Impact of Exercise Modalities on Peripheral and Central Components of Cardiorespiratory Capacity in Heart Transplantation Patients: A Systematic Review and Meta-Analysis

Background and Objectives: To analyze the effects of aerobic, resistance, and combined training on peripheral and central components related to cardiorespiratory capacity after HTx. Materials and Methods: No time restriction was applied for study inclusion. MEDLINE/PubMed; EMBASE, CENTRAL, and PEDro databases were investigated. Studies reporting heart transplanted patients older than 19 years following aerobic, resistance, and combined training according. The outcomes included: V′O2 peak, VE/V’CO2 slope, heart rate (HR peak), systolic and diastolic blood pressure (SBP and DBP peak), maximum repetition test(1RM), sit-to-stand test, and flow-mediated dilation (FMD). The studies were selected by consensus. Four hundred ninety-two studies initially met the selection criteria. Cochrane handbook was used for abstracting data and assessing data quality and validity. Independent extraction by two observers was applied. Results: Isolated aerobic training leads to a greater increase in V′O2 peak than combined training compared to the control group (p < 0.001, I2 = 0%). However, no significant differences were found in the subgroup comparison (p = 0.19, I2 = 42.1%). HR peak increased similarly after aerobic and combined training. High-intensity interval training (HIIT) was better than moderate continuous intensity to increase the V′O2 after long term in HTx. Still, there is scarce evidence of HIIT on muscle strength and FMD. No change on VE/V’CO2 slope, FMD, and SBP, DBP peak. 1RM and the sit-to-stand test increased after resistance training (p < 0.001, I2 = 70%) and CT (p < 0.001, I2 = 0%) when compared to control. Conclusions: Aerobic and combined training effectively improve VO2 peak and muscle strength, respectively. HIIT seems the better choice for cardiorespiratory capacity improvements. More studies are needed to examine the impact of training modalities on VE/V’CO2 slope and FMD.


Introduction
Despite the improvements in pharmaceutical and resynchronization treatments and even considering the advent of the left-ventricular assist device [1], heart transplantation (HTx) remains a notable treatment for advanced heart failure [2][3][4]. HTx gives a new life opportunity for such patients improving peak oxygen uptake (V O 2 peak) [5][6][7], a well-recognized prognostic variable assessed via cardiopulmonary exercise test and a

Selection, Evaluation of Studies, and Quality Assessment
The initial search identified 2712 studies and the present systematic review included 15 studies based on the inclusion criteria, and 13 studies considered for metanalysis. The reasons for exclusion are in Figure 1.

Studies Included in the Systematic Review
The publication period of the included studies ranged from 1998 to 2019, involving a total of 453 patients undergoing HTx with 407 (72%) males, with a mean age of 51 years in the intervention group and 47 years in the control group. Two studies reported only the RT protocol and three studies, included combined training (resistance + aerobic training). Surprisingly, only one of those CT studies reported a detailed prescription of the RT [30]. Also, only two studies presented a comparison between aerobic intensities, high-intensity interval training (HIIT) vs. moderate continuous training (MCT-AE) [31,38], and one study compared hospital-based versus home-based exercise following the same training prescription. Table 1 illustrates the studies summary. PEDRo score is included in Table 1 to provide the quality score from each study individually (widely used in the rehabilitation field). Nine studies involving AT [23][24][25][26][27][28][29][30][31], three studies involving CT [32][33][34], and two involving RT [35,36] were included. Only two studies compared exercise intensities  Outpatient-clinic-supervised. AT: HIIT on a treadmill (10 min warm-up + 4 min exercise bouts at 85-95% of HR peak , separated by 3 min active pauses at Borg scale 11-13, 6-20 RPE).
Additionally, the patients were encouraged to continue any physical activity on their own. CG: No intervention was given to the control group other than basic. Outpatient-clinic-supervised. AT: Supervised: 5 min warm-up + 40 min walking/jogging on a treadmill (80% HR of the RCP -69.0% ± 1.9 % of VO 2 max . Endurance Exercise Intensity was continually adjusted) + 5 min cool down. Non-supervised: Same exercise protocol following exercise intensity of 11-13 on the rate of Borg scale (range: [16][17][18][19][20]; CG: maintain their daily activities without AE during the 12-week period. Muscle strength Outpatient-clinic-supervised. RT: 5 min of warm-up walking on a treadmill + lumbar extensor training 1 day/week and upper and lower body resistance training 2 days/week. A single set of 10-15 repetitions was completed for each exercise: lumbar extension, duo-decline chest press, knee extension, pullover, knee flexion, triceps extension, biceps flexion, shoulder press, and the abdominal machine. The initial training weight represented 50% of the one-repetition maximum  test. The transplant recipients were not permitted to exceed 15 repetitions. Rather, when 15 repetitions were successfully achieved, the weight was increased by 5-10% at the next training session. CG: No resistance training intervention Not described 12 24 4 CG 7 51 ± 8 Outpatient-home-based-supervised RT: standard care home-based walking program (not supervised) associated with resistance training. 5 min warming-up + a single set from 10 to 15 repetitions were completed for each exercise: chest press, knee extension, pulldown, seated leg curl, shoulder press, seated triceps dip, biceps curl, and lumbar extension at 50% of 1 RM. The resistance was increased by 5% to 10% at the next training session when 15 repetitions were successfully achieved.
Upper body exercises were alternated with lower body exercises. CG: standard care home-based walking program (not supervised) AT-MCT-AE: The CON sessions consisted of biking for 45 min with an intensity corresponding to 60-70% of VO2peak. All sessions began with a 10 min warm-up and ended with a 10-min cooldown.
One study utilized the higher patient's tolerance sustained for 30 min [33]. Four studies applied HIIT [26,27,30,31] and one applied HIIT alternating with continuous training [32]. HIIT intensities ranged from 80-90% of the V O 2 peak, 80-95% of the HR peak, or 90-100% [32,39] of the baseline peak power output. Interval duration varied from 30 s to 4 min, alternated by low-intensity phases with an intensity ranging from 11 to 13 [27] according to the BORG scale or a recovery rest phase [26,32,39]. Recovery duration varied from 30 s to 3 min, while some studies adopted passive rest recovery.
Heterogeneity was low for V O 2 peak, slope VE/V CO 2 , and sit-to-stand test (I 2 < 50%). SBP, DBP and HR peak, FMD, and 1RM indicated high heterogeneity (I 2 > 50%). No study reported adverse effects. The agreement level between the reviewers, by Kappa coefficient, was 0.95 (95% CI: 0.75 to 1.03. 3 Exercise effects on peak oxygen consumption). Exercise

Exercise Effects on Peak Heart Rate
The analysis of peak heart rate was separated according to time post-HTx: de novo and long therm. Comparing the pooled effect analysis of aerobic and combined training versus control group after long term post HTx (≥1 year), jointly both training induced a slightly favorable effect in HR peak (4 studies, n = 164 patients), MD = 8.10, 95% CI: 1.98 to 14.22 bpm, I 2 = 87%, p = 0.009, with no subgroup differences (p = 0.12) (Figure 3.) Figure 3. Impact of exercise training on Heart Rate peak (bpm) in HTx: exercise versus the control group Two studies have explored exercise versus control in de novo HTx (˂1 year) and both aerobic [29] and combined training [33] did not indicate improvement on HR peak. Two studies compared HIIT versus moderate continuous training in de novo [30] and long term [31] post-HTx and the delta comparison indicated better results for HIIT only in long term post-HTx (p = 0.027) [31].  Two studies have explored exercise versus control in de novo HTx (<1 year) and both aerobic [29] and combined training [33] did not indicate improvement on HR peak. Two studies compared HIIT versus moderate continuous training in de novo [30] and long term [31] post-HTx and the delta comparison indicated better results for HIIT only in long term post-HTx (p = 0.027) [31].

Exercise Effects on Flow-Mediated Dilation
Three studies compared the exercise treatment with control in HTx patients. Exercise training did not demonstrate positive effect on FMD (3 studies, n = 86 patients), MD = 3.48%, 95% CI: −0.29 to 7.25%, p = 0.07). However, the studies presented a high heterogeneity (I 2 = 80%)- Figure 6. From those included studies, one study applied MCT-AE [25], one study applied CT [32], and one HIIT [26]. Only HIIT [28] presented an expressive improvement in FMD. However, a subgroup analysis was not possible due to the small number of studies in each exercise modality. Nytroen 2019 compared HIIT vs moderate continuous training and did not indicated differences for this parameter between modalities (n = 78), MD = −1.5 (−4 to 0.9), p = 0.208 [30].

Exercise Effects on Flow-Mediated Dilation
Three studies compared the exercise treatment with control in HTx patients. Exercise training did not demonstrate positive effect on FMD (3 studies, n = 86 patients), MD = 3.48%, 95% CI: −0.29 to 7.25%, p = 0.07). However, the studies presented a high heterogeneity (I² = 80%)- Figure 6. From those included studies, one study applied MCT-AE [25], one study applied CT [32], and one HIIT [26]. Only HIIT [28] presented an expressive improvement in FMD. However, a subgroup analysis was not possible due to the small number of studies in each exercise modality. Nytroen 2019 compared HIIT vs moderate continuous training and did not indicated differences for this parameter between modalities (n = 78), MD = −1.5 (−4 to 0.9), p = 0.208 [30].

Exercise Effects on Muscle Strength
Six studies analyzed the exercise impact on muscle strength, but two were not included in the metanalysis forest plots due to different muscle strength assessments (Isokinetic and isotonic evaluations). Two studies evaluated the maximum repetition test (1RM) and two evaluated the sit-to-stand test. Isolated resistance training (RT) was associated with a significant improvement in the 1RM for both chest press and leg extension movements, MD = 35.50 Kg, 95% CI: 19.42 to 51.59, I² = 70, p < 0.0001) ( Figure  7A) while the other two studies involving combined training showed increases on the sitto-stand test, MD: 5.54, 95% CI 3.07 to 8.01; I² = 0% ( Figure 7B). Isokinetic and isotonic evaluations not included in the metanalysis forest plots, also suggested an increase in muscle strength after CT and AT, respectively [27,34].

Exercise Effects on Muscle Strength
Six studies analyzed the exercise impact on muscle strength, but two were not included in the metanalysis forest plots due to different muscle strength assessments (Isokinetic and isotonic evaluations). Two studies evaluated the maximum repetition test (1RM) and two evaluated the sit-to-stand test. Isolated resistance training (RT) was associated with a significant improvement in the 1RM for both chest press and leg extension movements, MD = 35.50 Kg, 95% CI: 19.42 to 51.59, I 2 = 70, p < 0.0001) ( Figure 7A) while the other two studies involving combined training showed increases on the sit-to-stand test, MD: 5.54, 95% CI 3.07 to 8.01; I 2 = 0% ( Figure 7B). Isokinetic and isotonic evaluations not included in the metanalysis forest plots, also suggested an increase in muscle strength after CT and AT, respectively [27,34].

Discussion
The novelty of this meta-analysis is the analysis of different exercise modalities and intensities in clinical outcomes related to cardiorespiratory capacity after HTx, strengthening information about training prescription for clinical practice. A superior effect of aerobic training to improve V′O2 peak in HTx patients was demonstrated with a

Discussion
The novelty of this meta-analysis is the analysis of different exercise modalities and intensities in clinical outcomes related to cardiorespiratory capacity after HTx, strengthening information about training prescription for clinical practice. A superior effect of aerobic training to improve V O 2 peak in HTx patients was demonstrated with a moderate level of certainty of evidence. Interesting, high-intensity interval training (HIIT) demonstrated a higher effect on the V O 2 peak than moderate continuous training (MCT-AE) with no adverse effect [30,31].
Considering the low level of certainty of the evidence, resistance exercise training (RT) led to improvements in muscle strength. The skeletal muscle weakness, vasodilatory capacity impairment, and muscle capillary density reduction are the main peripheral factors related to exercise capacity reductions after HTx and partially explain the V O 2 peak impairment [40]. Nytrøen et al. pointed to the association between muscular deconditioning and V O 2 peak reduction [27], recognizing the peripheral limitations in HTx patients. Peripheral adaptation such as mitochondrial volume density, oxidative enzyme capacity, and the percentage of type 1 muscle fibers distribution increase, are associated with the cardiorespiratory capacity increase [10,36,41]. These results indicate RT, isolated, or in combination with AT, increases muscle strength and attenuates V O 2 peak impairment post-HTx.
There was no evidence that exercise affect DBP and SBP peak in HTx [27,28,33] and like HR peak, all indicated a very low level of certainty of evidence. HR peak increased after all training modalities, especially after AE. Compared to MCT-AE, a higher increase in HR peak occurred after HIIT with a moderate level of certainty of evidence. However, the magnitude seems not equivalent to the exercise capacity improvement [6,42,43], possibly due to the chronotropic incompetence [24,44]. Autonomic nervous system improvement may explain it [24,43,44]. The average increase of SBP peak should be 50% of the resting value and an insufficient increase has been associated with left ventricular systolic dysfunction [45]. Nevertheless, the relationship between SBP response and V O 2 peak is unclear.
The absence of improvements after CT [33], HIIT [27,46], or MCT-AE [47] on LV end-diastolic or end-systolic volume, stroke volume, or ejection fraction after HTx [10] contributes to the rationale that the improvements in peak V O 2 seem not only related to central (cardiac) adaptations [10].
Exercise training was not associated with FMD benefits [25,26,32] (very low level of certainty of the evidence), but an expressive improvement was when HIIT was compared to a control group [26]. More studies are needed, but recognizing that endothelial dysfunction predicts cardiac allograft vasculopathy [14], HIIT seems a promising approach post-HTx. The unique study that compared HIIT vs. moderate continuous training did not indicate differences between those modalities, although the statistical difference was seen only within the HIIT group. Although an endothelial function recovery occurs post-HTx [57], peripheral endothelial dysfunction remains after 1 to 13 years [58]. The primary mechanism of the endothelial dysfunction post-HTx relates to cyclosporin therapy [10,57]. Exercise training can counteract it by enhancing nitric oxide (NO) production [48,49].
According to the two included studies and with a very low level of evidence, there is no effect on VE/V CO 2 slope after exercise training post-HTx when compared to a control group. Nytroen, 2019 when comparing HIIT vs. moderate continuous training also did not reveal any difference between modalities [30]. VE/V CO 2 slope is a strong independent predictor of mortality in HF patients [59] as accurate as V O 2 peak for HTx. Although the VO 2 peak is related to mortality like VE/VCO 2 slope, none of the included studies addressed morbidity or mortality rates after their intervention or control period. The positive association between VE/V CO 2 slope reduction and functional capacity improvement was identified in 40% of the patients post-HTx, even five years later [60]. VE/V CO 2 slope increase has been associated with peripheral factors, such as muscle deconditioning, peripheral oxygen transport problems, and type IIb-muscle-fiber increased, leading to a primary lactic acidosis during exercise demanding high ventilatory response [60]. In our meta-analysis, only aerobic training (27) (28) explored VE/V CO 2 slope post-HTx.
The small number of studies available and the low level of certainty of evidence from many outcomes is the major limitation of this systematic review with meta-analysis. Additionally, the lack of studies reporting comparison between modalities has limited the results of exercise training post-HTx. More research is required, mainly for the outcomes VE/V CO 2 slope and FMD. In addition, the absence of an increase in heart rate peak after exercise training may have been influenced by the poor autonomic response due to cardiac denervation. Another limitation is the absence of information regarding the age mismatch between donors and recipients of the included studies. Finally, it is also a problem that patients with left ventricular assistant systems before transplantation often present good exercise tolerance and are not scrutinized from those without left ventricular assistant systems [61] which could influence the final meta-analysis results. However, even considering the limitations, this review is important to demonstrate the state of the art on training prescription in HTx, revealing the need for new clinical trials with higher quality.

Conclusions
Cardiac rehabilitation is essential after HTx to improve the training performance of the patients. From this review, aerobic training seems the best training modality after HTx, mainly high-intensity interval training with the biggest effect on peak oxygen consumption 7265677. Isolated resistance training or combined training improve muscle strength. More studies are needed.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/medicina58010032/s1, Figure S1: Risk of bias assessed via Rob 2 tool, Figure S2: Summary of Findings Tables with quality of evidence for exercise training compared to control, Figure S3: Summary of Findings Tables with quality of evidence for HIIT compared to moderate continous training, Table S1: Physiotherapy Evidence Database (PEDro) scores for each of the 12 included studies in the metanalysis.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
All authors from this manuscript have no conflicts of interest or financial ties to disclose.