Chronotropic incompetence (CI) is common in advanced heart failure and is associated with worse functional capacity.1 This impaired heart rate (HR) response during exercise is ameliorated but persists after left ventricular assist device (LVAD) implantation.2
Patients with continuous flow LVAD (CF-LVAD) suffer from significant exercise limitation despite restoration of resting cardiac output.3 Whether CI contributes to exercise limitation in this setting is unknown. We examined the role of CI and the effect of rate response pacing (RRP) on functional capacity in a group of stable patients with LVAD.
Patients with CF-LVAD in our heart failure clinic were screened. Those with permanent atrial fibrillation, recent hospitalization for decompensated heart failure or physical inability to exercise were excluded.
After written informed consent, participants underwent maximal exercise testing with treadmill (modified Naughton protocol) or cycle ergometer (15 W/min ramp) based cardiopulmonary exercise test (CPX), and 6 minute walk test (6 MWT) before and, if CI was present during baseline testing, after activation of RRP in their cardiac implantable electronic devices (CIEDs). Pacer upper tracking rate was set at 80% of age-predicted maximum heart rate (APMHR).
Timing and pattern HR response to exercise were determined from surface electrocardiogram (ECG) tracings obtained during CPX. Continuous gas analysis was performed using a metabolic cart, and test results were interpreted by an experienced cardiologist.
Thirty patients participated in the study, age was 49 ± 15 yr (mean ± standard deviation [SD]), 30% were women and 70% had non-ischemic cardiomyopathy (NICMP). Time on LVAD support was 20 ± 17 months. LVADs were HeartWare in three patients and HeartMate II in 27. Twenty-six had CIED; 19 had single-lead implantable cardiac defibrillators (ICDs) and seven had cardiac resynchronization therapy (CRT). Manufacturers were Medtronic 47%, Boston Scientific 23%, St Jude’s Medical 13%, and Biotronik 3%.
Baseline CPX’s were treadmill-based for 20 patients and cycle ergometer based for 10 patients. The prevalence of CI (peak HR < 80% APMHR) was 70% independent of exercise modality.
Patients with CI were more likely to have a peak oxygen consumption (pVO2) < 14 ml·kg·min–1 (p = 0.01) and a lower average oxygen consumption during peak stages of exercise: pVO2 value 14 ± 2 versus 11.5 ± 3 ml·kg·min–1; p = 0.02. During submaximal exercise (ventilatory threshold), oxygen consumption (VO2) trended to lower values: 8.7 ± 2.7 versus 10.3 ± 2.5 ml·kg·min–1; p = 0.10.
Intrinsic HR response during baseline CPX showed a weak positive correlation with VO2, both at ventilator threshold (VT) and at peak exercise (VO2 at VT; r = 0.5, r2 = 0.24, p = 0.005 and pVO2; r = 0.6, r2 = 0.36, p < 0.001).
Patient’s age, sex, body mass index (BMI), and diabetic status had no relation to CI. A similar number of patients with and without CI were on Beta-blocker therapy, although daily Carvedilol equivalent doses were higher among patients with CI (18 ± 23 vs. 37 ± 27 mg; p = 0.03). Time with LVAD had no relation to CI after correction for beta-blocker dosage.
Resting right ventricular (RV) systolic function was assessed qualitatively in 16 patients, by RV fractional area change in 7 and was indeterminate in 7. It was graded as normal in 3, mildly reduced in 8 and moderately reduced in 11 patients. Peak VO2 was not influenced by RV function: normal = 13 ± 1.7 versus mildly reduced = 12.6 ± 1.0 versus moderately reduced = 12.7 ± 0.9 ml·kg·min–1 (p = 0.8). Baseline 6 MWT distance was also unaffected by RV function: normal 299.7 ± 54 m versus mildly reduced 354.3 ± 47 m versus moderately reduced 276.2 ± 33.5 m (p = 0.5).
Average resting blood pressure was 78.7 mmHg. Resting mean arterial pressures appeared to have no correlation to Peak VO2 values (R2 = 0.44; p = 0.9). Blood pressure response to exercise was not measured during this study.
Thirteen patients in the treadmill group, those with CI during baseline testing, underwent repeat testing after RRP activation. RRP significantly increased average 6 MWT distance (6 MWD) from 307 ± 84 to 353 ± 86 m; p = 0.001 (Figure 1).
The effect of RRP during CPX was less uniform due to lack of efficacy in a majority of patients. In seven patients (54%) RRP did not commence despite previous activation.
In six patients (46%) RRP was evidenced by a pacer-mediated increase in HR during the test leading to a peak of at least 80% of APMHR. Cardiac implantable electronic device manufacturers in these patients were: Boston Scientific in 2, St Jude in 1, and Medtronic in 3.
In two of these patients, RRP was inadequate, starting after ventilatory threshold in one and beginning early but ceasing before maximal effort in another patient. No improvement in aerobic capacity was seen in either.
In four patients RRP was considered adequate, starting early during exercise and remaining sustained through the completion of the test. Two of these patients saw a substantial improvement in exercise capacity: VO2 at VT increase of 22% and 44%, and pVO2 increase of 37% and 25%, respectively (Figure 2).
Our findings demonstrate the association between CI and poor functional capacity in patients with advanced heart failure and CF-LVAD, in line with one small prior study.4
Findings in this cohort point out the inadequacy of current RRP technologies for sensing signals other than atrial rate during different types of physical activity. When RRP increased the HR promptly and in a sustained manner, replicating the activity of the sinus node, the effect on aerobic capacity was substantial, but this occurred in only a minority of patients.
In contrast to the heterogeneous effect of RRP during treadmill-based CPX, its effect on 6 MWD was more homogeneous. This could represent a difference in CIED sensing efficacy since all of the employed devices in this study have an accelerometer-based RRP system. Ambulation, producing linear displacement of the body during 6 MWT could be more easily sensed by accelerometer-based systems than the more static motion during treadmill exercise.
These findings warrant further exploration to allow for improvement of current RRP technologies. An efficacious sensing system that can replicate sinus nodal function during different types of physical and emotional activity may lead to important improvements in the quality of life of the growing durable CF-LVAD population.
1. Sims DB, Mignatti A, Colombo PC, et al. Rate responsive pacing using cardiac resynchronization therapy in patients with chronotropic incompetence
and chronic heart failure. Europace 2011.13: 1459–1463.
2. Grosman-Rimon L, McDonald MA, Bar-Ziv SP, et al. Chronotropic incompetence
, impaired exercise capacity, and inflammation in recipients of continuous-flow left ventricular assist devices. J Heart Lung Transplant 2013.32: 930–932.
3. de Jonge N, Kirkels H, Lahpor JR, et al. Exercise performance in patients with end-stage heart failure after implantation of a left ventricular assist device
and after heart transplantation: An outlook for permanent assisting? J Am Coll Cardiol 2001.37: 1794–1799.
4. Dimopoulos S, Diakos N, Tseliou E, et al. Chronotropic incompetence
and abnormal heart rate recovery early after left ventricular assist device
implantation. Pacing Clin Electrophysiol 2011.34: 1607–1614.