Novel Device-Based Algorithm Provides Optimal Hemodynamics During Exercise in Patients With Cardiac Resynchronization Therapy

Abstract

Background: An adaptive cardiac resynchronization therapy (aCRT) algorithm has been described for synchronized left ventricular (LV) pacing and continuous optimization of cardiac resynchronization therapy (CRT). However, there are few algorithmic data on the effect of changes during exercise.

Methods and Results: We enrolled 27 patients with availability of the aCRT algorithm. Eligible patients were manually programmed to optimal atrioventricular (AV) and interventricular (VV) delays by using echocardiograms at rest or during 2 stages of supine bicycle exercise. We compared the maximum cardiac output between manual echo-optimization and aCRT-on during each phase. After initiating exercise, the optimal AV delay progressively shortened (P<0.05) with incremental exercise levels. The manual-optimized settings and aCRT resulted in similar cardiac performance, as demonstrated by a high concordance correlation coefficient between the LV outflow tract velocity time integral (LVOT-VTI) during each exercise stage (Ex.1: r=0.94 P<0.0008, Ex.2: r=0.88 P<0.001, respectively). Synchronized LV-only pacing in patients with normal AV conduction could provide a higher LVOT-VTI as compared with manual-optimized conventional biventricular pacing at peak exercise (P<0.05).

Conclusions: The aCRT algorithm was physiologically sound during exercise by patients.

Cardiac resynchronization therapy (CRT) is an efficient therapeutic strategy for heart failure (HF) patients with reduced left ventricular (LV) ejection fraction (EF) and cardiac dyssynchrony who are unresponsive even with adequate medication. CRT improves symptoms, reduces hospitalizations for HF, and increases survival in such patients, but approximately one-third of patients may not obtain any improvement in their clinical status even after initiating biventricular (BiV) pacing.^1–3 There can be many post-implant factors that contribute to a suboptimal response, among which suboptimal atrioventricular (AV) timing could be a common reason.⁴ Although optimization of the AV and interventricular (VV) delay during BiV pacing could be 1 solution to maximize the acute hemodynamic effect of CRT,^5,6 a single AV and VV optimization at rest could be considered to have only limited meaning for improving the clinical prognosis.⁷ One explanation could be that the physiological conduction may vary dramatically depending on the patient’s exercise level,^8,9 and therefore, the optimal AV and VV timings are frequently changing. Recently, several studies used echocardiography to examine the effect of exercise on the optimal AV delay in CRT patients and their results suggest that AV optimization during exercise improves the cardiac performance in CRT patients.^8,10,11 Based on these findings, an algorithm that can provide ambulatory CRT optimization and more physiologic pacing has been developed.

The effectiveness of the AdaptivCRT algorithm (aCRT, Medtronic, Mounds View, MN, USA) has been published.¹² The algorithm automatically adjusts the pacing mode or AV/VV pacing intervals every minute based on the intrinsic AV conduction. When normal AV conduction occurs, synchronized LV-only pacing is provided and withholds right ventricular (RV) pacing. During periods of prolonged AV conduction, the algorithm continuously optimizes the AV and VV intervals. The aCRT algorithm has been demonstrated to decrease the risk of HF hospitalization as compared with echocardiography-optimized BiV pacing, and especially when the percent synchronized LV-only pacing exceeds more than 50%, a significant reduction in death or HF hospitalizations has been shown.¹³ Furthermore, the aCRT is also reported to reduce the risk of developing atrial fibrillation.¹⁴ To date, few studies have been conducted regarding the effect of its validity during exercise-induced changes. Thus, the goal of this study using exercise stress echocardiography was to identify whether this algorithm could provide hemodynamically reasonable and beneficial effects when compared with conventional echocardiographic AV and VV optimization during exercise.

Methods

Study Population

Patients with HF previously implanted with a CRT device were recruited. The eligibility for CRT was based on international guidelines, including advanced HF (NYHA functional class II, III, or IV), depressed LV function (LVEF ≤35%), and a wide QRS complex (>120 ms) on ECG.^15,16 Inclusion criteria were as follows: able to complete a supine bicycle exercise stress protocol, in sinus rhythm at baseline, and received a CRT device equipped with the aCRT algorithm. Exclusion criteria were patients with atrial fibrillation, prior valvular surgery, significant sinus node dysfunction, and absence or lack of intrinsic AV conduction or pacing dependence. The study protocol was approved by the institutional research ethics committee and all patients provided written informed consent.

The aCRT Algorithm

The aCRT algorithm is designed to automatically adjust the pacing mode by evaluating the intrinsic AV conduction every minute. During a normal AV conduction interval (intrinsic AV ≤200 ms during sinus rhythm, or AV ≤250 ms during atrial pacing) and a heart rate that did not exceed 100 beats/min, the synchronized LV-only pacing provides an atrial to RV sense interval of ≥40 ms (adaptive LV). Contrarily, during a prolonged AV conduction interval, the BiV pacing adjusts the AV and VV timing based on the intervals of the atrial to RV sense, atrial to P-wave end, and RV sense to QRS end (adaptive BiV).^12,14 Once the algorithm is activated, the algorithm-determined pacing mode or AV/VV delays are delivered immediately.

Echocardiographic Imaging and CRT Optimization

The echocardiographic images were recorded with a Vivid E9 ultrasound system (VingMed Ultrasound, General Electric, Milwaukee, WI, USA) equipped with a 3.5-MHz imaging probe and offline cine-loop analysis software. The images were acquired in a semi-supine position at rest and supine during exercise.

Pulsed-wave Doppler images of the transmitral flow velocity were recorded by placing the sample volume at the tips of the mitral valve leaflets and using a sweep speed of 100 mm/s. Pulsed-wave Doppler sampling of the LV outflow tract (LVOT) was recorded by placing the sample volume 1 cm below the aortic valve in the apical 5-chamber view or 3-chamber view, and the LVOT velocity time integral (LVOT-VTI) was measured at a sweep speed of 100 mm/s. The iterative method was used for AV optimization.¹⁷ The optimal AV delay was defined as the maximal E-wave and A-wave separation without truncation of the A wave. Subsequently, VV optimization was performed. The optimal VV delay was defined as the delay that produced the maximum LVOT-VTI. The echocardiographic images were recorded over 3 consecutive cardiac cycles and the measurement values were averaged. All echocardiographic data were analyzed by 2 experienced sonographers with disagreements settled by consensus, and they were blinded to the algorithm being on or off.

Exercise and the Study Protocol

Prior to acquiring the echocardiographic images, a device interrogation was performed and the intra-device ECG/marker signals were continuously viewed and recorded throughout the study. Blood pressure and 12-lead ECG were recorded at 1-min intervals. The 12-lead ECGs were assessed for an appropriate BiV or synchronized LV capture throughout the test. In all eligible patients the devices were programmed to the atrial tracking mode (DDD), the aCRT algorithm (aCRT-on) was activated before the study, and a full baseline echocardiographic examination was performed. Next, the aCRT algorithm was de-activated (aCRT-off) and resting AV/VV echo-optimization was applied according to the previously mentioned method. The heart rate at the end of this phase was defined as the baseline heart rate and an exercise test was initiated while remaining in the aCRT-off mode.

The patients underwent supine bicycle exercise at a fixed workload (10 W) and the resistance of the bicycle was continually adjusted; the patients were coached to remain alert to the pedaling speed as necessary in order to obtain the target heart rate. After achieving a first target heart rate of 10 beats/min above baseline or at 2 min from the beginning of the exercise (Ex.1), echocardiographic AV optimization was performed by the same method as performed at rest. After optimizing the AV delay, the VV intervals were programmed starting with an interval of 0 ms, which was increased iteratively by 10 ms (=LV pacing preceding the RV pacing by 10 ms) intervals up to 50 ms and the LVOT-VTI at each setting was recorded. Next, the aCRT algorithm was activated (aCRT-on) and pulsed-wave Doppler images of the transmitral flow velocity and LVOT-VTI were recorded. After completing the first exercise stage and once more turning off the aCRT algorithm (aCRT-off), the patients began the second stage of the exercise protocol with a target heart rate of 20 beats/min above baseline (Ex.2) by adjusting the workload until the patient developed symptoms. Similar to the first exercise stage, after achieving the target heart rate, the reprogramed AV echocardiographic optimizations and subsequent LVOT-VTI evaluation for each VV delay interval were performed. Finally, the aCRT algorithm was again activated (aCRT-on) in the patient and consequently the transmitral flow velocity and LVOT-VTI were recorded (Figure 1). A representative case is illustrated in Figure 2. The phase at the maximum exercise workload was described as peak exercise in this study.

Figure 1.

Flowchart of the exercise and study protocol. aCRT, adaptive cardiac resynchronization therapy algorithm; AV, atrioventricular; Ex, exercise; VV, interventricular.

Figure 2.

Echocardiographic images of a representative patient. (A) Transmitral Doppler flow at baseline with a shorter AV delay (90 ms) shows partial truncation of the A-wave, and LV filling time of 291 ms (Upper panel). With a longer AV delay (150 ms), the E-wave end and A-wave head partially overlap and the LV filling time is 247 ms (Lower panel). The maximal E-wave and A-wave separation without truncation of the A wave is shown in this setting when the AV delay is 100 ms, and the LV filling time is the longest (322 ms). This AV delay time was defined as the optimal value (Middle panel). (B) After initiating exercise, the optimal AV delay shortened to 70 ms, and as compared to that with shorter and longer AV delay patterns, the corresponding LVOT-VTI was largest in this setting (yellow frame). Of note, during exercise, the optimal AV delay at baseline was no longer optimal and the LVOT-VTI increased after AV optimization (Lower panel). (C) The LVOT-VTI with an optimal AV and VV delay during exercise and with aCRT-on. The aCRT algorithm provided a higher LVOT-VTI as compared with the manual echo-optimized setting. HR, heart rate; LV, left ventricle/ventricular; LVOT-VTI, left ventricular outflow tract velocity time integral. Other abbreviations as in Figure 1.

During the study protocol, as the pacing mode was changed from aCRT-on to aCRT-off, a ≥30-s waiting period was provided to allow for the change in function to stabilize. The real-time VV delay echo-optimization was often inaccurate because of the difficulty in obtaining precise LVOT-VTI measurements during exercise. Thus, we analyzed the optimal VV delay by reviewing the recorded echocardiographic images after completing the study. Of note, the aCRT algorithm is designed to provide synchronized LV-only pacing only when the heart rate is <100 beats/min, and therefore our exercise protocol was intended to keep the patients’ heart rate under 100 beats/min. To verify reproducibility, 7 CRT patients in whom the exercise stress echo-optimization had been repeated with the same protocol were preliminarily examined by 2 different independent observers. The inter- and intraobserver variabilities, expressed as the intraclass correlation coefficient regarding the post-optimized LVOT-VTIs, was more than 0.80 throughout the phase.

Statistical Analysis

The continuous data are expressed as mean±SD for a normal distribution, or the median (25th and 75th percentiles) for a skewed distribution. Dichotomous data are presented as numbers and percentages. A comparison of the continuous data was performed with Student’s t-test for paired and unpaired data. Nonparametric data were compared using the Wilcoxon rank sum test.

Correlations between continuous variables were calculated with a Pearson correlation coefficient. A Spearman rank correlation coefficient was used to calculate the continuous variables that were not normally distributed. Hazard ratios (HRs) with 95% confidence intervals (CIs) and 2-sided P values were reported. A 2-tailed P value of <0.05 was considered statistically significant. All statistical analyses were performed with JMP 12 software (SAS Institute, Inc., Cary, NC, USA).

Results

Study Population

As our cohort of CRT patients, a total of 27 patients were enrolled. Their baseline characteristics are summarized in Table 1. Most were men with moderate HF, a prolonged QRS with left bundle branch block (LBBB), equivalent representation of ischemic and non-ischemic, and a severely reduced LVEF. The majority of the patients were prescribed negative chronotropic agents such as β-blockers or amiodarone. At the 6-month follow-up, the mean NYHA functional class improved from 3.0±0.3 to 2.0±0.7 (P<0.001) and the LVEF from 25.6±8.4 to 32.9±14.1 (P<0.003). All eligible patients had their atrial leads placed in the right atrial appendage and the RV lead in the RV apex. The LV lead was placed in an anterolateral (n=8), lateral (n=11) or posterior (n=8) cardiac vein.

Table 1. Baseline Characteristics

	All patients (n=27)
Baseline characteristics
Age (years)	69±11
Male, n (%)	22 (81)
Ischemic heart disease, n (%)	13 (48)
Hypertension, n (%)	16 (56)
Diabetes, n (%)	7 (26)
CKD, n (%)	14 (52)
Treatments
β-blocker, n (%)	26 (96)
ACE inhibitor, n (%)	20 (74)
Diuretics, n (%)	17 (63)
Amiodarone n (%)	7 (26)
Pre-CRT
NYHA functional class (II/III/IV) (%)	7/89/4
QRS duration (ms)	154±26
LBBB, n (%)	22 (81)
LVEF (%)	26±8
LVEDV (mL)	186±89
LVESV (mL)	136±79
Post-CRT
ΔNYHA class	−1.0±0.9
LVEF (%)	33±14
LVEDV (mL)	175±59
LVESV (mL)	122±51

Data are n (%) or mean±SD. ACE, angiotensin-converting enzyme; CKD, chronic kidney disease; CRT, cardiac resynchronization therapy; LBBB, left bundle branch block; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LVESV, left ventricular end-systolic volume; NYHA, New York Heart Association.

Exercise Tolerance and the Pacing Mode

All patients were able to successfully complete the protocol of the first exercise stage (Ex.1), but 3 patients could not achieve the target heart rate in the second exercise stage (Ex.2) because of reported symptoms of fatigue and inability to increase their heart rate. The resting heart rate was 74.3±9.0 beats/min, which increased to 80.8±9.4 beats/min during Ex.1 and 89.6±8.3 beats/min during Ex.2. Before initiating the exercise, 18 (66%) patients were programmed to the adaptive BiV mode and 9 (33%) subjects to the adaptive LV mode. Interestingly, during exercise, 2 patients converted to an adaptive LV from an adaptive BiV, and conversely 3 patients converted to an adaptive BiV from an adaptive LV. The pacing mode of the aCRT algorithm during each exercise phase is shown in Table 2.

Table 2. Pacing Mode of the aCRT Algorithm During Each Phase

Pacing mode	Rest (n, %)	Ex. 1 (n, %)	Ex. 2 (n, %)
Adaptive BiV	18 (66)	19 (70)	17 (71)
		BiV at rest to BiV 16 (59)	BiV at rest to BiV 15 (63)
		LV at rest to BiV 3 (11)	LV at rest to BiV 2 (8)
Adaptive LV	9 (33)	8 (30)	7 (29)
		BiV at rest to LV 2 (7)	BiV at rest to LV 1 (4)
		LV at rest to LV 6 (22)	LV at rest to LV 6 (25)

aCRT, adaptive cardiac resynchronization therapy algorithm; BiV, biventricular pacing; LV, synchronized left ventricular pacing.

Optimal AV/VV Delay and the LV Filling Time at Rest and During Exercise

At baseline, the mean optimal AV delay was 116.3±27.2 ms. During exercise, the optimal AV delay was significantly reduced to 95.8±23.9 ms during Ex.1 (mean difference 20.7 ms; 95% CI 29.6–11.9 ms; P<0.0001), and progressively shortened to 90.9±18.0 ms during Ex.2 (mean difference from Ex.1: 8.2 ms; 95% CI 15.8–0.6 ms; P=0.018) (Figure 3). There was no significant difference in the LV filling time at rest (pre-optimization 442±88 ms; post-optimization 438±19 ms; P=0.6). After initiating the exercise, the LV filling time significantly decreased from 438±19 ms to 365±25 ms (P<0.01); however, a corresponding increase was observed in the mean LV filling time from 365±75 ms to 386±84 ms following the AV optimization at peak exercise (P<0.01).

Figure 3.

Changes in the optimal atrioventricular (AV) delay assessed at baseline and during each exercise phase. Ex., exercise protocol.

The median VV delay at rest was 0 ms (0–20 ms) and exercise induced a change in the VV delay programming in 13 (48%) patients. The rest-to-peak exercise variations in the optimal VV delay ranged from 0 to 50 ms, with a median value of 20 ms. There were no concordant correlations between the heart rates and optimal VV delays (r=−0.04, P=0.9). Importantly, 2 patients developed ineffective LV pacing at a VV delay of 0 ms and successfully converted to an effective LV pacing at a VV delay of up to 20 ms during exercise.

Correlation Between Manual Echo-Optimization and aCRT

In the overall patients, the baseline mean LVOT-VTI for the manual echo-optimized settings and aCRT was 12.5±3.0 cm and 12.5±3.1 cm, respectively (P=0.8). During Ex.1, the LVOT-VTI for the manual echo-optimized settings and aCRT was 14.2±3.5 cm and 14.2±4.0 cm, respectively (P=0.9), and 13.8±3.0 cm and 14.5±3.3 cm (P=0.005) during Ex.2. There were no significant differences in the heart rate between that at the time of the maximal LVOT-VTI with a manual echo-optimized setting and that of the aCRT both during Ex.1 (80.8±9.9 beats/min and 81.2±9.2 beats/min, respectively; P=0.9) and Ex. 2 (89.6±8.5 beats/min and 89.6±7.6 beats/min, respectively; P=0.95). Surprisingly, the manual echo-optimized setting and aCRT resulted in a similar cardiac performance, as demonstrated by a high concordance correlation coefficient between the LVOT-VTI at each exercise stage (baseline: r=0.95 P<0.001, Ex.1: r=0.94 P<0.0008, Ex.2: r=0.88, P<0.001, respectively) (Figure 4A). Similarly, only among the patients with an adaptive BiV pacing mode, there was a high concordance correlation coefficient between the manual echo-optimized setting and aCRT for each exercise stage regarding the LVOT-VTI (baseline: r=0.90 P<0.001, Ex.1: r=0.94 P<0.001, Ex.2: r=0.89, P<0.001, respectively) (Figure 4B).

Figure 4.

Comparison of cardiac performance with the aCRT and echo-optimized settings at baseline and during exercise. Plot of the LVOT-VTI at the settings calculated by the aCRT algorithm and settings obtained using the echocardiographic optimization protocol in all subjects (A) and only in those with adaptive BiV pacing (B). aCRT, adaptive cardiac resynchronization therapy algorithm; BiV, biventricular; Ex., exercise protocol; LVOT-VTI, left ventricular outflow tract velocity time integral.

When compared with the manual echo-optimized conventional BiV pacing, synchronized LV-only pacing (adaptive LV) had a tendency for a higher LVOT-VTI at rest (median 13.2 cm, range 12.2–16.9 cm vs. median 11.7 cm, range 10.8–13.4 cm, P=0.08). Interestingly, the adaptive LV at peak exercise provided a higher value compared with manual echo-optimized BiV pacing at peak exercise (17.6 cm, range 15.4–19.7 cm vs. 14.7 cm, range 9.8–15.9 cm, P=0.016) (Figure 5).

Figure 5.

Comparison of stroke volume during synchronized LV pacing (adaptive LV) and at the echo-optimized settings at baseline and during exercise. Boxplot (25–75th percentile) and median values (open circles) showing the distribution of the LVOT-VTI during synchronized LV pacing and echo-optimized conventional BiV pacing, both at rest and during exercise. The adaptive LV shows a higher value compared with manual echo-optimized BiV pacing at peak exercise. BiV, biventricular; LVOT-VTI, left ventricular outflow tract velocity time integral.

Discussion

The main findings of our study on the validity of the aCRT algorithm during exercise can be summarized as follows: (1) the optimal AV delay progressively shortened with incremental exercise levels; (2) AV optimization during exercise could increase LV filling time; (3) the optimal VV delay varied from rest to exercise without a concordant correlation; (4) a significant correlation was shown between the aCRT and echo-optimization regarding stroke volume, both at rest and during exercise; and (5) synchronized LV-only pacing in CRT patients with normal AV conduction could provide a higher stroke volume during exercise as compared with manual-optimized conventional BiV pacing. In the present study, we showed the hemodynamic validity of the aCRT algorithm during exercise.

Effects of Optimization Performed During Exercise

Normal physiological adaptation to exercise leads to an increase in the heart rate, shortening of the AV node conduction time, and increased myocardial contractility, and therefore, generally results in a shorter AV interval at higher heart rates. Increasing the heart rate plays a major role in producing a cardiac output increase with exercise and the maximum cardiac output depends on the maximum heart rate. However, AV synchrony contributes to the cumulative effects for heart rates up to approximately 80 beats/min and increases the myocardial contractility dose above 80 beats/min in normal hearts.¹⁸ Accordingly, the significance of AV synchrony for maximizing cardiac performance is well described and is more pronounced in patients with systolic HF.^5,8,10,11 To date, several studies have shown that AV and VV optimization improves LV systolic performance even during exercise.^8,10,11 We demonstrated that the optimal AV delay progressively shortened during exercise, and similar findings have been reported by Sun et al,⁸ who showed an almost 40% reduction in the programmed AV delay during exercise. They also suggested that the magnitude of the change in the AV delay with increasing heart rates in CRT patients appeared to be significantly greater than the physiological exercise-induced change in the PR intervals observed in healthy subjects. Similarly, Shanmugam et al¹⁰ also reported that over 90% of patients with CRT showed a significant reduction (mean 50 ms) in the optimal AV delay during exercise and they demonstrated a corresponding increase in the LV filling time following AV optimization during exercise as well. Moreover, they successfully showed a 10% improvement in exercise capacity with a programmable rate-adaptive AV delay. Our findings concur with those reports and led us to believe that it may be beneficial to program the device to automatically adjust the AV delay according to the activity level.

Furthermore, we demonstrated significant variation in the optimal VV delay between rest and exercise. Unlike the AV delay, the optimal VV delay cannot be considered as correlated to the heart rate and a large variation can be seen between individual subjects. Notably, some patients developed ineffective LV pacing during exercise and this phenomenon may be found in the general population of patients with CRT. Valzania et al⁹ showed that the optimal VV delay varied considerably from rest to exercise and a re-assessment during exercise yielded an additional hemodynamic benefit. They also demonstrated the validity of their algorithm, which was based on the intracardiac electrogram method (QuickOpt^TM), as having a strong agreement with the echo-optimization during exercise regarding the LVOT-VTI. However, the Frequent Optimization Study Using the QuickOpt Method study (FREEDOM) did not show its superiority to conventional methods of optimization when the clinical composite score was compared.¹⁹ Moreover, the SMART-AV study that compared device-based, echo-based, and empirical programing of the AV delay did not show any significant changes in the LV volume and LVEF.⁷ Much evidence has suggested that the optimal AV and VV delay varies between rest and exercise, but it could still be challenging to discover an algorithm that can provide an adaptive AV/VV delay for general patients with CRT.

The aCRT algorithm can provide better clinical outcomes, or at least those as effective as comprehensive echo-optimized BiV pacing, through synchronized LV pacing and continuous optimization.^12,13 Importantly, this novel algorithm is based on clinical research conducted essentially during the resting phase,^20–23 and thus it has been unclear whether these results can be applied during exercise. As mentioned before, the major role of AV synchrony in contributing to the hemodynamics is during exercise with a heart rate of less than 100 beats/min,¹⁸ and the aCRT algorithm can only provide synchronized LV-only pacing when the heart rate is under 100 beats/min. Thus, our study protocol was designed to target a heart rate under 100 beats/min. In the present study, we provide new information regarding the validity of this algorithm even during low-intensity exercise. Through all the phases, a high correlation was shown between aCRT and echo-optimization regarding the stroke volume among patients with prolonged AV conduction. Furthermore, when normal AV conduction was observed, synchronized LV-only pacing could also provide a better stroke volume compared with echo-optimized BiV pacing during exercise. Synchronized LV-only pacing may prevent RV pacing-induced dyssynchrony and allow for simultaneous RV and LV electrical activation. A previous study on CRT in patients with LBBB and preserved RV conduction showed that the RV dP/dtmax was greater when CRT produced synchronized LV-only pacing.²⁴ Avoidance of RV pacing can provide superior RV function in patients with sufficiently preserved conduction to the RV. The results of our study indicated that aCRT offered a similar or greater stroke volume compared with real-time echo-optimization both at rest and during low-intensity exercise, which could be a possible explanation for the better clinical outcomes in some clinical trials of this novel algorithm.

Study Limitations

A potential weakness of this study was that it was conducted at a single center with a relatively small patient cohort and only included sinus rhythm patients with availability of the aCRT algorithm, which could expose the study to a myriad of biases, particularly selection bias and statistical power limitations. A higher prevalence of patients with prolonged AV conduction and moderate to severe HF could be considered as a limitation. Moreover, it is important to emphasize that the stroke volume is affected by multiple factors such as the cardiac pre- or afterload or exercise-induced mitral regurgitation, which make it difficult to clearly distinguish whether or not this additional hemodynamic contribution was only based on pacing settings. Also, it was unclear how the duration of exercise, passage of time, and order of the measurements from aCRT-off to aCRT-on affected the results. Finally, the study design did not intend to determine whether other hemodynamic benefits, excepted for stroke volume, would be observed when using the aCRT algorithm, as well as whether the same was true with much harder exercise or during tachycardia. Therefore, the current data need to be confirmed in a larger number of participants with a wider HF status and AV conduction range, and further studies are needed to evaluate the validity of the aCRT algorithm for other hemodynamic factors or during harder exercise.

Conclusions

Among the study patients with CRT, the optimal AV delay progressively shortened with incremental exercise levels. The aCRT algorithm is a novel device-based algorithm that has a significant correlation to echo-optimization regarding the stroke volume both at rest and during exercise. Once the patients developed a normal AV conduction following exercise, the aCRT algorithm provided an additional hemodynamic benefit through synchronized LV-only pacing.

Conflicts of Interest

Dr. Nakajima received a scholarship from the Japanese Heart Rhythm Society and received speaking honoraria from Medtronic Japan Co., Ltd.

References

• 1.   Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002; 346: 1845–1853.
• 2.   Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, et al; Cardiac Resynchronization-Heart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352: 1539–1549.
• 3.   Moss AJ, Hall WJ, Cannom DS, Klein H, Brown MW, Daubert JP, et al. Cardiac-resynchronization therapy for the prevention of heart-failure events. N Engl J Med 2009; 361: 1329–1338.
• 4.   Mullens W, Grimm RA, Verga T, Dresing T, Starling RC, Wilkoff BL, et al. Insight from a cardiac resynchronization optimization clinic as part of a heart failure disease management program. J Am Coll Cardiol 2009; 53: 765–773.
• 5.   Auricchio A, Stellbrink C, Block M, Sack S, Vogt J, Bakker P, et al. Effect of pacing chamber and atrioventricular delay on acute systolec function of paced patients with congestive heart failure. The Pacing Therapies for Congestive Heart Failure Study Group. The Guidant Congestive Heart Failure Research Group. Circulation 1999; 99: 2993–3001.
• 6.   Sawhney NS, Waggoner AD, Garhwal S, Chawla MK, Osborn J, Faddis MN. Randomized prospective trial of atrioventricular delay programing for cardiac resynchronizeation therapy. Heart Rhythm 2004; 1: 562–565.
• 7.   Ellenbogen KA, Gold MR, Meyer TE, Fernndez Lozano I, Mittal S, Waggoner AD, et al. Primary results from the SmartDelay determined AV optimization: A comparison to other AV delay methods used in cardiac resynchronization therapy (SMART-AV) trial: A randomized trial comparing empirical, echocardiography-guided, and algorithmic atrioventricular delay programming in cardiac resynchronization therapy. Circulation 2010; 122: 2660–2668.
• 8.   Sun JP, Lee AP, Grimm RA, Hung MJ, Yang XS, Delurgio D, et al. Optimisation of atrioventricular delay during exercise improves cardiac output in patients stabilised with cardiac resynchronisation therapy. Heart 2012; 98: 54–59.
• 9.   Valzania C, Eriksson MJ, Boriani G, Gadler F. Cardiac resynchronization therapy during rest and exercise: Comparison of two optimization methods. Europace 2008; 10: 1161–1169.
• 10.   Shanmugam N, Prada-Delgado O, Campos AG, Campos AG, Grimster A, Valencia O, et al. Rate-adaptive AV delay and exercise performance following cardiac resynchronization therapy. Heart Rhythm 2012; 9: 1815–1821.
• 11.   Grimm RA, Sun JP, Agler D, Fitzgerald B, Wilkoff B, Hilpisch K, et al. Shorter AV delay provide improved echocardiographic hemodynamics during exercise in patients receiving cardiac resynchronization therapy. Pacing Clin Electrophysiol 2009; 32: 457–465.
• 12.   Martin DO, Lemke B, Birnie D, Krum H, Lee KL, Aonuma K, et al. Investigation of a novel algorithm for synchronized left-ventricular pacing and ambulatory optimization of cardiac resynchronization therapy: Results of the adaptive CRT trial. Heart Rhythm 2012; 9: 1807–1814.
• 13.   Birnie D, Lemke B, Aonuma K, Kurm H, Lee KL, Gasparini M, et al. Clinical outcomes with synchronized left ventricular pacing: Analysis of the adaptive CRT trial. Heart Rhythm 2013; 10: 1368–1374.
• 14.   Birnie D, Hudnall H, Lemke B, Aonuma K, Lee KL, Gasparini M, et al. Continuous optimization of cardiac resynchronization therapy reduces atrial fibrillation in heart failure patients: Results of the Adaptive Cardiac Resynchronization Therapy Trial. Heart Rhythm 2017; 14: 1820–1825.
• 15.   Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA 3rd, Freedman RA, Gettes LS, et al. ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities-executive summary; a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2008; 51: 2085–2105.
• 16.   Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA 3rd, Freedman RA, Gettes LS, et al. 2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2013; 61: e6–e75.
• 17.   Kerlan JE, Sawhney NS, Waggoner AD, Chawla MK, Garhwal S, Osborn JL, et al. Prospective comparison of echocardiographic atrioventricular delay optimization methods for cardiac resynchronization therapy. Heart Rhythm 2006; 3: 148–154.
• 18.   Benditt DG, Milstein S, Buetikofer J, Gornick CC, Mianuli M, Fetter J. Senso-triggered, rate-variable cardiac pacing: Current technologies and clinical implications. Ann Intern Med 1987; 107: 714–724.
• 19.   Abraham WT, Gras D, Yu CM, Guzzo L, Gupta MS; FREEDOM Steering Committee. Rationale and design of a randomized clinical trial to assess the safety and efficacy of frequent optimization of cardiac resynchronization therapy: The Frequent Optimization Study Using the QuickOpt Method (FREEDOM) trial. Am Heart J 2010; 159: 944–948.
• 20.   van Gelder BM, Bracke FA, Meijer A, Pijls NH. The hemodynamic effect of intrinsic conduction during left ventricular pacing as compared to biventricular pacing. J Am Coll Cardiol 2005; 46: 2305–2310.
• 21.   Lee KL, Burnes JE, Mullen TJ, Hettrick DA, Tse HF, Lau CP. Avoidance of right ventricular pacing in cardiac resynchronization therapy improves right ventricular hemodynamics in heart failure patients. J Cardiovasc Electrophysiol 2007; 18: 497–504.
• 22.   Jones RC, Svinarich T, Rubin A, Levin V, Phang R, Murillo J, et al. Optimal atrioventricular delay in CRT patients can be approximated using surface electrocardiography and device electrogram. J Cardiovasc Electrophysiol 2010; 21: 1226–1232.
• 23.   Khaykin Y, Exner D, Birnie D, Sapp J, Aggarwal S, Sambelashvili A. Adjusting the timing of left-ventricular pacing using electrocardiogram and device electrograms. Europace 2011; 13: 1464–1470.
• 24.   Lee KL, Burnes JE, Mullen TJ, Hettrick DA, Tse HF, Lau CP. Avoidance of right ventricular pacing in cardiac resynchronization therapy improves right ventricular hemodynamics in heart failure patents. J Cardiovasc Electrophysiol 2007; 18: 497–504.