American Journal of Transplantation 2014; 14: 2391–2399 Wiley Periodicals Inc.

 C

Copyright 2014 The American Society of Transplantation and the American Society of Transplant Surgeons doi: 10.1111/ajt.12873

Brief Communication

Effect of High-Intensity Training Versus Moderate Training on Peak Oxygen Uptake and Chronotropic Response in Heart Transplant Recipients: A Randomized Crossover Trial C. H. Dall1,2,*, M. Snoer1, S. Christensen3, T. Monk-Hansen1, M. Frederiksen1, F. Gustafsson4, H. Langberg5 and E. Prescott1 1

Department of Cardiology, Bispebjerg Hospital, University of Copenhagen, Copenhagen, Denmark 2 Institute of Sports Medicine Copenhagen, Bispebjerg Hospital, University of Copenhagen, Copenhagen, Denmark 3 Department of Cardiology, Gentofte Hospital, University of Copenhagen, Copenhagen, Denmark 4 Heart Center, Department of Cardiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 5 CopenRehab, Department of Public Health, University of Copenhagen, Copenhagen, Denmark
Corresponding author: Christian H. Dall, [email protected]

HIIT are partly a result of improved chronotropic response. Abbreviations: absolute VO2peak, peak oxygen uptake (mL/min); bpm, beats per minute; CON, continued moderate exercise; DBP, diastolic blood pressure; HIIT, high-intensity interval training; HR, heart rate; HRpeak, peak heart rate; HRrecovery, heart rate recovery; HRreserve, heart rate reserve; HRrest, resting heart rate; HTx, heart transplant; RER, respiratory exchange ratio; SBP, systolic blood pressure; SET, strength and endurance training; VO2peak, peak oxygen uptake (mL/min/kg) Received 31 January 2014, revised and accepted for publication 16 June 2014

Introduction In heart transplant (HTx) recipients, there has been reluctance to recommend high-intensity interval training (HIIT) due to denervation and chronotropic impairment of the heart. We compared the effects of 12 weeks’ HIIT versus continued moderate exercise (CON) on exercise capacity and chronotropic response in stable HTx recipients >12 months after transplantation in a randomized crossover trial. The study was completed by 16 HTx recipients (mean age 52 years, 75% males). Baseline peak oxygen uptake (VO2peak) was 22.9 mL/kg/min. HIIT increased VO2peak by 4.9  2.7 mL/min/kg (17%) and CON by 2.6  2.2 mL/ kg/min (10%) (significantly higher in HIIT; p < 0.001). During HIIT, systolic blood pressure decreased significantly (p ¼ 0.037) with no significant change in CON (p ¼ 0.241; between group difference p ¼ 0.027). Peak heart rate (HRpeak) increased significantly by 4.3 beats per minute (p ¼ 0.014) after HIIT with no significant change in CON (p ¼ 0.34; between group difference p ¼ 0.027). Heart rate recovery (HRrecovery) improved in both groups with a trend toward greater improvement after HIIT. The 5-month washout showed a significant loss of improvement. HIIT was well tolerated, had a superior effect on oxygen uptake, and led to an unexpected increase in HRpeak accompanied by a faster HRrecovery. This indicates that the benefits of

Heart transplantation is the best treatment for selected end-stage heart failure patients, conferring significant improvement in quality-of-life, physical function and exercise capacity (1,2). Nevertheless, exercise capacity remains below that of healthy individuals (1). In HTx recipients, the heart is denervated with some functional cardiac allograft reinnervation (3,4), which results in a resting heart rate (HRrest) 15–20 beats above that of a healthy individual, a reduced peak heart rate (HRpeak) and to some extent a delayed chronotropic response (1–4). The slow rise of heart rate (HR) when initiating exercise depends primarily on the increase in circulating catecholamines (1,2,4,5). Exercise training is central in cardiac rehabilitation, benefiting well being, exercise capacity and prognosis of cardiac patients in general. Recommendations for cardiac rehabilitation have been conservative with intensities below 60% of peak oxygen uptake (VO2peak) (5). Several studies have used a protocol for HTx recipients consisting of continued moderate exercise (CON) (intensity of 60–70% of HRpeak) (2,5–7). Despite chronotropic impairment, a few studies, including one by our group, have indicated that highintensity interval training (HIIT) is feasible and safe in HTx 2391

Dall et al

recipients (8–10). The effect of HIIT may be superior to that of CON, but this has not been tested in a direct comparison study in HTx recipients.

After baseline testing, a person not involved in the study randomized patients (envelope randomization) into two groups: HIIT-washout-CON or CON-washout-HIIT (see Figure 1 for consort flow diagram).

We hypothesized that VO2peak is improved in an intensitydependent manner in stable HTx recipients. The aim of this study was thus to compare the effect of HIIT versus CON in stable HTx recipients.

All participants gave written consent regarding participation. The study reporting follows the CONSORT statement. The study conforms to the Declaration of Helsinki and was approved by the scientific research ethics committee for the Capital Region of Denmark (H-3-2010 030) and registered at clintrial.gov: (NTC01914406).

Intervention

Materials and Methods Study design The study was a randomized controlled crossover trial. Clinically stable HTx recipients (12 months or more after HTx, age >18 years) from the Heart Centre, Rigshospitalet, University of Copenhagen physically capable of participating in HIIT were included. Exclusion criteria were unstable condition defined as recent moderate or severe rejection episodes (>H1R 80% of VO2peak, separated by a 2-min active rest period (approx. 60% of VO2peak). Each session lasted 32 min. The >80% of VO2peak was used as a minimum exercise effort in the 4-, 2-, and 1-min interval blocks. If the HTx recipients were able to work at an even higher exercise intensity in the shorter timeframes (e.g., in the 2- or 1-min interval blocks) they were told to do so. See Figure 2 describing the HIIT exercise protocol and HR curves from one HTx recipient with major chronotropic impairment and one HTx recipient with normalized chronotropic response. The CON sessions consisted of biking for 45 min with an

Assessed for eligibility (n=20) Heart transplant (HTx) Copenhagen University Hospital 20 HTx gave wrien consent of parcipaon (2010-2012) Excluded (n=3) • Not meeng inclusion criteria (n=2) • Declined to parcipate (n=0) • Other reason (n=1) Randomized (n=17)

Allocated to intervenon (n= 9) • Received allocated intervenon (n=9) 12 weeks of connued moderate exercise

Allocated to intervenon (n= 8) • Received allocated intervenon (n=8) 12 weeks of high-intensity interval training 5-month washout (n=17) Crossover

Allocated to intervenon (n=8) • Received allocated intervenon (n=8) 12 weeks of connued moderate exercise

Allocated to intervenon (n=9) • Received allocated intervenon (n=9) 12 weeks of high-intensity interval training

Analyzed (n=8) • Excluded from analysis (n=0)

Analyzed (n=9) • Excluded from analysis (n=1)

N=16 HTx completed the study

Figure 1: Consort flow diagram depicting the flow of participants through the crossover trial. In the inclusion period (2010–2012), 20 HTx recipients where referred to trial from The Heart Centre, Rigshospitalet, Copenhagen. Two excluded due to exclusions criteria; one multi-organ transplantation and one painful osteoporosis. One dropout, due to long transport time to cardiac rehabilitation but before randomizing. In total, 17 participants were tested 4 times, one was excluded due to insufficient exercise test (see text) leaving 16 HTx participants completed the study on whom the analyses are based.

2392

American Journal of Transplantation 2014; 14: 2391–2399

Exercise and Heart Transplantation

Figure 2: Heart rate (HR) curve and high-intensity interval training (HIIT) protocol. Illustration of the HIIT training protocol (10-min warm-up, 16 min of HIIT training þ recovery and 10-min cool down). The two HR curves represent the difference in chronotropic response between to heart transplant recipients following the same protocol.

intensity corresponding to 60–70% of VO2peak. All sessions began with a 10min warm-up and ended with a 10-min cool down. Blood lactate was monitored to control for the intensity during the training in both groups (11). The 12-week training interventions were separated by a 5month washout in which patients were instructed to resume their usual lifestyle.

Outcome measures The main outcome measure was VO2peak and secondary outcomes related to exercise testing, including blood pressure, HRpeak, HRrest, heart rate reserve (HRreserve), HRrecovery and workload. All measures were performed at baseline, at 12 weeks, after 5 months’ washout, and again after the second 12-week intervention.

Peak oxygen uptake All patients performed an upright bicycle (Sprint 150; Ergoline, Ho¨chberg, Germany) exercise test with breath-by-breath gas exchange measurement (Jaeger, CPX; Cardinal Health, Ho¨chberg, Germany). HR was monitored continuously before, during and after the exercise. Ventilatory oxygen uptake, carbon dioxide production, minute ventilation and respiratory

American Journal of Transplantation 2014; 14: 2391–2399

exchange ratio (RER) were calculated online. We aimed at an RER  1.05 as the criterion for an adequate maximal exercise test (12). HRpeak was measured at VO2peak and HRrecovery was measured at timepoint 30 s, 1 min and 2 min postpeak performance. HRrecovery was defined as HRpeak minus HR at a specified time period after exercise and represented the drop in HR during that time interval (13,14). Predicted VO2peak was calculated using Wasserman/Hansen equation (12). Resting blood pressure was measured in a quiet room with the participant lying down for 5 min. The outcome assessors were blinded to the intervention allocation, and the patients were asked not to discuss exercise habits with the test staff.

Analysis and statistics Change in VO2peak was used as the primary efficacy criterion. Sample size was based on an expected increase in VO2peak after HIIT of 4.5 mL/kg/min. Based on a previous study of HIIT in HTx recipients (9), the expected SD of the within patient change was 2.3 mL/kg/min. With power (beta) of 90% and two-sided alpha of 0.05, it was calculated that 16 patients were needed in this crossover study. Due to the risk of dropout we aimed to include 20 patients. All other efficacy variables as well as safety parameters are presented by descriptive statistics and compared between treatment groups using suitably chosen tests for paired or unpaired observations. Data

2393

Dall et al were analyzed as 2  2 crossover design using the pkcross command in Stata with formal testing for carryover and period effects. A p-value of less than 0.05 was considered statistically significant. All analyses were performed in Stata 11.1 (STATA statistical Software: Release 11, College Station, TX).

Results Before randomization, three patients were excluded (two did not meet the inclusions criteria, and one with no reason). The remaining 17 patients completed the study. One patient had insufficient exercise tests (RER < 0.85) and was an outlier on several parameters (HRpeak, HRreserve, VO2peak) and was thus excluded from the analyses; however, all the analyses repeated with this patient yielded similar results to those presented. Thus, analyses are based on 16 patients: 4 women and 12 men with mean age 52 years (range 33–70 years) and mean of 6.4 years after transplantation (range 1–17 years). The baseline VO2peak was 22.9 mL/kg/min (range 12.5–35.6) corresponding to 83% of predicted VO2peak. HRpeak was 138 corresponding to 82% of age-predicted maximum. All patients had a coronary angiography within 12 months of inclusion showing no significant coronary lesions (>50% stenosis) and in the initial VO2peak test, electrocardiogram recordings showed no signs of ischemia. Of the patients, nine were on nondihydropyridine calcium channel antagonists, nine on angiotensin converting enzyme inhibitors/ angiotensin receptor antagonists and one patient was on beta blockers. There were no changes in the immune therapy, antihypertensive medications or other change in medication in the 2  12 weeks of exercise intervention. Mean level of calories used per session was calculated based on HR with the Polar RS400 (Polar electro, Kempele, Finland). In CON, the number of kilocalories used after exercise was a little higher than in the HIIT intervention. Due to increase use of calories immediately following HIIT, the two exercise interventions were regarded as isocaloric. The mean blood lactate levels after exercise were 5.6 mmol/L (range 4.3–9.2) in HIIT and 3.7 mmol/L (range 2.6–4.9) in CON, respectively (Table 1). Effect of intervention on exercise capacity For the main outcome VO2peak there were no indications of period effect (p ¼ 0.89) or carryover effect (p ¼ 0.78). None of the other variables showed any significant period or carryover effects. There was a significant improvement in VO2peak after both training interventions, and patients with a near normal chronotropic response and those with a greater degree of chronotropic impairment had effect on improved VO2peak (Table 2). VO2peak in HIIT increased by 17%, from 23.2 mL/ kg/min to 28.1 mL/kg/min (p < 0.001), and in CON by 10%, from 23.0 mL/kg/min to 25.6 mL/kg/min (p < 0.001). The 2394

Table 1: Baseline patient characteristic Patients characteristics (n ¼ 16) Age, years Years after transplantation Sex (male/female) Diabetes Hypertension Smoking status (now/before HTx) BMI, kg/m2 SBP, mmHg DBP, mmHg Total cholesterol, mmol/L High density lipoprotein (HDL), mmol/L Low density lipoprotein (LDL), mmol/L Triglycerides, mmol/L Creatinine, mmol/L Cyclosporine/tacrolimus MMF/everolimus Prednisolone Antihypertensive medication Normal EF > 55% VO2peak, mL/kg/min % of predicted VO2peak Absolute VO2peak, mL/min HRrest, bpm HRpeak, bpm % of age-predicted HRpeak Respiratory exchange ratio Workload, W

51.9 (33–70) 6.4 (1–17) 75/25% 6% 81% 0/56% 25.5 (18.9–33.5) 128.6 76.8 5.0 (3.5–6.4) 1.3 (0.8–2.3) 2.9 (1.7–4.1) 1.6 (0.6–2.6) 106.2 (55–161) 81/6% 81/50% 44% 94% 100% 22.9 (12.5–35.6) 83.2% (40–155%) 1872.1 (1614.1–2130.2) 84.6 (76–99) 138.2 (105–165) 82.2% (60–100%) 1.19 (1.03–1.24) 155.0 (134.7–175.3)

Absolute VO2peak, peak oxygen uptake (mL/min); bpm, beats per minute; DBP, diastolic blood pressure; EF, ejection fraction; HRpeak, peak heart rate; HRrest, resting heart rate; HTx, heart transplant; MMF, mycophenolate mofetil; SBP, systolic blood pressure; VO2peak, peak oxygen uptake (mL/kg/min); workload (W), peak power output. Data are mean (range) or proportions as appropriate. Predicted VO2peak was calculated using Wasserman/Hansen equation (12). Cyclosporine, tacrolimus, MMF and everolimus are immunosuppressive drugs.

improvement was significantly higher with HIIT (p < 0.001) (Table 3, Figure 3A and B). In the HIIT intervention, workload improved by 9% (p ¼ 0.004), with no significant change in CON (Table 3). RER did not increase in the follow-up, indicating that all tests performed were on maximum individual effort. Blood pressure During HIIT, systolic blood pressure (SBP) decreased significantly from 131 mmHg to 125.8 mmHg (p ¼ 0.037) with no significant change in diastolic blood pressure (DBP) (p ¼ 0.094). During CON, no significant change in SBP (p ¼ 0.241) or DBP (p ¼ 0.285) was seen. Effect of intervention on heart rate HRrest was 84.6 ( 5.9) with a marginal decrease of 1.0 beats per minute (bpm) (p ¼ 0.037) after HIIT intervention, American Journal of Transplantation 2014; 14: 2391–2399

Exercise and Heart Transplantation Table 2: Baseline and follow-up data (data presented as mean  SD and % change) Variable (n ¼ 16) HIIT VO2peak, mL/kg/min Absolute VO2peak, mL/min SBP DBP HRpeak, bpm Workload, W RER HRrest, bpm HRreserve, bpm HRrecovery% after 30 s HRrecovery% after 1 min HRrecovery% after 2 min Weight, kg CON VO2peak, mL/kg/min Absolute VO2peak, mL/min BPsystolic BPdiastolic HRpeak, bpm Workload, W RER HRrest, bpm HRreserve, bpm HRrecovery% after 30 s HRrecovery% after 1 min HRrecovery% after 2 min Weight, kg

% Mean change

p-Value


28.1  8.1 2202  578.9 125.8  11.2 77.8  4.5 143.6  16.8 162.0  34.3 1.15  0.1 83.9  6.7 59.6  19.1 3.2  3.6 6.5  3.6 13.0  6.2 78.1  13.3

17.4% 15.3% 4% 3% 3.0% 8.6% 2.5% 1.2% 8.9% 45.3% 28.0% 24.1% 1.0%