Article Text

Download PDFPDF
Original article
Cardiac resynchronisation therapy in patients with heart failure and a normal QRS duration: the RESPOND study
  1. Paul W X Foley1,
  2. Kiran Patel2,
  3. Nick Irwin3,
  4. John E Sanderson1,
  5. Michael P Frenneaux1,
  6. Russell E A Smith1,
  7. Berthold Stegemann4,
  8. Francisco Leyva1
  1. 1Centre for Cardiovascular Sciences, University of Birmingham, Queen Elizabeth Hospital, Birmingham, UK
  2. 2Sandwell and City Hospitals NHS Trust, Birmingham, UK
  3. 3Good Hope Hospital, Sutton Coldfield, West Midlands, UK
  4. 4Medtronic Inc, Bakken Research Center, Maastricht, The Netherlands
  1. Correspondence to Dr Francisco Leyva, Center for Cardiovascular Sciences, University of Birmingham, Queen Elizabeth Hospital, Birmingham B15 2TH, UK; cardiologists{at}hotmail.com

Abstract

Objectives To evaluate the clinical response to cardiac resynchronisation therapy (CRT) in patients with heart failure and a normal QRS duration (<120 ms).

Setting Single centre.

Patients 60 patients with heart failure and a normal QRS duration receiving optimal pharmacological treatment (OPT).

Interventions Patients were randomly assigned to CRT (n=29) or to a control group (OPT, n=31). Cardiovascular magnetic resonance was used in order to avoid scar at the site of left ventricular (LV) lead deployment.

Main outcome measures The primary end point was a change in 6 min walking distance (6-MWD). Other measures included a change in quality of life scores (Minnesota Living with Heart Failure questionnaire) and New York Heart Association class.

Results In 93% of implantations, the LV lead was deployed over non-scarred myocardium. At 6 months, the 6-MWD increased with CRT compared with OPT (p<0.0001), with more patients reaching a ≥25% increase (51.7% vs 12.9%, p=0.0019). Compared with OPT, CRT led to an improvement in quality-of-life scores (p=0.0265) and a reduction in NYHA class (p<0.0001). The composite clinical score (survival for 6 months free of heart failure hospitalisations plus improvement by one or more NYHA class or by ≥25% in 6-MWD) was better in CRT than in OPT (83% vs 23%, respectively; p<0.0001). Although no differences in total or cardiovascular mortality emerged between OPT and CRT, patients receiving OPT had a higher risk of death from pump failure than patients assigned to CRT (HR=8.41, p=0.0447) after a median follow-up of 677.5 days.

Conclusions CRT leads to an improvement in symptoms, exercise capacity and quality of life in patients with heart failure and a normal QRS duration.

(ClinicalTrials.gov number, NCT00480051.)

  • Cardiac resynchronisation therapy
  • heart failure
  • mortality
  • normal QRS duration
  • pacemakers
  • cardiac remodelling
  • cardiomyopathy dilated

Statistics from Altmetric.com

Introduction

Cardiac resynchronisation therapy (CRT) is an established treatment for selected patients with heart failure. In the Cardiac Resynchronization Heart Failure (CARE-HF) study, CRT was associated with 36% reduction in all-cause mortality.1 Other benefits of CRT include an improvement in symptoms, exercise capacity and quality of life.1–4

A QRS complex duration ≥120 ms has been adopted as the cut-off point for electrical dyssynchrony in CRT trials1 4 and, consequently, by treatment guidelines. The emerging recognition that mechanical dyssynchrony is present in patients with a QRS<120 ms5–7 has provided a conceptual basis for extending CRT to this patient population, which comprises the majority of patients with heart failure.8 Several observational studies have so far shown a benefit from CRT in patients with a QRS<120 ms.9–11 The only randomised controlled trial of CRT in patients with a QRS<130 ms, however, failed to show a benefit for peak oxygen uptake on cardiopulmonary exercise testing.12

We sought to determine whether CRT leads to a symptomatic improvement in patients with heart failure and a QRS<120 ms. The 6 min walking distance (6-MWD) using the 6 min walk test13 is sensitive to changes in the symptoms of heart failure14 and reflects activities of daily living, particularly after CRT.15 In addition, it is highly reproducible16 and easily quantifiable. Its utility has been extensively evaluated in systematic reviews.17 It is on this basis that a change in 6-MWD was adopted as the primary end point in this study.

Patients and methods

Patients

The Resynchronization in Patients with Heart Failure and a Normal QRS Duration (RESPOND) study was a single-centre (Good Hope Hospital) randomised controlled trial comparing the efficacy of CRT-pacing (CRT-P) with optimised pharmacological treatment (OPT) in patients with moderate-to-severe heart failure and a normal QRS duration (<120 ms). Eligibility criteria were as follows: sinus rhythm; New York Heart Association (NYHA) class III or IV; attendance at a dedicated heart failure clinic for optimisation of pharmacological treatment; a QRS duration <120 ms, and; a left ventricular ejection fraction (LVEF) <35%. Exclusion criteria were indications for conventional, right ventricular pacing; contraindications to cardiac pacing; myocardial infarction or acute coronary syndrome within the previous month; severe structural valvular heart disease; presence of comorbidities likely to threaten survival for 12 months and inability to perform a 6 min walk test. In the UK, primary prevention implantable cardioverter-defibrillator (ICD) therapy in patients with a normal QRS duration is indicated in the event of either a LVEF <35% (no worse than NYHA class III), non-sustained ventricular tachycardia on Holter monitoring and inducible ventricular tachycardia on electrophysiological testing.18 19 All patients were screened accordingly and excluded if these criteria were satisfied.

The diagnosis of heart failure was made on the basis of echocardiographic evidence of left ventricular (LV) systolic dysfunction. All participants had undergone coronary angiography. The diagnosis of ischaemic cardiomyopathy was made if LV systolic dysfunction was associated with a history of myocardial infarction20 and if there was significant coronary artery disease. The findings of late gadolinium enhancement cardiovascular magnetic resonance were also used to ascertain the aetiology of heart failure.21 The study conforms with the Declaration of Helsinki and was approved by the local ethics committee. It was registered at http://www.clinicaltrials.gov (ClinicalTrials.gov number, NCT00480051.)

Study design

This study consisted of 60 patients with heart failure with maximum-tolerated medical treatment who were randomised to CRT-P or no CRT-P in a 1:1 ratio. Randomisation assignments, generated using S-plus software version 6.2 (Insightful), were provided for the investigators in sealed envelopes numbered 1 to 60. These envelopes were opened following patient consent. Implantations were performed between August 2007 and September 2009. Patients underwent a clinical assessment and echocardiography at baseline and at 3 and 6 months after randomisation. Cardiovascular magnetic resonance was also undertaken at baseline in order to guide LV lead deployment.

The study was conceived and initiated by the investigators, all of whom contributed to the execution of the study and writing of the manuscript. Investigators had unrestricted access to the data, with complete freedom from the sponsor. The data were presented to the sponsor once fully analysed and written up by the investigators.

Clinical assessment

Patient assessment included evaluation of NYHA class and an unencouraged 6 min walk test, undertaken according to the American Thoracic Society guidelines.22 In addition, quality of life was assessed using the Minnesota Living with Heart Failure questionnaire, according to which scores range from 0 to 105, with higher scores denoting a poorer quality of life.13 Clinical response was also assessed with a composite clinical score, according to which improvement was defined as survival for 6 months after implantation free from hospitalisations for heart failure plus improvement by ≥1 NYHA class or by ≥25% in the 6-MWD. Sudden cardiac death was defined as a ‘natural, unexpected death due to cardiac causes, heralded by an abrupt loss of consciousness within 1 h of the onset of acute symptoms.’23 Death from pump failure was defined as ‘death after a period of clinical deterioration in signs and symptoms of heart failure despite medical treatment’.24

Device therapy

A transvenous (cephalic, subclavian or femoral) approach was used for all CRT device implantations. Following the demonstration of a reduced benefit from deploying the LV lead over scarred myocardium,25 26 implanters (FL, KP and REAS) had access to late gadolinium-enhancement cardiovascular magnetic resonance scans at the time of implantation and were asked to deploy the LV lead over non-scarred myocardium, if technically possible. Scar imaging was undertaken before implantation. Images were acquired on a 1.5 Tesla (General Electric, Slough, England) scanner, as previously described.25 The cardiovascular magnetic resonance scans and fluoroscopic images acquired at implantation were retrospectively analysed to determine whether the LV lead tip subtended scarred or non-scarred myocardium. To this end, the longitudinal distance from the atrioventricular plane to the lead tip, in a base-to-apex direction, was quantified in millimetres using the 45° right anterior oblique fluoroscopic view. This longitudinal distance was transposed to the four-chamber cardiovascular magnetic resonance view, so as to determine the short-axis slice that corresponded to the LV lead tip position. The 45° left anterior oblique fluoroscopic view was then used to determine the circumferential position. The longitudinal and circumferential coordinates thus permitted localisation of the LV lead tip in relation to any myocardial scar. The lead tip was considered to be over myocardial scar even if there was limited—for example, non-transmural, scar.

The anatomical position of the LV lead tip was determined according to the venous tributary in which the lead tip was deployed as well as to its longitudinal and its circumferential position. In the right anterior oblique view at an angle of 30°, the left ventricle is divided into basal, middle and apical sections, from the atrioventricular plane to the apex. The atrioventricular plane is assumed to lie at the same level as the coronary sinus. In the left anterior oblique view at an angle of 30°, the LV free wall is divided into anterolateral, lateral and posterior sectors, with the anterolateral wall extending from 1 to 2 o'clock, the lateral wall from 2 o'clock to 4 o'clock, and the posterior wall from 4 o'clock to 5 o'clock.27 Lead positions were assessed retrospectively by a senior radiographer who was blinded to the clinical outcome data.

The Medtronic Insync III generator (Medtronic Inc, Watford, England) was used in 27 cases and the Medtronic Concerto generator in one case. One patient died before implantation. Atrioventricular optimisation was undertaken within 24 h of device implantation and at 1 and 3 months, using Doppler echocardiography and the iterative method. Accordingly, the atrioventricular delay was set just below the intrinsic PR interval (to ensure capture) and then sequentially shortened by 20 ms increments, until the emergence of A-wave truncation. Atrioventricular delay was then lengthened by 10 ms increments until A-wave truncation disappeared and maximum E- and A-wave separation was achieved. The V–V delay was programmed at a device default of 4 ms.

Echocardiography

Two-dimensional echocardiography was performed using Vivid Systems 5 and 7 scanners (General Electric Healthcare Worldwide, Slough, UK). LV volumes were assessed using the biplane Simpson's method. The LVEF was calculated as the difference between the end-diastolic and end-systolic volumes divided by the end-diastolic volume and expressed as a percentage. An echocardiographic response, denoting LV reverse remodelling, was defined as a ≥15% reduction in LV end-systolic volume (LVESV) at follow-up.

With respect to dyssynchrony measures, the interventricular mechanical delay was computed as the absolute difference between the aortic and the pulmonary pre-ejection delay, both of which were quantified as the delay in milliseconds between the QRS onset and the onset of ejection, detected using pulsed-wave Doppler imaging. The septal-to-posterior wall delay was computed as the difference between the septal wall delay (interval between the QRS onset and the maximal septal-to-posterior displacement) and the posterior wall delay (interval between the QRS onset and the maximal posterior-to-septal displacement). An interventricular delay of ≥40 ms28 and septal-to-posterior wall motion delay of ≥130 ms29 were defined as significant interventricular or intraventricular dyssynchrony, respectively. Dyssynchrony measures and LV volumes at follow-up were assessed retrospectively by echocardiographers who were blinded to clinical data.

Statistical analysis

The predefined primary end point was the change in 6-MWD at 6 months' follow-up. Secondary end points included change in quality of life and in NYHA class at 6 months. We have previously observed an SD of 112 m for 6-MWD in 209 patients in sinus rhythm and a QRS>120 ms who underwent CRT.30 In these patients, CRT was associated with an increase in 6-MWD of 96.8 m at 783 days. For this study, we considered that CRT might lead to a smaller response in patients with a QRS<120 ms, perhaps 80% of 96.8 m (77.4 m). In sample size calculations, a number of 25 in each group achieved 91% power to detect a difference of 77.4 between the null hypothesis mean of 230.7 and the alternative hypothesis mean of 308.1, with an SD of 112.0 and a significance level (α) of 0.05, using a two-sided Wilcoxon test and assuming that the actual distribution was uniform. Five patients were added to each group to allow for attrition.

The intention-to-treat principle was adopted in all analyses. Continuous variables are expressed as mean±SD. Normality was tested using the Shapiro–Wilk test. Variables which were not normally distributed are expressed as median and IQR. Comparisons of changes from baseline in the CRT and the OPT groups were analysed by the Wilcoxon rank-sum test. Categorical variables were analysed using χ2 tests and Fisher's exact test. Kaplan–Meier survival curves and the log-rank test as well as Cox proportional hazards analyses were used to analyse clinical events. No censoring was carried out at 6 months. As in similar studies,12 values for clinical variables, such as 6-MWD, NYHA class and quality of life were assumed to be the worst values possible if patients died, withdrew or were unable to carry out a test. The values imputed were 0 m for 6-MWD, class IV for NYHA class and 105 for quality of life score. Statistical analyses were performed using Statview (Cary, North Carolina, USA) and SPSS 13.0 (Chicago, Illinois, USA). Sample size calculations were performed using PASS version 08.0.5 (Kaysville, Utah, USA). A two-tailed p value of <0.05 was considered statistically significant.

Results

Over the period June 2007 to September 2009, 110 patients were screened and 60 recruited. Of these, 29 were randomised to CRT-P and 31 to OPT. One patient randomised to CRT died before device implantation, one withdrew from the study and one patient crossed over from control to CRT-with defibrillator backup after the development of sustained ventricular tachycardia. Patients underwent transvenous biventricular pacemaker implantation using a cephalic vein/subclavian vein (n=26) or femoral vein (n=2) approaches. In all but two of the 28 implantations, the LV lead had been deployed over non-scarred myocardium. The LV lead tip positions, with respect to the coronary vein in which the LV lead tip was finally deployed, were as follows: 19/28 (67.9%) in posterolateral veins; 4/28 (14.3%) in anterolateral veins, and; 6/28 (21.4%) in the middle cardiac vein. The LV lead tip positions achieved with respect to circumferential position were as follows: 18/28 (64.3%) in lateral positions; 6/28 (21.4%) in anterolateral positions, and; 5/28 (17.9%) in posterior positions. The LV lead tip positions achieved with respect to longitudinal position (base-to-apex) were as follows: 23/28 (82.1%) in middle positions, and; 6/28 (22.4%) in apical positions.

Clinical response

As shown in table 1, the CRT group and the OPT group were similar with respect to 6-MWD at baseline. At 6 months, the 6-MWD increased significantly in the CRT group (146.3 [204.3] m to 249.0 [228.4] m, mean (IQR)), but not in the OPT group (120 (135) m to 90.0 (150) m) (p<0.0001). A ≥25% increase in 6-MWD was reached by more patients in the CRT group than in the OPT group at 6 months (51.7% vs 12.9%, p=0.0019) and at 3 months. (40.7% vs 11.5%, p=0.0276).

Table 1

Characteristics of the study group at baseline*

At 6 months, quality of life scores also decreased significantly in the CRT group (56.8±20.6 to 44.6±30.0) compared with the OPT group (63.5±18.8 to 62.1±27.7, p=0.0265) (figure 1). Improvements in NYHA class were also observed in the CRT group (3.34±0.48 to 2.24±0.99, p<0.0001), but not in the OPT group (3.12±0.43 to 3.32±0.79). Changes from baseline in 6-MWD (p=0.0030), quality of life (p=0.0074) and NYHA class (p<0.0001) were evident at 3 months. For the composite clinical score, CRT was associated with a more favourable response than OPT (83% vs 23%, respectively; p<0.0001).

Figure 1

Effect of cardiac resynchronisation therapy (CRT) on clinical variables. The change from baseline for 6 min walking distance and quality of life scores (Minnesota Living with Heart Failure questionnaire) at 3 and 6 months are shown in box-and-whisker plots, in which the five horizontal lines represent the 10th, 25th, 50th, 75th and 90th centiles, from bottom to top. For quality of life, scores range from 0 to 105, with high scores denoting the poorest quality of life. A reduction denotes improvement in quality of life. For New York Heart Association (NYHA) class, the distribution of classes at baseline, 3 months and 6 months is shown. p Values refer to differences in NYHA class, expressed as a continuous variable. OPT, optimal pharmacological treatment.

Clinical events

Among the 60 patients randomised, four patients, all of whom were in the OPT group, died before the scheduled 6-month follow-up. After a median follow-up of 677.5 days for events, six (20.7%) deaths occurred in the CRT group and 10 (32.3%) in the OPT group (p=NS). For cardiovascular deaths, there 10 (32.3%) occurred in the OPT group and five (17.2%) in the CRT group. No significant differences emerged between the OPT and the CRT groups with respect to these end points, With respect to deaths from pump failure, however, nine (29%) occurred in the OPT group and one (3.4%) in the CRT group. In Kaplan–Meier survival analyses (p=0.0161) and Cox proportional hazards analyses (HR=8.41; 95% CI 1.05 to 67.3, p=0.0447), the OPT group had a higher risk of death from pump failure than the CRT group (figure 2).

Figure 2

Kaplan–Meier estimates of the time to death from pump failure. Results of univariate Cox proportional hazards analyses are expressed as hazards ratio (HR) and 95% confidence limits (in parentheses).

Echocardiography

As shown in table 1, the CRT group and the OPT group had similar LV volumes, ejection fraction and measures of dyssynchrony at baseline. A shown in table 2, 55.6% of patients in the CRT group were echocardiographic responders. This, however, was not significant compared with the OPT group. The CRT group, however, had reduced intraventricular dyssynchrony (septal-to-posterior wall motion delay, p=0.0043) compared with the OPT group. No group differences emerged with respect to interventricular mechanical delay.

Table 2

Effect of cardiac resynchronisation therapy on echocardiographic variables*

Discussion

In this randomised controlled study of CRT in patients with heart failure and a QRS<120 ms, we have shown that compared with OPT CRT was associated with an improvement in 6-MWD, NYHA class and quality of life. The clinical benefit was particularly evident on the basis of the composite clinical score.

Major outcome trials of CRT, such as CARE-HF1 adopted a QRS>120 ms as an arbitrary surrogate marker of mechanical dyssynchrony. It is well recognised, however, that mechanical dyssynchrony is also present in patients with a normal QRS duration. Studies using tissue Doppler imaging have shown evidence of LV dyssynchrony in 36%6 to 51%7 of patients with heart failure and a QRS≤120 ms, depending on the method used. Similar findings have emerged from strain echocardiography31 and radionuclide phase analysis.32 Cardiovascular magnetic resonance, on the other hand, has shown that intraventricular dyssynchrony is almost universal in patients with heart failure.33 34 In this study, interventricular and intraventricular dyssynchrony, assessed retrospectively, were present in 23.3% and 43.3% of recruited patients, respectively. We also observed a significant reduction in the septal-to-posterior wall motion delay in the CRT group but not in the OPT group, indicating that intraventricular resynchronisation occurred after CRT. It would appear, therefore, that patients with a normal QRS duration have cardiac dyssynchrony and that this is potentially treatable. We should also consider that the benefit of CRT may also be mediated via other mechanisms. Relief of the diastolic ventricular interaction has recently been implicated in the haemodynamic benefit of CRT in patients with a normal QRS complex and no detectable dyssynchrony.35

Our findings have emerged in the context of the RethinQ study, in which 172 patients with heart failure and a QRS duration <130 ms were randomly assigned to CRT defibrillation or ICD treatment only.12 In that study, CRT was not associated with an improvement in the change in peak oxygen uptake of at least 1.0 ml/kg/min during cardiopulmonary exercise testing. This finding, however, should be interpreted with caution. While peak oxygen consumption is useful in prognostic stratification, it bears little relation to symptoms. Wilson et al, for example,36 found no relation between peak oxygen uptake and a dyspnoea/fatigue index: up to 45% of patients with a peak oxygen uptake ≤14 ml/kg/min had few or no exertional symptoms. In the Multicenter InSync ICD Randomised Clinical Evaluation II (MIRACLE ICD II), a randomised, double-blind, parallel-controlled trial of CRT in patients with NYHA class II heart failure, CRT led to an improvement in NYHA class, but not in peak oxygen uptake.37 On this basis, peak oxygen uptake is a doubtful surrogate measure of symptomatic benefit in patients with heart failure. It is noteworthy that although the RethinQ study found no benefit for peak oxygen uptake, it did demonstrate an improvement in NYHA class.12

We found significant LV reverse remodelling in 55.6% of patients in the CRT group. This, however, was not significant compared with the OPT group. It is well recognised, however, that although LV reverse remodelling predicts clinical outcome, it does not necessarily predict symptomatic response. Yu et al, for example, found no relationship between reductions in LVESV and changes in NYHA class, 6-MWD or quality-of-life scores in 141 patients undergoing CRT.38 Recently, Ypenburg et al also found similar improvements in NYHA class, quality-of-life scores and 6-MWD in patients exhibiting ≥15% reduction in LVESV, compared with those exhibiting a reduction in LVESV of <14%.39 Lafitte et al also found that 63% of patients who did not show a ≥15% reduction in LVESV after CRT nevertheless derived a clinical benefit.40 The discordance between the effects of CRT on symptoms and LV remodelling might be expected since LV reverse remodelling is only one of multiple mechanisms of action of CRT. The relief of the diastolic ventricular interaction41 and reductions in mitral regurgitation42 have been shown to relate to clinical benefit and not to depend on LV reverse remodelling.43

In contrast to other CRT studies, we used cardiovascular magnetic resonance to image scar before LV lead deployment. This approach was adopted on the basis of a previous study showing that pacing LV scar is associated with a dramatically higher risk of cardiovascular death or heart failure hospitalisation (HR=2.30, p=0.0014), compared with pacing non-scar.44 A 201Tl single photon emission CT had also shown that scar density in the segments in the vicinity of the LV lead was lower in responders than in non-responders.26 Mechanistically, these findings are supported by the finding that pacing a myocardial scar leads to a prolonged and fragmented QRS complex45 46 as well as electrical and mechanical dyssynchrony. Furthermore, it has been shown that myocardial scars47 effectively reduce the volume of myocardium available to a LV pacing stimulus.48

This study was not powered for mortality end points but nevertheless, clinical events and mode of death were recorded. As expected, no differences emerged between the OPT and the CRT groups with respect to total mortality or cardiovascular mortality. With respect to mode of death, however, the OPT group had a considerably higher risk of death from pump failure (HR=8.41, p=0.0447). This finding is also unexpected, particularly in such a small study. Further studies are needed to determine whether pacing away from scarred myocardium contributes to this effect of CRT in the normal QRS population.

Limitations

Importantly, this study was not double blinded. The possibility of a placebo effect cannot, therefore, be discounted. With reference to the Multisite Biventricular Pacing in Patients with Heart Failure and Intraventricular Conduction Delay (MUSTIC) study,2 a single-blind, crossover trial, the 6-MWD decreased from a mean of 350 m to 316 m during the inactive CRT phase (p<0.0001). This suggests that an inactive, implanted CRT device does not have a ‘placebo’ effect on 6-MWD. Moreover, the fact that the changes after CRT in our study were sustained at 3 and 6 months would vie against a placebo effect. Although the numbers in this study are small, the study achieved the primary end point addressed in power calculations. In this study, 93% of LV leads were deployed in non-scarred myocardium. Although this raises the possibility that the avoidance of pacing myocardial scars produces a more favourable response to CRT, no other studies have employed such a technique. It is therefore difficult to ascertain whether our implantation technique accounts for the favourable response to CRT observed in this study. Further studies are needed to determine whether the observed symptomatic response to CRT is sustained in the longer term.

Conclusions

We conclude that CRT leads to an improvement in symptoms, exercise capacity and quality of life in patients with moderate-to-severe heart failure and a normal QRS complex (<120 ms) who are not selected on the basis of mechanical dyssynchrony. The association of this clinical response with echocardiographic evidence of resynchronisation provides a mechanistic basis for the application of CRT to patients with heart failure and a normal QRS duration. The use of an implantation technique that avoids scar at the site of LV lead deployment is likely to be relevant in achieving a response.

Acknowledgments

We are grateful to Medtronic Inc for sponsoring this study. We are also grateful to Jane Tipping, Lisa Ball, Janet Brashaw-Smith and Emily Mulch for their input into the follow-up of patients included in this study.

References

Footnotes

  • See Editorial, p 1029

  • Funding Medtronic Inc. Bakken Research Centre, The Netherlands.

  • Competing interests PWXF held a research fellowship sponsored by Medtronic Inc. BS is an employee of Medtronic Inc. REAS and FL have received sponsorship from Medtronic Inc. MPF has received a grant in aid funding for investigator-led studies from Medtronic Inc and is on the End Points Committee for a study funded by Biotronik.

  • Ethics approval This study was conducted with the approval of the North Birmigham Ethics Committee.

  • 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.

Linked Articles