Skip main navigation

Personalized Rate-Response Programming Improves Exercise Tolerance After 6 Months in People With Cardiac Implantable Electronic Devices and Heart Failure

A Phase II Study
Originally published 2020;141:1693–1703



Heart failure with reduced ejection fraction (HFrEF) is characterized by blunting of the positive relationship between heart rate and left ventricular (LV) contractility known as the force-frequency relationship (FFR). We have previously described that tailoring the rate-response programming of cardiac implantable electronic devices in patients with HFrEF on the basis of individual noninvasive FFR data acutely improves exercise capacity. We aimed to examine whether using FFR data to tailor heart rate response in patients with HFrEF with cardiac implantable electronic devices favorably influences exercise capacity and LV function 6 months later.


We conducted a single-center, double-blind, randomized, parallel-group trial in patients with stable symptomatic HFrEF taking optimal guideline-directed medical therapy and with a cardiac implantable electronic device (cardiac resynchronization therapy or implantable cardioverter-defibrillator). Participants were randomized on a 1:1 basis between tailored rate-response programming on the basis of individual FFR data and conventional age-guided rate-response programming. The primary outcome measure was change in walk time on a treadmill walk test. Secondary outcomes included changes in LV systolic function, peak oxygen consumption, and quality of life.


We randomized 83 patients with a mean±SD age 74.6±8.7 years and LV ejection fraction 35.2±10.5. Mean change in exercise time at 6 months was 75.4 (95% CI, 23.4 to 127.5) seconds for FFR-guided rate-adaptive pacing and 3.1 (95% CI, −44.1 to 50.3) seconds for conventional settings (analysis of covariance; P=0.044 between groups) despite lower peak mean±SD heart rates (98.6±19.4 versus 112.0±20.3 beats per minute). FFR-guided heart rate settings had no adverse effect on LV structure or function, whereas conventional settings were associated with a reduction in LV ejection fraction.


In this phase II study, FFR-guided rate-response programming determined using a reproducible, noninvasive method appears to improve exercise time and limit changes to LV function in people with HFrEF and cardiac implantable electronic devices. Work is ongoing to confirm our findings in a multicenter setting and on longer-term clinical outcomes.


URL:; Unique identifier: NCT02964650.

Clinical Perspective

What Is New?

  • Rate-adaptive cardiac implantable electronic device programming taking into account the abnormal force-frequency relationship in patients with heart failure with reduced ejection fraction is associated with improved exercise time.

  • Standard age-related rate-adaptive programming might contribute to deteriorating left ventricular function.

What Are the Clinical Implications?

  • Out-of-the-box age-guided rate-adaptive pacing might be a suboptimal choice in patients with heart failure.

  • An assessment of the force-frequency relationship might be of clinical benefit in patients with heart failure by facilitating personalized rate-adaptive programming.


The hallmark of chronic heart failure with reduced ejection fraction (HFrEF) is impaired exercise tolerance: it is the most common presenting symptom, the basis for poor quality of life (QoL),1 and related to prognosis.2,3 Exercise intolerance in HFrEF is commonly thought to be exacerbated by limited heart rate (HR) rise.4 However, we have shown previously that HR rise programming in people with HFrEF and cardiac implantable electronic devices (CIEDs; cardiac resynchronization therapy or implantable defibrillators) using standa.rd age-related algorithms does not improve exercise capacity.5 This is likely the result of weakening of the usually close relationship among HR, left ventricular (LV) contractility, and stroke volume known as the force-frequency relationship (FFR).6 In HFrEF, this physiologic response is attenuated and characterized by a decline in LV contractility above a certain HR (the critical HR [CHR]).7 We have previously demonstrated that the FFR can be assessed reliably in people with HFrEF and CIEDs using echocardiography.8 In a randomized, controlled, crossover study, we showed that personalized HR programming, guided by these noninvasive FFR data, acutely improves exercise capacity in people with HFrEF, persistent symptoms, and a CIED.8

The present study aimed to explore whether device-based HR rise programming, tailored to an individual’s FFR, compared with conventional age-guided programming, is associated with improved exercise time in the longer term.


The data that support the findings of this study are available from the corresponding author on reasonable request.

Study Design

The study was a double-blind, single-center, randomized, controlled, parallel-group phase II trial comparing FFR-guided rate-adaptive programming with standard age-related rate-adaptive programming in patients with HFrEF taking optimal guideline-directed medical therapy and implanted with a CIED (cardiac resynchronization therapy [CRT] or implantable cardioverter-defibrillator) according to international guidelines. The primary outcome of this study was change in treadmill exercise walk time at 6 months from baseline, with key secondary outcomes of change in peak oxygen consumption, cardiac function, and QoL. Patients were eligible if they had stable (>3 months) symptomatic HFrEF, were willing and able to provide written consent and walk on a treadmill, and were prepared to complete QoL questionnaires. Patients were ineligible if they had substantial cognitive impairment, had a life expectancy <6 months, had angina pectoris limiting exercise tolerance, or were taking calcium channel blockers.

Baseline Study Procedures


Study procedures have been described in detail previously.8 Briefly, from June 22, 2017, all participants were invited to the Cardiovascular Clinical Research Facility at Leeds Teaching Hospitals National Health Service Trust, United Kingdom, and underwent full baseline echocardiography with 2-dimensional grayscale and tissue Doppler images recorded in 2- and 4-chamber views at resting HRs and at each 15 beats per minute increase during an incremental pacing protocol. Images were stored in the Echopac digital imaging system and analyzed offline (GE, Milwaukee, WI). This analysis included a calculation of LV end-diastolic and end-systolic volumes using the biplane discs (modified Simpson) method.9 A mean of 3 measurements was used in the final analysis. The frame at the R-wave was taken as end diastole, and the frame with the smallest LV cavity was taken as end systole. The LV end-systolic volume index was calculated at each stage as end-systolic volume/body surface area, where body surface area was calculated using the Mosteller equation.10

Blood pressure measurement

Calculation of the end-systolic pressure–volume relation requires measurement of the LV pressure at end-systole.11 Systolic blood pressure (SBP) measured using a manual blood pressure cuff was used as a surrogate for end-systolic LV pressure. Blood pressure recordings were made using a sphygmomanometer and a standard stethoscope to coincide with echocardiographic images at each HR stage. SBP was recorded at the point where the first tapping sound (phase 1 Korotkoff) occurred for 2 consecutive beats.12

Pacing protocol

Echocardiographic images were collected at rest with intrinsic atrial rhythm or a base rate of 40 beats per minute, after which atrial pacing was initiated in the DDD mode (or VVI in patients with atrial fibrillation) for patients with CRT and AAI mode (or DDD with long atrioventricular delays to avoid right ventricular pacing, or VVI for those in atrial fibrillation) for patients without CRT, at the lowest multiple of 10 above baseline. After 4 minutes, another set of echocardiographic images was recorded, and subsequently the pacing rate increased in stepwise intervals of 15 beats per minute with images recorded after every 4 minutes. This was repeated until the maximum HR predicted by the Åstrand equation (220 − age) was reached. At this point, peak data were collected, and pacing was returned to baseline settings. For safety, participants were asked to remain in the research facility for 30 minutes after testing.

Force frequency calculation

Dividing the SBP by the LV end-systolic volume index provides a surrogate of contractility,13–15 which has been validated versus invasive methods.13,15 Our protocol allowed us to plot the FFR for each participant. We defined the CHR as the HR at which, in a biphasic pattern, the SBP/LV end-systolic volume index reached maximum value or beyond which the SBP/LV end-systolic volume index declined by 5%. In case of negative FFR (in cases where there was no increase in contractility with increments in HR), baseline HR was deemed the CHR.16

Laboratory arrangement and exercise protocol

Patients exercised using the ramping treadmill protocol.17 Expired air was collected and metabolic gas exchange analysis performed (Ultima CardO2, Medgraphics, St Paul, MN) throughout the test. HR (beats per minute), oxygen uptake (Vo2; mL/kg/min), and carbon dioxide output (Vco2; mL/kg/min) were recorded as 15-second averages. Anaerobic threshold was calculated using the V-slope method. Stroke volume can be estimated during cardiopulmonary exercise testing through the calculation of an exercise oxygen pulse (O2/HR [units = milliliters of O2 per beat]).18

Cardiopulmonary exercise testing equipment was recalibrated using manufacturer-recommended volume and gas calibration techniques before each test. All participants were encouraged to exercise to exhaustion and no motivation or instructions were given. The arrangement of the laboratory to ensure double blinding has been described previously.5,8 To maintain blinding, the continuous 12-lead ECG monitor was obscured throughout the test and recovery phase from participants and the supervising physician. Only the unblinded cardiac physiologist was aware of the programming mode or testing arm. The cardiac physiologist monitored the ECG throughout the study and communicated with the other team members only if there were safety concerns. The effective delivery of biventricular stimulation in patients with CRT was confirmed from electrocardiographic traces at peak exercise at both time points by absence, fusion, or other QRS morphology changes.

Quality of life

At baseline and 6-month follow-up, participants were asked to complete 3 QoL assessments: the Minnesota Living With Heart Failure Questionnaire, the EQ-5D, and a visual analogue scale of overall QoL.19


After baseline testing, patients were allocated randomly, using random number generation, to 1 of 2 groups, and their device was programmed by the unblinded cardiac physiologist to either conventional age-related rate-adaptive pacing20 or rate-adaptive pacing guided by FFR assessment, specifically limiting the upper sensor rate to the CHR. Atrioventricular delay programming was optimized to avoid fusion and maintain consistent biventricular stimulation at higher HRs in patients with CRT devices, and device-specific pacing avoidance algorithms were activated in patients in sinus rhythm without CRT devices. VVIR programming was the default for patients in atrial fibrillation. Patients were blinded to their allocation.


Patients were telephoned at 1 week to assess safety and were invited back at 6 months for repeat cardiopulmonary exercise testing, transthoracic echocardiography, and QoL assessment.

Primary Outcome Measure

Treadmill walk time was our primary outcome measure. Exercise incapacity is the key driver of impaired QoL in HF,1,21 and treadmill walk time is a patient-oriented outcome of direct clinical relevance that can be converted to distance easily. In a previous randomized clinical trial, we observed significant variability in 6-minute corridor walk testing,22 and therefore elected to use treadmill walk time as our primary outcome measure for the present study.

Sample Size

The trial was designed as a single-center phase II trial. Because there was an absence of data describing variability in outcomes, a robust definitive trial could not be designed. This trial aimed to make an initial unbiased comparison of groups and inform variability in outcomes in the target population of patients. As such, the target sample size was based on achieving a sample size appropriate to estimate variability in the 6-month primary outcome measure according to published guidance for phase II trials.23 We aimed to have outcome data at 6 months for a minimum of 70 patients and the recruitment target was inflated to 85 patients to account for anticipated dropout.

Statistical Analyses

Analyses followed a predefined plan as set out in the trial protocol (URL:; Unique identifier: NCT02964650). As a phase II trial, the aim was to describe outcomes and variability in outcomes. Normality for continuous variables was explored visually by distribution plots, tested using the Shapiro-Wilk test, and skewness and kurtosis levels were confirmed at <1 for all key variables. After testing for normality, continuous baseline characteristics were reported as mean±SD. Analysis of covariance was used to assess intergroup differences in outcome variables. All across–treatment group comparisons were 2-sided and presented as mean change (95% CI). Exploratory post hoc univariable analyses of candidate variables against the primary end point as predictors of change were undertaken with the aim of adding all variables with a P value <0.10 to a multivariable model. Data were analyzed using the Statistical Package for the Social Sciences (SPSS v.23; IBM).

Ethical Considerations

After ethical review by East Midlands–Derby Research Ethics Committee, the trial was approved by the Health Research Authority of the United Kingdom (17/EM/0004). Written informed consent was obtained from all participants. The trial was registered prospectively (URL:; Unique identifier: NCT02964650).


A total of 83 patients were recruited between November 2, 2017, and January 16, 2019: 38 randomized to FFR-guided rate-adaptive programming and 45 to conventional age-guided programming (Figure 1). The 2 randomized groups were balanced for important baseline variables, including FFR variables (CHR [beats per minute] and peak contractility) and exercise testing variables (resting HR [65.9±9.4 versus 66.4±9.1] and peak exercise HR [116.2±21.8 versus 120.4±19.5]; Table 1).

Table 1. Patient Demographic and Baseline Characteristics at Randomization: Intention-to-Treat Population

CharacteristicTotal (n=83)Tailored (n=38)Conventional (n=45)
Male sex58 (71)27 (73)31 (69)
Age, y74.6±8.773.7±10.675.4±6.8
Ischemic heart disease52 (63)26 (68)26 (58)
Atrial fibrillation30 (36)13 (34)17 (38)
Hypertension39 (47)16 (42)23 (51)
Creatinine, μmol/L108.3±37.2107.7±27.5108.9±45.5
New York Heart Association class
 II58 (70)27 (71)31 (69)
 III25 (30)11 (29)14 (31)
 β-blockers80 (96)38 (100)42 (93)
 Bisoprolol equivalent dose, mg/d*7.4 (3.8)7.7 (3.2)7.0 (4.2)
 Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker78 (94)34 (90)37 (82)
 Ramipril equivalent dose, mg/d*6.2 (3.5)6.3 (3.3)6.2 (3.6)
 Loop diuretic50 (60)24 (63)26 (58)
 Statin69 (83)30 (79)39 (87)
Device allocation
 Cardiac resynchronization therapy with defibrillator35 (42)13 (34)22 (49)
 Cardiac resynchronization therapy with pacemaker23 (28)9 (24)14 (31)
 Dual-chamber implantable cardioverter-defibrillator25 (30)16 (42)9 (20)
Resting hemodynamics
 Resting heart rate, beats per minute66.1±9.265.9±9.466.4±9.1
 Systolic blood pressure, mm Hg123.8±20.5122.0±21.4125.2±19.9
Echocardiography and force-frequency relationship data
 Left ventricular ejection fraction, %35.2±10.535.3±10.635.1±10.6
 Left ventricular end-diastolic volume, mL142.7±62.5141.2±73.0144.0±52.8
 Left ventricular end-systolic volume, mL95.9±62.595.7±60.796.1±45.1
 Peak contractility4.4±2.84.6±3.14.2±2.5
 Critical heart rate, beats per minute104.6±17.8106.7±18.0102.8±17.7
Exercise test results
 Treadmill walk time, s397.1±185.9376.4±172.1414.9±197.2
 Peak oxygen consumption, mL/kg/min15.5±5.615.1±6.415.9±4.9
 VE/Vco2 slope32.7±9.632.3±9.433.2±9.8
 Peak exercise heart rate, beats per minute118.5±20.6116.2±21.8120.4±19.5
 O2 pulse, mL/beat10.89±3.6910.46±3.6711.28±3.71
Quality of life scores at baseline
 Visual analogue scale66.4±16.264.7±17.267.8±15.1
 Minnesota Living With Heart Failure Questionnaire31.1±19.732.8±18.729.7±20.5

Continuous variables expressed as mean±SD and categorical variables as n (%). EQ-5D-5L indicates 5-level EQ-5D; and VE/Vco2 slope, slope relating ventilation and carbon dioxide output.

*Mean doses calculated as described previously.31

Figure 1.

Figure 1. CONSORT (Consolidated Standards of Reporting Trials) diagram demonstrating patient enrollment, randomization, and disposition during the study.

Of the 83 patients enrolled, 38 were allocated to FFR-guided HR rise programming and 45 to conventional age-guided programming. There were 3 patients in each group who did not tolerate the intervention. These patients were reprogrammed to their original settings and remained in the intention-to-treat analysis (Figure 1).24 Six-month follow-up data were available for 69 patients. Of the 14 patients who were not reassessed, 1 had died and 13 declined to attend or were lost to follow-up. Of those who attended, 3 declined to undergo a cardiopulmonary stress test.

Device interrogation at follow-up confirmed high mean±SD rates of biventricular pacing in both groups at baseline (98.12±2.07% versus 98.26±1.73%) and follow-up (97.80±2.17% versus 98.13±2.18%), with no difference in change between baseline and follow-up between the FFR-guided group (mean, 0.37 [95% CI, −0.58 to 0.26]) and the age-guided group (mean, −0.31 [95% CI, −0.34 to 0.71]). Patients allocated to FFR-guided rate-adaptive programming had lower mean±SD peak HRs during the follow-up exercise test (98.6±19.4 versus 112.0±20.3 beats per minute).

Review of the baseline and follow-up exercise ECG traces confirmed that effective CRT was delivered without fusion in patients with CRT devices. In patients without CRT devices, the percentage of right ventricular pacing was not different between the 2 groups at baseline or follow-up, and there was no across–randomized group change in right ventricular pacing percentage between baseline and follow-up within the FFR-guided (mean, 1.36 [95% CI, −3.3 to 6.1]) and standard age-guided (mean, −1.81 [95% CI, −8.4 to 4.8]) groups.

Primary Outcome Measure

At 6 months, patients allocated to FFR-guided programming experienced a greater improvement in treadmill walk time compared with patients randomized to conventional age-guided programming. Changes in the FFR-guided arm versus the conventional arm from baseline to 6 months were as follows: mean±SD 376±172 to 468±252 seconds versus 414±197 to 423±217 seconds, respectively (Figure 2). The mean difference between groups on 2-sample analysis of covariance was 72.3 seconds (95% CI, 2.0–142.7 seconds; P=0.044, test statistic=4) in favor of FFR-guided programming (Table 2), an increase of 20% (or around 54 meters) from baseline. In the post hoc univariable analysis of predictors of change, no baseline variable reached a P value <0.10 (data not shown).

Table 2. Change in Primary and Secondary Outcome Variables After 6 Months of Tailored Versus Conventional Pacemaker Heart Rate Rise Programming: Intention-to-Treat Population

End PointFinal Value, Mean (95% CI)Change After 6 Months, Mean (95% CI)Analysis of Covariance Difference, Mean Change (95% CI)
Primary outcome
 Treadmill walk time, s
483.15 (431.10, 535.20)75.40 (23.35, 127.45)72.31 (1.94, 142.67)*
401.84 (363.62, 458.07)3.09 (−44.14, 50.31)
Secondary outcomes
 Left ventricular ejection fraction, %
  Tailored35.79 (33.80, 37.78)−0.23 (−2.31,1.84)3.46 (0.61, 6.30)*
  Conventional32.13 (30.28, 33.00)−3.69 (−5.62, −1.76)
 Left ventricular end-diastolic volume, mL
  Tailored146.63 (135.87, 157.39)3.24 (7.52, 14.00)−2.29 (−16.99, 12.42)
  Conventional148.92 (138.91, 158.92)5.52 (−4.48, 15.53)
 Left ventricular end-systolic volume, mL
  Tailored97.36 (89.36, 105.35)1.78 (−6.22, 9.77)−8.26 (−19.18, 2.67)
  Conventional105.61 (98.18, 113.05)10.03 (2.60, 17.47)
 pVO2, mL/kg/min
  Tailored16.72 (15.81, 17.64)0.84 (−0.07, 1.75)1.14 (−0.10, 2.38)
  Conventional15.59 (14.75, 16.43)−0.30 (−1.14, 0.54)
 Peak O2 pulse, mL/beat
  Tailored13.84 (12.75, 14.94)2.74 (1.64, 3.83)2.02 (0.53, 3.51)*
  Conventional11.82 (10.82, 12.83)0.72 (−0.29, 1.73)
 VE/Vco2 slope
  Tailored32.81 (30.30, 35.32)−0.73 (−3.24, 1.78)0.50 (−2.91, 3.91)
  Conventional32.31 (30.00, 34.62)−1.23 (−3.54, 1.08)
0.73 (0.68, 0.78)−0.02 (−0.07, 0.04)0.04 (−0.04, 0.12)
0.66 (0.62, 0.71)−0.06 (−1.11, −0.01)
  Tailored67.13 (62.18, 72.07)−1.16 (−6.72, 4.40)2.19 (−5.44, 9.82)
  Conventional62.84 (58.28, 67.41)−3.35 (−8.56, 1.86)
 Minnesota Living With Heart Failure Questionnaire
  Tailored31.68 (27.45, 35.91)−0.04 (−4.33, 4.26)0.59 (−5.32, 6.49)
  Conventional30.25 (26.34, 34.15)−0.63 (−4.65, 3.40)

EQ-5D-5L indicates 5-level EQ-5D; EQ-VAS, EQ–visual analogue scale; pVo2, peak oxygen consumption; and VE/Vco2 slope, slope relating ventilation and carbon dioxide output.

*95% Significance.

Figure 2.

Figure 2. Change in treadmill walk time after 6 months of conventional versus force-frequency relationship–guided rate-adaptive pacing programming, presented as mean (95% CI).

Secondary Outcome Measures

Six months of FFR-guided rate-adaptive pacing was not associated with a change in LV ejection fraction (LVEF; mean±SD 35.3%±10.6 at baseline and 35.9%±10.6 at follow-up). Patients allocated conventional programming experienced a reduction in LVEF from 35.1%±10.6% to 32.1%±11.7% with a mean difference between groups of 3.7% (95% CI, 0.9–6.4; P=0.009) in favor of FFR-guided programming (Table 2 and Figure 3). Mean difference between randomized groups on 2-sample analysis of covariance was 3.5% (95% CI, 0.6–6.3%) in favor of FFR-guided programming (Table 2, Figure 3). There was no significant change in LV end-diastolic volume in either randomized group, with changes from baseline of 3.2 mL (95% CI, −7.5 to 14.0 mL) and 5.5 mL (95% CI, −4.5 to 14.0 mL; Table 2) in the FFR-guided and conventional groups, respectively. There was also no significant change in LV end-systolic volume from baseline to follow-up in the FFR-guided group (1.8 mL [95% CI, −6.2 to 9.7 mL]) or the conventional group (10.0 mL [95% CI, 2.6 to 17.5 mL]; Table 2 and Figure 4A and 4B).

Figure 3.

Figure 3. Change in left ventricular ejection fraction after 6 months of conventional versus force-frequency relationship–guided rate-adaptive pacing programming, presented as mean (95% CI).

Figure 4.

Figure 4. Change in left ventricular end-diastolic and end-systolic volumes. Change in left ventricular end-diastolic volumes (A) and left ventricular end-systolic volumes (B) after 6 months of conventional versus force-frequency relationship–guided rate-adaptive pacing programming, presented as mean (95% CI).

In the group randomized to FFR-guided rate adaptation, there was a trend to favorable changes in cardiopulmonary exercise testing variables from baseline compared with the conventional programming group, with a mean difference between groups of 1.14 mL/kg/min (95% CI, −0.1 to 2.4 mL/kg/min). Peak O2 pulse increased in the FFR-guided group, with a mean difference between the groups of 2.02 mL/beat (95% CI, 0.5 to 3.5 mL/beat; Table 2).

There was no difference in change in QoL observed over the follow-up period between the randomized groups. Mean change from baseline in EQ-5D score was −0.02 (95% CI, −0.07 to 0.04) and −0.06 (95% CI, −1.11 to −0.01) for the FFR-guided and conventional groups, respectively. Mean change in visual analogue scale score from baseline was −1.16 (95% CI, −6.72 to 4.40) and −3.35 (95% CI, −8.56 to 1.86) for the FFR-guided versus conventionally programmed groups. Mean change in baseline in Minnesota Living With Heart Failure questionnaire score was −0.04 (95% CI, −4.33 to 4.26) for the FFR-guided group and −0.63 (95% CI, −4.65 to 3.40) for the conventionally programmed group (Table 2).


Although disease-modifying treatments have substantially improved life expectancy in HFrEF,24 they have had much less of an effect on exercise capacity, the main driver of QoL and a key target for patients.1,21 Millions of patients with HFrEF worldwide have a persistently poor QoL and are unable to undertake activities of daily living. We have shown an acute improvement in exercise time with FFR-guided rate-adaptive pacing previously. The present study aimed to explore whether long-term tailored HR programming could improve exercise time in patients with HFrEF receiving guideline-directed optimal medical and device therapy.

The key results of the present trial are that in people with HFrEF receiving optimal guideline-directed medical and device therapy, 6 months of tailored FFR-guided HR programming leads to improved exercise time and prevents the decline in LVEF seen in patients randomized to conventional age-related programming.

HR Rise and Exercise Capacity

Cardiac output is a function of HR and stroke volume. Poor HR rise during exercise could contribute to exercise intolerance by affecting cardiac output adversely. The relationship among HR rise, cardiac output, and exercise capacity, and how their interaction changes with age, sex, fitness, and disease, is poorly understood. For example, despite >80 years of research, there is no clear target HR in healthy adults.25–27 Data sets are small and of limited generalizability.28 The most frequently quoted Åstrand formula (220 − age) is the benchmark by which a diagnosis of poor HR rise (chronotropic incompetence) is made.

HR Rise and Device Therapy in HFrEF

CIEDs have a programming option that can detect movement or increased ventilation and increase HR accordingly, known as rate-adaptive pacing. In patients without HF, compared with fixed rate programming (rate-adaptive option deactivated), rate-adaptive programming increases cardiac output during exercise29 and improves QoL30,31 but improves exercise capacity inconsistently.32

The situation is unclear in HFrEF, where HR limitation is a cornerstone of therapy. Even without β-blockers, patients with HFrEF frequently fail to achieve their age-predicted maximal HR during exercise.3,33,34 This is commonly perceived to contribute to reduced exercise tolerance.35,36 This paradox, where HR limitation using β-blockers reduces hospitalization and mortality37,38 but is proposed to exacerbate exercise intolerance, possibly contributes to poorly defined HR targets in guidelines39 and low rates of achievement of optimal β-blocker doses,40,41 even when a CIED could provide HR support,42 despite overwhelming data on their dose-related prognostic benefit.43

Treatment for 30 to 40% of people with HFrEF includes a CIED. Rate-adaptive pacing during exercise through an age-related algorithm in HFrEF unreliably improves exercise capacity.44 Our previous work and that of others suggests that higher HRs caused by imprecise rate-adaptive pacing could be disadvantageous.45,46 Despite those findings, proven benefits of HR limitation, and a limited evidence base to guide when rate-adaptive pacing should be used, the standard age-related rate-adaptive algorithm is active in >68% of 210 000 CRT devices in the United States.47

Cardiac Contractility During Exercise

HR contributes to cardiac output, a key determinant of exercise capacity.48 In the healthy state, cardiac output is positively coupled with LV contractility (the power of contraction) by the FFR.7 The FFR ensures that LV contractility and thereby stroke volume increase with HR to compensate for reduced filling time. We and others have shown that this critical physiologic response is flattened in patients with HFrEF, with a decline in LV contractility above a certain HR.8,15 In HFrEF, therefore, the close relationship between HR and stroke volume is perturbed such that maximal HR is not synonymous with optimal HR.

In line with this result, we showed that programming CIEDs to increase the HR to the standard age-related maximal HR does not improve exercise time.5 We hypothesized that this was because conventional rate-adaptive algorithms do not take into account the altered FFR in HFrEF, and that there might be an HR beyond which there is lower contractility, and therefore reduced cardiac output. Using echocardiography, we confirmed an abnormal FFR with impaired contractility across the entire HR range in HFrEF, with a lower slope in response to HR increases, lower peak contractility, and a lower HR for peak contractility (the CHR), above which contractility worsens.8 We were able to determine an ideal range of HR over which contractility increased in all patients.

In an unselected group of patients with HFrEF, we then carried out a randomized, double-blind, crossover study of acute programming of the rate-adaptive algorithm based on an individual’s FFR (limiting HR rise to below the CHR) versus standard settings (increasing to age-predicted maximum). We found an acute improvement in mean exercise time despite a much lower peak HR. FFR-guided programming was also associated with greater exercise time than fixed-rate pacing (rate-adaptive algorithm programmed off).8 That crossover study informed the design of the current study.


This study was designed as a single-center phase II trial and the data should be interpreted in light of this. However, participants were approached consecutively, with no selection bias in terms of response to CRT. For example, it is possible that so-called responders to CRT may have less to gain from tailored rate-adaptive programming, but our selection criteria were deliberately inclusive, requiring ongoing symptomatic heart failure. The trial was randomized and undertaken in a double-blind fashion to reduce bias as described previously.5,8 Our approach was informed by results from our randomized, placebo-controlled crossover study in 52 patients showing a favorable effect of FFR-guided HR rise programming on exercise capacity.8

Whereas the changes we observed in LVEF are smaller than the published patient-level measurement error of echocardiography, it is possible to detect smaller margins confidently by studying groups of people. Scans were analyzed offline by an experienced operator blinded to treatment and time point. Any systematic error in measurement will have been accounted for by our randomized design. The difference in change in LVEF was statistically significant by our predefined criteria. If persistent, a reduction in LVEF in the standard-care arm of this magnitude is likely to be associated with an adverse outcome.49

On the basis of our previous work, we a priori chose exercise time as our primary outcome as one of direct patient relevance. Our study was not powered to detect differences in measures of QoL and we did not measure concentrations of BNP (B-type natriuretic peptide) at baseline or follow-up. Future mechanistic work could include changes in neurohormones and sympathetic activation and the effects of higher HRs on stroke volume and LV perfusion.

Fewer than half of the patients in this study had participated previously in a double-blind crossover study of FFR-guided HR programming versus standard adaptive rate-response programming.8 In that study, patients’ CIEDs were programmed according to their FFR or using standard age-related settings only for the duration of the visit in random order. Between the 2 study visits, which were a week apart, the device was programmed to usual care. These 2 exercise tests are unlikely to stimulate a training effect and the last patient in the acute crossover study completed follow-up 7 months before the first patient was enrolled into the present study. Patients in the current study were allocated randomly to the intervention and the primary outcome of the study is change in exercise time. There is no potential for a carryover or training effect that could influence the present results.

Our approach to establish the FFR in patients with CIEDs requires skills in echocardiography and CIED programming and is likely to add time to echocardiographic-based optimization of CRT devices.


In a novel, single-center, randomized, double-blind, controlled trial, we have shown for the first time that CIED HR rise programming guided by noninvasive FFR data appears to be associated with improved exercise time and reduced progressive deterioration in LV function. Future work is planned to explore the mechanisms of our finding, to determine whether the greater increase in exercise time than peak oxygen consumption represents greater efficiency, to determine whether the CHR can be predicted from clinical variables, and to confirm our findings in a multicenter setting and on longer-term clinical outcomes including hospitalization. Future research will need to explore whether patients with less severe LV systolic dysfunction and those with persistent symptoms but no existing indication for a CIED might also benefit from this approach.


*Dr Gierula and J.E. Lowry contributed equally.

Sources of Funding, see page 1701

Continuing medical education (CME) credit is available for this article. Go to to take the quiz.

Klaus K. Witte, MD, Leeds Institute of Cardiovascular and Metabolic Medicine, LIGHT Building, University of Leeds, Clarendon Way, Leeds, UK LS2 9JT. Email


  • 1. Hobbs FD, Kenkre JE, Roalfe AK, Davis RC, Hare R, Davies MK. Impact of heart failure and left ventricular systolic dysfunction on quality of life: a cross-sectional study comparing common chronic cardiac and medical disorders and a representative adult population.Eur Heart J. 2002; 23:1867–1876. doi: 10.1053/euhj.2002.3255CrossrefMedlineGoogle Scholar
  • 2. Hsich E, Gorodeski EZ, Starling RC, Blackstone EH, Ishwaran H, Lauer MS. Importance of treadmill exercise time as an initial prognostic screening tool in patients with systolic left ventricular dysfunction.Circulation. 2009; 119:3189–3197. doi: 10.1161/CIRCULATIONAHA.109.848382LinkGoogle Scholar
  • 3. Al-Najjar Y, Witte KK, Clark AL. Chronotropic incompetence and survival in chronic heart failure.Int J Cardiol. 2012; 157:48–52. doi: 10.1016/j.ijcard.2010.11.018CrossrefMedlineGoogle Scholar
  • 4. Brubaker PH, Kitzman DW. Chronotropy: the Cinderella of heart failure pathophysiology and management.J ACC Heart Fail. 2013; 1:267–269. doi: 10.1016/j.jchf.2013.03.009CrossrefMedlineGoogle Scholar
  • 5. Jamil HA, Gierula J, Paton MF, Byrom R, Lowry JE, Cubbon RM, Cairns DA, Kearney MT, Witte KK. Chronotropic incompetence does not limit exercise capacity in chronic heart failure.J Am Coll Cardiol. 2016; 67:1885–1896. doi: 10.1016/j.jacc.2016.02.042CrossrefMedlineGoogle Scholar
  • 6. Bowditch HP. Uber die Eigenthϋmlichkeiten der Reizbarkeit, welche die Muskelfasern des Herzens zeigen.Ber Sachs Ges (Akad) Wiss. 1871; 23:652–689.Google Scholar
  • 7. Cotton JM, Kearney MT, MacCarthy PA, Grocott-Mason RM, McClean DR, Heymes C, Richardson PJ, Shah AM. Effects of nitric oxide synthase inhibition on basal function and the force-frequency relationship in the normal and failing human heart in vivo.Circulation. 2001; 104:2318–2323. doi: 10.1161/hc4401.098515LinkGoogle Scholar
  • 8. Gierula J, Paton MF, Lowry JE, Jamil HA, Byrom R, Drozd M, Garnham JO, Cubbon RM, Cairns DA, Kearney MT, et al. Rate-adaptive programming tailored to the force-frequency relationship improves exercise tolerance in chronic heart failure.JACC Heart Fail. 2018; 6:105–113. doi: 10.1016/j.jchf.2017.09.018CrossrefMedlineGoogle Scholar
  • 9. Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, Flachskampf FA, Foster E, Goldstein SA, Kuznetsova T, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.J Am Soc Echocardiogr. 2015; 28:1–39.e14. doi: 10.1016/j.echo.2014.10.003CrossrefMedlineGoogle Scholar
  • 10. Mosteller RD. Simplified calculation of body-surface area.N Engl J Med. 1987; 317:1098. doi: 10.1056/NEJM198710223171717CrossrefMedlineGoogle Scholar
  • 11. Grossman W, Braunwald E, Mann T, McLaurin LP, Green LH. Contractile state of the left ventricle in man as evaluated from end-systolic pressure-volume relations.Circulation. 1977; 56:845–852. doi: 10.1161/01.cir.56.5.845LinkGoogle Scholar
  • 12. Nutter D. Measuring and recording systemic blood pressure.Hurst JW, Logue RB, Schlant R, Wenger NK, eds. In: The Heart. 4th ed. New York: McGraw-Hill; 1978:220–222.Google Scholar
  • 13. Bombardini T, Agrusta M, Natsvlishvili N, Solimene F, Pap R, Coltorti F, Varga A, Mottola G, Picano E. Noninvasive assessment of left ventricular contractility by pacemaker stress echocardiography.Eur J Heart Fail. 2005; 7:173–181. doi: 10.1016/j.ejheart.2004.04.019CrossrefMedlineGoogle Scholar
  • 14. Ginzton LE, Laks MM, Brizendine M, Conant R, Mena I. Noninvasive measurement of the rest and exercise peak systolic pressure/end-systolic volume ratio: a sensitive two-dimensional echocardiographic indicator of left ventricular function.J Am Coll Cardiol. 1984; 4:509–516. doi: 10.1016/s0735-1097(84)80094-7CrossrefMedlineGoogle Scholar
  • 15. Bombardini T, Correia MJ, Cicerone C, Agricola E, Ripoli A, Picano E. Force-frequency relationship in the echocardiography laboratory: a noninvasive assessment of Bowditch treppe?J Am Soc Echocardiogr. 2003; 16:646–655. doi: 10.1016/s0894-7317(03)00221-9CrossrefMedlineGoogle Scholar
  • 16. Inagaki M, Yokota M, Izawa H, Ishiki R, Nagata K, Iwase M, Yamada Y, Koide M, Sobue T. Impaired force-frequency relations in patients with hypertensive left ventricular hypertrophy: a possible physiological marker of the transition from physiological to pathological hypertrophy.Circulation. 1999; 99:1822–1830. doi: 10.1161/01.cir.99.14.1822LinkGoogle Scholar
  • 17. Porszasz J, Casaburi R, Somfay A, Woodhouse LJ, Whipp BJ. A treadmill ramp protocol using simultaneous changes in speed and grade.Med Sci Sports Exerc. 2003; 35:1596–1603. doi: 10.1249/01.MSS.0000084593.56786.DACrossrefMedlineGoogle Scholar
  • 18. Whipp BJ, Higgenbotham MB, Cobb FC. Estimating exercise stroke volume from asymptotic oxygen pulse in humans.J Appl Physiol (1985). 1996; 81:2674–2679. doi: 10.1152/jappl.1996.81.6.2674CrossrefMedlineGoogle Scholar
  • 19. Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol Group.Ann Med. 2001; 33:337–343. doi: 10.3109/07853890109002087CrossrefMedlineGoogle Scholar
  • 20. Fox SM, Naughton JP, Haskell WL. Physical activity and the prevention of coronary heart disease.Ann Clin Res. 1971; 3:404–432.MedlineGoogle Scholar
  • 21. Kitzman DW. Exercise intolerance.Prog Cardiovasc Dis. 2005; 47:367–379. doi: 10.1016/j.pcad.2005.02.002CrossrefMedlineGoogle Scholar
  • 22. Witte KK, Byrom R, Gierula J, Paton MF, Jamil HA, Lowry JE, Gillott RG, Barnes SA, Chumun H, Kearney LC, et al. Effects of vitamin D on cardiac function in patients with chronic HF: the VINDICATE study.J Am Coll Cardiol. 2016; 67:2593–2603. doi: 10.1016/j.jacc.2016.03.508CrossrefMedlineGoogle Scholar
  • 23. Teare MD, Dimairo M, Shephard N, Hayman A, Whitehead A, Walters SJ. Sample size requirements to estimate key design parameters from external pilot randomised controlled trials: a simulation study.Trials. 2014; 15:264. doi: 10.1186/1745-6215-15-264CrossrefMedlineGoogle Scholar
  • 24. Cubbon RM, Gale CP, Kearney LC, Schechter CB, Brooksby WP, Nolan J, Fox KA, Rajwani A, Baig W, Groves D, et al. Changing characteristics and mode of death associated with chronic heart failure caused by left ventricular systolic dysfunction: a study across therapeutic eras.Circ Heart Fail. 2011; 4:396–403. doi: 10.1161/CIRCHEARTFAILURE.110.959882LinkGoogle Scholar
  • 25. Robinson S. Experimental studies of physical fitness in relation to age.Arbeitsphysiol. 1938; 10:251–323.Google Scholar
  • 26. Åstrand P. Experimental Studies of Physical Working Capacity in Relation to Sex & Age. Copenhagen: Musksgaard; 1952.Google Scholar
  • 27. Astrand I, Astrand PO, Hallbäck I, Kilbom A. Reduction in maximal oxygen uptake with age.J Appl Physiol. 1973; 35:649–654. doi: 10.1152/jappl.1973.35.5.649CrossrefMedlineGoogle Scholar
  • 28. Robergs RA, Landwehr R. The surprising history of the “HRmax=220-age” equation.J Exerc Physiol Online. 2002; 5:1–10.Google Scholar
  • 29. McMeekin JD, Lautner D, Hanson S, Gulamhusein SS. Importance of heart rate response during exercise in patients using atrioventricular synchronous and ventricular pacemakers.Pacing Clin Electrophysiol. 1990; 13:59–68. doi: 10.1111/j.1540-8159.1990.tb02004.xCrossrefMedlineGoogle Scholar
  • 30. Lau CP, Rushby J, Leigh-Jones M, Tam CY, Poloniecki J, Ingram A, Sutton R, Camm AJ. Symptomatology and quality of life in patients with rate-responsive pacemakers: a double-blind, randomized, crossover study.Clin Cardiol. 1989; 12:505–512. doi: 10.1002/clc.4960120907CrossrefMedlineGoogle Scholar
  • 31. Smedgård P, Kristensson BE, Kruse I, Ryden L. Rate-responsive pacing by means of activity sensing versus single rate ventricular pacing: a double-blind cross-over study.Pacing Clin Electrophysiol. 1987; 10:902–915. doi: 10.1111/j.1540-8159.1987.tb06046.xCrossrefMedlineGoogle Scholar
  • 32. Osswald S, Leiggener C, Buser PT, Pfisterer ME, Burckhardt D, Burkart F. Benefits and limitations of rate adaptive pacing under laboratory and daily life conditions in patients with minute ventilation single chamber pacemakers.Pacing Clin Electrophysiol. 1996; 19:890–898. doi: 10.1111/j.1540-8159.1996.tb03384.xCrossrefMedlineGoogle Scholar
  • 33. Witte KK, Clark AL. Beta-blockers and inspiratory pulmonary function in chronic heart failure.J Card Fail. 2005; 11:112–116. doi: 10.1016/j.cardfail.2004.07.007CrossrefMedlineGoogle Scholar
  • 34. Witte KK, Clark AL. Chronotropic incompetence does not contribute to submaximal exercise limitation in patients with chronic heart failure.Int J Cardiol. 2009; 134:342–344. doi: 10.1016/j.ijcard.2008.02.014CrossrefMedlineGoogle Scholar
  • 35. White M, Yanowitz F, Gilbert EM, Larrabee P, O’Connell JB, Anderson JL, Renlund D, Mealey P, Abraham WT, Bristow MR. Role of beta-adrenergic receptor downregulation in the peak exercise response in patients with heart failure due to idiopathic dilated cardiomyopathy.Am J Cardiol. 1995; 76:1271–1276. doi: 10.1016/s0002-9149(99)80355-5CrossrefMedlineGoogle Scholar
  • 36. Colucci WS, Ribeiro JP, Rocco MB, Quigg RJ, Creager MA, Marsh JD, Gauthier DF, Hartley LH. Impaired chronotropic response to exercise in patients with congestive heart failure: role of postsynaptic beta-adrenergic desensitization.Circulation. 1989; 80:314–323. doi: 10.1161/01.cir.80.2.314LinkGoogle Scholar
  • 37. Maurer MS, Sackner-Bernstein JD, El-Khoury Rumbarger L, Yushak M, King DL, Burkhoff D. Mechanisms underlying improvements in ejection fraction with carvedilol in heart failure.Circ Heart Fail. 2009; 2:189–196. doi: 10.1161/CIRCHEARTFAILURE.108.806240LinkGoogle Scholar
  • 38. McAlister FA, Wiebe N, Ezekowitz JA, Leung AA, Armstrong PW. Meta-analysis: beta-blocker dose, heart rate reduction, and death in patients with heart failure.Ann Intern Med. 2009; 150:784–794. doi: 10.7326/0003-4819-150-11-200906020-00006CrossrefMedlineGoogle Scholar
  • 39. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JGF, Coats AJS, Falk V, González-Juanatey JR, Harjola VP, Jankowska EA, et al; ESC Scientific Document Group. 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) developed with the special contribution of the Heart Failure Association (HFA) of the ESC.Eur Heart J. 2016; 37:2129–2200. doi: 10.1093/eurheartj/ehw128CrossrefMedlineGoogle Scholar
  • 40. Greene SJ, Fonarow GC, DeVore AD, Sharma PP, Vaduganathan M, Albert NM, Duffy CI, Hill CL, McCague K, Patterson JH, et al. Titration of medical therapy for heart failure with reduced ejection fraction.J Am Coll Cardiol. 2019; 73:2365–2383. doi: 10.1016/j.jacc.2019.02.015CrossrefMedlineGoogle Scholar
  • 41. Fowler MB, Lottes SR, Nelson JJ, Lukas MA, Gilbert EM, Greenberg B, Massie BM, Abraham WT, Franciosa JA; COHERE Participant Physicians. Beta-blocker dosing in community-based treatment of heart failure.Am Heart J. 2007; 153:1029–1036. doi: 10.1016/j.ahj.2007.03.010CrossrefMedlineGoogle Scholar
  • 42. Martens P, Verbrugge FH, Nijst P, Bertrand PB, Dupont M, Tang WH, Mullens W. Feasibility and association of neurohumoral blocker up-titration after cardiac resynchronization therapy.J Card Fail. 2017; 23:597–605. doi: 10.1016/j.cardfail.2017.03.001CrossrefMedlineGoogle Scholar
  • 43. Witte KK, Drozd M, Walker AMN, Patel PA, Kearney JC, Chapman S, Sapsford RJ, Gierula J, Paton MF, Lowry J, et al. Mortality reduction associated with β-adrenoceptor inhibition in chronic heart failure is greater in patients with diabetes.Diabetes Care. 2018; 41:136–142. doi: 10.2337/dc17-1406CrossrefMedlineGoogle Scholar
  • 44. Tse HF, Siu CW, Lee KL, Fan K, Chan HW, Tang MO, Tsang V, Lee SW, Lau CP. The incremental benefit of rate-adaptive pacing on exercise performance during cardiac resynchronization therapy.J Am Coll Cardiol. 2005; 46:2292–2297. doi: 10.1016/j.jacc.2005.02.097CrossrefMedlineGoogle Scholar
  • 45. Thackray SD, Ghosh JM, Wright GA, Witte KK, Nikitin NP, Kaye GC, Clark AL, Tweddel A, Cleland JG. The effect of altering heart rate on ventricular function in patients with heart failure treated with beta-blockers.Am Heart J. 2006; 152:713.e9–713.13. doi: 10.1016/j.ahj.2006.07.007CrossrefGoogle Scholar
  • 46. Nägele H, Rödiger W, Castel MA. Rate-responsive pacing in patients with heart failure: long-term results of a randomized study.Europace. 2008; 10:1182–1188. doi: 10.1093/europace/eun213CrossrefMedlineGoogle Scholar
  • 47. Källner N, Nishimura M, Birgersdotter-Green U, Hoffmayer KS, Han FT, Krummen DE, Raissi F, Feld GK, Hsu JC. Predictors of rate-adaptive pacing in patients implanted with implantable cardioverter-defibrillator and subsequent differential clinical outcomes.J Interv Card Electrophysiol. 2019; 55:83–91. doi: 10.1007/s10840-019-00536-9CrossrefMedlineGoogle Scholar
  • 48. Clark AL, Poole-Wilson PA, Coats AJ. Exercise limitation in chronic heart failure: central role of the periphery.J Am Coll Cardiol. 1996; 28:1092–1102. doi: 10.1016/S0735-1097(96)00323-3CrossrefMedlineGoogle Scholar
  • 49. Kramer DG, Trikalinos TA, Kent DM, Antonopoulos GV, Konstam MA, Udelson JE. Quantitative evaluation of drug or device effects on ventricular remodeling as predictors of therapeutic effects on mortality in patients with heart failure and reduced ejection fraction: a meta-analytic approach.J Am Coll Cardiol. 2010; 56:392–406. doi: 10.1016/j.jacc.2010.05.011CrossrefMedlineGoogle Scholar


eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.