Journal Logo


Overreaching Attenuates Training-induced Improvements in Muscle Oxidative Capacity


Author Information
Medicine & Science in Sports & Exercise: January 2020 - Volume 52 - Issue 1 - p 77-85
doi: 10.1249/MSS.0000000000002095
  • Free


Intensified and/or progressive increases in training load (i.e., increases in training intensity and/or volume) are considered necessary to induce performance improvements and physiological adaptations in trained endurance athletes (1). Of the possible training-induced muscular adaptations that relate to improvements in exercise performance, skeletal muscle oxidative capacity has been identified as one of the strongest determinants of endurance performance in highly trained endurance athletes (2). Previous research has shown that skeletal muscle oxidative capacity and endurance performance can be improved after both intensified training (i.e., implementing a short period of high-intensity interval training) (3,4) and increased training volume (5) in healthy, active men. However, these findings have not been replicated in already well-trained athletes, whereby 7 wk of intensified training did not alter markers of mitochondrial content or pulmonary V˙O2 kinetics in trained cyclists (6). In addition, there is currently no published literature investigating the effects increasing training volume on skeletal muscle oxidative capacity in trained endurance athletes.

A recent review (7) highlighted the increased emphasis on manipulating the training prescription variables to promote mitochondrial adaptations. Bishop and colleagues (7) reported the existence of a strong positive correlation (r2 = 0.80) between training volume and increases in citrate synthase activity, suggesting that increases in training volume may be an important determinant of changes in mitochondrial content. However, it is not fully understood how alterations in training volume may affect mitochondrial adaptations in trained endurance athletes. Indeed, for athletes with already high training volumes, further increases in training volume may result in acute fatigue (AF) and the development of functional overreaching (FOR) or nonfunctional overreaching (NFOR) (8). FOR results from an accumulation of training and/or nontraining stress leading to a short-term decrement in exercise performance, in which restoration (9), and sometimes supercompensation (10), may occur after a taper period (~1–3 wk) (8). Although FOR may be considered a necessary component of the overtraining continuum (1,8), FOR is associated with an increased incidence of illness (11) and sleep disturbance (11), and it precedes the more severe state of NFOR, which is associated with maladaptation to training (12) and a decrease in performance that will not resume for several weeks or months (8). Indeed, it was recently demonstrated (12) that a period of intensified training attenuated the training-induced increase in citrate synthase and mitochondrial complex IV activity in the red gastrocnemius muscle of NFOR rats compared with FOR rats. Whether skeletal muscle oxidative capacity and endurance performance would be altered in response to intensified training in endurance athletes who already have substantially high training volumes, where the risk of overreaching and maladaptation to training is substantially high (8,12), remains to be investigated. Therefore, the aim of the present study was to determine whether the measurements of skeletal muscle oxidative capacity derived by near-infrared spectroscopy (NIRS) are influenced by increases in training volume and a resultant taper period in highly trained middle-distance runners.



Twenty-four highly trained middle-distance runners participated in this study. Eight runners were females (mean ± SD: age 21.3 ± 3.2 yr, stature 171.2 ± 4.9 cm, body mass 53.1 ± 6.0 kg, maximal oxygen uptake (V˙O2peak) 63.2 ± 3.4 mL·kg−1·min−1), and 16 runners were males (age 21.0 ± 3.6 yr, stature 181.3 ± 5.1 cm, body mass 70.6 ± 7.9 kg, V˙O2peak 73.3 ± 4.3 mL·kg−1·min−1). Our inclusion criteria specified that participants were competitive middle-distance runners (800 and 1500 m) who had a consistent training history of at least 2 yr in these events and were without major injury interruption for the previous 3 months. The male runners had personnel best running times for the 800 and 1500 m of 119.4 ± 7.8 s (range, 108.3–133.4 s) and 238.0 ± 16.8 s (225.2–279.1 s), respectively, whereas the females had times of 135.0 ± 8.6 s (124.1–153.4 s) and 284.1 ± 18.8 s (257.4–321.4 s), respectively. In the 3 wk preceding the study, the male runners had a mean running training volume of 73.9 ± 19.2 km·wk−1, whereas the female runners ran 53.9 ± 16.0 km·wk−1. Five of the female runners were taking oral contraception, whereas the other three had regular menstrual cycles. The study was completed in the final months (December–February) of the preparation phase and the first month of the competitive season, whereby the runners competed in some minor events/races during the study period. All runners provided written informed consent before participating in this study, which was approved by the Griffith University Human Research Ethics Committee.

General design

The study period lasted a total of 7 wk, which was divided into three distinct training phases. The runners first completed 3 wk of normal training (NormTr) that was prescribed from their coach, 3 wk of high-volume training (HVTr; weekly stepwise increase in training volume by 10%, 20%, and 30% during each successive week from NormTr), and a 1-wk taper (TapTr; 55% exponential reduction in training volume from HVTr week 3). Before and immediately after each training phase, the recovery rate of skeletal muscle oxygen consumption (mV˙O2) in the gastrocnemius medialis after short-duration (~10 s) plantarflexion exercise was measured using NIRS and subsequently fit to a monoexponential curve, with the rate constant (k; min−1) being used as an index of skeletal muscle oxidative capacity. In addition, runners also performed a submaximal running test to determine running economy and a maximal incremental running test to determine gas exchange threshold (GET), respiratory compensation threshold (RCT), time to exhaustion (TTE), maximal heart rate (HRmax), and V˙O2peak. Throughout the training period, runners completed a training diary and a weekly subjective fatigue questionnaire and wore a GPS-equipped running watch and heart rate monitor for each running training session. Runners who had a decreased performance in the maximal incremental running test after HVTr that was greater than the test–retest typical error determined from the pre- and post-NormTr were classified as FOR, others as AF.


The training of each participant was monitored for a period of 10 wk in total (including the 3 wk preceding NormTr). NormTr consisted of coach-prescribed training. After this, training volume was increased in a weekly stepwise manner by 10%, 20%, and 30% during each successive week while training intensity was maintained (i.e., HVTr). The increase in training volume during this 3-wk period was prescribed from the initial NormTr period whereby the participants completed the same weekly distribution, type, and content of running training sessions but with the prescribed increased volume. For example, in the third week of HVTr (i.e., +30% training volume), a track session that included 10 × 200-m repetitions at 800-m race pace would become 13 × 200-m repetitions at the same prescribed pace. After HVTr, participants undertook a 1-wk taper (i.e., TapTr), which reduced the training volume of each runner by 55% from week 3 HVTr in an exponential manner. Training volume was reduced each day beginning with a 20% reduction, followed by a further 15%, 10%, 10%, 10%, 5%, and 5% reduction for each successive day. This tapering strategy was used because a taper of this duration, volume reduction, and nature has been shown to induce a performance super compensation after an overload training period (13–16).

Testing procedures

Participants attended the laboratory on five occasions in total: twice before NormTr (one familiarization and one performance trial), once before HVTr and TapTr, and once again after TapTr. Participants arrived at the laboratory between 5:00 and 7:30 am after an overnight fast and not having undertaken strenuous exercise for at least 24 h. Laboratory temperature and humidity (22°C–23°C and 45%–50%, respectively) were consistent throughout all tests.

Dietary standardization

To minimize variability in exercise metabolism attributable to dietary intake, participants were provided with a standardized dinner (~55 kJ·kg−1 body mass, providing 2.0 g carbohydrate·kg−1 body mass, 0.3 g fat·kg−1 body mass, and 0.6 g protein·kg−1 body mass) and breakfast (~40 kJ·kg−1 body mass, providing 1.8 g carbohydrate·kg−1 body mass, 0.2 g fat·kg−1 body mass, and 0.1 g protein·kg−1 body mass), which were consumed approximately 12 and 1.5 h before the exercise tests. To support the HVTr, participants were provided with a posttraining beverage to facilitate recovery, which included ~1.5 g carbohydrate·kg−1 body mass and ~0.2 g whey protein isolate·kg−1 body mass (Bulk Nutrients, Grove, AU) to consume within 1 h after each training session.

Submaximal running test

After a warm-up (5 min at 8–10 km·h−1), participants completed two 4-min submaximal incremental stages on a motorized treadmill (HP cosmos Saturn, Traunstein, Germany), which was set at a speed equivalent to 100% and 120% of the GET, which was determined in the familiarization testing session. The treadmill belt was also set at 1% gradient to reflect the energetic cost of running overground at these speeds (17). The two stages were briefly interrupted by a 30-s rest period to allow earlobe blood sampling for the determination of blood lactate concentration ([La]b) with a Lactate Pro 2 device (Arkray Inc., Japan). Pulmonary gas exchange was measured on a breath-by-breath basis throughout each stage using a calibrated metabolic system (Cosmed Quark b2, Rome, Italy).

Calculation of running economy

Oxygen uptake (V˙O2) during the final minute of each submaximal stage was used to determine the oxygen cost (mL·kg−1·km−1), whereas both the V˙O2 and the carbon dioxide output (V˙CO2) during the same period were used to calculate the energy cost (kcal·kg−1·km−1) of exercise. Updated nonprotein respiratory quotient equations (18) were used to estimate substrate use (g·min−1) during the final minute. The energy derived from each substrate was then calculated by multiplying fat and carbohydrate usage by 9.75 and 4.07 kcal, respectively, reflecting the mean energy content of the metabolized substrates during moderate to high-intensity exercise (19).

Maximal incremental running test

After a 5-min rest period after the submaximal running test, each participant performed an incremental treadmill run to volitional exhaustion starting at 10 km·h−1 and 1% gradient, with the speed increased by 1 km·h−1 each minute until a speed of 20 km·h−1. After 1 min at 20 km·h−1, the treadmill gradient was increased by 1% each minute until volitional exhaustion. Gas exchange variables (V˙O2, V˙CO2, and expired ventilation [E]) were measured as described for the submaximal exercise test and subsequently averaged into 30-s bins. The GET was determined using the V-slope method described by Beaver et al. (20), whereas the RCT was determined using the E-versus-V˙CO2 relationship also described by Beaver et al. (20). Two investigators performed threshold determinations independently, and a third investigator was consulted if there was disagreement between the two. Heart rate was recorded each second (H10, Polar Electro Oy, Kempele, Finland) to determine the heart rates corresponding to GET and RCT, as well as HRmax. V˙O2peak was determined as the average of the two highest consecutive 30-s V˙O2 values, whereas TTE was used as a measure of running capacity. [La]b was measured from the earlobe at 1, 3, 5, and 7 min after the completion of the test with the highest [La]b value obtained at the end of exercise considered peak blood lactate concentration ([La]bmax).

NIRS assessment of skeletal muscle oxidative capacity

NIRS signals were obtained using a noninvasive, near-infrared spectrometer Oxymon MKIII (Artinis Medical Systems b.v., Zetten, the Netherlands), which measured the relative concentrations of deoxyhemoglobin and deoxymyoglobin (HHb) and oxyhemoglobin and oxymyoglobin (HbO2). The transmitting and receiving optodes were set to a detector distance of 40 mm, which allowed for approximately 20 mm of tissue penetration. Subcutaneous adipose thickness at the site of NIRS placement was measured with Harpenden calipers (Baty International, RH15 9LB United Kingdom) to ensure that skeletal muscle was being interrogated. NIRS-derived assessment of muscle oxidative capacity was performed according to the protocol described in detail by Ryan et al. (21), which has been validated against phosphorus magnetic resonance spectroscopy (22) and in vitro measurements of high-resolution respirometry in permeabilized muscle fibers (23). Each participant laid in a supine position with both legs extended while their right foot was placed into a custom-built plantarflexion exercise device. The foot of the participant was strapped firmly to the device using nonelastic Velcro straps. The NIRS optodes were placed at the level of the largest circumference of the triceps surae, over the medial head of the gastrocnemius muscle, and secured with biadhesive tape and covered with a nonstick opaque crepe bandage. A blood pressure cuff (Hokanson SC12D, Bellevue, WA), which was attached to a rapid-inflation system (Hokanson E20), was placed proximal to the NIRS optodes and above the knee joint. The test protocol consisted of the measurements of resting mV˙O2 by way of inflation of the pressure cuff (~300 mm Hg) to arrest blood flow for two 30-s periods. After this, an ischemia/hyperemia calibration was performed to normalize the NIRS signals. To do this, each participant performed ~10 s of voluntary plantarflexion exercise against manual resistance, followed by inflation of the pressure cuff for ~5 min, until the NIRS signals plateaued. Upon release of the cuff, a 1- to 3-min period of hyperemia occurred. This calibration was used to scale the NIRS signals to this range (i.e., minimum value of the plateau and maximum value of the hyperemia). After this, a brief exercise protocol was used to increase mV̇O2 followed by a series of 15 intermittent cuff inflations (arterial occlusions) to quantify mV˙O2 recovery by assessing the change in the slope of the HbO2 signal. The exercise protocol consisted of ~15 s of plantarflexion exercise (two contractions per second) against manual resistance to desaturate the muscle to reach a target of 50% of the calibrated NIRS signal range (21,24). The duty cycles of the subsequent arterial occlusions were as follows: cycle 1–5, 5 s on/5 s off; cycle 6–10, 5 s on/10 s off; cycle 11–15, 10 s on/20 s off). After 2 min, the exercise protocol was repeated, with the mean of the paired ensemble slopes being used when curve fitting. We also corrected NIRS signals for changes in blood volume, as previously described (21). For each intermittent arterial occlusion, the slope of the blood volume–corrected HbO2 signal (%·s−1) was used to express mV˙O2. The mV˙O2 exponential recovery rate constant (k, min−1) was estimated using nonlinear least-squares regression and used as an index of skeletal muscle oxidative capacity (22,23).

Training load monitoring

To monitor training volume and intensity, each participant wore either an M430 GPS running watch (n = 14; Polar Electro Oy) or a Garmin Forerunner 235 (n = 8; Garmin Ltd., Canton of Schaffhausen, Switzerland) during every running session. Training intensity distribution was quantified from running speed using the total time-in-zone approach quantified by a training analysis software (TrainingPeaks WEEKO+, Boulder, CO). From this, the percentage of training time spent with a running speed in each of the three training zones was quantified for each individual training session. The relative training time in each zone for all sessions was then determined. The three training zones according to the reference running speed values that corresponded to physiological thresholds obtained during the maximal running assessment were used: zone 1 (<GET), zone 2 (between GET and the RCT), and zone 3 (>RCT). Participants were also provided with a training diary, which instructed them to rate the global intensity, distance, and duration of all training sessions and races using a modified category ratio scale ranging from 0 to 10 (25). In addition, at the end of each week, participants were also asked to rate their levels of perceptual fatigue on a Likert scale questionnaire consisting of a 100-mm long printed black line with 0 at the left end and the words “extremely fresh” and a 10 at the right end of the scale with the words “extremely tired.” The subjects were instructed to place a mark on the line corresponding to the severity of fatigue they were experiencing.

Assessment of overreaching

In line with previous research (9,10), the smallest worthwhile change (SWC) was used as an FOR threshold. The SWC was calculated as 0.5 × the coefficient of variation (CV) of TTE from the incremental running tests performed before and after NormTr. To be diagnosed as FOR after HVTr, participants had to report an elevated subjective fatigue rating after HVTr and had to show an individual performance (i.e., TTE) decrement larger than the SWC. The remaining participants who maintained or improved their performance but also showed an elevated subjective fatigue rating after HVTr were considered to be AF.

Statistical analysis

Results are expressed as mean ± SD unless stated otherwise. A two-way repeated-measures ANOVA (group [i.e., FOR]–time point [i.e., pre-NormTr]) was used with Bonferroni post hoc comparisons to identify differences in performance and physiological variables between FOR and AF groups. The effect size (d) statistic with upper and lower 95% confidence intervals (CI) were also calculated to assess the magnitude of difference between groups. The magnitude of difference was classified as small (0.2 to 0.6), moderate (0.6 to 1.2), large (1.2 to 2.0), and very large (2.0 to 4.0). All statistical analyses were performed using SPSS 25.0 (SPSS Inc., Chicago, IL), with statistical significance accepted as P < 0.05. Test–retest reliability of running TTE, V˙O2peak, and k values for skeletal muscle oxidative capacity were analyzed using the CV and intraclass correlation coefficients (ICC).


Incidence of overreaching and subjective fatigue

After HVTr, 12 participants developed signs of FOR (8 males, 4 females), and 12 others showed AF (8 males, 4 females). On a group level, subjective fatigue scores were significantly higher each week during HVTr compared with both NormTr and TapTr (Fig. 1). Although there was no significant between-group differences at any time point (group–time interaction: P = 0.13), in comparison with the 3-wk mean of NormTr the FOR group had moderately higher effect size differences (d [95% CI]) in subjective fatigue ratings after the first (0.67 [−0.20 to 1.54]) and second week of HVTr (0.70 [−0.21 to 1.61]). These effect size differences were unclear after the third week of HVTr (0.14 [−0.64 to 0.92]) and after TapTr (0.22 [−0.61 to 1.05]).

Subjective fatigue rating (mean ± SD) of both AF and FOR groups for the 3-wk throughout NormTr, 3 wk throughout HVTr, and taper training (TapTr). aSignificantly different from NormTr.

Training volume

Adherence to the specific training content in each training session and compliance with the program were 99%, as inferred from the training analysis software, training diaries, and verbal communication with the participants. Figure 2 shows the mean training volume completed by the participants during the study. There were no between- or within-group differences in the mean 3-wk training volume completed before the beginning of the study (mean ± SD, 67.1 ± 20.4 km) and during NormTr (69.7 ± 20.8 km; +3.9%; P = 0.56). Training volume was increased from NormTr throughout HVTr week 1 (76.6 ± 21.6; +9.9%), week 2 (85.2 ± 23.6; +22.3%), and week 3 (91.8 ± 25.0 km; +31.8%; all P < 0.01) and was reduced during TapTr (42.6 ± 11.6 km; −38.9% compared with NormTr; −53.6% compared with HVTr; P < 0.01). The training intensity distribution, quantified as the percentage of total time spent in each of the three training zones based on running speed, did not change throughout the study period when comparing the prestudy period (zone 1, 80.2% ± 8.3%; zone 2, 4.1% ± 2.1%; zone 3, 15.7% ± 7.4%), NormTr (81.2% ± 9.1%; 4.5% ± 1.4%; 14.3% ± 7.9%), HVTr (81.8% ± 9.0%; 4.2% ± 0.9%; 14.0% ± 8.4%) and TapTr (80.0% ± 9.4%; 3.6% ± 1.2%; 16.4% ± 9.6%). Furthermore, there were no differences in training volume (group–time, P = 0.89) or training intensity distribution (P = 0.79) between AF and FOR groups at any time point. There was no difference in the number of resistance training sessions performed between AF (0.7 ± 0.8 sessions per week) and FOR groups (0.8 ± 0.7 sessions per week) throughout the study.

Mean training volume (distance run [km]; mean ± SD) during the 3 wk before the study and for each week during NormTr, HVTr, and during TapTr for the AF and FOR groups. aSignificantly higher than prestudy, NormTr and TapTr. bSignificant lower than all other weeks.

Running performance

The CV (expressed as a CV% between pre- and post-NormTr maximal incremental running test) for incremental running TTE (s) and absolute (L·min−1) and relative V˙O2peak (mL·kg−1·min−1) was 6.3%, 3.0%, and 2.8%, respectively. Figures 3 and 4 show the percentage change in TTE and V˙O2peak (mL·kg−1·min−1) relative to pre-NormTr. Although there was no change in TTE or V˙O2peak in the AF group after HVTr, there was a significant increase after TapTr (40 ± 25 s, P > 0.01; 0.11 ± 0.08 L·min−1, P = 0.01; 1.88 ± 1.0 mL·kg−1·min−1, P = 0.008). The FOR group had a decrease in TTE (−49 ± 14 s, P < 0.001) and V˙O2peak (−2.33 ± 2.0 mL·kg−1·min−1, P = 0.04) after HVTr, which was reversed after TapTr (67 ± 22 s, P < 0.001; 5.11 ± 2.50 mL·kg−1·min−1, P = 0.006). Compared with the FOR group, the AF group had substantially larger improvements in TTE from pre-HVTr to post-TapTr (absolute difference score, 37 ± 31 s; P = 0.04), whereas the improvements in V˙O2peak were similar between groups (AF, 3.52 ± 1.40 mL·kg−1·min−1; FOR, 2.78 ± 1.80 mL·kg−1·min−1; P = 0.45).

Percentage change in TTE (s) relative to pre-NormTr (percent change ± 95% CI) during the maximal incremental running test after normal training (post-NormTr), after high-volume training (post-HVTr), and after taper training (post-TapTr) for the AF group (squares) and the FOR group (circles). Shaded area represents the SWC (0.5 CV). aSignificantly different from pre- and post-NormTr. bSignificant between-group difference.
Percentage change in relative V˙O2peak (mL·min−1·kg−1) relative to pre-NormTr (percent change ± 95% CI) during the maximal incremental running test after normal training (post-NormTr), after high-volume training (post-HVTr), and after taper training (post-TapTr) for the AF group (squares) and the FOR group (circles). Shaded area represents the SWC (0.5 CV). aSignificantly different from pre- and post-NormTr. bSignificant between-group difference.

The FOR group had a significant reduction in peak HR and [La]bmax after HVTr (−4 ± 3 bpm, P = 0.02; −4.30 ± 1.80 mmol·L −1, P = 0.01), with both parameters returning to pre-NormTr values after TapTr, whereas there was no change at any time point for the AF group (Table 1). There was no change in running economy (oxygen cost nor energy cost) or the ventilatory thresholds at any time point for either group.

Values of TTE V˙O2peak, GET, and RCT speed, running economy, and peak blood lactate concentration measured during the incremental running test to exhaustion conducted before and after NormTr, and after the HVTr and TapTr period, for the FOR and AF groups.

Skeletal muscle oxidative capacity (mV˙O2k)

Figure 5 shows the mV˙O2k of both AF and FOR groups for the gastrocnemius medialis before and after NormTr, and after HVTr and TapTr. The test–retest values for repeat measurements performed pre-NormTr without movement of the NIRS probe for k were CV = 9.6% and ICC = 0.90, whereas the measurements performed pre- and post-NormTr were CV = 13.1% and ICC = 0.82. The r2 value was >0.95 for all measurements at pre-NormTr (mean, range; r2 = 0.981, 0.954–0.997), post-NormTr (r2 = 0.980, 0.959–0.996), post-HVTr (r2 = 0.984, 0.952–0.995), and post-TapTr (r2 = 0.984, 0.957–0.996). The AF group demonstrated an increase in skeletal muscle oxidative capacity after HVTr (0.39 ± 0.25 min−1; 15.1% ± 9.7%; P = 0.009), with no further improvement after TapTr (P = 0.89). For the FOR group, there was no change in skeletal muscle oxidative capacity at any time point (P > 0.05). Furthermore, the change in mV˙O2k from pre-HVTr to post-HVTr was significantly greater in the AF group compared with the FOR group (0.40 ± 0.35 min−1; P = 0.03).

Skeletal muscle oxidative capacity (mean ± 95% CI) of both AF and FOR groups for gastrocnemius medialis before (pre-NormTr) and after (post-NormTr) normal training, and after high-volume training (post-HVTr) and taper training (post-TapTr). Data are expressed as rate constants for the postexercise recovery of mV˙O2. aSignificantly different from pre- and post-NormTr.


The purpose of this study was to determine whether skeletal muscle oxidative capacity is influenced by increases in training volume and a resultant taper period in highly trained middle-distance runners. The results from this study indicate three key points: 1) skeletal muscle oxidative capacity is increased in response to HVTr in runners who do not develop FOR but is unchanged in those who do; 2) despite improvements in running TTE and V˙O2peak after a taper period, there is not a concomitant increase in skeletal muscle oxidative capacity; and 3) runners who did not develop FOR had substantially larger improvements in running capacity after a taper period.

In the present study, 12 of the 24 middle-distance runners were classified as FOR after HVTr on the basis of an increased subjective fatigue rating and performance decrement (running TTE) larger than the SWC after HVTr. The remaining subjects (AF) either maintained or improved their performance. Runners who were classified as AF after HVTr demonstrated improvements in skeletal muscle oxidative capacity, which remained elevated after TapTr, whereas there was no change in FOR at any time point. These findings indicate that runners who developed FOR had impaired training adaptations relating to improvements in skeletal muscle oxidative capacity. Although no previous study has assessed mitochondrial adaptations in athletes that develop FOR, a recent training study in rats (12) reported impaired training-induced alterations in citrate synthase activity and mitochondrial complex IV activity in NFOR rats. However, it must be noted that the classification of FOR and NFOR rats in this study (12) is not consistent with the consensus statement defining FOR (8) or the classification of FOR that is typically used in human studies (9–11,26), including the present study. Ferraresso et al. (12) classified all rats who experienced a performance decline as NFOR, whereas rats who had a performance increase were classified as FOR. From this classification system, NFOR rats may well have been FOR, which suggests that this training state was associated with blunted training-induced mitochondrial adaptations. This revised classification lends supporting evidence to the findings from the present study, but this needs further experimental confirmation in human studies. One hypothesis that may explain the attenuated training adaptations in the FOR group in the present study relates to the reduction in [La]bmax after HVTr. Lactate has been identified as an important signaling molecule that can mediate exercise-induced adaptations relating to mitochondrial biogenesis (27). Indeed, one recent study (28) demonstrated that decreased lactate accumulation in response to chronic dichloroacetate administration reduced the mitochondrial adaptations to high-intensity interval training in mice. Translating these findings to the present study, the reduction in [La]bmax in the FOR group may have been indicative of reduced [La]b accumulation during HVTr and lead to an attenuation in mitochondrial adaptations, although this hypothesis required further clarification. Furthermore, it is difficult to speculate as to the nature of the mitochondrial adaptations, which would explain the improvements in skeletal muscle oxidative capacity in the present study. Given that improvements in skeletal muscle oxidative capacity may result from alterations in mitochondrial biogenesis, and although there is contention as to what defines mitochondrial biogenesis (29), this may arise from changes mitochondrial content, structure, quality, and/or respiratory function (7,29).

In the present study, the AF group demonstrated significant improvements in skeletal muscle oxidative capacity after HVTr, which were maintained but not further improved after TapTr, despite further improvements in running capacity (i.e., running TTE) and V˙O2peak. This agrees with most (30–32) but not all reports (33) that have assessed changes in markers of skeletal muscle oxidative capacity and performance before and after a taper period in well-trained endurance athletes. In support of the findings of the present study, Neary et al. (30) reported substantial improvements in oxidative enzyme activity (cytochrome oxidase and citrate synthase) after a progressive overload training program, but there were no further improvements after a 4- and 8-d taper protocol despite further improvements in the power output at the ventilation threshold in well-trained cyclists. Skovgard et al. (32) reported that 10 and 18 d of tapering after a period of intensified training resulted in improvements in short-duration running capacity, but this was not accompanied by any changes in muscle oxidative enzyme (citrate synthase and 3-hydroxyacyl-CoA dehydrogenase) activity. Similarly, Luden et al. (31) found no change in gastrocnemius lateralis citrate synthase activity after a 3-wk volume reduced taper in collegiate distance runners despite improvements (3% ± 1%) in 8-km cross-country race performance. Collectively, the findings from these studies (30–32), in conjunction with those of the present, indicate that a taper period may not induce further improvements in skeletal muscle oxidative capacity, but exercise performance is typically improved after an appropriate taper period. Interestingly, a later study by Neary et al. (33) reported that there were fiber type–specific increases in markers of skeletal muscle oxidative capacity (cytochrome oxidase, succinate dehydrogenase, and 3-hydroxyacyl-CoA dehydrogenase) that were only evident in type II muscle fibers after a progressive overload program and subsequent taper period. This may indicate that studies quantifying changes in mitochondrial enzymes in mixed fiber muscle samples after a taper period may be masking potential fiber type–specific adaptations that could only be revealed in single muscle fiber analyses. Indeed, although the NIRS-derived measure of skeletal muscle oxidative capacity used in the current study has its advantages compared with invasive and costly muscle biopsies (34), this approach is unable to detect fiber type–specific changes in skeletal muscle oxidative capacity.

In agreement with other studies in runners (31,35,36), we found that running economy and physiological threshold speeds (GET and RCT) were unchanged after the increase in training volume and resultant taper period. By contrast, there was a significant decrease in running TTE (−7.4%) and V˙O2peak (−3.4%) in the FOR group after HVTr, whereas there was a nonsignificant change in running TTE (+2.5%) and V˙O2peak (+2.4%) during this same period for the AF group. Importantly, the performance super compensation (change in performance from pre-HVTr to post-TapTr) was substantially larger in the AF group (absolute difference score, 37 ± 31 s; P = 0.04), whereas the changes in V˙O2peak during this period were similar. When this is coupled with the absence of improvements in skeletal muscle oxidative capacity in the FOR group, these findings are suggestive of impaired physiological and performance adaptations in response to an increase in training volume in highly trained middle-distance runners who develop FOR. This notion is supported by the findings of one previous study that reported greater performance improvements (peak incremental cycling test power output) and physiological adaptations (V˙O2peak) after an overload period in triathletes who did not develop FOR (9). Aubrey et al. (9) prescribed a 3-wk, 30% increase in training volume to triathletes and reported that lower taper-induced improvements in V˙O2peak and cycling capacity were associated with FOR subjects. The results of the present study provide empirical evidence to further support this notion and suggest that impaired adaptations in skeletal muscle oxidative capacity may be one contributing factor to the attenuated performance supercompensation in FOR runners. There are some notable differences between the current study and that of Aubrey et al. (9), who recruited triathletes who undertook a graded cycling test to exhaustion, which was assessed each week during a 4-wk, 40% step taper whereby 83% and 73% of peak performance supercompensation occurred within the first 2 wk of the taper period in AF and FOR triathletes, respectively. Although we used a 1-wk exponential decay taper model consisting of a 55% reduction in training volume with maintained training intensity and frequency, we acknowledge that an optimal taper period may be individualized for each athlete, and it has been suggested that longer taper periods may be required after an overload training period because of greater stress and fatigue (37). However, contrary to this supposition, the taper characteristics in the present study are in line with recommendations based on a meta-analysis (13), modeling (14), and experimental studies (16), suggesting that a ~50% reduction in training volume in an exponential decay fashion over a period of 1–2 wk can elicit peak performance improvements in endurance athletes. In addition, there was no differences in subjective fatigue scores between the FOR and the AF groups after the taper period in the current study, suggesting that both groups perceived their recovery to be of a similar magnitude.

In the present study, 12 of the 24 middle-distance runners were classified as FOR after HVTr, which agrees with the incidence of overreaching reported in other studies using similar overload periods (i.e., increases of 30%–40% of training volume for 3–4 wk) with these studies reporting 69% (10), 50% (11), 33% (26), and 48% (9) of endurance athletes being diagnosed as FOR after increases of this magnitude in training volume. There is currently no evidence that clearly demonstrates the reason why there is a large between-athlete variability in response to increases in training volume while others display signs and symptoms of fatigue and overreaching. It could be postulated that gender, training status, or habitual training volume, among other factors, may all influence the likelihood of developing FOR. In the present study, gender did not seem to influence the incidence of overreaching as half of the male and female runners were diagnosed as FOR. Furthermore, there was no difference in V˙O2peak or training volume between FOR and AF groups at any time point. Given that FOR precedes the more severe and persistent states of the overtraining continuum such as NFOR and overtraining (8), future studies should aim to gain a better understanding of the individual athlete characteristics that may predispose athletes to developing FOR. It should also be noted that subgroup analysis is a topical issue in relation to longitudinal intervention-based research (38). It has been suggested that dividing the participant group of a study into smaller subgroups (i.e., FOR and AF) leads to overstated or misleading results (39). However, in this instance, the subgroup analysis of FOR and AF groups helps to provide insight into the complex issue of the individual responses to exercise training.

The present study was able to demonstrate that skeletal muscle oxidative capacity was increased in response to a short period of increased training volume, but only in runners who did not develop FOR. However, there was not a compensatory increase after the taper period in either group, despite improvements in running capacity. Other findings from the present study indicate that runners who did not develop FOR had substantially larger performance improvements after a taper period. Collectively, the results from this study maintain that training volume is an important training program variable to consider when periodizing and planning an endurance training program for middle-distance runners.

The authors have no financial relationships relevant to this article to disclose and no competing interests to disclose. The results of the present study are presented clearly, honestly, and without fabrication and do not constitute endorsement by the American College of Sports Medicine.


1. Halson SL. Monitoring training load to understand fatigue in athletes. Sports Med. 2014;44(2 Suppl):S139–47.
2. Jacobs RA, Rasmussen P, Siebenmann C, et al. Determinants of time trial performance and maximal incremental exercise in highly trained endurance athletes. J Appl Physiol. 2011;111(5):1422–30.
3. Christensen PM, Jacobs RA, Bonne T, Flück D, Bangsbo J, Lundby C. A short period of high-intensity interval training improves skeletal muscle mitochondrial function and pulmonary oxygen uptake kinetics. J Appl Physiol. 2016;120(11):1319–27.
4. Gunnarsson TP, Brandt N, Fiorenza M, Hostrup M, Pilegaard H, Bangsbo J. Inclusion of sprints in moderate intensity continuous training leads to muscle oxidative adaptations in trained individuals. Physiol Rep. 2019;7(4):e13976.
5. Granata C, Oliveira RSF, Little JP, Renner K, Bishop DJ. Mitochondrial adaptations to high-volume exercise training are rapidly reversed after a reduction in training volume in human skeletal muscle. FASEB J. 2016;30(10):3413–23.
6. Christensen PM, Gunnarsson TP, Thomassen M, Wilkerson DP, Nielsen JJ, Bangsbo J. Unchanged content of oxidative enzymes in fast-twitch muscle fibers and kinetics after intensified training in trained cyclists. Physiol Rep. 2015;3(7):e12428.
7. Bishop DJ, Granata C, Eynon N. Can we optimise the exercise training prescription to maximise improvements in mitochondria function and content? Biochim Biophys Acta. 2014;1840(4):1266–75.
8. Meeusen R, Duclos M, Foster C, et al. 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. 2013;45(1):186–205.
9. Aubry A, Hausswirth C, Louis J, Coutts AJ, Le Meur Y. Functional overreaching: the key to peak performance during the taper? Med Sci Sports Exerc. 2014;46(9):1769–77.
10. Le Meur Y, Pichon A, Schaal K, et al. Evidence of parasympathetic hyperactivity in functionally overreached athletes. Med Sci Sports Exerc. 2013;45(11):2061–71.
11. Hausswirth C, Louis J, Aubry A, Bonnet G, Duffield R, Le Muer Y. Evidence of disturbed sleep patterns and increased illness in functionally overreached endurance athletes. Med Sci Sports Exerc. 2014;46(5):1036–45.
12. Ferraresso RL, de Oliveira R, Macedo DV, et al. Interaction between overtraining and the interindividual variability may (not) trigger muscle oxidative stress and cardiomyocyte apoptosis in rats. Oxid Med Cell Longev. 2012;2012:935483.
13. Bosquet L, Montpetit J, Arvisais D, Mujika I. Effects of tapering on performance: a meta-analysis. Med Sci Sports Exerc. 2007;39(8):1358–65.
14. Banister EW, Carter JB, Zarkadas PC. Training theory and taper: validation in triathlon athletes. Eur J Appl Physiol Occup Physiol. 1999;79(2):182–91.
15. Houmard JA, Scott BK, Justice CL, Chenier TC. The effects of taper on performance in distance runners. Med Sci Sports Exerc. 1994;26(5):624–31.
16. Neary JP, Bhambhani YN, McKenzie DC. Effects of different stepwise reduction taper protocols on cycling performance. Can J Appl Physiol. 2003;28(4):576–87.
17. Jones AM, Doust JH. A 1% treadmill grade most accurately reflects the energetic cost of outdoor running. J Sports Sci. 1996;14(4):321–7.
18. Peronnet F, Massicotte D. Table of nonprotein respiratory quotient: an update. Can J Sport Sci. 1991;16(1):23–9.
19. Jeukendrup AE, Wallis GA. Measurement of substrate oxidation during exercise by means of gas exchange measurements. Int J Sports Med. 2005;26(1 Suppl):S28–37.
20. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol. 1986;60(6):2020–7.
21. Ryan TE, Erickson ML, Brizendine JT, Young H-J, McCully KK. Noninvasive evaluation of skeletal muscle mitochondrial capacity with near-infrared spectroscopy: correcting for blood volume changes. J Appl Physiol. 2012;113(2):175–83.
22. Ryan TE, Southern WM, Reynolds MA, McCully KK. A cross-validation of near-infrared spectroscopy measurements of skeletal muscle oxidative capacity with phosphorus magnetic resonance spectroscopy. J Appl Physiol. 2013;115(12):1757–66.
23. Ryan TE, Brophy P, Lin C-T, Hickner RC, Neufer PD. Assessment of in vivo skeletal muscle mitochondrial respiratory capacity in humans by near-infrared spectroscopy: a comparison with in situ measurements. J Physiol. 2014;592(15):3231–41.
24. Adami A, Cao R, Porszasz J, Casaburi R, Rossiter HB. Reproducibility of NIRS assessment of muscle oxidative capacity in smokers with and without COPD. Respir Physiol Neurobiol. 2017;235:18–26.
25. Foster C. Monitoring training in athletes with reference to overtraining syndrome. Med Sci Sports Exerc. 1998;30:1164–8.
26. Mackinnon LT, Hooper SL, Jones S, Gordon RD, Bachmann AW. Hormonal, immunological, and hematological responses to intensified training in elite swimmers. Med Sci Sports Exerc. 1997;29(12):1637–45.
27. Hashimoto T, Hussien R, Oommen S, Gohil K, Brooks GA. Lactate sensitive transcription factor network in L6 cells: activation of MCT1 and mitochondrial biogenesis. FASEB J. 2007;21(10):2602–12.
28. Hoshino D, Tamura Y, Masuda H, Matsunaga Y, Hatta H. Effects of decreased lactate accumulation after dichloroacetate administration on exercise training–induced mitochondrial adaptations in mouse skeletal muscle. Physiol Rep. 2015;3(9).
29. Short KR. Measuring mitochondrial protein synthesis to assess biogenesis. Am J Physiol Endocrinol Metab. 2012;302(9):E1153–4.
30. Neary JP, Martin TP, Reid DC, Burnham R, Quinney HA. The effects of a reduced exercise duration taper programme on performance and muscle enzymes of endurance cyclists. Eur J Appl Physiol Occup Physiol. 1992;65(1):30–6.
31. Luden N, Hayes E, Galpin A, et al. Myocellular basis for tapering in competitive distance runners. J Appl Physiol. 2010;108(6):1501–9.
32. Skovgaard C, Almquist NW, Kvorning T, Christensen PM, Bangsbo J. Effect of tapering after a period of high-volume sprint interval training on running performance and muscular adaptations in moderately trained runners. J Appl Physiol. 2018;124(2):259–67.
33. Neary JP, Martin TP, Quinney HA. Effects of taper on endurance cycling capacity and single muscle fiber properties. Med Sci Sports Exerc. 2003;35(11):1875–81.
34. Adami A, Rossiter HB. Principles, insights, and potential pitfalls of the noninvasive determination of muscle oxidative capacity by near-infrared spectroscopy. J Appl Physiol. 2018;124(1):245–8.
35. Houmard J, Kirwan J, Flynn M, Mitchell J. Effects of reduced training on submaximal and maximal running responses. Int J Sports Med. 1989;10(1):30–3.
36. Lehmann M, Baumgartl P, Wiesenack C, et al. Training-overtraining: influence of a defined increase in training volume vs training intensity on performance, catecholamines and some metabolic parameters in experienced middle- and long-distance runners. Eur J Appl Physiol Occup Physiol. 1992;64(2):169–77.
37. Thomas L, Busso T. A theoretical study of taper characteristics to optimize performance. Med Sci Sports Exerc. 2005;37(9):1615–21.
38. Wang R, Lagakos SW, Ware JH, Hunter DJ, Drazen JM. Statistics in medicine—reporting of subgroup analyses in clinical trials. N Engl J Med. 2007;357:2189–94.
39. Yusuf S, Wittes J, Probstfield J, Tyroler HA. Analysis and interpretation of treatment effects in subgroups of patients in randomized clinical trials. JAMA. 1991;266:93–8.


Copyright © 2019 by the American College of Sports Medicine