, Jun L, et al. QT dispersion has no prognos-

7 138 Padmanabhan et al American Heart Journal January 2003 tic information for patients with advanced congestive heart failure and reduced left ventricular systolic function. Circulation 2001;103: Pinsky DJ, Sciacca RR, Steinberg JS. QT dispersion as a marker of risk in patients awaiting heart transplantation. J Am Coll Cardiol 1997;29: Fei L, Goldman JH, Prasad K, et al. QT dispersion and RR variations on 12-lead ECGs in patients with congestive heart failure secondary to idiopathic dilated cardiomyopathy. Eur Heart J 1996;17: Galinier M, Vialette JC, Fourcade J, et al. QT interval dispersion as a predictor of arrhythmic events in congestive heart failure: importance of aetiology. Eur Heart J 1998;19: Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. J Am Soc Echocardiogr 1989;2: van Royen N, Jaffe CC, Krumholz HM, et al. Comparison and reproducibility of visual echocardiographic and quantitative radionuclide left ventricular ejection fractions. Am J Cardiol 1996;77: van t Hof AW, Schipper CW, Gerritsen JG, et al. Comparison of radionuclide angiography with three echocardiographic parameters of left ventricular function in patients after myocardial infarction. Int J Card Imaging 1998;14: Xue Q, Reddy S. Algorithms of computerized QT analysis. J Electrocardiol 1998;30: Murray A, McLaughlin NB, Campbell RW. Measuring QT dispersion: man versus machine. Heart 1997;77: Savelieva I, Yi G, Guo X, et al. Agreement and reproducibility of automatic versus manual measurement of QT interval and QT dispersion. Am J Cardiol 1998;81: The following article is an AHJ Online Exclusive. Full text of this article is available at no charge at our website: Reduction of oxidative stress augments natriuretic effect of furosemide in moderate heart failure Hirofumi Tomiyama, MD, a Gohki Watanabe, MD, b Hideo Yoshida, MD, b Nobutaka Doba, MD, b and Akira Yamashina, MD a Tokyo and Ichihara, Japan Background The significance of antioxidant therapy in heart failure has not been fully examined. This study evaluated whether vitamin C has beneficial effects on renal function or augments the renal effects of furosemide in patients with heart failure. Methods There were 2 protocols. In protocol 1, plasma level of thiobarbituric acid-reactive substances (TBARS) and renal function were assessed before and after intravenous infusion of vitamin C or placebo in 8 patients with moderate congestive heart failure (CHF) treated with enalapril. In protocol 2, a randomized crossover study was performed in patients with moderate CHF treated with either an ACE inhibitor (enalapril) (n 10) or an angiotensin II receptor antagonist (losartan) (n 9) and in asymptomatic patients with impaired left ventricular function treated with enalapril (n 8). TBARS and renal function were assessed before and after intravenous infusion of furosemide alone, coinfusion of furosemide with placebo and vitamin C, or coinfusion of furosemide with vitamin C and a kallikrein inhibitor (nafamostat mesilate). Results In protocol 1, although vitamin C reduced TBARS, it did not affect renal function. In protocol 2, TBARS was higher in patients with moderate CHF than in asymptomatic patients. Vitamin C augmented natriuretic effect of furosemide (from to mol/min, P.01) only in patients with moderate CHF treated with enalapril but not in the other 2 groups. Nafamostat mesilate prevented this augmentation. Conclusions In patients with CHF treated with enalapril, counteraction of the increased oxidative stress by vitamin C may contribute to the augmented natriuretic effect of furosemide through the renal kinin-nitric oxide pathway. (Am Heart J 2003;145:e2.) 42