SEARCH

SEARCH BY CITATION

Keywords:

  • QT;
  • QT dispersion;
  • QT corrected interval;
  • central fat deposition;
  • body fat

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Objective: Because obese patients generally may be prone to ventricular arrhythmias, this study was designed to measure the interval between Q- and T-waves of the electrocardiogram (QT) interval dispersion (QTD) in uncomplicated overweight and obese patients. QTD is an electrocardiographic parameter whose prolongation is thought to be predictive of the possibility of sudden death caused by ventricular arrhythmias. To better evaluate the association between obesity per se and QTD, the study population was intentionally selected because they were free of complications.

Research Methods and Procedures: QTD (defined as the difference between the maximum and the minimum QT corrected interval [QTc] across the 12-lead electrocardiogram) was measured manually in 54 obese patients (Group A: mean body mass index [BMI] of 38.1 ± 0.9 kg/m2 [SEM], 15 males and 39 females), 35 overweight patients (Group B: mean BMI of 27.3 ± 0.2 kg/m2, 10 males and 25 females), and 57 normal weight healthy control subjects (Group C: mean BMI of 21.9 ± 0.2 kg/m2, 17 males and 40 females). The obese and overweight patients had no heart disease, hypertension, diabetes, or impaired glucose tolerance and did not have any hormonal, hepatic, renal or electrolyte disorders. The study subjects were matched in terms of age (mean age 38.4 ± 1.2 years) and sex.

Results: The QTDs were comparable among the three groups: Group A, 56.4 ± 2.6 ms; Group B, 56.7 ± 2.1 ms; and Group C, 59.4 ± 2.1 ms; not significant. The QTc intervals of Group A and Group B were similar to that of Group C (411.8 ± 3.3, 407.2 ± 3.9, and 410.3 ± 3.9 ms, respectively [not significant]) and did not correlate with BMI. An association was found between QTD and QTc (r = 0.24, p < 0.005). Using multivariate stepwise regression analysis of the study population, QTD did not correlate with age, BMI, waist circumference, or abdominal sagittal diameter.

Discussion: These data suggest that QTD in uncomplicated obese or overweight subjects is comparable with that in age- and sex-matched normal weight healthy controls. In this study population, no association was found between QTD and anthropometric parameters reflecting body fat distribution.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Epidemiological studies indicate that obesity may be an independent risk factor for a number of cardiovascular complications, including hypertension, cardiac failure, and ischemic heart disease (1). Moreover, the occurrence of sudden death has been reported in obese subjects who had no apparent heart abnormalities (2) (3), although its mechanism is still unclear. The risk of fatal arrhythmias in obese people (especially in those with morbid obesity) is also thought to be correlated with a prolonged electrocardiographic interval between Q- and T-waves (QT). The relationship between obesity and QT length was first described more than 10 years ago (4), but subsequent studies on the degree and the meaning of the QT interval in obese people have led to conflicting results (5) (6) (7) (8) (9) (10).

The QT dispersion (QTD), defined as the difference between the maximum and minimum QT corrected (QTc) interval occurring in any of the 12 electrocardiogram (ECG) leads, has been suggested to be an electrocardiographic index reflecting the physiological variability of regional ventricular repolarization. An increase in QTD (in milliseconds) is a possible substrate for ventricular arrhythmias and sudden death in patients with chronic heart failure (11), hypertension, and left ventricular hypertrophy (12), in individuals with a prolonged QT interval (13), and in subjects who are >55 years of age (14). However, the prognostic value of QTD as a risk factor for arrhythmias has been questioned in patients with recent myocardial infarction (15). An increase in QTD has also been described in diabetic patients, especially those with autonomic dysfunction (16). It is therefore of interest to study QTD in obesity, a condition that may be accompanied by subclinical left ventricular hypertrophy (17) (18), cardiac arrhythmias, prolonged QT interval, and sudden death (2) (3). To the best of our knowledge only one very recent study has addressed this issue in Japanese obese patients (19), and no data are currently available concerning obese white patients.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Subjects

The study involved 146 subjects (42 men, 104 women), with a mean age (±SE) of 38.4 ± 1.2 years (range 20 to 60 years). A total of 54 individuals were obese (Group A: 15 men, 39 women, mean body mass index [BMI] of 38.1 ± 0.9 kg/m2, range 30.1 to 54.2 kg/m2), 35 subjects were overweight (Group B: 10 men, 25 women, mean BMI of 27.3 ± 0.2 kg/m2, range 25 to 29.1 kg/m2), and 57 participants were healthy normal weight volunteers belonging to the medical and student staff of our department (Group C: 17 men, 40 women, mean BMI of 21.9 ± 0.2 kg/m2, range 16 to 24.9 kg/m2). The three groups were matched for age and sex. The obese and overweight patients were consecutively recruited from the outpatients attending our Metabolic Unit. Subjects with a history of hypertension (systolic and diastolic blood pressure of ≥140/90), diabetes or impaired glucose tolerance (IGT), ischemic heart disease, previous arrhythmic episodes, or hepatic, renal, or thyroid diseases were excluded. None of the subjects were taking medication known to affect electrocardiographic intervals. The obese and overweight patients had been weight-stable (within 2%) during the 3 months preceding the study. To exclude diabetes or IGT, all of the subjects underwent a 75-g oral glucose tolerance test according to World Health Organization criteria. Blood pressure was measured using a mercury sphygmomanometer with the subjects in a sitting position after a 10-minute rest period, with the mean of three determinations being recorded; diastolic pressure was measured at the fifth Korotkoff sound. For the obese subjects, an arm cuff of appropriate size was used. The clinical characteristics of the study subjects are shown in Table 1. The patients and healthy controls gave their informed consent to participate in the study, which was carried out between June of 1998 and April of 2000.

Table 1.  Clinical characteristics of the study population
 Group A (obese)Group B (overweight)Group C (normal weight)p
  • NS, not significant.

  • Data are expressed as mean values ± SEM.

  • *

    χ2 test.

  • ANOVA.

  • Group C and Group B versus Group A.

n (males/females)54 (15/39)35 (10/25)57 (17/40) 
Men (%)27.728.529.8NS*
Age (years)40.3 ± 1.538.5 ± 1.837.5 ± 1.9NS
BMI (kg/m2)38.1 ± 0.927.3 ± 0.221.9 ± 0.2<0.001
Waist circumference (cm)104.6 ± 2.184.1 ± 1.673.7 ± 1.4<0.005
Waist-to-hip ratio0.86 ± 0.010.83 ± 0.010.82 ± 0.01<0.01
Sagittal diameter (cm)25.6 ± 0.520.0 ± 0.517.01 ± 0.2<0.02
Smokers (%)48.147.644.2NS*

Measurements

The study subjects underwent a physical examination and an anthropometric evaluation, which included waist circumference, waist-to-hip ratio, and abdominal sagittal diameter. With the subjects in a supine position, sagittal diameter was measured using a suitable abdominal caliper applied ∼5 cm above the umbilicus, and the distance between the lumbar plane and upper caliper bar was calculated in centimeters. Using standard laboratory methods, blood samples were drawn after an overnight 12-hour fast to determine levels of electrolytes (Na+, K+, Ca2+, and Mg2+), thyroid-stimulating hormone and thyroid hormones (freeT3 and freeT4), hemoglobin and blood cell counts, total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, creatinine, uric acid, and fibrinogen. Fasting plasma insulin was also measured by means of a radioimmunoassay commercial kit in 22 obese subjects and 22 control subjects. Information on smoking was collected by interviews.

The ECGs were analyzed by one reader (R.E.) who was unaware of the characteristics of the subject. Simultaneous 12-lead ECGs were recorded (Marquette Electronic Mac PC, Milwaukee, WI) at a paper speed of 25 mm/s and an amplitude of 10 mm/mV. The QT intervals were measured manually from the onset of the interval between Q- and S-waves of the electrocardiogram to the end of the T wave at the level of the interval between the end of the T-wave and the subsequent P-wave isoelectric baseline and corrected according to Bazett's formula

  • image

QTD was defined as the difference between the maximum and the minimum QTc across the 12-lead ECG. In the presence of a U wave, the QT interval was measured using the tangent method, by which the end of the T wave is defined at the intersection with the baseline of the tangent on the descending limb of the T wave. If the T wave could not be reliably determined, the lead was excluded from the analysis. QT interval was measured in at least 10 leads in each subject. The reader was trained to obtain the minimum of intravariability of measurements. Intraobserver variability for QTD was 5.7 ms (95% confidence interval 4.3 to 7.2), and the correlation of the measurements on separate days was significant (r = 0.94, p < 0.0001).

Statistical Analysis

Statistical analyses were performed by using the Student's t test for unpaired data, Pearson correlation coefficients, ANOVA, and the χ2 test when appropriate. Stepwise regression analysis was used to assess the relative contribution of independent variables, with the QTD as the dependent variable. The variables were adjusted for potential confounders by means of the analysis of covariance; p values of ≤0.05 were considered significant. The data are expressed as mean values ± SEM and were analyzed using the Statistix Version 4.1 statistical package of Analytical Software (Tallahassee, FL).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Table 2 lists the electrocardiographic variables. Obese and overweight patients and the normal weight healthy controls had similar QTD values: Group A, 56.4 ± 2.6 ms; Group B, 56.7 ± 2.1 ms; and Group C, 59.4 ± 2.1 ms (not significant, ANOVA). The QTD distribution is shown in Figure 1 and the scattered plots of QTD in each patient are shown in Figure 2. There was no significant difference in QTD between the males and females in the study population as a whole (57.7 ± 3.1 vs. 58.8 ± 1.7 ms, respectively), or in the three groups (ANOVA, data not shown), and no correlation between QTD and the RR interval using the univariate linear regression model (r = 0.006; not significant).

Table 2.  Electrocardiographic intervals in the study population
  • NS, not significant.

  • Data are expressed as mean values ± SEM.

  • *

    ANOVA.

  • χ2 test.

  • Student's t test.

 All (n = 146)Group A: obese (n = 54)Group B: overweight (n = 35)Group C: normal weight (n = 57)p
QTc     
(ms)410.3 ± 2.2411.8 ± 3.3407.2 ± 3.9410.3 ± 3.9NS*
(range)(340–467)(343–464)(340–443)(350–467) 
n QTc > 440 ms 3/54 (5.5%)2/35 (5.7%)8/57 (14.0%)NS
QT dispersion (ms)57.7 ± 1.356.4 ± 2.656.7 ± 2.159.4 ± 2.1NS*
Median56.055.356.758.0 
(range)(22–107)(22–107)(28–98)(23–103) 
RR     
(s)0.83 ± 0.010.83 ± 0.040.82 ± 0.020.84 ± 0.02NS*
(range)(0.56–1.25)(0.61–1.25)(0.56–1.05)(0.62–1.06) 
  Men (n = 42)Women (n = 104)  
QT dispersion (ms) 57.7 ± 3.158.8 ± 1.7 NS
image

Figure 1. Distribution of QTD in the study population (n = 146).

Download figure to PowerPoint

image

Figure 2. Scattered plots of QTD in each patient of the three groups. Means and SEM are represented by horizontal lines.

Download figure to PowerPoint

The QTc interval was comparable in the three groups: Group A, 411.8 ± 3.3 ms; Group B, 407.2 ± 3.9 ms; and Group C, 410.3 ± 3.9 ms (not significant, ANOVA). Three of the 54 obese patients (5.5%), 2 of the 35 overweight subjects (5.7%), and 8 of the 57 healthy normal weight controls (14.0%) had a prolonged QTc interval (i.e., >440 ms; not significant, χ2 test). Simple linear regression analysis revealed an association between QTc and QTD (r = 0.24, p < 0.005).

Pearson's correlation analysis of the study population as a whole showed that QTD did not correlate with age, BMI, waist circumference, abdominal sagittal diameter, or the waist-to-hip ratio (Table 3). When each group was studied separately, QTD seemed to be inversely associated with age only in the group of obese patients (Table 4). Multiple stepwise regression analysis of the study population as a whole was performed with QTD as the dependent variable; none of the independent variables selected from the Pearson analysis (including those with a p value of <0.1) entered the model (Table 5).

Table 3.  Pearson correlation coefficients of QTD with demographic and anthropometric variables in the study population
Variablesrp
Age−0.160.12
BMI−0.140.18
Waist circumference−0.180.08
Sagittal diameter−0.180.09
Waist-to-hip ratio−0.170.10
Table 4.  Pearson correlation coefficients of QTD with demographic and anthropometric variables in the obese, overweight, and normal weight subjects of the study population
 Group A: obese (n = 54)Group B: overweight (n = 35)Group C: normal weight (n = 57)
Variablesrprprp
Age−0.340.03−0.140.570.190.27
BMI−0.060.670.200.420.010.93
Waist circumference−0.190.22−0.050.830.070.68
Sagittal diameter−0.180.24−0.070.780.140.41
Waist-to-hip ratio−0.190.23−0.350.150.0050.97
Table 5.  Multiple stepwise regression analysis coefficients for the association of QTD and selected independent variables in the study population
Independent variablesrβConstantp
  1. The variables were selected from the Pearson analysis. Waist circumference and abdominal sagittal diameter were included because their p value was <0.1 (see Table 3); their p value remained unchanged even when the stepwise regression model was restricted to these two independent variables.

Age−0.16−1.5657.75850.18
BMI−0.14−1.34 0.18
Waist circumference−0.18−1.75 0.08
Sagittal diameter−0.18−1.70 0.09

There was no correlation between fasting plasma insulin levels and QTD in the obese patients and controls undergoing this test (r = −0.20 and 0.03, respectively; not significant). The values of fasting blood glucose, total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, creatinine, uric acid, fibrinogen, electrolytes, hemoglobin, blood cell counts, thyroid-stimulating hormone, and thyroid hormones were within the normal range, with no significant differences between the groups (data not shown) except for fasting plasma insulin, which was higher in the obese patients than in the normal weight subjects (15.8 ± 1.6 vs. 7.8 ± 0.8 mU/liter; p < 0.005); HDL cholesterol showed a trend toward lower values and fibrinogen showed a trend toward higher values in the obese patients.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Obesity is a condition that may be associated with early electrocardiographic and/or echocardiographic abnormalities even in the absence of clinical symptoms (17) (18); however, the sudden deaths and/or arrhythmias in obese people (2) (3) (4) are similar to those occurring in subjects suffering from a variety of heart diseases (11) (12) (13). It has been suggested that these events may be linked to abnormalities in repolarization (20) (21).

Our study considered QTD (the most simple index of ventricular repolarization variability), because although routine QTD measurement has methodological limitations and the relevance of this parameter is still questioned (15) (22) (23), a prolonged QTD is thought to be a possible risk factor for arrhythmias and sudden death (22) (24). The QT interval has been widely analyzed with conflicting results in obese people; however, QTD has only been studied in 36 Japanese obese patients (19).

Our main finding is that QTD in an obese and overweight white population without complications is comparable with that observed in healthy age- and sex-matched normal weight healthy controls. Although relating to a relatively small number of subjects, the QTD values we found were similar to those reported by some studies of normal European and American individuals (13) (23) (24) (25) (26) but slightly higher than those reported by other investigators (27) (28) (29) (30). The mean higher QTD values we observed may be due to the fact that, in the majority of our study subjects, we measured QTD in 12 leads, very rarely in 10 to 11 leads. It has been reported that QTD in 12 leads is larger than that measured in a smaller number of leads (14) (29), as performed in some studies (27) (29) (30) (31). Nevertheless, although QTD is rather difficult to measure manually, the level of reproducibility in our study was satisfactory, according to low intraobserver variability.

Our results are different from those of the Japanese study, in which the obese patients had a 36% greater QTD than the normal weight controls (19); in particular, 38% of them had a QTD of >60 ms. This discrepancy is difficult to explain. Even when without any apparent complications, obesity may be accompanied by minimal asymptomatic myocardial dysfunctions (17) (18), but the abnormalities in regional myocardial repolarization may be too slight to be recognized by the relatively rough method of manual QTD measurement. In the absence of myocardial disease and/or hypertension or other metabolic dysfunctions, the QTD pattern corresponds to the normal regional distribution of ventricular repolarization.

The difference between the two results could also be due to the fact that the obese patients in the Japanese study were free only of heart disease and diabetes; plasma electrolyte measurements were limited to potassium and calcium, no hormonal measurements were made, and the percentage of smokers was higher in the group of obese patients. It cannot be excluded that at least some of the Japanese patients had vascular, metabolic, hormonal, or partial electrolyte alterations; therefore, they may not truly represent a simply obese population. Our overweight and obese subjects were carefully chosen because they were free of cardiovascular disease, hypertension, diabetes or IGT, and hormonal and electrolyte dysfunction (including Mg2+); in addition, they were not receiving treatments that would modify the characteristics of repolarization. In addition, smokers and nonsmokers were equally distributed in the three groups. However, apart from all of these considerations, some racial differences in the two obese populations cannot be excluded.

Unlike the data published by Fei et al. (28), which relate to a limited number of normal subjects, we did not find any difference in QTD values between men and women regardless of the presence of abnormal body weight. The similarity of QTD in both sexes of a large white population has been reported recently (29). The lack of correlation we found between QTD and age in the study population as a whole is in keeping with recent studies carried out in large general populations and in healthy people (28) (29) (32).

When the study population was analyzed by means of univariate and multivariate regression models, QTD did not correlate with BMI or with the anthropometric parameters of fat distribution, such as waist circumference and abdominal sagittal diameter.

An additional finding of our study is the lack of a correlation between QTc and BMI. This result is in disagreement with some studies (4) (7) (8) (9) (33) but in accordance with others (5) (6) (19). It is worth noting that the obese populations in the majority of these studies were not homogeneous and were mainly represented by morbidly obese subjects; in addition, the presence of hypertension was not assessed. Furthermore, the presence of diabetes or IGT was excluded in only one study (7), whereas it is well known that patients with type 1 and type 2 diabetes (the latter being frequently obese) have a higher prevalence of a prolonged QTc interval (16) (34) (35) (36) (37). The number of obese patients with an abnormal QTc (>440 ms) was lower (5.5%) in our study than in other studies in which 10% to ≥20% showed a long QTc (4) (8) (9) (10). This is probably due to the fact that our obese patients were selected because they were without any complications, as indicated by their QTDs, which were comparable with those of controls.

We did not find any relationship between QTD and fasting insulin plasma levels in the obese and normal subjects undergoing this test. Watanabe et al. (38) have found recently that QTD significantly correlated with plasma insulin concentrations measured during the oral glucose tolerance test in 20 healthy adult normoinsulinemic volunteers, regardless of any variations in plasma glucose. We cannot compare our findings with those of Watanabe et al. (38) because we only analyzed fasting insulin concentrations.

In conclusion, QTD seems to be within the normal range and seems comparable with that of age- and sex-matched healthy normal weight controls in a population of obese and overweight patients that have a wide range of BMIs but are without any cardiovascular, hormonal, metabolic, or electrolyte complications and are without pharmacological treatments. In addition, QTD does not depend on BMI but is associated with the QTc interval. Further studies are needed to investigate QTD in complicated obesity, especially in relation to echocardiographic parameters.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This research was supported by a 60% grant from the Italian Ministry for the Universities and Scientific and Technical Research. We thank Prof. L. De Ambroggi (Department of Cardiology) for criticisms and suggestions.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References