Applied Psychophysiology and Biofeedback (2021) 46:215–226 https://doi.org/10.1007/s10484-020-09495-8 Can HRV Biofeedback Improve Short-Term Effort Recovery? Implications for Intermittent Load Sports Mauro Perez‑Gaido1 · Jaume F. Lalanza1,2 · Eva Parrado1,3 · Lluis Capdevila1,3 Received: 21 May 2020 / Revised: 31 August 2020 / Accepted: 20 November 2020 / Published online: 5 January 2021 © The Author(s) 2021 Abstract As intensity and physical demands continue to rise in sport competition, faster and better recovery becomes essential. The aim of this study was to assess the effects of HRV biofeedback (HRVB) while recovering from a submaximal aerobic exer- cise. Ten physically-active graduate students participated in the study, which was conducted in four sessions: exercise with free-breathing recovery, first resonance frequency (RF) detection, second RF detection, and exercise applying HRVB during recovery. Measurements included time spent running and recovering, HRV parameters, and recovery/exertion perceptions. The results indicate that using HRVB during recovery improves cardiac variability (RRmean, SDNN, RMSSD and LF; p < 0.01). HRVB also lowers recovery time (p < 0.05) and seems to be improving the perception of recovery (p = 0.087). Moreover, time spent exercising (p < 0.01) and perceived physical exertion (p < 0.05) were higher when applying HRVB. The improvement in the psychophysiological adaptation after intensive aerobic exercise provided by the HRVB is a valuable benefit, not only for competition-driven athletes, but also for the general population. Keywords Recovery · Breathing · Resonance frequency · HRV · Biofeedback · HRVB In modern sport competition, intensity and physical demands et al. 2005), active recovery and stretching (Bahnert et al. continue to rise, making recovery a precious resource to 2013). Current methods of recovery assessment include minimize overtraining, injury occurrence and increase per- physiological (heart rate variability, HRV; level of lactate formance. In order to continue performing at maximum on blood; etc.) and psychological measures (self-report level, it becomes paramount to develop and apply strategies questionnaires, e.g. POMS, REST-Q-Sport; Total Recovery that lead athletes to regain an optimal psychophysiological Questionnaire, TQR; Kenttä and Hassmén 1998). state as fast as possible. Kellman and Kallus (2001) defined HRV parameters are currently widely used as recovery recovery as a process that restores the abilities related to markers (Buchheit 2014; Djaoui et al. 2017; Michael et al. performance. It has intra- and inter-individual components, 2017). HRV could be defined as the variation in time inter- such as psychological, physiological, and social processes. vals (R–R intervals) between adjacent heartbeats (Shaffer Several strategies has been reported to achieve a better and Ginsberg 2017). Higher levels of HRV are associated recovery state, including an individual and sport-specific with health, adaptability, resilience, and self-regulatory diet (Wardlaw and Smith 2006), proper hydration, passive capacity (McCraty and Shaffer 2015). HRV parameters recovery, sufficient sleep, physical and mental relaxation reflect regulation of autonomic balance, blood pressure techniques (Kudlackova et al. 2013), massages (Weerapong (BP), gas exchange, heart, and vascular tone. Different HRV parameters reflect different phenomena, and they are * Lluis Capdevila divided into time domain and frequency domain parameters. lluis.capdevila@uab.cat Among the time domain parameters, the standard deviation of all R–R intervals (SDNN) reflects all the cyclic compo- 1 Department of Basic Psychology, Universitat Autónoma de nents responsible for variability in a given period (Laborde Barcelona, Bellaterra, Barcelona, Spain et al. 2017). The root mean square of successive differences 2 Department of Psychology, UiT The Arctic University (RMSSD) represents the vagal tone, and is more influenced of Norway, Tromsø, Norway by the parasympathetic nervous system (PNS Hill et al. 3 Sport Research Institute UAB, Universitat Autònoma de 2009). Frequency domain parameters mainly include low Barcelona, Bellaterra, Barcelona, Spain 13 Vol.:(0123456789) 216 Applied Psychophysiology and Biofeedback (2021) 46:215–226 frequency (LF), high frequency (HF), and the ratio of LF to physical effort to allow a reduction in the time spent to HF (LF/HF). LF indexes a mix of activity from the sympa- recover and provide a better psychophysiological state for thetic nervous system (SNS), vagus nerve, and baroreflexes. the next bout of exercise. It is hypothesized that partici- LF is highly influenced by slow paced breathing (less than pants, by using HRVB during recovery, will have a better 8.5 breaths per min; Sowder et al. 2010). LF/HF represents psychophysiological adaptation during the recovery-stress a mix of SNS and vagal activity, whereas HF power is gen- process. erated by the PNS. In short-term recordings of SDNN, spe- cifically, the source of its fluctuation is respiratory sinus arrhythmia (RSA; Shaffer et al. 2014). In addition to the time and frequency parameters of the HRV, there are non- Method linear parameters, such as approximate entropy, which are also widely used (Huikuri et al. 2003). Participants Continuing on Vaschillo’s original studies in Russia, Lehrer et al. (2000) published a manual for training in reso- Fifteen physically active graduate university students (7 nant frequency biofeedback (currently called HRV biofeed- males), ranging from 23 to 35 years (25.80 ± 4.18 years), back, HRVB) to increase cardiac variability. Such work was participated in this study. The level of exercise was the inclu- the underpinning of almost two decades worth of research sion criteria, and it was measured using the Spanish adapted and applied techniques. By stimulating the cardiovascular version of the Physical Activity Stages of Change Ques- system at a specific frequency, HRVB produces very large tionnaire (PASC-Q; Marcus et al. 1992). This only included sinusoidal oscillations in heart rate (HR) and in BP, a phe- participants that exercised a minimum of three times a week, nomenon called RSA. This is a resonance effect triggered at least 30 min each session, and were at least in action by breathing at a certain frequency, usually called resonance phase (according to the Transtheoretical Model; Marcus and frequency (RF). Simkin 1994). All participants that answered positively to Several applications using HRVB have been discovered any item of the Physical Activity Readiness Questionnaire over the years. A recent systematic review and meta analysis (PAR-Q, Thomas et al. 1992) or were unable to run past the (Lehrer et al. 2020) focused on biological, psychological warm-up phase, were excluded from the study. Of the 15 and performance-related issues, and found small to moderate participants that started the study, 10 were finally considered effect sizes in HRVB treatments. However, valuable results for data analysis (see Table 1) and 5 were excluded: 1 did were found among clinical applications on asthma, chronic not complete all of the sessions, 1 was excluded because of a obstructive pulmonary disease, functional gastrointestinal positive item in PAR-Q, 2 were disregarded for not complet- disorders, fibromyalgia, hypertension, chronic muscle pain, ing the warm-up phase, and 1 for coughing during a recovery gynecology, depression, anxiety disorders, and sleep (for phase. Privacy was assured for all participants as regards a review: Gevirtz 2013). HRVB has been used to measure all data collected. All participants were volunteers and pro- training status and adaptive physical processes (Buchheit vided written informed consent. The study was reviewed and 2014; Nakamura et al. 2017). Moreover, a review by Jiménez approved by the local ethics committee and meets the Ethical and Molina (2017) reported that this technique has shown Standards in Sport and Exercise Science Research (Harris improvement in performance in areas such as baseball, golf, and Atkinson 2009). dance, music, basketball, volleyball, long-distance running, and soccer. Within different sports, HRVB has been used to improve concentration, anxiety, stress, and emotional control Table 1  Descriptive statistics for participants (Lagos et al. 2011; Paul and Garg 2012; Paul et al. 2012; Mean SD Whited et al. 2014; Gross et al. 2016; Rijken et al. 2016). Moreover, in aerobically fit individuals, HRV recovers faster Age 25.80 4.18 after exercise (Stanley et al. 2013). As indicated by Peçanha Height 174.85 8.69 et al. (2013), it is necessary to foster interventions that have Women 167.00 3.00 a positive impact on HRV, which is known to decrease after Men 178.21 8.13 exercise. Weight 75.54 8.37 A thorough search of the relevant literature yielded no Women 64.93 5.40 results regarding any HRVB application for short-term Men 80.09 3.87 recovery after an effort or physical stress. Improving Exercise per week 430.50 199.51 recovery would be useful for athletes and any person inter- Time exercising 38.90 20.14 ested in practicing sports or physical exercise. Therefore, Height expressed in centimeters, weight in kilograms, exercise per the aim of this study was to apply HRVB after making a week in minutes and time exercising in weeks. n = 10 13 Applied Psychophysiology and Biofeedback (2021) 46:215–226 217 Measures Procedure Cardiac Measures A within-subject design was used in this study. Participants were contacted via email. After completing an online ver- The PolarBand H7 (Polar Electro, Finland) was used to sion of the PASC-Q, they were selected according to the detect the R–R intervals signal. HealthSportLab software inclusion criteria. The study was conducted in four sessions, (HealthSportLab S.L., Barcelona, Spain) was used to in chronological order (see Fig. 1): (1) exercise with free- detect and collect the R–R interval signal and calculate breathing recovery (Free Session); (2) first RF detection; (3) the cardiorespiratory variables. For time domain analysis, second RF detection; and (4) exercise with HRVB recov- HR average (HRmean), R–R intervals average (RRmean), ery (RF Session). The sessions were approximately 1 week SDNN, and RMSSD were calculated, and for frequency separated from each other. Before every session participants domain, low frequency (0.04–015 Hz) and LF/HF ratio completed a HRV control booklet (Parrado et al. 2010) to were computed. During recovery phases, measurements assess behaviors that could affect HRV measurement (car- were made in bouts of 90 s (Ultra Short Term HRV; Esco dioactive medication, physical activity practice, caffeine, et al. 2018; Melo et al. 2018; Nakamura et al. 2017). HRV tobacco, alcohol, food intake, sleeping quality and quantity, analysis complies with all guidelines recommended by the and taking of contraceptives), as recommended by Laborde Task Force of the European Society of Cardiology and the et al. (2017). Participants were asked to control such behav- North American Society of Pacing and Electrophysiology iors to the maximum of their capability in the hours before (1996) for measurement of HRV. to the sessions. In the first session, participants completed an informed consent form and the PAR-Q questionnaire. Sub- maximal HR (80%) was calculated for each participant in Self‑reported Measures of Recovery and Physical Exertion order to control their effort during exercise, using Tanaka’s formula (2001): HRmax = 208 − 0.7 × age. This formula was Perception of recovery was assessed using a modified validated in runners Nikolaidis et al. (2018). version of the Total Quality Recovery perceived scale The objective of the first session (Free Session, see (TQRper; Kenttä and Hassmén 1998). Scores on this Fig. 2a) was to perform and recover twice from a sub- version range from 1 to 10, where “1” corresponds to no maximal aerobic exercise test by the participants, without recovery at all, and “10” to maximal recovery. Perception giving any instructions regarding how they should breath of exertion was evaluated using the Borg’s Category Ratio during recovery. First of all, a 5-min HRV baseline (PRE) Scale (CR-10; Borg and Löllgen 2001). Scores range from was measured while the participant was seated on a chair. 1 to 10, where “1” corresponds to no exertion, and “10” to Afterwards, the exercise started with a warm-up, running maximal exertion. for 3 min at a speed of 6 km/h and an inclination of 1%. The speed was then increased by1 km/h each minute (e.g., Fig. 1  Diagram of the different sessions 13 218 Applied Psychophysiology and Biofeedback (2021) 46:215–226 Fig. 2  a Example of HR plot for a Free Exercise session (Partici- cise session (Participant 9). Below the graph there are the different pant 9). Below the graph there are the different phases of the ses- phases of the session. Heart Rate (HR) expressed in beats per minute; sion. Heart Rate (HR) expressed in beats per minute; Time in X axis Time in X axis expressed in minutes and seconds expressed in minutes and seconds. b Example HR plot of a RF Exer- 3rd–4th min 7 km/h, 4th–5th min 8 km/h, etc.). When the calculated as the time taken to reach 50% HR, immediately participants reached 80% of their HR, the speed of the after sitting down on the chair, in each recovery phase. treadmill was slowly decreased until full stop. The partici- In the second and third session, participants attended pant was instructed to sit on a chair for 2 min. No breath- the RF detection sessions to assess their RF and then, learn ing instruction was given. Immediately afterwards, the and practice the individualized breathing protocol. RF exercise test and subsequent recovery period was repeated was assessed following a modified version of the protocol according to the same procedure. After the second recov- of Lehrer et al. (2000). The modification consisted of the ery, the subjective measurements of TQRper and CR-10 breathing frequencies used (7, 6.5, 6, 5.5, and 5 breaths per were taken using a written scale. Finally, a 5-min HRV min). Following an auditory pacer, the participants breathed (POST) assessment was performed. Exercise performance at every frequency for 2 min each without a pause in- was calculated as the time spent running from the start between and at the same inspiration/expiration ratio (without of each exercise (including the warm-up) until the end of holding). The RF was selected automatically by the FitLab a 15-s period with at least 80% HR. Recovery time was Team app (HealthSportLab S.L., Barcelona, Spain), taking 13 Applied Psychophysiology and Biofeedback (2021) 46:215–226 219 into account the highest RSA amplitude during the different maximum signal error for RR records of 11% was accepted frequencies. Afterwards, participants practiced following and filtered. The mean of signal error for all RR records their RF for 10 min using Breath Pacer app (T.C. Appli- was 0.67%. One participant was discarded from the PRE cations Inc., Toronto, Canada), while measuring HRV, and assessment data due to software malfunction. Thus, all finally, completed a breathing checklist to assess whether it analyses that included a variable of the PRE phase were was comfortable to follow the RF. performed using n = 9. Wilcoxon’s test was used to analyze The last session (RF session, see Fig. 2b) was aimed at the overall 2 × 2 design (session × exercise), and to com- using the RF while recovering after the exercise bouts. Tak- pare related variables. Data gathered during the exercises ing into account that RF can be not stable, the frequency was analyzed by a repeated measures ANOVA with exer- used in the last session was calculated as the mean between cise (2) as a within-subject factor and gender as a between- the two frequencies obtained in both RF detection sessions. subject factor. The software used to perform the statistical The mean of the RF used by the participants during the analysis was the SPSS predictive analytics, version 23.0 application of HRVB was 5.7 breaths/min (0.6). This session (SPSS Inc., Chicago, IL, United States). was conducted in the same manner as the first session: first exercise test and first recovery phase; second exercise test and second recovery phase. The difference was in the recov- ery phases in which the HRVB protocol was applied. The Results HRVB protocol consisted of instructing the participant to breathe first at double the individual RF for 30 s and then at Recovery Time RF for 90 s (always using the same auditory and visual pacer as the RF detection sessions). The period of 30 s breathing As regards the recovery phase, the recovery time (time at double the RF was meant to provide a better adaptation to spent until 50% of HR) in the RF session was significantly the paced breathing, since RF is low and difficult to follow, lower than in the Free session (Z = − 2.073, p = 0.038). after a sub-maximal physical exercise test. Furthermore, the recovery time in the first exercise of the RF session had a tendency to be significantly lower Statistical analysis than in the first exercise of the Free Session (Z = − 1.897, p = 0.058). Similarly, recovery time in the second exercise All data were reported as means (standard deviation, of the RF session was significantly lower than in the sec- SD). The standard error of the mean (SEM) is reported ond exercise of the Free Session (Z = − 1.956, p = 0.05) in the figures. The significance level used is p < 0.05. A (see Table 2 and Fig. 3). Table 2  Data gathered during Variables Free session RF session the exercises Exercise 1 Exercise 2 Exercise 1 Exercise 2 Recovery Time to 50% HR 56.40 (30.0) 66.90 (32.8) 45.80 (30.2) 55.40 (28.2)* HR mean 94.30 (12.3) 99.40 (11.0) 88.50 (13.3)* 94.60 (12.3)* RR mean 654.09 (86.1) 619.47 (70.4) 723.10 (110.3)** 669.42 (85.7)** SDNN 63.21 (16.6) 62.67 (21.4) 115.57 (32.5)** 99.86 (32.4)** RMSSD 35.39 (21.7) 24.57 (20.1) 65.87 (36.1)* 47.50 (26.5)** LF 1311.69 (1134.9) 1835.30 (2410.7) 8305.09 (5297.1)** 5930.47 (5122.7)** Exercise performance Time spent running 369.10 (127.2) 353.20 (127.3) 417.10 (151.4)** 385.30 (139.1)** Subjective measures TQRper 6.90 (0.5) 7.80 (0.2) CR-10 3.30 (0.4) 4.80 (0.6)* Values are expressed as mean (SD); n = 10 (7 men and 3 women). Time expressed in seconds for Time spent running and Time to (achieve) 50% HR; values in milliseconds (ms) for RR mean, SDNN and RMSSD; values in m­ s2 for LF *p < 0.05, **p < 0.01 vs the corresponding exercise of the free session 13 220 Applied Psychophysiology and Biofeedback (2021) 46:215–226 Fig. 3  Time spent from the beginning of recovery until reaching 50%HR (Recovery time). Rec means recovery and Ex means exercise. Comparison between the mean of each ses- sion using Wilcoxon’s test (RF vs Free). Comparison between the corresponding recovery of the other session using Wil- coxon’s test (RF session Rec Ex 1 vs Free session Rec Ex 1; RF session Rec Ex 2 vs Free session Rec Ex 2). *p < 0.05. n = 10 HR and HRV Parameters During Recovery Phase second exercise of the Free Session (Z = − 2.666, p = 0.008; Z = − 2.668, p = 0.008). Within the recovery phase (Table 2 and Fig. 4), the mean HR was significantly lower in the RF Session when compared to Subjective Measures the Free Session (Z = − 2.346, p = 0.019). Moreover, in the RF Session the heart rate average (HRmean) in the first exer- In the subjective measures (see Table 2), the TQRper scores cise and in the second exercise was significantly lower when in the RF session had a tendency to be significantly higher compared to HRmean in the respective first exercise and than in the Free Session (Z = − 1.710, p = 0.087). The CR-10 second exercise of the Free Session (Z = − 2.176, p = 0.03; scores were significantly higher in the RF session than in the Z = − 2.321, p = 0.021). Regarding time domain HRV param- Free Session (Z = − 2.539, p = 0.011). eters, during the recovery phase of the RF session, higher values were obtained for RRmean (Z = − 2.803, p = 0.005), Gender Comparison for the Effects of HRVB SDNN (Z = − 2.803, p = 0.005) and RMSSD (Z = − 2.803, p = 0.005) compared to recovery phase of the Free session. The results show no gender differences by any HRV param- Furthermore, significantly higher values were obtained in eter as effects of HRVB intervention, comparing Free an RF the first exercise and second exercise of the RF Session com- sessions. There are also no gender differences in recovery pared to the values obtained in the respective first exercise and performance times. and second exercise for RRmean (Z = − 2.803, p = 0.005; Z = − 2.803, p = 0.005), SDNN (Z = − 2.803, p = 0.005 and PRE and POST Heart Rate Variability Comparisons Z = − 2.701, p = 0.007) and RMSSD (Z = − 2.497, p = 0.013 and Z = − 2.599, p = 0.009). Regarding frequency domain No significant differences were found between the PREs and parameters analysis, in the recovery phases, LF was signifi- the POSTs of each session in any HRV parameters. Simi- cantly higher in the RF session when compared to the Free larly, no significant differences were found in HRV param- Session (see Fig. 5; Z = − 2.803, p = 0.005). Moreover, the eters when comparing the mean of PRE and POST of RF LF in the first exercise and second exercise of the RF Session session against the mean of PRE and POST of Free session was higher when compared to LF in the respective first exer- (see Table 3). cise and second exercise of the Free Session (Z = − 2.803, p = 0.005 and Z = − 2.803, p = 0.005). Discussion Exercise Performance This study was aimed at assessing the effects of HRV bio- feedback (HRVB) on effort recovery. Participants under- As regards exercise performance (see Table 2 and Fig. 6), went a 4-session procedure that started with the first ses- the time spent running was significantly longer in the RF sion performing an aerobic exercise (running on a treadmill). Session than in the Free Session (Z = − 2.666, p = 0.008). Measurements included HRV parameters, running time, and Furthermore, the time spent running during the first exer- recovery time, as well as subjective recovery and exertion. cise and during the second exercise of the RF Session was The following two sessions served to establish and prac- significantly longer than the respective first exercise and tice HRVB. The final session was almost identical to the 13 Applied Psychophysiology and Biofeedback (2021) 46:215–226 221 Fig. 4  HRV parameters during PRE, recovery of exercises and POST. Wilcoxon’s test (RF session Rec Ex 1 vs Free session Rec Ex 1; RF Rec means recovery and Ex means exercise. Comparison between the session Rec Ex 2 vs Free session Rec Ex 2). RR mean, SDNN and mean of each session using Wilcoxon’s test (RF vs Free). Compari- RMSSD expressed in milliseconds. *p < 0.05, **p < 0.01. n = 10 son between the corresponding recovery of the other session using first one, with the single difference of applying HRVB dur- HR amplitudes produced by breathing at RF, which activates ing recovery. Results show improvement in all main HRV the PNS faster than in normal recovery. Breathing at RF parameters, with lower recovery times and higher exertion. makes HR rise and fall with more extreme peaks and valleys, From the moment that the participant sits down and starts making it more likely to achieve the 50% of the maximum recovering from a physical exertion, HR naturally starts fall- HR faster (Lehrer 2013; Lehrer and Gevirtz 2014). Differ- ing (Peçanha et al. 2017), as the autonomous nervous system ences in SDNN show a clear effect on HRV when breathing shifts from the predominantly sympathetic to the parasym- at RF. Differences in RMSSD imply differences in vagal pathetic branch (Michael et al. 2017). The faster recovery tone, indicating higher activation of the PNS (Shaffer and shown in this study can be explained as a result of higher Ginsberg 2017). According to Thayer et al. (2009), higher 13 222 Applied Psychophysiology and Biofeedback (2021) 46:215–226 Fig. 5  LF during recovery from exercises. Rec means recovery and Ex means exercise. Com- parison between the mean of each session using Wilcoxon’s test (RF vs Free). Comparison between the corresponding recovery of the other session using Wilcoxon’s test (RF ses- sion Rec Ex 1 vs Free session Rec Ex 1; RF session Rec Ex 2 vs Free session Rec Ex 2). LF expressed in milliseconds squared ­(ms2). **p < 0.01. n = 10 Fig. 6  Time spent running from the beginning of the warm- up until reaching 80%HR. Ex means exercise. Comparison between the mean of each ses- sion using Wilcoxon’s test (RF vs Free). Comparison between the corresponding exercise of the other session using Wil- coxon’s test (RF session Rec Ex 1 vs Free session Rec Ex 1; RF session Rec Ex 2 vs Free session Rec Ex 2). **p < 0.01. n = 10 Table 3  Data gathered before (PRE) and after (POST) sessions recovery according to Chen et al. (2011). The improvement Variables Free session RF session in recovery can be explained by the previously mentioned RF breathing effect that produces higher variability in the PRE POST PRE POST HR and it has two main effects. Firstly, aiding the cardio- HR mean 72.9 (10.0) 83.0 (7.4) 69.2 (10.7) 82.4 (9.7) vascular system in adapting to the changes in exterocep- RR mean 836.9 (118.4) 728.1 (62.2) 884.4 (133.4) 736.7 (85.2) tive and interoceptive demands (in this case from a stressful SDNN 71.8 (23.8) 49.7 (9.8) 75.9 (15.1) 45.2 (11.9) environment to a non-stressful environment): thus return- RMSSD 42.4 (15.0) 23.3 (7.7) 46.1 (14.2) 21.7 (11.8) ing to a faster and better adjusted rest (Lehrer and Eddie LF/HF 4.4 (3.6) 10.5 (10.3) 8.0 (13.6) 10.3 (10.5) 2013). Secondly, increasing efficiency in gas exchange that takes place in the alveoli (Hayno et al. 1996; Yasuma and Time expressed in milliseconds for RR mean. Values are expressed as Hayano 2004; Ramos-Campo et al. 2017), therefore provid- mean (SD). n = 9 ing an enhancement in the oxygen uptake and the reduction of blood carbon dioxide levels, characteristic of the process vagal tone relates to better executive cognitive performance called Excess Post-Exercise Oxygen Consumption (EPOC; (key in sports), emotional, and health regulation. Further- LaForgia et al. 2006). more, in view of the fact that vagal activity protects the Participants were able to correctly assess their internal cardiovascular system (Peçanha et al. 2013), prompt vagal load, evidenced by the fact that both perceived exertion and recovery means protection against cardiac injuries for the the time-measured exercise performance increased (Borre- athlete. Also, higher PNS activity correlates positively with sen and Lambert 2008). Although exercise performance and 13 Applied Psychophysiology and Biofeedback (2021) 46:215–226 223 perceived exertion of that session was significantly higher, sufficiently large sample and balancing it according to the the perception of recovery after the RF session was close level of physical activity. to significance when compared to the Free session. This The main limitation of the study was the difference meant that the sample of participants, thanks to the HRVB, shown in exercise performance between the first exercises had a tendency to perceive that they were more recovered. of each session. The first exercises of each session should This impact on a subjective measurement suggests that not have resulted in significant differences, because up to more psychological measures related to recovery should be these instances both sessions were equal. Only after the first taken into consideration when assessing the quantitative and exercise, Free session and RF session were different: the qualitative consequences of HRVB on recovery, such as the first had no breathing instruction and the latter did. This Acute Recovery and Stress Scale and the Short Recovery effect has different explanations. Improvements in fitness and Stress Scale (Nässi et al. 2017). in some participants could have been noticeable in the time As regards the comparison of PRE and POST measure- spent running, but not in the HRV parameters during PRE ments between Free and RF session, the most interesting assessment (as indicated above). This mismatch has been result is the absence of significant differences between the reported in previous literature, and the fitness-related link PREs of both sessions. This result is indicative of simi- between distance-based performance and HRV parameters lar levels of physical fitness (Buchheit 2014) and stress has still to be clarified (Buchheit 2014). Another explana- (Morales et al. 2013), meaning that the participants arrived tion could be that the participants got used to running on the at both sessions in a similar state. Moreover, this result treadmill. Fixing this limitation could be done by adding a shows that there is no long-lasting or chronic effects from specific session to accustom participants to the instruments the RF detection sessions that could have improved HRV and experimental context, or comparing participants to a parameters, therefore disregarding common persistency control group, or adding a second RF session at the end, effects from HRVB training (Deschodt-Arsac et al. 2018) or using an incremental maximal running test (Weberruss and implying only acute effects on HRV. et al. 2018). The latter suggestion would be rather invasive, The design of the procedure was robust because of the and we did not want to change the daily routine of the par- differences seen in both SDNN and RMSSD. While both ticipants. The use of a control group is not the best option, are parasympathetically influenced, SDNN is more reli- because of difficulties in interpersonal comparisons of HRV able against artifacts (because abnormal beats have been (Laborde et al. 2017). removed); and even though RMSSD is more susceptible to There are several applications suggested by the results artifacts, it correlates highly with vagal tone (Shaffer and of this study. Every time an athlete performs, they exert a Ginsberg 2017). Moreover, the signal error average for all physical and psychological effort that needs to be recovered RR records was 0.67%, therefore measurements of both from; a process that takes time and can be improved by fol- parameters are valid. lowing these specific instructions. To the best of our knowl- Differences in LF showed that the participants were edge, HRVB has never been used before during short-term breathing at a low frequency within the recovery period, recovery. The effective decrease in recovery times and the as they were supposed to (see Fig. 5). LF is composed by improvement in the quality of such recovery would allow frequencies that lie between 0.04 and 0.15 Hz, and the par- the athletes to return to competition faster and in a better ticipants were instructed to breath in frequencies from within psychophysiological state. Sports with intermittent loads that range (0.08–0.12 Hz; Laborde et al. 2017). This meant such as basketball, handball, volleyball, tennis, etc., would that if the value of LF was higher in the session in which particularly benefit from this, because of their characteris- participants had to breath at a low frequency (RF session) in tic exertion-pause-exertion nature that provides intervals to comparison to the session in which they had no instruction apply HRVB. During competition, many periods of time are (Free session), then they were indeed breathing within the available for applying HRVB, such as during substitutions, range that they were supposed to. This effect is due to RSA in between points, games and sets, during halftime, etc. (Steffen et al. 2017). (Lagos et al. 2011; Shaw et al. 2012). Emotional regulation Vaschillo et al. (2006) find gender differences in HRVB, is another benefit that the application of HRVB can bring, but we did not find any gender difference in the effects of such as demonstrated in a study by Gross et al. (2016) using HRVB. Possibly in our case it is due to the small size of similar principles. Recovery after exercises in training ses- the sample (there are only 3 women and 7 men in the final sions would also be shorter (therefore freeing training time) sample). Our goal was not gender comparison, but the nor- and prepare the athlete better for the next activity. Amateur mal differences in running performance between women and athletes could benefit from HRVB too, for example by using men could affect the results. Therefore, it would be interest- it after an aerobic bout of exercise. ing to analyze possible gender differences in HRVB with a Further research can provide a wider and deeper knowl- edge of the application of HRVB during recovery. First of 13 224 Applied Psychophysiology and Biofeedback (2021) 46:215–226 all, improving the HRVB protocol by making the partici- Ethical Approval The study was reviewed and approved by the local pants breathe longer at RF should provide more noticeable ethics committee and meets the Ethical Standards in Sport and Exercise Science Research (Harris and Atkinson 2009). The procedures used in effects and even benefits. Using different physical tests, this study adhere to the tenets of the Declaration of Helsinki. such as an incremental maximal running test, could check the HRVB effectiveness in different situations. Conduct- Informed Consent All participants were volunteers and provided writ- ing research on a sport-specific sample would shed light on ten informed consent. distinguishing appropriate sport settings that would benefit from using this HRVB protocol and its effects on sport per- Open Access This article is licensed under a Creative Commons Attri- formance. Applying it on physically non-active participants bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long may corroborate assumptions on whether HRVB works in as you give appropriate credit to the original author(s) and the source, such population. Assessing more psychological variables, provide a link to the Creative Commons licence, and indicate if changes such as relaxation (a stress management protocol with were made. The images or other third party material in this article are similar theoretical foundations have proven effective, Gross included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in et al. 2018), concentration, emotion regulation, anxiety, and the article’s Creative Commons licence and your intended use is not mood, could show advantages not previously thought of. permitted by statutory regulation or exceeds the permitted use, you will Enlarging the sample may give rise to increasing differences need to obtain permission directly from the copyright holder. To view a that, in the current sample, were not large enough to result copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. in significance. It would be beneficial to try to integrate our HRVB protocol to competition and training environments, in order to assess specific effects and obstacles. HRVB should also be applied in early beginners. It is important that young References athletes become aware of the availability of such a benefi- Bahnert, A., Norton, K., & Lock, P. (2013). Association between post- cial technique that bring improvements in wellbeing and game recovery protocols, physical and perceived recovery, and performance. Finally, researching the effects of the HRVB performance in Elite Australian Football League players. Journal protocol on sport-related outcomes (such as scoring, passes of Science and Medicine in Sport, 16(2), 151–156. https​://doi. completed, shots taken, etc.) would be of great interest. org/10.1016/j.jsams​.2012.05.008. Borg, G., & Löllgen, H. (2001). Borg’s perceived exertion and pain scales. Deutsche Zeitschrift Fur Sportmedizin, 52(9), 252. https​ ://doi.org/10.1097/00005​768-19980​9000-00018​. Conclusion Borresen, J., & Lambert, M. I. (2008). Quantifying training load: A comparison of subjective and objective methods. International Journal of Sports Physiology and Performance, 3(1), 16–30. https​ Our results show that using HRVB during recovery improves ://doi.org/10.1123/ijspp​.3.1.16. cardiac variability and the time spent exercising, lowers Buchheit, M. (2014). Monitoring training status with HR measures: Do recovery time, and improves the perceived physical exer- all roads lead to Rome? Frontiers in Physiology, 5, 73. https:​ //doi. org/10.3389/fphys​.2014.00073​. tion and the recovery perception. Faster and better recovery Chen, J. L., Yeh, D. P., Lee, J. P., Chen, C. Y., Huang, C. Y., Lee, S. is an advantageous resource in current competitive sport D., et al. (2011). Parasympathetic nervous activity mirrors recov- environments. The benefit–cost ratio of applying HRVB dur- ery status in weightlifting performance after training. Journal of ing recovery would make it highly valuable to a handful of Strength and Conditioning Research, 25(6), 1546–1552. https​:// doi.org/10.1519/JSC.0b013​e3181​da785​8. sport-related situations. The rapid establishment of personal Deschodt-Arsac, V., Lalanne, R., Spiluttini, B., Bertin, C., & Arsac, L. RF and the effortless change in the inherent physiological M. (2018). Effects of heart rate variability biofeedback training in process of breathing, give rise to powerful effects on recov- athletes exposed to stress of university examinations. PLoS ONE, ery and, therefore, to the practice of sport itself. 13, 1–13. https​://doi.org/10.1371/journ​al.pone.02013​88. Djaoui, L., Haddad, M., Chamari, K., & Dellal, A. (2017). Monitor- ing training load and fatigue in soccer players with physiologi- cal markers. Physiology and Behavior, 181, 86–94. https​://doi. Funding This study was supported by the grants DEP2015-68538-C2- org/10.1016/j.physb​eh.2017.09.004. 1-R and PID2019-107473RB-C21 from the “Ministerio de Ciencia e Esco, M. R., Williford, H. N., Flatt, A. A., Freeborn, T. J., & Naka- Innovación” of the Spanish Government. mura, F. Y. (2018). Ultra-shortened time-domain HRV parameters at rest and following exercise in athletes: An alternative to fre- Compliance with Ethical Standards quency computation of sympathovagal balance. European Jour- nal of Applied Physiology, 118, 175–184. https:​ //doi.org/10.1007/ s0042​1-017-3759-x. Conflict of interest The authors declare that the research was con- Gevirtz, R. (2013). The promise of heart rate variability biofeedback: ducted without any commercial or financial relationships that could be Evidence-based applications. Biofeedback, 41(3), 110–120. https​ construed as a potential conflict of interest. ://doi.org/10.5298/1081-5937-41.3.01. Gross, M. J., Bringer, J. D., Kilduff, L. P., Cook, C. J., Hall, R., & Shearer, D. A. (2018). A multi-modal biofeedback protocol to 13 Applied Psychophysiology and Biofeedback (2021) 46:215–226 225 demonstrate physiological manifestations of psychological stress Lehrer, P. M., Vaschillo, E., & Vaschillo, B. (2000). Resonant fre- and introduce heart rate variability biofeedback stress manage- quency biofeedback training to increase cardiac variability: ment. Journal of Sport Psychology in Action, 9(4), 216–226. https​ Rationale and manual for training. Applied Psychophysiology ://doi.org/10.1080/21520​704.2018.14962​10. Biofeedback, 25, 177–191. https:​ //doi.org/10.1023/A:100955​ 4825​ Gross, M. J., Shearer, D. A., Bringer, J. D., Cook, C. J., & Kilduff, L. P. 745. (2016). Abbreviated resonant frequency training to augment heart Marcus, B. H., Rossi, J. S., Selby, V. C., Niaura, R. S., & Abrams, D. B. rate variability and enhance on-demand emotional regulation in (1992). The stages and processes of exercise adoption and mainte- elite sport support staff. Applied Psychophysiology and Biofeed- nance in a worksite sample. Health Psychology, 11(6), 386–395. back, 41, 263–274. https​://doi.org/10.1007/s1048​4-015-9330-9. https​://doi.org/10.1037/0278-6133.11.6.386. Harriss, D. J., & Atkinson, G. (2009). Ethical standards in sport and Marcus, B. H., & Simkin, L. R. (1994). The transtheoretical model: exercise science research. International Journal of Sports Medi- Applications to exercise behavior. Medicine & Science in Sports cine, 30(10), 701–702. https​://doi.org/10.1055/s-0029-12373​78. & Exercise, 26(11), 1400–1404. Hayano, J., Yasuma, F., Okada, A. M., Mukai, S., & Fujinami, T. Mccraty, R., & Shaffer, F. (2015). Heart rate variability: New perspec- (1996). Respiratory sinus arrhythmia. A phenomenon improving tives on physiological mechanisms, assessment of self-regulatory pulmonary gas exchange and circulatory efficiency. Circulation, capacity, and health risk. Global Advances in Health and Medi- 94, 842–847. https​://doi.org/10.1161/01.CIR.94.4.842. cine, 4(1), 46–61. https​://doi.org/10.7453/gahmj​.2014.073. Hill, L. B. K., Siebenbrock, A., Sollers, J. J., & Thayer, J. F. (2009). Melo, H. M., Martins, T. C., Nascimento, L. M., Hoeller, A. A., Are all measures created equal? Heart rate variability and respira- Walz, R., & Takase, E. (2018). Ultra-short heart rate variability tion. Biomedical Sciences Instrumentation, 45, 71–76. recording reliability: The effect of controlled paced breathing. Huikuri, H. V., Mäkikallio, T. H., & Perkiömäki, J. (2003). Meas- Annals of Noninvasive Electrocardiology, 23(5), 1–9. https​://doi. urement of heart rate variability by methods based on nonlinear org/10.1111/anec.12565​. dynamics. Journal of Electrocardiology, 36, 95–99. https​://doi. Michael, S., Graham, K. S., & Davis, G. M. (2017). Cardiac autonomic org/10.1016/j.jelec​troca​rd.2003.09.021. responses during exercise and post-exercise recovery using heart Jiménez Morgan, S., & Molina Mora, J. A. (2017). Effect of heart rate rate variability and systolic time intervals—A review. Frontiers variability biofeedback on sport performance, a systematic review. in Physiology, 8, 301. https​://doi.org/10.3389/fphys​.2017.00301​. Applied Psychophysiology Biofeedback, 42(3), 235–245. https​:// Morales, J., Garcia, V., García-Massó, X., Salvá, P., Escobar, R., & doi.org/10.1007/s1048​4-017-9364-2. Buscà, B. (2013). The use of heart rate variability in assessing Kellman, M., & Kallus, K. W. (2001). Recovery-stress questionnaire precompetitive stress in high-standard judo athletes. Interna- for athletes: User manual. Champaign, IL: Human Kinetics. tional Journal of Sports Medicine, 34(2), 144–151. https​://doi. Kenttä, G., & Hassmén, P. (1998). Overtraining and recovery: A con- org/10.1055/s-0032-13237​19. ceptual model. Sports Medicine, 26, 1–16. Nakamura, F., Pereira, L., Cal Abad, C., Cruz, I., Flatt, A., Esco, M., Kudlackova, K., Eccles, D. W., & Dieffenbach, K. (2013). Use of et al. (2017). Adequacy of the ultra-short-term HRV to assess relaxation skills in differentially skilled athletes. Psychology of adaptive processes in youth female basketball players. Journal Sport and Exercise, 14(4), 468–475. https​://doi.org/10.1016/j. of Human Kinetics, 56(1), 73–80. https​://doi.org/10.1515/hukin​ psych​sport​.2013.01.007. -2017-0024. Laborde, S., Mosley, E., & Thayer, J. F. (2017). Heart rate variability Nässi, A., Ferrauti, A., Meyer, T., Pfeiffer, M., & Kellmann, M. (2017). and cardiac vagal tone in psychophysiological research—Rec- Development of two short measures for recovery and stress in ommendations for experiment planning, data analysis, and data sport. European Journal of Sport Science, 17(7), 894–903. https​ reporting. Frontiers in Psychology, 8, 213. https:​ //doi.org/10.3389/ ://doi.org/10.1080/17461​391.2017.13181​80. fpsyg​.2017.00213​. Nikolaidis, P. T., Rosemann, T., & Knechtle, B. (2018). Age-predicted LaForgia, J., Withers, R. T., & Gore, C. J. (2006). Effects of exercise maximal heart rate in recreational marathon runners: A cross- intensity and duration on the excess postexercise oxygen con- sectional study on Fox’s and Tanaka’s equations. Frontiers in sumption. Journal of Sports Sciences, 24(12), 1247–1264. https​ Physiology, 9, 1–6. https​://doi.org/10.3389/fphys​.2018.00226​. ://doi.org/10.1080/02640​41060​05520​64. Parrado, E., García, M. Á., Ramos, J., Cervantes, J. C., Rodas, G., Lagos, L., Vaschillo, E., Vaschillo, B., Lehrer, P., Bates, M., & & Capdevila, L. (2010). Comparison of omega wave sys- Pandina, R. (2011). Virtual reality-assisted heart rate vari- tem and polar S810i to detect R-R intervals at rest. Interna- ability biofeedback as a strategy to improve golf perfor- tional Journal of Sports Medicine, 31(5), 336–341. https​://doi. mance: A case study. Biofeedback, 39(1), 15–20. https​://doi. org/10.1055/s-0030-12483​19. org/10.5298/1081-5937-39.1.11. Paul, M., & Garg, K. (2012). The effect of heart rate variability bio- Lehrer, P. (2013). How does heart rate variability biofeedback work? feedback on performance psychology of basketball players. Resonance, the Baroreflex, and other mechanisms. Biofeedback, Applied Psychophysiology Biofeedback, 37, 131–144. https​://doi. 41(1), 26–31. https​://doi.org/10.5298/1081-5937-41.1.02. org/10.1007/s1048​4-012-9185-2. Lehrer, P., & Eddie, D. (2013). Dynamic processes in regulation and Paul, M., Garg, K., & Sandhu, J. S. (2012). Role of biofeedback in some implications for biofeedback and biobehavioral interven- optimizing psychomotor performance in sports. Asian Journal tions. Applied Psychophysiology Biofeedback, 38(2), 143–155. of Sports Medicine, 3(1), 29–40. https​://doi.org/10.5812/asjsm​ https​://doi.org/10.1007/s1048​4-013-9217-6. .34722​. Lehrer, P. M., & Gevirtz, R. (2014). Heart rate variability biofeedback: Peçanha, T., Bartels, R., Brito, L. C., Paula-Ribeiro, M., Oliveira, R. How and why does it work? Frontiers in Psychology, 5, 756. https​ S., & Goldberger, J. J. (2017). Methods of assessment of the post- ://doi.org/10.3389/fpsyg​.2014.00756​. exercise cardiac autonomic recovery: A methodological review. Lehrer, P., Kaur, K., Sharma, A., Shah, K., Huseby, R., Bhavsar, J., International Journal of Cardiology, 227, 795–802. https​://doi. et al. (2020). Heart rate variability biofeedback improves emo- org/10.1016/j.ijcar​d.2016.10.057. tional and physical health and performance: A systematic review Peçanha, T., de Paula-Ribeiro, M., Nasario-Junior, O., & de Lima, J. and meta analysis. Applied Psychophysiology Biofeedback. https​ R. P. (2013). Post-exercise heart rate variability recovery: A time- ://doi.org/10.1007/s1048​4-020-09466​-z. frequency analysis. Acta Cardiologica, 68(6), 607–613. https​:// doi.org/10.2143/AC.68.6.80000​08. 13 226 Applied Psychophysiology and Biofeedback (2021) 46:215–226 Ramos-Campo, D. J., Rubio-Arias, J. A., Ávila-Gandía, V., Marín- variability: Standards of measurement, physiological interpreta- Pagán, C., Luque, A., & Alcaraz, P. E. (2017). Heart rate vari- tion and clinical use. Circulation, 93(5), 1043–1065. ability to assess ventilatory thresholds in professional basketball Thayer, J. F., Hansen, A. L., Saus-Rose, E., & Johnsen, B. H. (2009). players. Journal of Sport and Health Science, 6(4), 468–473. https​ Heart rate variability, prefrontal neural function, and cognitive ://doi.org/10.1016/j.jshs.2016.01.002. performance: The neurovisceral integration perspective on self- Rijken, N. H., Soer, R., de Maar, E., Prins, H., Teeuw, W. B., Peuscher, regulation, adaptation, and health. Annals of Behavioral Medicine, J., et al. (2016). Increasing performance of professional soccer 37(2), 141–153. https​://doi.org/10.1007/s1216​0-009-9101-z. players and elite track and field athletes with peak performance Thomas, S., Reading, J., & Shephard, R. J. (1992). Revision of the training and biofeedback: A pilot study. Applied Psychophysiol- Physical Activity Readiness Questionnaire (PAR-Q). Cana- ogy and Biofeedback, 41, 421–430. https:​ //doi.org/10.1007/s1048​ dian Journal of Sport Sciences, 17(4), 338–345. https​://doi. 4-016-9344-y. org/10.4141/S97-051. Shaffer, F., & Ginsberg, J. P. (2017). An overview of heart rate vari- Vaschillo, E., Vaschillo, B., & Lehrer, P. (2006). Characteristics of res- ability metrics and norms. Frontiers in Public Health, 5, 1–17. onance in heart rate variability stimulated by biofeedback. Applied https​://doi.org/10.3389/fpubh​.2017.00258​. Psychophysiology and Biofeedback, 31(2), 129–142. https​://doi. Shaffer, F., McCraty, R., & Zerr, C. L. (2014). A healthy heart is not org/10.1007/s1048​4-006-9009-3. a metronome: An integrative review of the heart’s anatomy and Wardlaw, G., & Smith, A. (2006). Contemporary nutrition (9th ed.). heart rate variability. Frontiers in Psychology, 5, 1–19. https:​ //doi. Boston, MA: McGraw Hill. org/10.3389/fpsyg​.2014.01040​. Weberruss, H., Maucher, J., Oberhoffer, R., & Müller, J. (2018). Recov- Shaw, L., Zaichkowsky, L., & Wilson, V. (2012). Setting the balance: ery of the cardiac autonomic nervous and vascular system after Using biofeedback and neurofeedback with gymnasts. Journal of maximal cardiopulmonary exercise testing in recreational athletes. Clinical Sport Psychology, 6(1), 47–66. European Journal of Applied Physiology, 118, 205–211. https​:// Sowder, E., Gevirtz, R., Shapiro, W., & Ebert, C. (2010). Restoration of doi.org/10.1007/s0042​1-017-3762-2. vagal tone: A possible mechanism for functional abdominal pain. Weerapong, P., Hume, P. A., & Kolt, G. S. (2005). The mechanisms Applied Psychophysiology Biofeedback, 35(3), 199–206. https​:// of massage and effects on performance, muscle recovery and doi.org/10.1007/s1048​4-010-9128-8. injury prevention. Sports Medicine, 35(3), 235–256. https​://doi. Stanley, J., Peake, J. M., & Buchheit, M. (2013). Cardiac parasympa- org/10.2165/00007​256-20053​5030-00004​. thetic reactivation following exercise: Implications for training Whited, A., Larkin, K. T., & Whited, M. (2014). Effectiveness of prescription. Sports Medicine, 43(12), 1259–1277. https​://doi. emWave biofeedback in improving heart rate variability reactivity org/10.1007/s4027​9-013-0083-4. to and recovery from stress. Applied Psychophysiology and Bio- Steffen, P. R., Austin, T., DeBarros, A., & Brown, T. (2017). The feedback, 39, 75–88. https​://doi.org/10.1007/s1048​4-014-9243-z. impact of resonance frequency breathing on measures of heart rate Yasuma, F., & Hayano, J. I. (2004). Respiratory sinus arrhythmia: Why variability, blood pressure, and mood. Frontiers in Public Health, does the heartbeat synchronize with respiratory rhythm? Chest, 5, 6–11. https​://doi.org/10.3389/fpubh​.2017.00222​. 125(2), 683–690. https​://doi.org/10.1378/chest​.125.2.683. Tanaka, H., Monahan, K. D., & Seals, D. R. (2001). Age-predicted maximal heart rate revisited. Journal of the American College Publisher’s Note Springer Nature remains neutral with regard to of Cardiology, 37(1), 153–156. https​://doi.org/10.1016/S0735​ jurisdictional claims in published maps and institutional affiliations. -1097(00)01054​-8. Task Force of the European Society of Cardiology and the North Amer- ican Society of Pacing and Electrophysiology. (1996). Heart-rate 13