Differences in QT interval determined from multi-lead ambulatory ECG during rest and physical exercise

Exercise and cardiac regulation: what can electrocardiographic time series tell us?

Electrocardiographic (ECG) monitoring allows temporal analysis of cardiac rhythm. We are usually interested in the variability of two components of the ECG: RR interval (a surrogate marker of cardiac interval) and QT interval (the duration of ventricular depolarization/repolarization). Quantification of RR rhythm, called heart rate variability (HRV) analysis, reflects the cardiac influences of the autonomic nervous system. QT variability provides insight regarding the risk of ventricular arrhythmia, and is at least partially independent of HRV. In this review, we consider the analysis of ECG time series during physical exercise. Our objectives were to show the variety of methods that can be used to characterize these time series data and to demonstrate "normal" changes in these characteristics during exercise and recovery. Attaining a comprehensive understanding of cardiac electrical conduction changes during exercise is not straightforward: analysis methods are numerous and results require careful interpretation. However, we recommend that assessment of both HRV and QT properties yields the most valuable information, the utility of which can be appreciated from the viewpoints of the athlete (e.g. providing feedback on the cardiac effects of training), the clinician (assisting in cardiovascular screening and exercise therapy evaluation) and the exercise physiologist (providing data for physiological process modelling).

Influence of smoking abstinence and nicotine replacement therapy on heart rate and QT time-series

SUMMARY

Many smokers attempt to quit without using nicotine replacement therapy (NRT) or pharmacotherapy, i.e. 'cold-turkey'. The cardiac implications of this are important but are incompletely understood. Previous studies have associated smoking cessation with improvements in heart rate (HR) and its variability, but its influence on QT time-series is unclear. Furthermore, the relative influence on these parameters of acute nicotine withdrawal and of NRT has not been adequately compared. Additional insight might come from analysing the dynamic (e.g. fractal) properties of electrocardiographic data during different levels of nicotine exposure. We examined the influence of smoking cessation, during cold-turkey and subsequent NRT, on HR and QT time-series during 30 days of smoking abstinence. Seven smokers and sixteen healthy non-smokers received ECG monitoring at baseline (Day 0). Smokers subsequently refrained from smoking without using NRT for 24 h, and then received NRT for 29 days. ECG monitoring was repeated at Days 1, 7, 30. Following smoking cessation we observed that: HR and rate-corrected QT were both reduced, heart rate variability (HRV) increased (improved), and QT variability index (QTVI) showed signs of improvement (trend only). Improvements in HR and QT were maintained throughout NRT use, whilst improvements in HRV and QTVI were sustained for at least the early stages of NRT. The dynamic (multifractal) properties of HR and QT were similar for smokers and non-smokers, and were unchanged by smoking abstinence or NRT. Our results provide tentative evidence that electrocardiographic improvements during a cold-turkey smoking quit attempt (acute nicotine withdrawal) are maintained during NRT pharmacotherapy.

Influence of bariatric surgery on indices of cardiac autonomic control

BACKGROUND

Obesity is associated with reduced heart rate variability (HRV), reflecting detrimental changes in cardiac regulation by the autonomic nervous system (ANS). Weight loss reverses this change and ANS dysfunction is thought to have a role in obesity-related cardiac pathology. Few studies have examined the influence of weight-reduction (bariatric) surgery on cardiac autonomic control. This study therefore sought to assess longitudinal changes in indices of cardiac autonomic control following two types of bariatric procedure, laparascopic gastric banding (LGB) and biliopancreatic diversion (BPD).

METHODS

Eleven morbidly obese subjects aged 47.8 +/- 7.9 years (mean+/-SD) with BMI 48.2 +/- 6.9 kg x m(-2) underwent weight-reduction surgery: five received BPD and six received LGB. Holter ECG was recorded and HRV was quantified together with a QT variability index (QTVI), a complexity index (SampEn), and a fractal (scaling) index (DFAalpha). Repeated measures ANOVA compared the indices for the two groups as a function of time (1, 6 and 12 months follow-up).

RESULTS

BMI was reduced by up to 24% (p=0.008) post-surgery despite patients remaining obese at one-year follow-up. Several indices showed prompt and persistent improvement with progressive weight loss, QTVI being the most sensitive discriminator of recovery time (F(3,216)=16.86; p<0.0005; eta(2)=0.190). Autonomic responsiveness was functionally normal throughout. The bariatric procedures induced similar changes in cardiac autonomic control, despite their differing mechanisms of action.

CONCLUSIONS

This pilot study suggests that the mechanism responsible for improving cardiac regulation following bariatric surgery might be the weight loss itself. Furthermore, post-surgery improvement in QTVI implies that weight loss reduces the risk of ventricular arrhythmic events.

Sample entropy of electrocardiographic RR and QT time-series data during rest and exercise

Sample entropy (SampEn) is a measure of the complexity of data. Few studies have compared the SampEn of electrocardiographic cardiac interval (RR) data (SampEn-RR) during differing physiological states, and none have examined SampEn for the corresponding QT interval (SampEn-QT). The aim of this study was to quantify SampEn-RR and SampEn-QT during rest and for a range of exercise workloads. Specific objectives were to assess both the utility of SampEn for discriminating between physiological states and the relationship of SampEn-RR with traditional measures of heart rate variability (HRV). Twelve males of similar age, mass and aerobic fitness participated. A three-lead ECG was recorded continuously during pre-exercise, progressive bicycle exercise and recovery, and beat-to-beat RR and QT intervals were quantified for sinus beats. SampEn and HRV were calculated within consecutive 1 min periods throughout. Consistent estimation of SampEn-RR and SampEn-QT was possible with an appropriate choice of SampEn parameters. SampEn-RR was sensitive to differing physiological conditions, but its discriminating ability was poorer than that of linear HRV indices. SampEn-RR was also negatively correlated with normalized LF and LF/HF parameters. We interpret changes in SampEn for RR and QT data in terms of the altered autonomic nervous system (ANS) control of either the atrial or the ventricular myocardium (or both) during discrete physiological states. We speculate that greater complexity in QT data might be explained by a direct ANS influence on the ventricular myocardium.