Content area
Background
Cardiac resynchronization therapy (CRT) is a cornerstone treatment for heart failure, yet the optimal pacing mode remains uncertain. This study investigates the impact of DDDR (rate-adaptive atrial support) versus VDD pacing modes on functional capacity and echocardiographic outcomes in patients with CRT-defibrillators.
MethodsIn a multicenter, double-blind, crossover trial, 26 sinus rhythm patients undergoing CRT-defibrillator implantation were randomized to DDDR at 60 beats-per-minute or VDD at 30 beats-per-minute. Each pacing mode was maintained for 6 months (period 1), followed by crossover to the alternate programming for additional 6 months (period 2). Distance covered in the 6-minute walk test (6MWT) and echocardiographic parameters were evaluated at baseline and after each period.
ResultsThe proportion of CRT responders was 73% at 6 months and 79% at 12 months. DDDR pacing yielded a median atrial pacing percentage of 55% (interquartile range, 48%-71%). No significant difference was found between DDDR and VDD modes in 6MWT distance (448 m [369-538] vs. 428 m [360-535], P = .71). The sequence of pacing modes was also not significant ( P = .17), while the study period ( P = .03) and the distance covered at baseline ( P < .01) were positively correlated with the distance walked at follow-up. Echocardiographic parameters, including left ventricular ejection fraction and ventricular volumes, showed no significant differences between the pacing modes.
ConclusionsIn this cohort, rate-adaptive atrial support did not offer any advantage over VDD pacing in terms of functional capacity or echocardiographic response. The findings support the recommendation to maximize intrinsic sinus rhythm sensing in CRT devices.
Trial registrationClinicalTrials.gov Identifier: NCT06592690.
Cardiac resynchronization therapy (CRT) is a cornerstone of device-based therapy for heart failure (HF). 1 , 2 The optimal pacing mode for CRT devices depends on the underlying atrial rhythm. In patients with sinus rhythm, an atrial-synchronous pacing mode with low rates (40-50 beats-per-minute [bpm]) is generally programmed to maximize the sensing of intrinsic sinus rhythm. 3 With this setting, the implanted atrial lead is primarily used to sense and track atrial activity, rather than pace. Recently, the feasibility of delivering CRT with a 2-lead system has been shown. 4 , 5 This system tracks atrial activity by a floating atrial dipole on the right ventricle lead with no pacing option (VDD mode). This approach reduces the amount of implanted hardware, potentially lowering the risk of device-related complications.
However, sinus bradycardia and chronotropic incompetence are common in this population and can be difficult to diagnose. 6 These conditions may also be exacerbated by HF medications. 7 Although atrial support pacing with rate-adaptive function (DDDR mode) did not demonstrate benefit in hard clinical endpoints in the randomized PEGASUS-CRT trial, 8 data from smaller studies have shown improvements in exercise capacity and quality of life with this pacing strategy. 9-11 The question remains whether a two-lead CRT technology without atrial pacing capability can be appropriately offered to a broad population of HF patients who require biventricular pacing but do not have a class I indication for atrial pacing.
To investigate this topic, we designed a randomized, cross-over, double-blind, multicenter study to compare DDDR and VDD pacing modes in sinus rhythm patients receiving a de-novo conventional CRT defibrillator (CRT-D) for standard indications. The primary outcome was functional capacity, measured by the distance walked during the 6-minute walking test (6MWT).
Materials and methodsThis multicentre, randomized, double-blind, crossover clinical trial included follow-up visits scheduled at 6 and 12 months after CRT-D implantation. The study aimed to compare the effects of 2 programming modalities: DDDR and VDD pacing. The study protocol received approval from the ethics committee of every centre and was conducted in accordance with the Declaration of Helsinki and local regulations (ClinicalTrials.gov Identifier: NCT06592690). All patients provided written informed consent.
Selection of patients and device programmingParticipants were selected based on the European Society of Cardiology/European Heart Rhythm Association guidelines for de novo CRT-D implantation. 2 Additional criteria included sinus rhythm and a resting heart rate above 50 bpm while receiving optimal pharmacological therapy for HF. Exclusion criteria were: (1) permanent atrial fibrillation (AF); (2) suboptimal HF drug therapy; (3) indication for atrial pacing due to significant chronotropic incompetence affecting exercise intolerance; (4) inability to perform the 6MWT; and (5) a life expectancy of less than 12 months.
Patients were enrolled prior to CRT-D implantation. Baseline sinus rate was measured using ECG and a pulse oximeter at the enrollment visit. Patients also underwent echocardiography and 6MWT to assess exercise intolerance and chronotropic response. The implantation procedure followed standard clinical practices using conventional 3-lead CRT-D systems with right atrial, right ventricular, and left ventricular leads. After successful implantation, patients were randomly assigned to either DDDR pacing (lower rate of 60 bpm) or VDD pacing (lower rate of 30 bpm) in a 1:1 allocation stratified by clinical site. The assigned pacing mode was maintained for 6 months (study period 1), after which the 6MWT and the echocardiographic evaluation were repeated. Patients were then crossed over to the alternative pacing mode for the next 6 months if they continued to meet eligibility criteria (study period 2), they underwent final 6MWT and echocardiographic assessments 12 months after implantation.
Throughout the study, both patients and the operators performing the tests were blinded to the pacing mode. The rate-adaptive sensor, when activated, was based on a standard motion-based accelerometer with nominal factor settings. Programming requirements were limited to the pacing mode and lower rate, with all other settings, including atrio-ventricular delay and antitachycardia programming, left to the investigator's discretion according to standard clinical practice. The 6MWT was performed as previously described. 12 Echocardiographic parameters were recorded with each center's available devices, aligned with secondary endpoint goals.
Study endpointsThe primary endpoint of the study was the distance walked during the 6MWT after a consecutive 6-month treatment period with the same pacing mode. Secondary study endpoints included echocardiographic parameters: left ventricle ejection fraction (LVEF), and left ventricle end-systolic and end-diastolic volumes (LVESV and LVEDV). All study endpoints were evaluated at baseline and during each scheduled follow-up visit.
StatisticsThe study was designed with 80% statistical power to detect a minimum 5% difference in walked distance during the 6MWT between DDDR and VDD pacing. Based on previous studies, 10 it was estimated that 26 were needed, accounting for a 10% dropout rate, for intraindividual comparison of the primary endpoint.
Sample distributions of continuous variables were reported as median (interquartile range [IQR]). Binary variables were reported as percentages. Generalized linear mixed models were used to analyze primary and secondary endpoints, including treatment (device programming), baseline values, period, and sequence as fixed effects, with patient as a random effect. Although a washout period was not included, the potential carryover effect was tested statistically as a sequence effect in the models. Statistical significance was defined as P < .05. Analyses were performed using Stata/MP 17.0 (StataCorp LLC, College Station, TX) and RStudio Version 1.4.1106 software (Posit, PBC, Boston, MA).
Results Population characteristics and study realizationA total of 26 patients who underwent CRT-D implantation were enrolled across three Italian sites: Ospedale Centrale, Bolzano; Ospedale Santa Maria del Carmine, Rovereto; and Ospedale Franz Tappeiner, Merano. Table 1 summarizes the cohort's characteristics, which are representative of a typical CRT-D population. The median age was 73 years (IQR: 68-77), with 77% of the patients being male. Most patients (73%) were classified as New York Heart Association functional class II and had ischemic cardiomyopathy (69%). Baseline electrocardiograms showed a resting heart rate of 70 bpm (IQR: 59-75), with all patients showing a chronotropic response, reaching a maximum heart rate of 87 bpm (IQR: 78-99) during testing. Median values for LVEF, LVESV, and LVEDV were 28% (IQR: 24%-31%), 167 mL (IQR: 137-196), and 220 mL (IQR: 181-286), respectively.
After excluding 3 patients (11%) who had unsuccessful LV lead implantation, 23 patients (89%) were randomized: 11 to DDDR -> VDD sequence and 12 to VDD -> DDDR sequence. All of them completed the study period 1, expect one patient who was explanted due to device infection. Finally, 19 patients (73%) complete the 12-month follow-up period ( Figure 1 ).
The proportion of CRT responders, defined as a decrease in LVESV of ≥15% and/or an absolute increase of 5% in LVEF, was 72.7% (16/22) at 6 months and 78.9% (15/19) at 12 months. The use of the rate-adaptive function yielded a median atrial pacing percentage of 55% (IQR: 48%-71%). The biventricular pacing percentage did not differ between study periods ( P = .63), with DDDR pacing at 98% (IQR: 96%-99%) compared to VDD pacing at 99% (IQR: 96%-99%) ( Table 2 ).
Functional capacity and echocardiographyThe distance covered during the 6MWT was 448 m (IQR: 369-538) with the DDDR pacing mode and 428 m (360-535) with the VDD pacing mode ( Table 3 ). Generalized linear mixed-model analysis showed no significant association between pacing mode and the primary endpoint ( P = .71). The sequence of pacing modes (carryover effect) was also not significant ( P = .17). However, the study period effect (estimate 33.2, standard error 14.3, P = .03) and the distance walked at baseline (estimate 0.84, standard error 0.14, P < .01) were positively correlated with the distance covered during the follow-up. Figure 2 illustrates the distances at baseline and after periods 1 and 2 for both study sequences.
No significant differences were observed between the DDDR and VDD pacing modes in the secondary endpoints ( Table 2). Echocardiographic results showed an LVEF of 41% (IQR: 31%-45%) during DDDR mode and 37% (IQR: 35%-46%) during VDD mode. LVESV was 102 mL (IQR: 82-159) and 117 mL (IQR: 82-145), while LVEDV was 174 mL (IQR: 141-232) and 183 mL (IQR: 148-226) for DDDR and VDD modes, respectively. The carryover effect was not significant for any variables of interest. However, the period effect significantly impacted LVEF (estimate 2.91, standard error 1.19, p=0.03) and LVESV (estimate -14.7, standard error 6.3, P = .03). Figure 3 displays the secondary endpoints at baseline and after periods 1 and 2 for both study sequences.
DiscussionThe main finding of our study is that rate-adaptive pacing (DDDR mode) did not result in any improvement in functional capacity when compared to VDD pacing with a low basic rate in patients with sinus rhythm receiving a CRT-D for HF treatment. Although the study was not powered for this endpoint, echocardiographic response was also not significantly better during the period with atrial support.
We found a high proportion of CRT responders, defined as a decrease in LVESV of ≥15% and/or absolute increase of 5% in the LVEF, in our population. The very scrupulous patient selection, with 85% of the patients having a complete left bundle branch block and more than 3 quarters of the patients having a wide (ie, >150 ms) QRS, could explain this result 13 It is important to highlight that there was no difference in terms of responders’ rate between the 2 programming modes. The positive effects of CRT in reducing mortality and hospitalizations for HF were demonstrated by randomized controlled trials two decades ago 14 , 15 Moreover, in many - but not all - patients, CRT can improve functional capacity by correcting impaired ventricular electromechanical coupling or dyssynchrony. In our population, this was evident at 6 months and slightly improved further after crossing over. Interestingly, a period effect and the distance walked at baseline were positively correlated with the distance covered at follow-up, but not the CRT pacing mode. These results contain important insights: first, rate-adaptive atrial support did not improve functional capacity in HF patients; second, the benefit of CRT persists and increases beyond 6 months from implantation; third, a worse baseline condition is associated with a poorer CRT response, suggesting that CRT devices may need to be implanted earlier and that further studies are needed to determine a 6MWT cutoff to predict CRT response. In a previous report, improvement of cardiac indices and biomarkers concentration were evident at 6 months, but functional indices (measured with New York functional class and cardiopulmonary exercise testing) improved after 12 months and were associated with restoration of heart rate variability and carotid baroreflex. 16 In this paper, responders presented less depressed hemodynamic, functional and neurohormonal indices at baseline.
Our findings support the strategy of preferring pacing modes that maximize the sensing of intrinsic sinus rhythm in CRT devices but also raise questions about the utility of atrial pacing capability in these devices. The implantation of an atrial lead in an ICD recipient is not in question when atrial pacing is needed for sinus node dysfunction. In a recent national study of US patients undergoing first-time ICD implantation without a clinical indication for an atrial lead, the use of dual-chamber devices decreased over 10 years but remained high (42.2%). 17 This is mainly due to the perception of a potential need for pacing in future, the effect of medications on pacing needs, arrhythmia discrimination algorithms, and specific scenarios in which pacing is beneficial. It is well known that the placement of any lead is burdened with a higher risk of complications. The crossover design of our study and the small sample size does not permit any observation regarding pneumothorax, lead dislodgement and infections because all patients received an atrial lead. However, the fact that there is no significant difference in the primary study endpoint allows us to speculate that using a 2 lead VDD CRT device could reduce the rate of complications when compared to a conventional 3-lead DDD device without affecting the benefits of the therapy. Lead related re-interventions (including dislodgement, malposition, subclavian crush syndrome, etc) have an incidence between 1% and 5.9% according to different studies and registries. 18-21 Other possible complications include pneumothorax (0.5%-2.2%), cardiac implantable electronic device related infection (0.7%-4.6%), cardiac perforation (0.3%-0.7%), haemothorax (0.1%), inadvertent arterial puncture generating haematoma, injury of the brachial plexus (<0.1%). In a large Danish cohort, right atrial leads were second in frequency of complications after left ventricular leads (2.3 vs 4.3%) 22. The same authors report that a CRT device (odds ratio 3.3), passive fixation right atrial lead (odds ratio 2.2) and chronic HF as indication (odds ratio 3.0) were the most important risk predictors for complications. Atrial lead placement requires dedicated vein isolation. Cephalic cutdown is strongly recommended to reduce the risk of pneumothorax; in the case of CRT devices, axillary or subclavian venipuncture is also often used. Subclavian vein access is associated with about 8-fold increased risk of pneumothorax. The use of ultrasound guided axillary vein puncture has been demonstrated to reduce this risk when compared with subclavian puncture. 23 Although pneumothorax is an uncommon complication, 24 it can have serious consequences. It is rarely life threatening but prolongs hospital stay and increases hospitalization costs in most cases. 25 Atrial lead dislodgement is more frequent when passive fixation leads are used. 22 , 26 Implantation time is longer for a DDD device when compared to VDD. 27 , 28 Infection rates increase with the number of implanted leads. 29-31 The use of an absorbable antibiotic-eluting envelope has proven to reduce, but not eliminate, major cardiac implantable electronic device infections within 12 months of CIED surgery in a randomized controlled trial and in real world settings. 32 , 33 Moreover, in the unfortunate case of endocarditis, the presence of three leads could represent a significant challenge.
One possible side effect of a single atrio-ventricular lead device is atrial undersensing. This was reported to be minimal, with no significant difference, in a small group of VDD-CRT patients when compared to a DDD-CRT population. 34 Potential drawbacks of this approach should be further tested in well-designed clinical trials.
Study limitationsThis study has several limitations that should be considered when interpreting the results. First, the small sample size (26 patients) may limit the generalizability of our findings, and a larger cohort would be necessary to confirm these results. Furthermore, it did not allow subgroup analyses, which could have provided additional insights into how different patient characteristics influence the response to pacing modes.
The follow-up duration was limited to 12 months, providing insights into short- to mid-term outcomes but potentially missing long-term effects and complications associated with each pacing mode. Although the crossover design allowed for intraindividual comparisons, the absence of a washout period raises the possibility of a carryover effect, despite statistical testing using sequence term in the models indicating it was not significant.
ConclusionIn our randomized crossover study, rate-adaptive pacing (DDDR) did not offer any advantage over VDD pacing in terms of functional capacity or echocardiographic response in patients with sinus rhythm receiving a CRT-D for HF treatment. The findings suggest that maximizing intrinsic sinus rhythm sensing may be preferable in CRT devices, potentially allowing for simpler, 2-lead systems without compromising clinical outcomes. Further studies with larger sample sizes and longer follow-up are needed to confirm these results.
CRediT authorship contribution statementLuca Donazzan: Writing – original draft, Validation, Project administration, Investigation, Conceptualization. Francesca Bulian: Writing – review & editing, Investigation. Massimiliano Maines: Writing – review & editing, Validation, Investigation. Martin Erckert: Writing – review & editing, Validation, Investigation, Data curation. Francesco Peruzza: Writing – review & editing, Validation, Investigation, Data curation. Werner Günther Rauhe: Writing – review & editing, Validation, Investigation, Data curation. Daniele Giacopelli: Writing – review & editing, Software, Methodology, Formal analysis. Massimiliano Manfrin: Writing – review & editing, Supervision, Investigation, Data curation, Conceptualization.
Declaration of competing interestD.G. is an employee of Biotronik Italia S.p.a. The remaining authors declare no conflicts of interest.
AcknowledgmentsThe authors are thankful to Filippo Suman for technical and organizational support.
FundingThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
| | |
| Sex, male (n, %) | 20 (77%) |
| Age (years) | 73 (68-77) |
| Height (cm) | 170 (165-175) |
| Weight (Kg) | 78 (65-85) |
| NYHA class | |
| | 19 (73%) |
| | 7 (27%) |
| Systolic blood pressure (mmHg) | 120 (110-125) |
| Diastolic blood pressure (mmHg) | 70 (60-71) |
| Cardiomyopathy (n, %) | |
| | 18 (69%) |
| | 7 (27%) |
| | 4 (15%) |
| | 4 (15%) |
| | 2 (8%) |
| Chemotherapy Induced | 1 (4%) |
| Comorbidities (n, %) | |
| | 21 (81%) |
| | 17 (65%) |
| | 12 (46%) |
| | 8 (31%) |
| | 4 (16%) |
| | 4 (15%) |
| | 3 (12%) |
| | 2 (8%) |
| ECG | |
| | 70 (59-75) |
| | |
| | 22 (85%) |
| | 3 (12%) |
| | 1 (4%) |
| | |
| | 6 (23%) |
| | 20 (77%) |
| | |
| | 28 (24-31) |
| | 167 (137-196) |
| | 220 (181-286) |
| | |
| | 365 (320-441) |
| | 87 (78-99) |
| | 96 (95-98) |
| | |
| | 24 (92%) |
| | 22 (85%) |
| | 19 (73%) |
| | 17 (65%) |
| | 15 (53%) |
| | 12 (46%) |
| | 11 (44%) |
| | 7 (27%) |
| | 6 (23%) |
| | 5 (18%) |
| Variable | DDDR | VDD | |
| Atrial pacing (%) | 55 (48-71) | - | - |
| Biventricular pacing (%) | 98 (96-99) | 99 (96-99) | .63 |
| Patients with device-detected AF (n) | 1 (4.3%) | 2 (8.7%) | .86 |
| Variable | DDDR | VDD | |
| Primary endpoint | |||
| | 448 (369-538) | 428 (360-535) | 0.71 |
| Secondary endpoints | |||
| | 41 (31-45) | 37 (35-46) | 0.44 |
| | 102 (82-159) | 117 (82-145) | 0.64 |
| | 174 (141-232) | 183 (148-226) | 0.20 |
©2024. Elsevier Inc.