Heart rate turbulence and heart rate variability in patients with mitral valve prolapse

Abstract

Aims Heart rate turbulence (HRT) and heart rate variability (HRV) have been shown to be independent and powerful predictors of mortality in a specific group of cardiac patients. However, the predictive valsue of HRV alone is modest and information on HRV in patients with mitral valve prolapse (MVP) has so far been conflicting. In addition, no studies have previously evaluated HRT in patients with MVP. To define better the effects of MVP on cardiac autonomic function, we assessed HRT and time-domain parameters of HRV in patients with MVP.

Methods and results Fifty patients with MVP and 70 controls without MVP were investigated. The diagnosis of MVP was confirmed by cross-sectional echocardiography in the parasternal long-axis view and apical 4-chamber view. The HRV and turbulence analysis were assessed from a 24-hour Holter recording. When HRT parameters were compared, the values of the HRT onset and slope were significantly lower in MVP patients than in the controls group (−0.109±0.207 vs. −0.289±0.170%, P=0.001 and 8.6±7.2 vs. 11.5±7.4 ms/RRI, P=0.043, respectively) and the number of patients who had abnormal HRT onset was significantly higher in the MVP group than in controls (15 vs. 8, P=0.011). In addition, HRV parameters were not statistically different between the two groups.

Conclusion Although we found that the decrease in HRV parameters was not significantly different between MVP patients and controls, HRT variables (especially HRT onset) were significantly lower in MVP patients. Therefore, in our opinion, HRT is an attractive, easily applicable, and better way of non-invasive risk prediction compared with another non-invasive risk predictor, HRV.

Introduction

Mitral valve prolapse (MVP) is mostly a benign condition; it is an uncommon cause of sudden cardiac death (SCD), but its propensity to cause sudden death has been of major concern mainly because of the prevalence of this abnormality in the general population.^1,2 It has been reported that MVP patients, especially when symptomatic, have sympathetic predominance and diminished cardiac vagal tone. The decreased cardiac vagal activity in these patients is linked to the pathogenesis of the disease and the sequence of emerging symptoms.³

Analysis of heart rate variability (HRV) has been used to assess autonomic function and/or to quantify risk in a wide variety of both cardiac and non-cardiac disorders.⁴ Decreased HRV is a reflection of enhanced sympathetic overdrive and decreased vagal activity, which has a strong association with the pathogenesis of ventricular arrhythmias and SCD in the general population, and especially in cardiac patients.⁵ However, the predictive value of HRV alone is modest⁵ and information on HRV in patients with MVP has so far been conflicting.^6–10 Recently, heart rate turbulence (HRT), which reflects a response of heart rate to a premature ventricular beat, has been introduced as another non-invasive tool for risk stratification. The disappearance of HRT implies the loss of normal autonomic nervous regulation.¹¹ Heart rate turbulence has been shown to be an independent and powerful predictor of mortality, with greater predictive power than HRV.¹²

No studies have previously evaluated HRT in patients with MVP. To define better the effects of MVP on cardiac autonomic function, we assessed HRT and time-domain parameters of HRV in these patients and compared the findings with a control group.

Methods

Patients

This study was performed at a single centre. All examinations were performed in İzzet Baysal University Hospital from June 2004 to March 2005. A total of 120 cases were investigated, including 50 patients who had been diagnosed as having MVP after evaluation by transthoracic echocardiography and 70 controls without evidence of structural heart diseases.

Patients who had disorders such as ischaemic or rheumatic heart disease, diabetes mellitus, thyroid disorders, stenotic valvular heart disease, severe left ventricular dysfunction, patients taking drugs affecting cardio-respiratory responses or alcohol were excluded. All subjects were asked to avoid smoking and drinking any caffeinated beverages during the study. In addition, all patients with MVP were questioned about the presence of symptoms such as atypical chest pain, dyspnoea, palpitations, syncope, dizziness, fatigue, migraine headaches and anxiety, depression or panic attacks.

The protocol was in agreement with the ethical guidelines of the Research Committee of Abant İzzet Baysal University and the relevant standards of the revised Declaration of Helsinki (1983). All participants gave informed consent.

Echocardiography

An experienced echocardiographer assessed the echocardiographic studies. In all subjects, transthoracic two-dimensional and Doppler echocardiographic examinations were performed, using a Vingmed Vivid 3 echocardiographic system (General Electric Vingmed Ultrasound, Israel), using 2.5–3.5 MHz transducers. The measurements were carried out according to the standards of the American Society of Echocardiography, using the parasternal long axis and apical 4-chamber windows.

The diagnosis of MVP was confirmed by cross-sectional echocardiography that showed definite prolapse (posterior/superior displacement) of anterior and/or posterior leaflet of the mitral valve beyond the plane of mitral annulus in the parasternal long-axis view and apical 4-chamber view. Prolapse of a leaflet in the apical 4-chamber view alone was excluded because of frequent false-positive diagnoses due to non-planar or saddle-shaped mitral annulus.

On the basis of prior clinical and prognostic studies, patients with MVP were divided into two groups: classic and non-classic prolapse. Classic prolapse was defined as superior displacement of the mitral leaflets, by more than 2 mm during systole, with maximal leaflet thickness of at least 5 mm during diastole. Non-classic prolapse was defined as displacement by more than 2 mm, with a maximal thickness of less than 5 mm.^13,14

The degree of mitral regurgitation (MR) was assessed as maximal regurgitant jet area/LA area ratio in the parasternal, apical long-axis, and apical four-chamber views. None-trace, mild, moderate, and severe MR were classified on the basis of jet area/LA area ratios of 0–10%, >10–20%, >20–40%, and >40%.¹⁵

Holter analysis

All patients and controls underwent 24-hour Holter monitoring. All patients were in sinus rhythm throughout the recording period. Holter ECGs were analysed using the Del Mar Reynolds Pathfinder Holter system. The first author, blinded to the diagnosis of the patients, conducted the analyses of Holter ECGs.

The HRV analysis was assessed over a 24-hour period and was performed in time domains according to European Society of Cardiology/North American Society of Pacing and Electrophysiology guidelines.⁵ The following time-domain parameters were calculated: mean of all normal RR intervals (mean RR); standard deviations of all NN intervals (SDNN); mean of the standard deviations of all NN intervals for all 5-minute segments of the entire recording (SDNNI); standard deviation of the averages of NN intervals in all 5-minute segments of the entire recording. (SDANN); the square root of the mean of the sum of the squares of differences between adjacent NN intervals (rMSSD); count of the total number of differences between adjacent RR intervals that were greater than 50 ms (sNN50 total); the number of pairs of adjacent NN intervals differing by more than 50 ms divided by the total number of all NN intervals (pNN50).

Heart rate turbulence parameters were calculated according to the original method reported by Schmidt et al.^11,12 Two numerical descriptors were estimated: turbulence onset reflecting the initial phase of sinus rhythm acceleration and turbulence slope describing deceleration phase. Heart rate turbulence onset was defined as the difference between the mean of the first two sinus-rhythm RR intervals following the compensatory pause after a premature ventricular complex (PVC) and the mean of the last two sinus-rhythm RR intervals preceding the PVC, expressed as a percentage of the former. The HRT slope was defined as the maximum positive slope of a regression line assessed over any sequence of five subsequent sinus-rhythm RR intervals within the first 20 sinus-rhythm intervals after a PVC, expressed as ms/beat. The HRT onset or slope was defined as abnormal if the onset was ≥0% or the slope was ≤2.5 ms/beat.

Patients with atrial fibrillation or without PVCs, PVCs preceded by noise, or artefacts were excluded from HRV and HRT analysis.

Statistical analysis

Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS, Chicago, IL, USA), version 11.0 software for Windows. Descriptive statistics were made and all data were expressed as mean±standard deviation and % ratio. The quantitative values between two groups were compared using Student's t-test, and the qualitative values were compared using the χ² test. In addition, the significance of difference quantitative values of more than two groups was estimated by means of one-way analysis of variance (ANOVA) with Scheffe post-hoc test. P value of <0.05 was considered as statistically significant in all cases.

Results

Fifty patients (22 men and 28 women, mean age 41±13 years) with MVP and 70 healthy controls without MVP (33 men and 37 women, mean age 43±9 years), 120 cases in total were included in this study. Age, sex, smoking, presence of hypertension, prevalence of PVCs, mean heart rate, left ventricle ejection fraction, and left atrium dimension were not significantly different between two groups (Table 1).

The values of the HRT onset and HRT slope were significantly lower in MVP patients than the control group (−0.109±0.207 vs. −0.289±0.170%, P=0.001 and 8.6±7.2 vs. 11.5±7.4 ms/RRI, P=0.043, respectively). When dichotomizing patients according to abnormal values (HRT onset >0% and HRT slope <2.5 ms/RR) as proposed by Schmidt et al.,¹² the number of patients who had abnormal HRT onset was significantly higher in the MVP group (15 vs. 8, P=0.011), but the number of subjects who had abnormal HRT slope was not different in the two groups (9 vs. 6, P=0.124). None of the HRV parameters were statistically different between the two groups. Table 2 summarizes the HRT and HRV parameters in the MVP and control patients.

Using current definitions and quantitative analysis of leaflet displacement and thickness in patients with MVP, 14 subjects were identified as having classic MVP, 36 as non-classic MVP, with the remaining 70 subjects acting as the control group. The values of the HRT onset were significantly lower in patients with classic MVP than the patients with non-classic MVP and the control group. The difference between the patients with non-classic MVP and the control group did not reach significance. When compared with the controls, more patients had abnormal HRT onset in the classic MVP group, but this difference was also not statistically significant between the classic and non-classic MVP groups. The values of the HRT slope were significantly lower in patients with classic MVP than the control group, which was also not significantly different between the subgroups of MVP patients. Furthermore, the prevalence of patients with abnormal HRT slope was not statistically different among the three groups. When HRV parameters were compared among the three groups, these parameters were not statistically different (Table 3).

In patients with MVP, age, gender, prevalence of PVCs, mean heart rate, left ventricular ejection fraction, left atrial dimension, and maximal leaflet displacement were not different between the classic and non-classic groups (Table 4). However, MR was more prevalent in the classic compared with the non-classic group. In all groups, none-trace and mild MR predominated (three patients in the classic MVP group had none-trace MR, six patients had mild, four patients had moderate, and one patient had severe MR, whereas in non-classic MVP group 24 patients had none-trace MR, 7 patients had mild, and 5 patients had moderate MR). In addition, the prevalence of symptoms was not statistically different between the two groups. Table 4 summarizes these findings in classic and non-classic MVP patients.

When HRV parameters were compared according to the presence of MR in MVP patients, no statistical difference was detected between the two groups (Table 5).

Discussion

Mitral valve prolapse is the most commonly diagnosed valvular heart disease, especially in the young, and according to the Framingham study, it affects up to 2.4% of the general population.^16,17 Mitral valve prolapse has a benign course and excellent prognosis for most patients.¹⁸ However, a minority of patients develops serious complications, including infective endocarditis, cerebrovascular events, progressive severe MR, and SCD.¹⁹ Sudden cardiac death is a significant complication in MVP cases; a 0.5% incidence of sudden death has been reported in different MVP studies, and it is suggested that the underlying mechanism of sudden death is arrhythmic.^3,20–22

Clinical and biochemical studies have indicated that the patients with MVP have autonomic dysfunction.^23–26 Non-invasive measures of cardiovascular responses in patients with MVP also suggested decreased parasympathetic activity in addition to increased ά and normal β adrenergic tone and responsiveness.²⁶ However, other studies have shown no evidence of abnormal autonomic or neuro-endocrine function, either at rest or during tilt testing in MVP patients.^27,28

Heart rate variability analysis has been used as a predic-tor of SCD or as a marker of the progression of cardiovas-cular disease in several high-risk populations, and diminished HRV is associated with increased sympathetic and decreased vagal modulation^5,29 in some studies. Analyses of HRV showed that patients with MVP have signifi-cantly decreased time-domain and frequency-domain indices when compared with age-matched controls.^6–8 However, several studies reported that HRV parameters are not changed in MVP patients.^9,10 In the present study, we could not find significant differences in HRV variables between patients and controls. Multiple studies have shown that leaflet thickening (greater than 5 mm, which defines classical prolapse) and presence of MR are associated with an increase risk of serious sequelae including sudden death, infective endocarditis, progressive valve dysfunction, or cerebral embolic events.^30–33 We also found that mean HRV variables were shorter in patients in the classical MVP group and patients in the MR than non-classic MVP and control groups and those without MR, which did not reach statistical significance.

If we had only investigated the HRV variables in MVP patients, we would have concluded that cardiac autonomic function was not altered in MVP patients. However, the prognostic value of these indices in patients with MVP remains unclear and conflicting. Moreover, as a univariate predictor, HRV has low sensitivity and low positive predictive accuracy.⁵

In 1999, Schmidt et al. introduced into electrocardiology a new non-invasive sudden death risk factor called heart rate turbulence, which is a physiological, biphasic reaction with early acceleration and subsequent deceleration of sinus rhythm in response to a premature ventricular complex. Heart rate turbulence is believed to reflect baroreflex sensitivity. It was shown in retrospective and prospective studies that HRT is an independent and powerful predictor of mortality after myocardial infarction, with greater predictive power than HRV.^11,12,34 But, no study has been published yet in the English literature, which investigated HRT variables in MVP patients.

In the present study, HRT onset, HRT slope, and the number of patients who had abnormal HRT were significantly different in MVP patients from controls, although the HRV parameters were similar in MVP patients and controls. Also, mean HRT onset and HRT slope were significantly lower and the number of patients with abnormal HRT onset was significantly higher, when the patients in the classic MVP subgroup were compared with controls. However, HRT variables of non-classical MVP patients were not significantly different from controls (Table 4).

As is well known, the most common complication of MVP is MR, and patients with MVP, complicated by severe mitral regurgitation, may have an increased risk of sudden death.^35–37 We also could not detect any significant differences between HRT variables in the MVP patients who were grouped according to the presence of MR (Table 5), which might be partially explained by most patients having mild MR (56%).

Conclusion

Although we found that HRV parameters were not significantly decreased between MVP patients and controls, HRT variables (especially HRT onset) were significantly decreased in MVP patients. Therefore, in our opinion, HRT is a simple and elegant way of measuring cardiac autonomic function in MVP patients, and measuring HRT is an attractive, easily applicable way of non-invasive risk prediction compared with HRV.

Limitations of the study

This study is a single centre, observational, and non-randomized comparative study; the number of cases included in the study is relatively low and the authors suggest confirmation of the findings and evaluation of the usefulness of HRT in risk stratification of patients in a larger number of MVP patients. Mitral regurgitation was mild (56%) in the majority, which might lead to the insignificant results when compared with patients who did not have MR, as MVP complications were mostly related to the severity of MR.

Heart rate turbulence cannot be measured in subjects without PVCs, as a minimum of 15–20 sinus beats after each PVC and 3–5 beats before the PVC are required for accurate calculation of HRT. Thus, we had to exclude the patients, who did not have any PVCs, from the study.

Because our Holter device (Del Mar Reynolds Pathfinder Holter system) analyses only time-domain parameters, we could not measure frequency domain parameters of HRV. But, it has been stated that each of the 24-hour frequency domain spectral measures has an equivalent time-domain variable, which is highly correlated with it, because both are influenced by the same physiological inputs and because of mathematical relationships.^4,5 In addition, we did not examine circadian variation in heart rate and HRV in our study.

References