Prolonged Corrected QT Interval Is Predictive of Future Stroke Events Even in Subjects Without ECG-Diagnosed Left Ventricular HypertrophyNovelty and Significance
We attempted to evaluate whether subjects who exhibit prolonged corrected QT (QTc) interval (≥440 ms in men and ≥460 ms in women) on ECG, with and without ECG-diagnosed left ventricular hypertrophy (ECG-LVH; Cornell product, ≥244 mV×ms), are at increased risk of stroke. Among the 10 643 subjects, there were a total of 375 stroke events during the follow-up period (128.7±28.1 months; 114 142 person-years). The subjects with prolonged QTc interval (hazard ratio, 2.13; 95% confidence interval, 1.22–3.73) had an increased risk of stroke even after adjustment for ECG-LVH (hazard ratio, 1.71; 95% confidence interval, 1.22–2.40). When we stratified the subjects into those with neither a prolonged QTc interval nor ECG-LVH, those with a prolonged QTc interval but without ECG-LVH, and those with ECG-LVH, multivariate-adjusted Cox proportional hazards analysis demonstrated that the subjects with prolonged QTc intervals but not ECG-LVH (1.2% of all subjects; incidence, 10.7%; hazard ratio, 2.70, 95% confidence interval, 1.48–4.94) and those with ECG-LVH (incidence, 7.9%; hazard ratio, 1.83; 95% confidence interval, 1.31–2.57) had an increased risk of stroke events, compared with those with neither a prolonged QTc interval nor ECG-LVH. In conclusion, prolonged QTc interval was associated with stroke risk even among patients without ECG-LVH in the general population.
The detection of a prolonged corrected QT (QTc) interval on an ECG has been reported to be a risk factor for cardiovascular mortality, which could be attributable to an increased risk of fatal arrhythmia. However, subjects with prolonged QTc intervals are often older and commonly exhibit higher blood pressure and left ventricular hypertrophy (LVH); thus, a prolonged QTc interval could be a marker of cardiovascular risk factor accumulation. Therefore, some studies have suggested that prolonged QTc intervals are associated with an increased risk of future stroke events.1–3
Individuals with prolonged QTc intervals often have LVH. The detection of LVH on ECG (ECG-LVH) is a strong risk marker for future stroke events, independently of blood pressure.4 However, it is not known whether the association between prolonged QTc intervals and the risk of stroke is derived from the associations with LVH.
The purpose of this study was to evaluate whether a prolonged QTc interval is a marker of future stroke events among subjects with or without ECG-LVH and to compare the effects of prolonged QTc intervals and ECG-LVH on the risk of such events.
The Jichi Medical School Cohort Study began in 1992 and aimed to clarify the risk factors for cardiovascular and cerebrovascular diseases in the Japanese general population. The details of the protocol of the Jichi Medical School Cohort Study have been reported elsewhere.5 Baseline data were collected between April 1992 and July 1995 in 12 rural districts using a government-sponsored mass screening system. In each community, a local government office sent personal invitations to all potential subjects by mail in accordance with the Health and Medical Service Law for the Aged. The individuals who participated in the mass screening examinations were aged 40 to 69 years in 8 areas (Iwaizumi, Tako, Kuze, Sakuma, Sakugi, Okawa, Ainoshima, and Akaike), were aged ≥30 years in 1 area (Wara), and belonged to other age groups in 3 areas (Hokudan, Yamato, and Takasu). At the baseline, the total number of subjects enrolled in the Jichi Medical School Cohort Study was 12 490 (4911 men and 7579 women). We initially excluded potential subjects for the following reasons: no ECG (n=1285), pacemaker implantation (n=7), atrial fibrillation (n=57), advanced atrioventricular block (n=1), complete atrioventricular block (n=1), dextrocardia (n=2), complete left bundle block (n=20), complete right bundle block (n=196), immeasurable QTc (n=27), heart rate of >150 bpm (n=1), incomplete data (n=5), or no follow-up data (n=84).6 In addition, we excluded subjects without data of Cornell product (n=7), with a history of stroke (n=104), or with myocardial infarction (n=50). Therefore, the number of subjects analyzed in this study was 10 643.
ECG Measurement and Interpretation
The ECG was performed at a paper speed of 25 mm/s and a gain of 10 mm/mV (or 5 mm/mV), using the ECG devices at each institution. Initially, a trained individual who was unaware of the subjects’ background data measured the QT interval manually from lead II (or lead I or III if the QT interval could not be measured from lead II), which is the best lead for depicting the end of the T wave on a 12-lead ECG, according to a previously described method for measuring the QTc interval.7 The ECG measurements were performed using a ruler with 0.01-mm graduations. The QTc interval for a single beat was measured from the beginning of the QRS complex to the end of the T wave. The end of the QT interval was taken as the last point of the T wave over leads I, II, and III, ie, where the downsloping limb joined the baseline, while we excluded U waves.8,9 RR intervals were also measured, and the mean of 3 RR intervals was calculated. Heart rate-adjusted QT intervals (QTc) were calculated using the Bazett10 formula: (QTc=QT/[RR1/2]). The intraobserver reproducibility of the Bazett QTc interval data was confirmed by comparing pairs of measurements for 98 ECGs. The mean relative error of the QTc interval data was 0.5%, and the mean intraobserver absolute error of the QTc interval data was 2.0 ms (SD, 17.0 ms).
A Cornell product of ≥244 mV×ms was used to diagnose ECG-LVH according to the Japanese Society of Hypertension Guidelines for the Management of Hypertension (2009)11 and the Losartan Intervention For End point (LIFE) Reduction in Hypertension Study.12 A prolonged QTc interval was defined as a Bazett QTc interval of ≥440 ms in men and ≥460 ms in women according to the method described in a previous report.6
Questionnaire and Other Measurements
Details of Questionnaire and Other Measurements are provided in the online-only Data Supplement.
The internal review board of the Jichi Medical University School of Medicine approved this study. Written informed consent for the study was obtained from each individual who participated in the mass screening health check-up examinations at the baseline.
Follow-Up and Diagnostic Criteria
Details of Follow-Up and Diagnostic Criteria are provided in the online-only Data Supplement.
Data are shown as mean±SD or percentage values. The nonpaired t test was performed to evaluate the difference in continuous variables between the subjects with and without prolonged QTc intervals. The χ2 test was used to detect differences in the frequencies of characteristics between the groups. Improvement of the predictive value of QTc intervals in addition to the traditional cardiovascular risk factors for stroke events was evaluated using net reclassification improvement (NRI) and integrated discrimination improvement.13 One-way ANOVA was performed to evaluate the overall differences among the subjects with neither prolonged QTc intervals nor ECG-LVH, those with prolonged QTc intervals but not ECG-LVH, and those with ECG-LVH; the Tukey honestly significant difference test was used for comparisons of mean values among the groups. Survival curves among the 3 groups were estimated using the Kaplan–Meier method. Adjusted hazard ratios and 95% confidence intervals for stroke events were calculated using Cox proportional hazards regression analysis adjusted for the following conventional cardiovascular risk factors: age, sex, body mass index, current smoking, alcohol intake of >20 g/d, systolic blood pressure, antihypertensive drug use, heart rate, the presence of hyperlipidemia, and the presence of diabetes mellitus. P values of <0.05 were considered to be statistically significant. The statistical software SPSS (version 18.0; Chicago, IL) was used for all analyses.
The characteristics of the study subjects are shown in Table 1. At the baseline, the subjects’ mean age was 55.4±11.2 years, and 37.6% of the subjects were men. The mean Bazett QTc interval was 388±27 ms.
QTc Interval as a Risk Marker of Stroke Events
During the mean follow-up period of 128.7±28.1 months (114 142 person-years), there were a total of 375 stroke events (85 cerebral hemorrhages, 242 ischemic strokes, 47 subarachnoid hemorrhages, and 1 stroke event of unknown cause).
The Stroke Risk in Subjects With Prolonged QTc Intervals (≥440 ms in Men and ≥460 ms in Women)
Characteristics of subjects with and without prolonged QTc intervals (≥440 ms in men and ≥460 ms in women) are shown separately in men and women (Table 1). The subjects with prolonged QTc interval were older and had higher percentage of hypertension and greater blood pressure level than those without prolonged QTc interval in both men and women.
In the analysis, combining men and women together, there were 354 stroke events (3.4%; 31.5 events per 10 000 person-years) in subjects without prolonged QTc intervals (n=10 481) and there were 16 stoke events (9.9%; 99.6 events per 10 000 person-years) in those with prolonged QTc intervals (n=162). The subjects with prolonged QTc intervals had a 2.13 times higher risk of stroke events than those without prolonged QTc intervals even after adjustment for the traditional risk factors and ECG-LVH (assessed as a Cornell product of >244 mV×ms; Table 2). The interaction between prolonged QTc interval and ECG-LVH was statistically insignificant (P=0.171) in the multivariate-adjusted Cox regression model. The multivariate-adjusted hazard risks stratified by the patients’ characteristics are shown in Figure 1. The stroke risk associated with prolonged QTc intervals was significant in subjects of older age and male sex and with hypertension and hyperlipidemia and in those without diabetes mellitus, obesity, or ECG-LVH.
NRI by Adding QTc Interval Into the Traditional Cardiovascular Risk Factors
The NRI for predictive values of stroke events was insignificant when the prolonged QTc interval (as a categorical variable) was included in the model in addition to the traditional risk factors (data not shown). The NRI of predictive values was significant when the QTc interval as a continuous variable was included in the model in addition to the traditional cardiovascular risk factors. The NRI was also significant when ECG-LVH alone was added and also when both the QTc interval and ECG-LVH were added to the traditional cardiovascular risk factors (Tables S1–S3 in the online-only Data Supplement). Moreover, NRI was significantly increased after adding the QTc interval (as a continuous variable) to the traditional cardiovascular risk factors that included ECG-LVH (NRI=0.014; P<0.001; Table 3).
Characteristics of the Subjects With Prolonged QTc Intervals but Not ECG-LVH
As the number of subjects with both prolonged QTc intervals and ECG-LVH was too small to allow a statistical analysis to be performed, we stratified the subjects into the following 3 groups: (1) those with neither prolonged QTc intervals nor ECG-LVH, (2) those with prolonged QTc intervals but not ECG-LVH, and (3) those with ECG-LVH. The characteristics of the subjects in each group are shown in Table 4. In total, 1.2% of the subjects had prolonged QTc intervals without ECG-LVH. The subjects with prolonged QTc intervals but not ECG-LVH were older, had higher percentage of men among them, and had higher blood pressure and heart rates than those with neither prolonged QTc intervals nor ECG-LVH. There were no significant differences in age and blood pressure level between the subjects with prolonged QTc intervals but not ECG-LVH and those with ECG-LVH. The propensity stroke risk score matched for traditional cardiovascular risk factors was significantly greater in subjects with prolonged QTc intervals but not ECG-LVH and in the subjects with ECG-LVH than that in the subjects with neither prolonged QTc intervals nor ECG-LVH (both P<0.001); however, it was insignificant between the subjects with prolonged QTc intervals but not ECG-LVH and the subjects with ECG-LVH (P=0.089).
Incidence of Stroke Events in the Subjects With Prolonged QTc Intervals Without ECG-LVH
The cumulative incidence of stroke events during the follow-up period is shown in Figure 2. The subjects with prolonged QTc intervals but not ECG-LVH exhibited a greater incidence of stroke events than those who exhibited ECH-LVH and those who exhibited neither prolonged QTc intervals nor ECG-LVH. Even after adjusting for traditional cardiovascular risk factors, the subjects with prolonged QTc intervals but not ECG-LVH were found to be at increased risk of stroke events in comparison with the subjects who exhibited neither prolonged QTc intervals nor ECG-LVH (Table 5). The hazard risks for stroke events of prolonged QTc interval but not ECG-LVH were greater than the hazard risks of ECG-LVH for cerebral hemorrhage and ischemic stroke but not for subarachnoid hemorrhage.
A prolonged QTc interval was found to be associated with an increased risk of future stroke events even among Japanese subjects without ECG-LVH, which was diagnosed based on the Cornell product. The subjects with prolonged QTc intervals but not ECG-LVH exhibited an increased stroke risk compared with the subjects who had neither prolonged QTc intervals nor ECG-LVH, and the hazard risk was greater than that in the subjects with ECG-LVH. These results suggest that the prolonged QTc interval was associated with stroke risk even among patients without ECG-LVH.
Three previous studies have demonstrated that prolonged QTc intervals are associated with an increased risk of stroke: (1) Cardoso et al1 detected this association in a Brazilian diabetic population, (2) Maebuchi et al2 observed it in the general Japanese population, and (3) Soliman et al3 found it in the general US population. In addition to supporting the findings of these previous studies, our results also suggest that a prolonged QTc interval is a predictor of future stroke events even in individuals without ECG-LVH.
In the analysis of NRI, adding the QTc interval as a continuous variable to the traditional cardiovascular risk factors that included ECG-LVH in the model improved the predictive value of stroke events; however, adding the prolonged QTc interval as a categorical variable did not increase the predictive value. There were no statistical differences in the propensity stroke risk score between the subjects with prolonged QTc interval but not ECG-LVH and those with ECG-LVH; therefore, the predictive value of prolonged QTc interval and ECG-LVH for stroke events did not differ significantly.
The mechanisms responsible for the association between prolonged QTc intervals and a greater risk of stroke events might involve increased arterial stiffness14 because prolonged QTc intervals have also been reported to be a marker of increased carotid intima media thickness.15 Therefore, the risk of subarachoid hemorrhage, which is susceptible to elevated blood pressure rather than atherosclerosis, might not be increased by prolonged QTc intervals, but it was increased by ECG-LVH. The QTc interval represents the time required for ventricular excitation and repolarization in the heart, but it was found to be related to stroke events rather than myocardial infarction in this study, probably because of the small number of myocardial infarction events experienced by our subjects (data not shown).
The relationship between prolonged QTc intervals and stroke events might be affected by heart rate. We used the Bazett formula to calculate QTc intervals in this study to make it possible to compare our results with those of a previous study involving a Japanese population.2 Compared with other QTc interval formulas, such as the Fridericia and Framingham formulas, the Bazett formula is considered to be markedly affected by heart rate.16 However, it was reported that the stroke risk associated with a prolonged QTc interval was not influenced by the formula used to calculate the interval,3 and in this study, the relationship between prolonged QTc intervals and stroke events remained significant after adjusting for heart rate.
The QTc interval has also been reported to be associated with the incidence of atrial fibrillation,17,18 a well-known risk factor for stroke events, and we previously reported that subjects who exhibited atrial fibrillation on ECG at the baseline were at an increased risk of stroke.18 Therefore, we excluded subjects who displayed atrial fibrillation at the baseline from the present analysis. In addition, the incidence of atrial fibrillation during the follow-up period was not evaluated in this study.
This study had the following limitations: (1) we measured the subjects’ QTc intervals manually using lead II (or lead I or III if it was difficult to measure using lead II). Although the mean QTc interval obtained in this study was shorter than that reported in previous studies involving Japanese patients who were referred to a university hospital19 or a general Japanese population,2 the Bazett QTc interval cutoff level that was found to be associated with a significantly increased risk of stroke was comparable to that described in a previous report.2 (2) The QTc interval could be affected by drug use and electrolyte levels; however, these factors were not evaluated in this study. (3) We did not obtain the incidence of atrial fibrillation during the follow-up period.
In the general Japanese population, a prolonged QTc interval is associated with an increased risk of future stroke events even in subjects without ECG-LVH, as diagnosed using the Cornell product. The presence of a prolonged QTc interval on ECG could be a risk marker of future stroke events in subjects without ECG-LVH.
Sources of Funding
This study was supported by a Grant-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and grants from the Foundation for the Development of the Community, Tochigi, Japan.
J.I., S.I., and K.K. are Jichi Medical School (JMS) Cohort Study Investigators.
Current address for J.S.: Division of Cardiovascular Medicine, Tokyo Metropolitan Geriatric Hospital and Institute of Gerontology, Tokyo 173-0015, Japan.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.114.04722/-/DC1.
- Received October 8, 2014.
- Revision received October 26, 2014.
- Accepted December 1, 2014.
- © 2014 American Heart Association, Inc.
- Cardoso CRL,
- Salles GF,
- Deccache W
- Maebuchi D,
- Arima H,
- Doi Y,
- Ninomiya T,
- Yonemoto K,
- Tanizaki Y,
- Kubo M,
- Hata J,
- Matsumura K,
- Iida M,
- Kiyohara Y
- Soliman EZ,
- Howard G,
- Cushman M,
- Kissela B,
- Kleindorfer D,
- Le A,
- Judd S,
- McClure LA,
- Howard VJ
- Ishikawa J,
- Ishikawa S,
- Kabutoya T,
- Gotoh T,
- Kayaba K,
- Schwartz JE,
- Pickering TG,
- Shimada K,
- Kario K
- Ishikawa J,
- Ishikawa S,
- Kario K
- Lepeschkin E,
- Surawicz B
- Algra A,
- Tijssen JG,
- Roelandt JR,
- Pool J,
- Lubsen J
- Bazett H
- Okin PM,
- Devereux RB,
- Gerdts E,
- Snapinn SM,
- Harris KE,
- Jern S,
- Kjeldsen SE,
- Julius S,
- Edelman JM,
- Lindholm LH,
- Dahlöf B
- Festa A,
- D’Agostino R Jr.,
- Rautaharju P,
- O’Leary DH,
- Rewers M,
- Mykkänen L,
- Haffner SM
- Rautaharju PM,
- Surawicz B,
- Gettes LS,
- et al
- Nielsen J,
- Graff C,
- Pietersen A,
- Lind B,
- Struijk J,
- Olesen M,
- Haunsø S,
- Gerds T,
- Svendsen J,
- Køber L,
- Holst A
Novelty and Significance
What Is New?
We evaluated whether prolonged corrected QT (QTc) interval was associated with an increased risk of stroke without ECG left ventricular hypertrophy.
What Is Relevant?
The prolonged QTc interval is not only a risk marker of fatal arrhythmia but also a marker of cardiovascular risk accumulation without ECG left ventricular hypertrophy.
The subjects with prolonged QTc interval had an increased risk of stroke without ECG left ventricular hypertrophy in the general population.