Share this article on:

Faster Heart Rate Recovery With Increased RPE: Paradoxical Responses After an 87-km Ultramarathon

Mann, Theresa N.,1; Platt, Cathrin E.,1; Lamberts, Robert P.1,2; Lambert, Michael I.1

The Journal of Strength & Conditioning Research: December 2015 - Volume 29 - Issue 12 - p 3343–3352
doi: 10.1519/JSC.0000000000001004
Original Research

Mann, TN, Platt, CE, Lamberts, RP, and Lambert, MI. Faster heart rate recovery with increased RPE: paradoxical responses after an 87-km ultramarathon. J Strength Cond Res 29(12): 3343–3352, 2015—The aim of this study was to determine the relationship between heart rate recovery (HRR) and an acute training “overload” by comparing HRR responses before and after an ultramarathon road race. Ten runners completed a standardized laboratory protocol ∼7 days before and between 2 and 4 days after participating in the 87-km Comrades Marathon. The protocol included muscle pain ratings, a 5-bound test, and 20 minutes of treadmill exercise at 70% of maximal oxygen uptake followed by 15 minutes of recovery. Respiratory gases and heart rate measurements were used to calculate steady-state exercise responses, HRR, and excess postexercise oxygen consumption (EPOC), and participants also provided a rating of perceived exertion (RPE) during exercise. The RPE was significantly increased (13 ± 2 vs. 11 ± 1) (p < 0.01), and HRR was significantly faster (35 ± 5 beats vs. 29 ± 4 beats) (p < 0.01) following the postrace vs. prerace submaximal exercise bout, with no significant changes in respiratory or heart rate parameters during exercise or in EPOC. Although previous studies have shown that faster HRR reflected an “adapted” state with enhanced training status, the current findings suggest that this may not always be the case. It follows that changes in HRR should be considered in the context of other factors, such as recent training load and RPE during submaximal exercise.

1Division of Exercise Science and Sports Medicine, Department of Human Biology, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa; and

2Division of Orthopaedic Surgery, Department of Surgery Sciences, Faculty of Medicine and Health Sciences, Stellenbosch University, Tygerberg, South Africa

Address correspondence to Theresa N. Mann,

Back to Top | Article Outline


Adaptation to exercise training occurs in response to the stimulus of an increased training load. However, high training loads must be accompanied by appropriate recovery periods if an individual is to avoid becoming overreached and, in time, overtrained (25,41 25,41). The deleterious effects of overtraining on performance, psychological parameters, and physiological parameters have been well described in the literature (25,41 25,41) and have prompted ongoing interest into measurements that could be used to detect fatigue or overreaching before progression into overtraining has occurred (8,34,37 8,34,37 8,34,37). Ideally, measurements of this kind would not only provide an indication of when to reduce training load but also provide an indication of when an individual was adapting well and could tolerate an increase in training load. In other words, a sensitive measurement that was objective, easy to administer, and inexpensive could potentially serve as a valuable tool for athlete monitoring, fine-tuning day-to-day training prescription, and avoiding unanticipated decrements in performance.

As yet, no single measurement can be relied on to reflect an individual's current response to training and “readiness” to adapt to further training (6). However, there has been increasing interest in various measures of heart rate (HR) in this role (9), including measures of postexercise heart rate recovery (HRR) (20). Longitudinal studies have demonstrated significant relationships between changes in HRR and changes in performance measured before and after a standardized training program of 4 or 8–9 weeks (10,12,35,36 10,12,35,36 10,12,35,36 10,12,35,36), and a recent systematic review concluded that faster HRR was indicative of an improvement in training status and slower HRR was related to a decrease in training status (20). However, practical application of a monitoring tool would typically involve more frequent measurements (e.g., to inform day-to-day adjustment of training load), and it is unclear whether faster HRR in the short term can be interpreted as “increased readiness to train” in the same way that faster HRR over a few weeks could be interpreted as an improvement in training status.

In fact, there is some evidence that faster HRR may coincide with fatigue and poor “readiness to train” under some circumstances. For example, Dupuy et al. (22) reported faster HRR responses after 2 weeks of overload training in endurance athletes, and Lamberts et al. (34) reported faster HRR despite an increased rating of perceived exertion (RPE) following weeks of high training load in a case study of an elite cyclo-cross athlete. However, in contrast, Hug et al. (27) reported no change in HRR after 3 weeks of overload training, and Borresen and Lambert (6) found that 1 week of increased training load was associated with slower HRR responses. It is likely that differences in study design, including the training status of the participants, the relative load of the training, the characteristics of the exercise bout preceding the HRR measurement, the timing of the HRR measurements relative to the preceding training sessions, and the form of HRR used, may to a large extent explain these contrasting findings. Nevertheless, it would seem that HRR responses following an acute increase in training load may require further investigation before short-term changes in HRR can be interpreted with confidence.

In the current study, we anticipated that an ultra-endurance event would represent an acute training overload and serve as an experimental model that would help to clarify the acute effect of increased training load on HRR. Therefore, the main aim of the current study was to compare HRR responses before and (3 days) after the Comrades ultramarathon, an 87-km road race between Durban and Pietermaritzburg in South Africa.

The start and finish of the Comrades race are alternated annually, and in 2011, the race was an “up-run,” starting at sea level in Durban and finishing in Pietermaritzburg at an elevation of 650 m. The Comrades ultramarathon is a highly strenuous event and postrace vs. control measurements have shown significant disturbances in metabolic, hematological, and hormonal parameters along with significant muscle damage (14,40,46,47,53 14,40,46,47,53 14,40,46,47,53 14,40,46,47,53 14,40,46,47,53). It is logical to assume that Comrades participants would be in a poor state to adapt to training in the days after the ultra-endurance, “overload” event, and we hypothesized that this state would result in significantly slower HRR measurements 3 days after the race compared with prerace measurements.

Back to Top | Article Outline


Experimental Approach to the Problem

In this prospective cohort study, participants were asked to visit the laboratory on 3 occasions as follows: ±14 days before the 87-km Comrades ultramarathon, ±7 days before the race, and 3 days after the race. During the first visit, participants completed a short training history questionnaire followed by anthropometric measurements and a maximal incremental treadmill test. The second and third laboratory visits were identical and included muscle pain ratings, a 5-bound test, and 20 minutes of treadmill exercise followed by 15 minutes of controlled recovery (Figure 1). In the prerace exercise bout, treadmill speed was set to elicit 70% of maximal oxygen uptake (V[Combining Dot Above]O2max), and the postrace exercise bout was completed at the same treadmill speed. By comparing HRR responses before and after the acute overload of an ultramarathon road race, the current study aimed to provide important insight into the relationship between acute changes in training load and acute changes in HRR.

Figure 1

Figure 1

Back to Top | Article Outline


Ten men and 3 women who were registered to take part in the 2011 Comrades ultramarathon “up-run” (87 km) volunteered to participate in the study. The study was approved by the University Human Research Ethics Committee, and all participants signed an informed consent document after receiving a full explanation of the laboratory tests involved. Furthermore, all participants were able to answer “no” to all the questions in a Physical Activity Readiness Questionnaire (2). All 13 participants completed the prerace laboratory testing and the Comrades ultramarathon; however, 1 participant was subsequently diagnosed with a severe viral infection and 2 other participants chose to withdraw from the study because they were no longer willing to complete postrace testing. The characteristics of the remaining 10 participants are shown in Table 1.

Table 1

Table 1

Back to Top | Article Outline


Visit 1

Training History and Anthropometry

During the first visit, participants completed a short, 3-month training history questionnaire, including the typical number of training sessions per week, typical total training distance per week, and a recent marathon time. This was followed by height and body mass measurements using a stadiometer (Seca model 708; Seca, Hamburg, Germany) and calibrated scale (Seca model 708), respectively. Body mass was remeasured on each subsequent visit to the laboratory. Finally, 7 skinfolds were measured (Lange Skinfold Calipers; Beta Technology, Cambridge, MD, USA), and body fat % was determined according to the procedures and formulae described by the American College of Sports Medicine (1).

Back to Top | Article Outline
Maximal Incremental Test and Verification Bout

Participants completed a self-paced warm-up on the treadmill followed by a maximal incremental treadmill test to determine V[Combining Dot Above]O2max, maximal heart rate (HRmax), and peak treadmill running speed (PTRS). The test started with participants running at 10 km·h−1 for 1 minute at 0% gradient, after which the treadmill speed was increased by 0.5 km·h−1 every 30 seconds until volitional exhaustion. Participants were verbally encouraged to produce a maximal effort. After an 8- to 10-minute rest period, participants ran to exhaustion at one stage higher than the highest stage completed in the incremental test. The purpose of this “verification” run was to ensure that a “true” V[Combining Dot Above]O2max had been obtained (42).

Respiratory gases were measured breath-by-breath during both the incremental test and the verification bout (Jaeger Oxycon Pro, Hoechberg, Germany) and averaged over 15-second intervals. The highest 15-second average V[Combining Dot Above]O2 from the incremental test and the verification bout were compared and V[Combining Dot Above]O2max was taken as the higher of these 2 values. Nevertheless, these differences were generally small (within 4 ± 4%) and consistent with the differences reported elsewhere (52). Beat-by-beat HR data were also collected (Suunto Oy, Vantaa, Finland), and HRmax was defined as the highest 2-second average during the incremental test. Peak treadmill running speed was defined as the highest completed stage during the incremental test.

Back to Top | Article Outline

Visits 2 and 3

Participants were asked to refrain from hard training the day before visits 2 and 3 and not to exercise before the laboratory visit on the day of the trial (although these conditions applied primarily to visit 2). Furthermore, participants were asked to abstain from all food and drink other than water for at least 2 hours before visits 2 and 3. The compliance with these requests was verbally confirmed with each participant upon their arrival at the laboratory. Although it was not possible to conduct all testing sessions for the study at the same time of day, the time of the laboratory visits was kept consistent for each participant (within 2 hours) so as to avoid variation as result of circadian rhythm.

Back to Top | Article Outline
Perceived Muscle Pain

At the start of visits 2 and 3, participants were asked to rate their perceived muscle pain during light stretching by placing a mark on a 10-cm visual analogue scale, ranging from “no pain” to “unbearable pain.” Participants provided separate ratings for the quadriceps, hamstrings, and calf muscles.

Back to Top | Article Outline
Warm-up and 5-Bound Test

Participants completed a standardized warm-up consisting of 4 minutes at 70% of PTRS and 1 minute at 90% PTRS. This was followed by a 10-minute break during which participants performed the 5-bound test on an indoor, synthetic surface. The objective of the 5-bound test was to cover the furthest possible horizontal distance in 5 alternating strides from a standing start (17). In the current study, postrace vs. prerace change in the distance covered over 5 bounds was used as a proxy for change in neuromuscular power (17). Participants completed a self-selected number of practice attempts at the 5-bound test (typically 2–3 attempts) followed by 3 official attempts. Horizontal distance covered was measured from the start line to the point of heel strike after the fifth stride, and 5-bound distance was taken as the best of the 3 official attempts.

Back to Top | Article Outline
Submaximal Treadmill Exercise and Recovery

The 5-bound test was followed by a 20-minute treadmill exercise at 70% V[Combining Dot Above]O2max. This intensity was intended to approximate race pace and was based on previous reports of %V[Combining Dot Above]O2max sustained over ultramarathon distances (21,43 21,43). The treadmill exercise was immediately followed by 15 minutes of controlled recovery, and breath-by-breath respiratory data (Jaeger Oxycon Pro) and HR (Suunto Oy) were recorded continuously throughout the exercise and recovery period. For visit 2, the treadmill speed to elicit 70% V[Combining Dot Above]O2max was estimated based on the relationship between V[Combining Dot Above]O2 and running speed during the incremental treadmill test in visit 1. If necessary, small adjustments to the treadmill speed were made within the first minutes of exercise to ensure that the target intensity was attained. For visit 3, treadmill speed was matched to the speed that had elicited 70% V[Combining Dot Above]O2max steady state in the prerace trial. With 1 minute of the 20-minute treadmill exercise remaining, participants were asked to provide an RPE on Borg's 6–20 scale (5). This scale was thoroughly explained to each participant at the start of the laboratory visit.

At the end of the 20-minute exercise bout, participants were given a countdown to take hold of the treadmill handrails and step to the side of the moving treadmill belt. The treadmill belt was stopped immediately and participants stepped back on to the stationary belt and stood upright and motionless for the next 1 minute and 30 seconds. At this point, participants were asked to be seated on a chair placed directly behind them on the treadmill belt and completed a further 13 minutes and 30 seconds of seated recovery. The combination of short period of standing recovery followed by seated recovery is similar to an approach that has been used elsewhere (7). Participants were asked to refrain from speaking and to remain as still as possible during the recovery measurements.

Back to Top | Article Outline

Data Analysis

Breath-by-breath respiratory data were analyzed as 15-second averages and beat-to-beat HR data as 2-second averages. Steady-state exercise V[Combining Dot Above]O2, minute ventilation (VE), respiratory exchange ratio (RER), and HR were taken as the average of the final 17 minutes of the 20-minute exercise bout. Energy expenditure (EE) for the 20-minute exercise bout was calculated based on the standard caloric equivalents for oxygen at different RER values (54).

Heart rate recovery was reported as both HRR60s and HRRτ. HRR60s was calculated as the difference between the mean HR during the final 16 seconds of exercise and the mean HR during the final 16 seconds of the first recovery minute as described previously (32). To calculate HRRτ, HR data over the 15-minute recovery period was modelled as a one phase exponential decay, with HRRτ taken as the time constant of the curve (GraphPad Prism 5; GraphPad Software, San Diego, CA, USA) (45). The span and plateau of the HR recovery curve was also recorded.

Recovery V[Combining Dot Above]O2 data were modelled as a one phase exponential decay curve (GraphPad Prism 5; GraphPad Software), an approach suitable for describing recovery V[Combining Dot Above]O2 kinetics following “moderate-intensity” or “heavy-intensity” exercise (44). The starting point of the curve made to equal the average V[Combining Dot Above]O2 over the last 3 minutes of exercise and the time constant (EPOCτ; where EPOC is excess postexercise oxygen consumption), span, and plateau of the curve were recorded.

The EPOCMAG was taken as the total area under the recovery curve rather than adjusting the area under the curve for baseline V[Combining Dot Above]O2. This approach avoided the methodological challenges of obtaining a meaningful resting V[Combining Dot Above]O2 measurement (19) and has a lower intraindividual variation than methods that incorporate a resting measurement (30). Average RER over the 15-minute recovery period was also reported.

Finally, each participant's finishing time for the race was used to calculate average speed over the 87-km course. Average race pace was reported as a % of PTRS.

Back to Top | Article Outline

Statistical Analyses

Muscle pain ratings included a number of zero values in the prerace measurements and were not normally distributed based on a D'Agnostino and Pearson Omnibus Normality test (GraphPad 5; GraphPad Software). As a result, muscle pain ratings are reported as median and range, and postrace vs. prerace changes in muscle pain were assessed using a Wilcoxon's t-test for paired observations that are not necessarily normally distributed (GraphPad Software). In contrast, 5-bound distance and all submaximal exercise and recovery parameters were normally distributed according to a D'Agnostino and Pearson Omnibus Normality test and are reported as mean ± SD. Postrace vs. prerace changes in these measurements were assessed using paired t-tests, and Pearson's correlation coefficients for changes in exercise measurements vs. changes in recovery measurements and for PTRS vs. race time were calculated (GraphPad 5; GraphPad Software). Statistical significance was accepted at p ≤ 0.05.

Postrace vs. prerace changes in 5-bound distance, exercise measurements, and recovery measurements were also assessed using Cohen's effect sizes (d) (18). Effect sizes were calculated as the difference in the postrace vs. prerace group mean divided by the pooled SD of the prerace and postrace measurements and were described as trivial = <0.2, small = ≥0.2 to <0.5, moderate = ≥0.5 to <0.8, or large = ≥0.8.

Finally, postrace vs. prerace measurement changes in individual participants were compared with the day-to-day variation or typical error as a coefficient of variation (CVTEM) for each measurement. These CVTEM values were determined using a separate group of 12 moderately trained runners (as described elsewhere (38)) and were as follows: V[Combining Dot Above]O2 CVTEM = 1.8%, VE CVTEM = 3.3%, RER CVTEM = 2.1%, EE CVTEM = 2.3%, HR CVTEM = 2.2%, RPE CVTEM = 7.3%, HRR60s CVTEM = 8.7%, HRRτ CVTEM = 10.0%, and HR plateau CVTEM = 3.0%. A postrace vs. prerace change that exceeded the CVTEM of the measurement was regarded as a meaningful increase or decrease for that individual, depending on the direction of change. These individual responses were rank ordered for each measurement to show the proportion of individuals who had a decrease larger than the day-to-day variation of the measurement, a change that fell within the day-to-day variation of the measurement or an increase that exceeded the day-to-day variation of the measurement.

We did not establish the day-to-day variation of perceived muscle soreness in our previous study. However, Burgess and Lambert showed that differences in perceived muscle pain of ∼3 units are significantly different at the group level for Comrades participants (14), and we elected to compare individual changes in perceived muscle soreness against a “threshold” change of 3 units.

We have previously observed that a sample size of 8–10 participants provides sufficient statistical power for the main outcome variables we measured in this study. In anticipation of participants dropping out of the study through injury, we recruited more participants (n = 13). A retrospective analysis on HRR60s data showed that for a difference of 6 ± 5 b·min−1 in HRR60s, which occurred in this study, there was sufficient statistical power to interpret the results (10 participants, alpha = 0.05 and power = 0.95).

Back to Top | Article Outline


Ultramarathon and Postrace Testing

All participants completed the 87-km race within the 12 hours cut-off, with finishing times between 8 hours 11 minutes and 11 hours 39 minutes. There was a strong correlation between PTRS and race finishing time (r = −0.89, p = 0.0005), and average speed for the race corresponded to 47–56% of each individual's PTRS. Although we had hoped to schedule the postrace trial 3 days after the race, in practice, it was necessary to conduct these trials 2 days (2 participants), 3 days (4 participants), or 4 days (4 participants) after the race because of the logistical constraints of testing. Nevertheless, anecdotal evidence suggests that runners take several weeks to recover from the Comrades ultramarathon. Therefore, it is reasonable to assume that all postrace testing took place over a period in which participants were still recovering from the race and a return to normal training would not have been advised.

Back to Top | Article Outline

Muscle Pain and Muscle Power

There was a significant increase in perceived pain in the quadriceps muscles (p < 0.01) and hamstring muscles (p ≤ 0.05) after the race, with a nonsignificant increase in perceived pain in the calf muscles (Table 2). However, the typical overall magnitude of perceived muscle pain 3–4 days after the race was relatively low at ratings of 2–3 out of a possible 10. There was no difference in 5-bound distance measured before and after the race (Table 3).

Table 2

Table 2

Table 3

Table 3

Back to Top | Article Outline

Submaximal Exercise Parameters

Differences between average V[Combining Dot Above]O2 and end-of-exercise V[Combining Dot Above]O2 and average HR and end-of-exercise HR were associated with trivial effect sizes for both prerace and postrace submaximal exercise bouts (change in V[Combining Dot Above]O2 = 0.4 ± 0.3 ml·kg−1·min−1, d = 0.01) (change in HR = 3 ± 3 b·min−1, d = 0.17). Furthermore, there were no significant differences in exercise V[Combining Dot Above]O2, VE, EE, RER, or HR during the submaximal exercise bout before and after the race, and postrace vs. prerace changes were associated with trivial-to-small effects (d = 0.0–0.3) (Table 3). In contrast, there was a significant increase in the RPE from 11 ± 1 in the prerace trial to 13 ± 2 in the postrace trial, a change associated with a large effect (d = 1.2) (p < 0.01).

Back to Top | Article Outline

V[Combining Dot Above]O2 and HR Recovery Parameters

There was no significant difference in EPOCMAG, EPOCτ, the plateau of the V[Combining Dot Above]O2 recovery curve, or the average RER over the recovery period in the postrace vs. prerace trials (Table 3). Nevertheless, changes in EPOCMAG and V[Combining Dot Above]O2 plateau were associated with moderate effects (d = 0.5), and there was a small, significant increase in the span of the V[Combining Dot Above]O2 recovery curve (p < 0.01). HRR60s and HRRτ reflected significantly faster recovery in the postrace trial compared with the prerace trial with large associated effects (d = 0.9–1.0) (p ≤ 0.05). There were, however, no significant changes in the span or plateau of the HR recovery curve.

Back to Top | Article Outline

Individual Responses

Individual responses for postrace vs. prerace changes in muscle pain, exercise responses, and HRR responses are shown in Figure 2. There was large interindividual variation in the pattern and magnitude of responses and correlations between postrace vs. prerace changes in exercise measurements, and postrace vs. prerace changes in recovery measurements showed no significant relationships (data not shown).

Figure 2

Figure 2

Back to Top | Article Outline


By self-report, participants had accumulated a typical weekly training distance of 72 ± 16 km, completed over 5 ± 1 training sessions per week, in the 3 months leading up to the race. Although an objective measure of typical training load would have been preferable, it is reasonable to assume that covering the 87-km race distance in a single session would have been a considerable acute training “overload” for all the participants. Furthermore, the strong correlation between PTRS and race time would suggest that participants attempted to complete the course in the fastest possible time. The main finding of this study was the HRR60s and HRRτ reflected significantly faster recovery ∼3 days after an ultramarathon when compared with prerace measurements, an observation that was contrary to the slower HRR responses we had hypothesized.

Previous studies have reported faster (22,33 22,33), slower (6), or no change in (27) HRR responses following 1–3 weeks of increased training load, hence the relationship between short-term changes in HRR and “readiness to train” is not yet clear. It has previously been suggested that slower HRR responses may predict decreased endurance capacity and the accumulation of fatigue (35), and we had expected that participants in the current study would be somewhat fatigued and show slower HRR responses in the postrace trial. Although the large, significant increase in RPE could be interpreted as the anticipated increase in fatigue, HRR responses were faster in the postrace trial—a change typically interpreted as an improvement in training status (20) or what we have described as “readiness to train.”

One explanation for faster HRR responses could be decreased sympathetic activity in the days after the race. For example, Sagnol et al. (51) monitored catecholamine levels in the days after a 10-hour triathlon and reported that plasma-free adrenalin levels were below prerace levels after 3 days of recovery. The authors speculated that this may reflect depleted adrenalin reserves or a suppressed response from the adrenal medulla after high levels of catecholamine release during the triathlon itself. In contrast, Fry et al. (23) reported a nonsignificant increase in epinephrine excretion after 2 weeks of overload training. However, this observation was accompanied by significant decrease in β2 adrenergic receptor density in the muscle, implying decreased sensitivity of the β-adrenergic receptors in the overloaded state. Either of these mechanisms might explain a faster HRRτ response after the race because this form of HRR is associated with a combination of parasympathetic reactivation and sympathetic withdrawal (13,45 13,45). However, the current findings show a very similar effect size in both HRRτ and HRR60s, which is somewhat contrary to this premise. The HRR60s has been attributed primarily to parasympathetic reactivation (13,28,31 13,28,31 13,28,31), and reduced sympathetic activity would be expected to improve HRRτ to a greater extent than HRR60s. For example, Dupuy et al. (22) reported faster HRRτ responses but no change in HRR60s responses associated with maximal exercise after 2 weeks of overload training in endurance athletes. The authors discussed reduced sympathetic activity or reduced β-adrenergic receptor sensitivity as possible mechanisms for this effect.

Another mechanism that might help to explain the faster HRR responses observed is plasma volume expansion. Plasma volume has been reported to be significantly increased in the days following a 42-km (39) and 56-km marathon (29), an effect attributed to an influx of serum albumin and electrolytes into the intravascular space over this period. In these examples, plasma volume expansion peaked on the second day of recovery but remained significantly elevated on the third day (29,39 29,39), and it is reasonable to speculate that plasma volume may have been increased above prerace levels 3–4 days after the 87-km Comrades ultramarathon. Buchheit et al. (11) investigated the effect of plasma volume on postexercise HR variability and HRR by comparing these measurements before and 2 days after a strenuous, supramaximal exercise bout designed to stimulate plasma volume expansion. The authors reported that while HR variability measurements were sensitive to changes in plasma volume, there was no association between plasma volume and HRRτ. This would suggest that increased plasma volume did not contribute to the faster HRRτ observed in the current study. Conversely, Buchheit et al. (11) reported a significant improvement in HRR60s with increased plasma volume. However, this finding was attributed to a decrease in exercise HR rather than a true alteration in HRR60s response. In the current study, there was no change in exercise HR, hence this effect could not easily account for the faster HRR60s response. Furthermore, a similar exercise HR may argue against significant plasma volume expansion (11,50 11,50), despite what has previously been observed following 42 and 56-km races (29,39 29,39).

One further possible explanation for faster HRR responses 3–4 days after the ultramarathon is faster parasympathetic reactivation following the postrace submaximal exercise bout, an effect that would account for improvements in both HRR60s and HRRτ responses at a similar exercise HR. A “rebound” toward increased parasympathetic activity in the days after an ultra-endurance event is in keeping with the work of Hautala et al. (26), who monitored HR variability continuously for 48 hours following a 56-km cross-country skiing race. The authors reported that the high frequency component of HR variability was reduced on the first day after the race but returned to or exceeded prerace levels on the second day after the race. A rebound shift toward increased parasympathetic activity during a period of rest or lighter training following a period of heavier training is also supported by previous studies that monitored resting HR variability (13,16,48,49 13,16,48,49 13,16,48,49 13,16,48,49) or HRR (13) over the course of several weeks.

Previous studies have demonstrated a relationship between increased parasympathetic activity and improved performance at an individual level (3,16,24 3,16,24 3,16,24) and a relationship between increased parasympathetic activity and reduced fatigue (3). However, in the current study, increased HRR60s and increased RPE would suggest increased parasympathetic activity accompanied by increased fatigue. We did not measure performance in the current study and hence cannot be sure whether postrace performance would have been impaired. Nevertheless, a dissociation between faster HRR and improved performance is compatible with recent work from Hug et al. (27), who found that improved performance coincided with slower HRR in the second week of a pre-marathon tapering period. Furthermore, the current observation of faster HRR in the presence of an increase in RPE is in keeping with a monitoring-based case study by Lamberts et al. (34). These authors measured weekly training load and HRR60s responses in a world-class cyclo-cross athlete over 10 weeks and found that peaks in weekly training load were associated with increased RPEs, faster HRR60s responses from 90% HRmax, and increased scores on the Daily Analysis of Life Demands for Athletes (DALDA) questionnaire (indicative of increased lifestyle stress) (34). Although increased RPE and DALDA scores might be expected to negatively affect performance, no maximal performance tests were conducted during this period (34).

These apparently conflicting findings may suggest that increased parasympathetic activity accompanied by an increase in stress, fatigue, or perceived exertion vs. increased parasympathetic activity accompanied by reduced stress, fatigue, or perceived exertion represent different physiological states. Although further investigation of these relationships is required, it seems that changes in HRR should be interpreted in combination with changes in other parameters when assessing an individual's readiness to train (32).

We are aware of only a few previous studies that incorporated submaximal exercise and recovery V[Combining Dot Above]O2 measurements in a baseline or control condition and after a strenuous exercise intervention (4,15 4,15). In these studies, there were significant changes in the physiological responses to the submaximal exercise bout and in EPOCMAG measurements, whereas in the current study, neither submaximal exercise responses nor recovery V[Combining Dot Above]O2 measurements were significantly different after the ultra-endurance race. These contrasting findings are likely to be related to the nature of the exercise intervention and the timing of the postintervention submaximal exercise bout. However, they would suggest that changes in EPOC may be determined by changes in the homeostatic stress of the preceding exercise.

Although HRR60s and HRRτ were significantly faster at the group-level 3 days after an ultramarathon road race, faster HRR responses were observed in only 7 (HRR60s) or 5 (HRRτ) of the 10 participants, with the remainder showing no apparent change in HRR responses. One explanation for this variation is that participants were at different stages of the return toward full recovery at the time of the postrace trial, which in turn may be explained by variation in the level of training going into the race and in the relative intensity at which the race was completed.

A significant limitation of the current study is that we did not measure performance before and after the race. Inclusion of a performance measurement would have allowed us to specify whether participants were overreached at the time of the postrace trial and generally aided interpretation of the findings. However, many of the participants found even the moderate intensity of the submaximal trial aversive in the days after the race, and the inclusion of, for example, a running time trial may have led to increased dropouts or lower initial participant recruitment. A further limitation is that we could not standardize the relative physiological stress of the race between individuals and, for practical reasons, had to allow some individual variation in recovery time before the postrace trial. The current findings are also somewhat limited by the relatively small sample size and the mixed gender participant group. Furthermore, although the standing and sitting components of the 15-minute recovery period were well-controlled, the change in posture constitutes a limitation for interpreting the one phase exponential decay recovery curves. It follows that maintaining a single posture during the postexercise period would have been preferable. Finally, it must also be acknowledged that the CVTEM values by which individual responses were assessed were based on day-to-day variation, and that typical CVTEM values between the prerace and postrace trials may be somewhat larger. Conversely, the current participants were more highly trained than the participants group from which the CVTEM values were calculated, a factor that would have contributed to lower typical measurement variation.

Back to Top | Article Outline

Practical Applications

It is well-accepted that faster HRR is indicative of improved training status when measured at a longitudinal level (20). However, the current study showed faster HRR responses just 2–4 days after an ultramarathon and in combination with increased RPE values. This finding suggests that faster HRR may not always be associated with a recovered or adapted state, particularly when HRR is measured at more frequent intervals. It follows that incorporation of other parameters, such as RPE and DALDA scores, may help to ensure that changes in HRR are interpreted correctly.

Back to Top | Article Outline


This research was supported financially by the Deutscher Akademischer Austausch Dienst, the Ernst and Ethel Eriksen Trust, and the University of Cape Town.

Back to Top | Article Outline


1. American College of Sports Medicine. Generalized skinfold equations. In: ACSM's Guidelines for Exercise Testing and Prescription. Whaley M.H., Brubaker P., Otto R., eds. Baltimore, MD: Lippincott Williams & Wilkins, 2006. pp. 62–63.
2. American College of Sports Medicine. Physical Activity Readiness Questionnaire (PAR-Q). In: ACSM's Guidelines for Exercise Testing and Prescription. Whaley M.H., Brubaker P.H., Otto R.M., eds. Baltimore, MD: Lippincott Williams & Wilkins, 2006. pp. 26.
3. Atlaoui D, Pichot V, Lacoste L, Barale F, Lacour JR, Chatard JC. Heart rate variability, training variation and performance in elite swimmers. Int J Sports Med 28: 394–400, 2007.
4. Bahr R, Opstad PK, Medbø JI, Sejersted OM. Strenuous prolonged exercise elevates resting metabolic rate and causes reduced mechanical efficiency. Acta Physiol Scand 141: 555–563, 1991.
5. Borg GA. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med 2: 92–98, 1970.
6. Borresen J, Lambert MI. Changes in heart rate recovery in response to acute changes in training load. Eur J Appl Physiol 101: 503–511, 2007.
7. Bosquet L, Gamelin FX, Berthoin S. Reliability of postexercise heart rate recovery. Int J Sports Med 29: 238–243, 2008.
8. Bosquet L, Merkari S, Arvisais D, Aubert AE. Is heart rate a convenient tool to monitor over-reaching? A systematic review of the literature. Br J Sports Med 42: 709–714, 2008.
9. Buchheit M. Monitoring training status with HR measures: Do all roads lead to Rome? Front Physiol 5: 73, 2014.
10. Buchheit M, Chivot A, Parouty J, Mercier D, Al Haddad H, Laursen PB, Ahmaidi S. Monitoring endurance running performance using cardiac parasympathetic function. Eur J Appl Physiol 108: 1153–1167, 2010.
11. Buchheit M, Laursen PB, Al Haddad H, Ahmaidi S. Exercise-induced plasma volume expansion and post-exercise parasympathetic reactivation. Eur J Appl Physiol 105: 471–481, 2009.
12. Buchheit M, Millet GP, Parisy A, Pourchez S, Laursen PB, Ahmaidi S. Supramaximal training and postexercise parasympathetic reactivation in adolescents. Med Sci Sports Exerc 40: 362–371, 2008.
13. Buchheit M, Papelier Y, Laursen PB, Ahmaidi S. Noninvasive assessment of cardiac parasympathetic function: Postexercise heart rate recovery or heart rate variability? Am J Physiol Heart Circ Physiol 293: H8–H10, 2007.
14. Burgess TL, Lambert MI. Differences in muscle pain and plasma creatine kinase activity after “up” and “down” comrades marathons. South Afr J Sport Med 20: 54–58, 2008.
15. Burt DG, Lamb K, Nicholas C, Twist C. Effects of exercise-induced muscle damage on resting metabolic rate, sub-maximal running and post-exercise oxygen consumption. Eur J Sport Sci 14: 337–344, 2014.
16. Chalencon S, Busso T, Lacour J-R, Garet M, Pichot V, Connes P, Gabel CP, Roche F, Barthélémy JC. A model for the training effects in swimming demonstrates a strong relationship between parasympathetic activity, performance and index of fatigue. PLoS One 7: e52636, 2012.
17. Chamari K, Chaouachi A, Hambli M, Kaouech F, Wisløff U, Castagna C. The five-jump test for distance as a field test to assess lower limb explosive power in soccer players. J Strength Cond Res 22: 944–950, 2008.
18. Cohen J. Statistical Power Analysis for Behavioural Sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.
19. Compher C, Frankenfield D, Keim N, Roth-Yousey L. Best practice methods to apply to measurement of resting metabolic rate in adults: A systematic review. J Am Diet Assoc 106: 881–903, 2006.
20. Daanen HAM, Lamberts RP, Kallen VL, Jin A, Van Meeteren NLU. A systematic review on heart-rate recovery to monitor changes in training status in athletes. Int J Sports Physiol Perform 7: 251–260, 2012.
21. Davies CT, Thompson MW. Aerobic performance of female marathon and male ultramarathon athletes. Eur J Appl Physiol Occup Physiol 41: 233–245, 1979.
22. Dupuy O, Bherer L, Audiffren M, Bosquet L. Night and postexercise cardiac autonomic control in functional overreaching. Appl Physiol Nutr Metab 38: 200–208, 2013.
23. Fry AC, Schilling BK, Weiss LW, Chiu LZF. beta2-Adrenergic receptor downregulation and performance decrements during high-intensity resistance exercise overtraining. J Appl Physiol 101: 1664–1672, 2006.
24. Garet M, Tournaire N, Roche F, Laurent R, Lacour JR, Barthélémy JC, Pichot V. Individual interdependence between nocturnal ANS activity and performance in swimmers. Med Sci Sports Exerc 36: 2112–2118, 2004.
25. Halson SL, Jeukendrup AE. Does overtraining exist? An analysis of overreaching and overtraining research. Sport Med 34: 967–981, 2004.
26. Hautala AJ, Tulppo MP, Mäkikallio TH, Laukkanen R, Nissilä S, Huikuri HV. Changes in cardiac autonomic regulation after prolonged maximal exercise. Clin Physiol 21: 238–245, 2001.
27. Hug B, Heyer L, Naef N, Buchheit M, Wehrlin JP, Millet GP. Tapering for Marathon and cardiac autonomic function. Int J Sports Med 35: 676–683, 2014.
28. Imai K, Sato H, Hori M, Kusuoka H, Ozaki H, Yokoyama H, Takeda H, Inoue M, Kamada T. Vagally mediated heart rate recovery after exercise is accelerated in athletes but blunted in patients with chronic heart failure. J Am Coll Cardiol 24: 1529–1535, 1994.
29. Irving RA, Noakes TD, Burger SC, Myburgh KH, Querido D, van Zyl Smit R. Plasma volume and renal function during and after ultramarathon running. Med Sci Sports Exerc 22: 581–587, 1990.
30. Jacobsen DJ, Bailey BW, LeCheminant JD, Hill JO, Mayo MS, Donnelly JE. A comparison of three methods of analyzing post-exercise oxygen consumption. Int J Sports Med 26: 34–38, 2005.
31. Kannankeril PJ, Le FK, Kadish AH, Goldberger JJ. Parasympathetic effects on heart rate recovery after exercise. J Investig Med 52: 394–401, 2004.
32. Lamberts RP. The Development of an Evidence-Based Submaximal Cycle Test Designed to Monitor and Predict Cycling Performance: The Lamberts and Lambert Submaximal Cycle Test (LSCT) (1st ed.). Enschede, the Netherlands: Ipskamp Drukkers B.V., 2009.
33. Lamberts RP, Maskell S, Borresen J, Lambert MI. Adapting workload improves the measurement of heart rate recovery. Int J Sports Med 32: 698–702, 2011.
34. Lamberts RP, Rietjens GJ, Tijdink HH, Noakes TD, Lambert MI. Measuring submaximal performance parameters to monitor fatigue and predict cycling performance: A case study of a world-class cyclo-cross cyclist. Eur J Appl Physiol 108: 183–190, 2010.
35. Lamberts RP, Swart J, Capostagno B, Noakes TD, Lambert MI. Heart rate recovery as a guide to monitor fatigue and predict changes in performance parameters. Scand J Med Sci Sports 20: 449–457, 2010.
36. Lamberts RP, Swart J, Noakes TD, Lambert MI. Changes in heart rate recovery after high-intensity training in well-trained cyclists. Eur J Appl Physiol 105: 705–713, 2009.
37. Le Meur Y, Hausswirth C, Natta F, Couturier A, Bignet F, Vidal PP. A multidisciplinary approach to overreaching detection in endurance trained athletes. J Appl Physiol 114: 411–420, 2013.
38. Mann TN, Lamberts RP, Lambert MI. Day-to-day variation in heart rate recovery and excess post-exercise oxygen consumption after a submaximal treadmill protocol. Int Sportsmed J 15: 352–364, 2014.
39. Maron MB, Horvath SM, Wilkerson JE. Blood biochemical alterations during recovery from competitive marathon running. Eur J Appl Physiol Occup Physiol 36: 231–238, 1977.
40. McKechnie JK, Leary WP, Noakes TD. Metabolic responses to a 90 km running race. S Afr Med J 61: 482–484, 1982.
41. Meeusen R, Duclos M, Foster C, Fry A, Gleeson M, Nieman D, Raglin J, Rietjens G, Steinacker J, Urhausen A. Prevention, diagnosis, and treatment of the overtraining syndrome: Joint consensus statement of the European College of Sport Science and the American College of Sports Medicine. Med Sci Sports Exerc 45: 186–205, 2013.
42. Midgley AW, Carroll S. Emergence of the verification phase procedure for confirming “true” VO2max. Scand J Med Sci Sports 19: 313–322, 2009.
43. Noakes T. Oxygen transport and running economy. In: Lore of Running. Champaign, IL: Human Kinetics, 2003. pp. 67.
44. Ozyener F, Rossiter HB, Ward SA, Whipp BJ. Influence of exercise intensity on the on- and off-transient kinetics of pulmonary oxygen uptake in humans. J Physiol 533: 891–902, 2001.
45. Perini R, Orizio C, Comandè A, Castellano M, Beschi M, Veicsteinas A. Plasma norepinephrine and heart rate dynamics during recovery from submaximal exercise in man. Eur J Appl Physiol Occup Physiol 58: 879–883, 1989.
46. Peters EM, Anderson R, Nieman DC, Fickl H, Jogessar V. Vitamin C supplementation attenuates the increases in circulating cortisol, adrenaline and anti-inflammatory polypeptides following ultramarathon running. Int J Sports Med 22: 537–543, 2001.
47. Peters EM, Robson PJ, Kleinveldt NC, Naicker VL, Jogessar VD. Hematological recovery in male ultramarathon runners: The effect of variations in training load and running time. J Sports Med Phys Fitness 44: 315–321, 2004.
48. Pichot V, Busso T, Roche F, Garet M, Costes F, Duverney D, Lacour JR, Barthélémy JC. Autonomic adaptations to intensive and overload training periods: A laboratory study. Med Sci Sports Exerc 34: 1660–1666, 2002.
49. Pichot V, Roche F, Gaspoz JM, Enjolras F, Antoniadis A, Minini P, Costes F, Busso T, Lacour JR, Barthélémy JC. Relation between heart rate variability and training load in middle-distance runners. Med Sci Sports Exerc 32: 1729–1736, 2000.
50. Roy BD, Green HJ, Grant SM, Tarnopolsky MA. Acute plasma volume expansion alters cardiovascular but not thermal function during moderate intensity prolonged exercise. Can J Physiol Pharmacol 78: 244–250, 2000.
51. Sagnol M, Claustre J, Cottet-Emard JM, Pequignot JM, Fellmann N, Coudert J, Peyrin L. Plasma free and sulphated catecholamines after ultra-long exercise and recovery. Eur J Appl Physiol Occup Physiol 60: 91–97, 1990.
52. Scharhag-Rosenberger F, Carlsohn A, Cassel M, Mayer F, Scharhag J. How to test maximal oxygen uptake: A study on timing and testing procedure of a supramaximal verification test. Appl Physiol Nutr Metab 36: 153–160, 2011.
53. Strachan AF, Noakes TD, Kotzenberg G, Nel AE, de Beer FC. C reactive protein concentrations during long distance running. Br Med J 289: 1249–1251, 1984.
54. Weir J. New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol 109: 1–9, 1949.

monitoring; training load; EPOC; Autonomic Nervous System

Copyright © 2015 by the National Strength & Conditioning Association.