Effect of High-Intensity Interval Training on Aerobic Capacity and Heart Rate Control of Heart Transplant Recipients: a Systematic Review with Meta-Analysis

Introduction Heart transplantation (HTx) is the gold standard procedure for selected individuals with refractory heart failure. High-intensity interval training (HIIT) is safe and allows patients to exercise in high intensity for longer time when compared to moderate-intensity continuous training (MICT). The primary aim of this study was to perform a systematic review and meta-analysis about the effect of HIIT compared to MICT on exercise capacity, peak heart rate, and heart rate reserve in HTx recipients. Secondarily, we pooled data comparing MICT and no exercise training in these patients. Methods This systematic review followed the standardization of the Preferred Reporting Items for Systematic Reviews and Meta-analyses statement and the Cochrane Collaboration Handbook. We presented the treatment effects of HIIT on the outcomes of interest as mean difference (MD) and 95% confidence interval (CI). Meta-analysis was performed using the random-effects, generic inverse variance method. Results HIIT improved peak oxygen consumption (peakVO2) (MD = 2.1; 95% CI 1.1, 3.1; P<0.0001), peak heart rate (MD = 3.4; 95% CI 0.8, 5.9; P=0.009), and heart rate reserve (MD = 4.8; 95% CI -0.05, 9.6; P=0.05) compared to MICT. Improvements on peakVO2 (MD = 3.5; 95% CI 2.3, 4.7; P<0.00001) and peak heart rate (MD = 5.6; 95% CI 1.6, 9.6; P=0.006) were found comparing HIIT and no exercise training. Conclusion Current available evidence suggests that HIIT leads to improvements on peakVO2, peak heart rate, and heart rate reserve compared to MICT in HTx recipients. However, the superiority of HIIT should be tested in isocaloric protocols.


INTRODUCTION
Heart transplantation (HTx) is the gold standard procedure for selected individuals with end-stage heart failure [1] . Even knowing that HTx improves patients' exercise tolerance, it is not restored to normal values [2] . The concern about exercise tolerance in HTx recipients is based on the association between peak oxygen consumption (peakVO 2 ), the gold standard method to assess cardiorespiratory fitness, and survival [3,4] .
Several studies have demonstrated the positive effects of aerobic exercise training on peakVO 2 in HTx recipients [5][6][7] . However, there is no consensus on how, when, or at what intensity exercise should be performed by HTx patients [8] . High-intensity interval training (HIIT) allows patients to exercise in higher intensity when compared to standard continuous training. A previous well performed meta-analysis [9] showed that HIIT was safe and effective in improving peakVO 2 in HTx recipients when compared to no training. However, the lack of studies that compared HIIT with moderate-intensity continuous training (MICT) limited the previous meta-analysis. Moreover,

Data Collection and Analysis
Two independent reviewers (COG and RESF) searched for relevant studies according to title and abstract. If at least one of the reviewers considered one reference eligible, the full text was obtained for a complete assessment. Then, the two reviewers assessed the full text according to the eligibility criteria. Thereafter, the following information from the studies were extracted:

Eligibility Criteria
To establish the eligibility criteria we used the following PICOT elements (standing for Participants, Intervention, Comparison, Outcomes, and Study type): Population, HTx recipients; Intervention, HIIT; Comparison, MICT or no exercise training; Primary outcome, peakVO 2 (ml/kg/min); Secondary outcomes, peak heart rate and heart rate reserve (bpm); Study type, RCTs.

Search Strategy
We searched for references in PubMed, Scopus, and Cochrane Central Register of Controlled Trials up to February 2019. A gray literature search included Google Scholar and OpenThesis. The first 100 results of the Google Scholar search were analyzed. The search was limited to studies published in full-text versions, without language restriction. Search strategy is provided in the online supplement. The reference lists of all eligible studies and reviews were also scanned to identify additional studies for inclusion. The authors were contact by e-mail for confirmation of any data or additional information if needed.

Quality of the Studies
Risk of bias was assessed according to the Cochrane guidelines for RCTs. Seven domains were assessed for evaluation: sequence generation and allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias), and other potential sources of bias ( Figure 2). Risk of bias was rated as low, unclear, or high according to established criteria ( Figure 3) [16] .

Fig. 2 -Risk of bias summary: review authors' assessments about each risk of bias item for each included study.
demographic characteristics of study participants, exercise protocols, adverse events, and outcome data. Values (mean difference [MD] and standard deviation [SD]) for peakVO 2 , peak heart rate, and heart rate reserve were extracted before and after intervention.
The meta-analysis was performed using the change between pre-and post-intervention means for each intervention group and the change SD calculated. If the SD of change [SDc]) for a given outcome was not reported, the formula: SDc= √ [(SD pre ) 2 + (SD post ) 2 -(2 × corr pre,post × SD pre × SD post )] was applied. SD pre , SD post , and corr pre,post represent the SD of the pre-intervention value, the SD of the post-intervention value, and the correlation coefficient between pre-and post-intervention values, respectively. The corr pre,post was conservatively set at 0.5. Mean changes were pooled using the random-effects, generic inverse variance method. A forest plot was used to graphically present the effect sizes and the 95% confidence intervals (CIs). A two-tailed P-value of <0.05 was used to determine significance. Statistical heterogeneity was assessed by using the Cochran Q test [17] and quantified by the I 2 index [18] .

Studies Characteristics
Five RCTs [5,12,13] (total of 212 HTx recipients with mean age of 57 years) were included in this systematic review. Exercise protocols were well reported (Tables 1 and 2). Follow-up periods were also well reported among studies: eight weeks [17,18] , 12 weeks [19,20] , and one year [8] . No serious adverse events were reported by the studies. In general, the studies presented a low/uncertain risk of bias. Figure 3 presents results of individual assessment by Cochrane risk of bias.

DISCUSSION
Our systematic review found that HIIT was superior than MICT on peakVO 2 , peak heart rate, and heart rate reserve in HTx recipients. In recent years, a growing number of studies have suggested that HIIT is similar or even superior to MICT in peakVO 2 , heart rate response, and quality of life improvements [8,13,21] .
A previous meta-analysis reported the efficiency of HIIT on peakVO 2 gains in HTx when compared to no exercise training [9] . However, our meta-analysis contributes to a better understanding of the effect size of HIIT when compared with the standard exercise intensity prescription (MICT). Additionally, some issues regarding the effects of exercise protocols must be addressed in relation to the apparent superiority of HIIT over MICT. In a recent review, Dun et al. [22] argued that the exercise duration and ratio of high and low-intensity bouts are key factors that differentiate

Fig. 4 -Forest plot showing meta-analyses of the high-intensity interval training (HIIT) vs. moderate-intensity continuous training (MICT) on a) peak oxygen consumption (peakVO2), b) peak heart rate, and c) heart rate reserve. Comparisons between HIIT vs. no exercise training on d) peakVO2 and e) peak heart rate were also shown. CI=confidence interval; SE=standard error
HIIT from MICT and may contribute to the patient's response. In general there are three main protocols of HIIT: long, medium, and short. The long-duration HIIT protocol (four minutes at high intensity -85-95% peakVO 2 -and three minutes at moderate intensity -60-70% of peakVO 2 ) is the most widely used in patients with cardiovascular diseases. Another metaanalysis demonstrated that the long-duration HIIT protocols were associated with larger improvements in peakVO 2 in healthy individuals [23] . In our meta-analysis, the studies comparing HIIT to MICT used long-duration HIIT protocol. This may be one of the main factors behind the superiority of HIIT over MICT [22] .
The concept of isocaloric protocol must also be considered when HIIT and MICT are compared. The superiority of HIIT over MICT disappeared when studies that used isocaloric protocols were analyzed in previous meta-analyses involving patients with coronary artery disease [11] and heart failure [10] . In our analysis, only one study [23] reported isocaloric protocol, which limits any pragmatic conclusion about HIIT superiority.
Additionally, our secondary analysis demonstrated that HIIT had superior effects to MICT in peak heart rate and heart rate reserve. In four of the five included studies, the mean time after transplantation was between four and six years [13,[16][17][18] . Initially, we supposed that this superiority would be associated with cardiac reinnervation. Long-term HTx recipients are expected to show some degree of cardiac reinnervation, which nearly normalizes heart rate control. In contrast, newly HTx recipients display a denervated status and the heart rate response is markedly reduced compared to health individuals. On the other hand, in one study, HTx recipients were followed 8-12 weeks after HTx [12] . The apparent superiority of HIIT may be associated with the volume of training [23] . Additionally, the small number of studies and of isocaloric protocols does not allow us to fully support the superiority of HIIT over MICT in cardiac dynamics.

Limitation
Caution is warranted in interpreting the results of this study. One important limitation is the low number of studies comparing HIIT to MICT and the lack of isocaloric protocols. Another important limitation is that in just one study [13] the patients were followed by a physical therapist in a 1:1 setting. This type of setting may guarantee that the patients reach the proper intensity prescribed. On the other hand, this type of setting may not be suitable in most cardiac rehabilitation centers around the world, especially in middle-and lower-income countries.

CONCLUSION
In conclusion, current available evidence suggests that HIIT is superior to MICT on peakVO 2 , peak heart rate, and heart rate reserve improvements in HTx recipients. However, new RCTs are necessary to analyze the influence of isocaloric protocols and different duration protocols on peakVO 2 in HTx recipients.