Article Text

Download PDFPDF

Beneficial effects of catheter ablation on left ventricular and right ventricular function in patients with frequent premature ventricular contractions and preserved ejection fraction
  1. Adrianus P Wijnmaalen,
  2. Victoria Delgado,
  3. Martin J Schalij,
  4. Carine F B van Huls van Taxis,
  5. Eduard R Holman,
  6. Jeroen J Bax,
  7. Katja Zeppenfeld
  1. Department of Cardiology, Leiden University Medical Center, Leiden, The Netherlands
  1. Correspondence to Dr K Zeppenfeld, Leiden University Medical Center, Postbus 9600, 2300 RC Leiden, The Netherlands; k.zeppenfeld{at}


Background Improvement of left ventricular ejection fraction (LVEF) after radiofrequency catheter ablation (RFCA) of frequent premature ventricular contractions (PVCs) has been reported. However, most patients with frequent PVCs have a normal LVEF. In these patients subtle and early forms of PVC-induced left and right ventricular (RV) impairment may not be detected by standard echocardiographic techniques.

Objective To assess the effect of frequent PVCs on ventricular function in patients with preserved LVEF.

Methods 49 patients (30 male, 49±16 years) with recent-onset (median 1.2 years), frequent PVCs (burden 26±13%) and 25 healthy controls were studied. Thirty-four patients with PVCs underwent successful RFCA. Two-dimensional echocardiography was performed at baseline and follow-up. LV volumes and LVEF were calculated by Simpson's rule. Tricuspid annulus plane systolic excursion and fractional area change were calculated to assess RV function. Multidirectional LV strain (radial, circumferential, longitudinal) and RV free-wall longitudinal strain were calculated by two-dimensional speckle tracking imaging. At baseline LVEF, volumes and RV dimensions were normal in patients and controls.

Results Speckle tracking imaging demonstrated reduced LV and RV strain in patients with PVC as compared with controls. At follow-up there were no changes in LVEF, LV volumes and RV dimensions and function in patients successfully treated by RFCA and untreated patients. However, radial, circumferential and longitudinal strain improved significantly in patients after RFCA but remained unchanged in untreated patients.

Conclusions Frequent PVCs can induce subtle cardiac dysfunction detected by speckle tracking imaging analysis in patients without apparent cardiomyopathy. RFCA can successfully eliminate PVCs and improve cardiac function.

  • Premature ventricular contraction
  • ventricular dysfunction catheter ablation
  • radiofrequency ablation (RFA)
  • ventricular tachycardia
  • echocardiography (three-dimensional)
View Full Text

Statistics from


Symptomatic ventricular ectopy is frequently encountered in clinical practice. The arrhythmia occurs commonly in patients without structural heart disease and is considered to be without prognostic significance.1

In some patients, frequent idiopathic premature ventricular contractions (PVCs) can cause moderate to severe left ventricular (LV) dysfunction.2–4 Elimination of these PVCs by radiofrequency catheter ablation (RFCA) may improve LV ejection fraction (LVEF).2–4 The majority of patients presenting with frequent symptomatic PVCs have normal LV volumes and LVEF. It has been shown, however, that impairment of LV function may evolve after several years of follow-up.5 Small series demonstrated a reduction of LV dimension after successful RFCA in patients with normal LV function using conventional echocardiography assessment.6 7 This finding suggests that frequent PVCs may already have an early subclinical detrimental effect on LV function which may not be unmasked with conventional echocardiography. Whether frequent PVCs have an adverse effect on right ventricular (RV) function is unknown.

Recently, two-dimensional (2D) speckle tracking strain imaging has been introduced.8 9 This novel technique enables accurate detection of subtle abnormalities in ventricular function that are not revealed by conventional echocardiographic parameters such as LVEF and RV fractional area change.10 In addition, 2D speckle tracking strain imaging allows angle-independent evaluation of multidirectional LV strain in radial, circumferential and longitudinal directions and RV longitudinal strain.11

The purpose of this study was to assess the effect of recent-onset, frequent PVCs on RV and LV function in patients with preserved LVEF using 2D speckle tracking strain imaging. In addition, the effect of successful RFCA of frequent PVCs on LV and RV function was evaluated.


Study population and study protocol

Forty-nine consecutive patients with symptomatic, frequent PVCs (>5% PVC on 24 h Holter monitoring) were included. All patients underwent extensive baseline evaluation to rule out structural heart disease, including a clinical history with regard to the onset of symptoms, 12-lead electrocardiogram (ECG), exercise testing, 24 h Holter monitoring and 2D transthoracic echocardiography.

The 12-lead ECG QRS duration, morphology and PVC coupling interval were evaluated. The PVC burden was calculated by dividing the number of PVCs by the total number of beats on Holter recording.

Echocardiographic reference values were obtained from a group of 25 people with structurally normal hearts. The controls were frequency matched to patients by age, gender, body surface area and LV function. The control group included individuals who were referred for evaluation of chest pain and who had normal echocardiograms. Those individuals referred for echocardiographic evaluation of known valvular heart disease, murmur, syncope, arrhythmias, congestive heart failure or cardiac transplantation were excluded.

Radiofrequency catheter ablation

After obtaining informed consent, catheter mapping and ablation was performed in the post-absorptive, non-sedated state. Antiarrhythmic drugs were discontinued for five half-lives, with the exception of amiodarone.

Mapping was facilitated by an electroanatomical mapping system (CARTO) using a transvenous or retrograde aortic approach. At the site of earliest activation, based on the onset of bipolar electrograms with a local unipolar QS pattern, pace-mapping was performed to confirm a ≥11/12 lead QRS pace-match.

Radiofrequency energy was delivered at the site of earliest activation and the best pace-map with the target temperature set at 60°C and maximum power output of 50 W. After ablation, programmed electrical stimulation was performed before and during the administration of isoproterenol (2–10 μ/min) to confirm that PVCs were not inducible by adrenergic stimulation. Complete procedural success was defined as the absence of spontaneous or inducible PVC for at least 45 min after ablation.


Two-dimensional echocardiography was performed using a commercially available system (Vivid-Seven, General Electric Vingmed, Horten, Norway) equipped with a 3.5 MHz transducer. Data acquisition was performed in the left lateral decubitus position at a depth of 16 cm in the parasternal and apical views. Standard M-mode, grey-scale 2D and colour-Doppler images, triggered to the QRS complex, were acquired and saved in cine-loop format for offline analysis (EchoPAC 7.0.0, GE Medical Systems, Horten, Norway). All echocardiographic measurements were performed on sinus beats, avoiding the first post-extrasystolic beat if feasible.

Left ventricular end-diastolic and end-systolic volumes were measured at the apical two- and four-chamber views by modified Simpson's rule and LVEF was derived.12 RV dimensions were measured according to current guidelines.12 From the apical four-chamber view, the mid-cavity and basal RV diameters as well as the RV longitudinal diameter were measured at end diastole. From the parasternal short-axis images, the end-diastolic RV outflow tract diameter was measured proximal to the pulmonary valve. RV systolic function was evaluated by calculating the fractional area change and measuring the tricuspid annular plane systolic excursion (TAPSE) index from the apical four-chamber view.12

Two-dimensional speckle tracking imaging analysis

Multidirectional analysis of LV strain (radial, circumferential and longitudinal directions) was performed using 2D speckle tracking imaging. Standard grey-scale 2D images were acquired at a high frame rate to assure adequate tracking of the speckles equally distributed within the myocardium. Myocardial strain can be calculated by measuring the change of the position of the speckles within a myocardial segment along the cardiac cycle.13 14

Applied to the short axis of the LV, radial strain assesses the thickening and thinning of the myocardial wall, whereas circumferential strain assesses the shortening and lengthening of the myocardium along the curvature of the LV wall. The mid-ventricular short axis of the LV is divided into six segments and the values of radial or circumferential strain are derived from the average of the six segmental peak systolic strain values (figure 1A and B).

Figure 1

Multidirectional strain analysis with two-dimensional speckle tracking imaging. In the left ventricle, myocardial strain can be measured in the three directions with two-dimensional speckle tracking imaging. Radial (A) and circumferential (B) strain are calculated at the mid-ventricular short-axis view of the left ventricle. The left ventricle is divided into six segments and the segmental peak systolic strain values with the corresponding time–strain curves along the cardiac cycle are obtained. Longitudinal strain of the left ventricle is calculated at the four-chamber, two-chamber and long-axis apical views (C). In each apical view, the left ventricle is divided into six segments, obtaining the segmental peak systolic strain values and the corresponding time–strain curves. Longitudinal strain of the right ventricular free wall (D) is measured at modified four-chamber view, focused on the right ventricle. The free wall is divided into three segments, providing the segmental peak systolic strain value and the time–strain curves along the cardiac cycle.

Applied to LV apical images (two-, four-chamber and long-axis views), longitudinal strain assesses the shortening of the LV resulting from the excursion of the mitral annulus plane towards the LV apex. Each LV apical view is divided into six segments and the global longitudinal strain value is derived from the average of the 18 segmental peak systolic strain values (figure 1C).

Myocardial deformation of the RV was assessed in the longitudinal direction.15 From a modified apical four-chamber view, focused on the right ventricle, longitudinal strain of the free wall was evaluated. The RV free wall is divided into three segments and the global longitudinal strain value is obtained from the average of the three segmental peak systolic strain values (figure 1D).

All echocardiographic analyses were performed by an independent observer blinded to the clinical history and the treatment. The interobserver agreement for 2D speckle tracking strain measurements has been previously reported,16 17 being 6.5±5.4 for radial strain, 2.3±2.4 for circumferential strain, 0.9±1.0 for longitudinal strain and 0.16±3.6% for RV longitudinal strain.

Statistical analysis

Continuous variables are expressed as mean±SD. Categorical variables are expressed as frequencies (%). Comparisons between controls and patients were performed by Student t test for unpaired data. Patients treated and not treated with RFCA were compared using Mann–Whitney U test for unpaired data. Changes in LV and RV dimensions and function at follow-up were compared with the Student t test and Wilcoxon signed ranks test for paired data, as appropriate. A p value <0.05 was considered statistically significant. All statistical analyses were performed with SPSS software (version 16.0, SPSS Inc).


Patient characteristics

Baseline characteristics of the 49 patients (30 men, mean age 49±16 years) are summarised in table 1. The median duration of symptoms before the first evaluation was 1.2 years (IQR 0.4–3 years). In patients who underwent RFCA, the 24 h Holter monitoring was performed 3.0 months (IQR 1.0–5.6) before the RFCA procedure.

Table 1

Baseline characteristics

The burden of PVCs was 26±13% of total beats. The PVCs had one dominant morphology in all patients. The mean QRS duration of PVC was 175±24 ms. The frontal plane axis was left inferior in 24 (49%), left superior in 1 (2%), right inferior in 23 (47%) and right superior in 1 (2%) patient. The average coupling interval was 498±115 ms. No evidence of ischaemia was found on exercise testing in any of the patients.

The control group included 25 people with normal echocardiograms. By definition, controls were comparable with patients for gender (14 (56%) men, p=0.665), age (46±9 years, p=0.333), body surface area (1.8±0.1 m2, p=0.080) and LVEF (58±5%, p=0.201).

Mapping and radiofrequency catheter ablation

Forty patients were scheduled for mapping and ablation; in nine no RFCA was attempted owing to patients' preference. Thirty-four patients could successfully be ablated with a median of three radiofrequency applications (range 1–14). The PVCs originated from the RV in 24 (71%) and the LV in 10 (29%) patients. The site of successful radiofrequency application was the RV outflow tract in 23, the pulmonary artery in one, the LV outflow tract in two, the mitral annulus in one, the anterior cardiac vein in three, the aortic cusps in three and the Purkinje system in one patient. In two of these patients two different morphologies could be successfully targeted in adjacent areas (right ventricular outflow tract in one and aortic cusps in one). In five of six patients with ablation failure, the site of origin could not be determined by an endocardial approach including mapping within the cardiac veins (>1 PVC morphology in two) and in one patient no ablation attempt was made owing to the proximity of the His bundle. Procedure and fluoroscopy times were 150±60 min and 17±15 min, respectively. In the successfully ablated patients the burden of PVCs was reduced to 0.2±0.8%, confirming freedom of PVCs at repeated Holter monitoring 4.1 months (interquartile range 1.4–9.2 months) after RFCA. In contrast, in the 15 patients in whom no mapping and/or ablation attempt was made the PVC burden was stable at follow-up. No complications were seen.

Baseline echocardiography and 2D speckle tracking strain analysis

At baseline, all patients had a non-dilated left ventricle with preserved LVEF (56±7%) and no differences were seen in comparison with controls (table 2). In addition, patients had comparable RV dimensions and function to controls (fractional area change 42±8% and TAPSE index 2.4±0.5 mm) (table 2).

Table 2

Baseline echocardiography and two-dimensional speckle tracking strain analysis

Multidirectional 2D speckle tracking assessment of the left ventricle demonstrated that patients had significantly impaired LV strain in all the three directions (radial, circumferential and longitudinal). Furthermore, patients had lower values of RV longitudinal strain than controls (table 2). In seven patients, 2D speckle tracking analysis was performed during sinusal post-extrasystolic beat. However, this had no significant impact on the values of multidirectional strain of the overall population.

Effects of PVC origin

To evaluate the impact of PVC origin on RV and LV function, patients with PVCs originating from the right ventricle confirmed by successful RFCA (n=24) were compared with patients with LV origin (n=10). No differences were seen in LVEF, LV end-systolic and diastolic function, RV fractional area change and TAPSE. Myocardial strain was decreased in all directions to a similar extent in patients with LV and RV site of origin: LV circumferential strain (−16.3±4.0% vs −15.6±3.3%, p=0.5), LV longitudinal strain (−18.3±3.1% vs −17.1±2.3%, p=0.7), LV radial strain (30.1±13.8% vs 33.1±15.6%, p=0.2) and RV longitudinal strain (−25.8±8.4% vs −22.1±2.3%, p=0.07).

PVC burden and echocardiographic findings

To account for the effect of PVC burden the 30 (61%) patients with a PVC burden of >20% (34±9.7%) were compared with 19 (39%) patients with a PVC burden <20% (13.3±4.7%). Patients with a high PVC burden tended to have a lower LVEF (55±8% vs 58±6%, p=0.07) and larger LV end-diastolic volume (130±37 ml vs 113±31 ml, p=0.13) and LV end-systolic volume (59±20 ml vs 49±17 ml, p=0.09) than patients with a moderate PVC burden, but these differences were not significant. Both groups showed comparable values of LV circumferential strain (−16.3±4.4% vs −15.8±4.0%, p=0.9), radial strain (−31.9±14.2% vs −32.1±16.0%, p=1.0), longitudinal strain (−17.1±3.5% vs −17.9±2.2%, p=0.9) and RV longitudinal strain (−23.3±6.0% vs −25.2±8.3, p=0.3).

Effect of RFCA on LV and RV function

After a median follow-up of 13 months (IQR 5–22 months), a second echocardiogram was obtained in all patients. At follow-up, the 34 patients successfully treated with RFCA showed a significant decrease in LV end-systolic volume (from 56±21 ml to 49±15 ml, p=0.018). Changes in LV end-diastolic volumes and LVEF (from 57±8 ml to 59±5%, p=0.245) were not significant. However, with 2D speckle tracking imaging, a significant improvement of strain values was seen in all the three directions, with normalisation of LV multidirectional myocardial strain. In addition, although there were no significant changes in RV dimensions and function, RV longitudinal strain also improved at follow-up, with normalisation of the longitudinal strain values in successfully treated patients. (table 3)

Table 3

Follow-up echocardiography study and two-dimensional speckle tracking strain analysis

In contrast, the group of 15 patients in whom RFCA could not successfully be performed or no RFCA attempt was made the PVC burden was stable and no changes in LV dimensions and function, assessed either by LVEF or multidirectional strain were found. Furthermore, no changes in RV dimensions, function and deformation properties were seen (table 3).


The major finding of this study is that patients with frequent PVCs of recent onset and normal LV volumes, LVEF and RV dimensions show subtle LV and RV dysfunction. This cardiac impairment normalised after successful ablation of PVCs but remained unchanged in non-treated patients and in patients after unsuccessful RFCA. These findings suggest that recent-onset, frequent PVCs can result in reversible biventricular dysfunction not detected by conventional echocardiographic parameters.

Effect of frequent PVCs on ventricular performance

Frequent PVCs are generally considered a benign condition when found in patients without structural heart disease. Small series have demonstrated a link between frequent PVCs and increased LV dimensions and impaired LVEF, which improved or normalised after abolishing PVCs by pharmacological treatment or RFCA. However, the majority of patients presenting with frequent PVCs have a preserved LVEF.1–5 This does not exclude the possibility of a negative effect of PVCs on systolic function as subtle or early impairment may not be detected by conventional parameters.

Niwano et al prospectively followed up 239 patients with frequent PVCs and only 13, the majority with a high PVC burden, developed a small but significant decline in LVEF after 4 years, suggesting that impaired LV function may develop only over long periods.5 However, 42 patients with significant symptoms and/or LV dysfunction at baseline were excluded. These patients might represent a subgroup of patients in whom LV dysfunction occurs earlier.5

In our study the median duration of symptoms was 13 months. Despite the short period, these patients had already decreased multidirectional strain as compared with controls, whereas LV volumes, RV dimensions and LV and RV function measured by conventional echocardiographic techniques were normal. This indicates that PVCs may induce early LV and RV dysfunction not detected by conventional 2D echocardiography measurements. The evaluation of multidirectional strain with 2D speckle tracking imaging enables the detection of functional abnormalities by exploring the mechanical properties of the myocardium.10 2D speckle tracking imaging may constitute a more sensitive tool to detect subtle ventricular dysfunction, in particular for RV function assessment since quantification of RV volumes and function with conventional echocardiography remains problematic owing to the complex geometry of the right ventricle. However, these results do not allow us to define robust baseline cut-off values of multidirectional myocardial strain to identify and predict which patients will develop cardiomyopathy if RFCA is not applied. Additional studies including larger populations and longer follow-up are warranted to elucidate this aspect.

An inverse relationship between PVC burden and cardiac function has been described and PVC frequency has also been suggested as risk factor for PVC-induced cardiomyopathy.4 6 7

In this study, patients with a PVC burden of >20% were compared with patients with a PVC burden <20%. Although we did not observe significant differences in LVEF and volumes, there was a tendency to larger LV volumes in patients with a higher PVC burden, still within normal range. LV and RV strain were decreased to a similar extent in patients with a high and moderate PVC burden, indicating that even a moderate burden of PVCs (mean 13.3%) can induce subclinical cardiac dysfunction. However, the proportion of patients with a moderate PVC burden was relatively small.

Previous studies reporting a reversal of cardiomyopathy after RFCA only included patients with PVCs from the RV outflow tract.3 7 More recently it was demonstrated that a reversible reduced EF was equally prevalent among patients who had PVCs with a right or left bundle branch block configuration.4 18 This was confirmed by our study demonstrating that subtle and reversible LV and RV dysfunction can be induced by PVCs regardless of the site of origin.

RFCA and improvement of LV and RV strain

This study extends the finding of prior reports, that RFCA for PVCs can reverse cardiac functional abnormalities induced by PVCs.3 4 6 7 While Yarlagadda et al3 and Bogun et al4 observed reversal of mild to severe dilated cardiomyopathy after ablation of PVCs, Takemoto et al7 and Sekiguchi et al6 studying patients with normal LVEF detected no functional abnormalities before RFCA, and demonstrated improvement of LVEF and diameters within the normal range during follow-up. Our study, however, shows that PVC induced LV and RV dysfunction detected at baseline by strain analysis in patients with normal ventricular dimensions and LVEF can be reversed after successful RFCA. In contrast, in the PVC RFCA control group decreased strain at baseline did not change during short-term follow-up. Additional studies with longer follow-up are warranted to elucidate whether ventricular strain is further impaired in patients with frequent PVC not treated with RFCA.

Clinical implications

Two-dimensional speckle tracking strain imaging detects cardiac functional abnormalities in symptomatic patients with PVCs not detected by conventional echocardiographic measurements. This relatively novel imaging technique has been validated against sonomicrometry and tagged MRI, and several studies have demonstrated its accuracy for measuring myocardial strain.13 14 By measuring the mechanical properties of the myocardium with this technique, subtle myocardial dysfunction can be detected. Indeed, despite preserved LV or RV functions as assessed with conventional echocardiography, 2D speckle tracking can demonstrate subtle myocardial dysfunction in different clinical conditions (ie, aortic stenosis, diabetic cardiomyopathy, coronary artery disease).16 19 20 In addition, subtle myocardial dysfunction as detected with 2D speckle tracking has important prognostic implications.21–23 As demonstrated in previous studies, LV longitudinal strain showed a better prognostic value than baseline clinical characteristics, LV ejection fraction or wall motion score index measured with conventional echocardiography.23 Therefore, 2D speckle tracking may constitute a valuable imaging tool to improve the risk stratification of patients with different clinical conditions, including those that potentially can cause cardiomyopathy at long-term follow-up if not treated.

RFCA for PVCs is associated with a high success rate and few complications.3 4 6 7 24 Beside its value in relieving patients' symptoms, RFCA can have an important role in the management of patients with PVC-induced ventricular dysfunction.

Since it has been suggested that advanced PVC-induced ventricular dysfunction may lead to irreversible cardiomyopathy in individual patients, the early detection of PVC-induced dysfunction might be of clinical relevance.25 Although there seems to be an association between PVC frequency, duration and decline in LV function, some patients with low PVC burden and/or short duration of symptoms have already deterioration of LV function. Therefore PVC burden and duration alone may not identify patients at risk.2 4 5 Until additional risk factors are identified, 2D speckle tracking strain imaging may constitute a valuable test to identify patients who need to be reassessed during long-term follow-up.


Frequent PVCs can induce subtle LV and RV dysfunction in patients without apparent cardiomyopathy. PVCs originating from the left and right ventricle have similar detrimental effects on ventricular function. RFCA can successfully eliminate PVCs and improve cardiac function. These changes can be evaluated by 2D speckle tracking imaging.


View Abstract


  • APW and VD contributed equally to this article and are joint first authors.

  • Funding JJB received research grants from GE Healthcare, BMS medical imaging, St Jude, Medtronic, Boston Scientific, Biotronik, and Edwards Lifesciences. MJS received research grants from Biotronik, Medtronic and Boston Scientific. The other authors have no disclosures to report.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.