The time needed to return to a fitness level close to the initial state after a prolonged training cessation (e.g., more than a month) is an important issue for many athletes. Detraining is characterized by marked alterations in the cardiorespiratory system (e.g., reduction in maximal oxygen uptake - V ˙ O₂max) and metabolic patterns during exercise, and depends on several factors, such as cessation duration, previous training load and initial level of the athletes (Mujika and Padilla, 2000; Barbieri et al., 2024).

Aging is associated with progressive reductions in cardiovascular and neuromuscular function, contributing to the decline in the physiological determinants of endurance performance with advancing age. For example, maximum cardiorespiratory fitness, measured by maximal oxygen uptake, decreases by 0.5%–1% per year starting at age 30, primarily due to lower cardiac output resulting from a reduction in maximum heart rate (0.7 beats per minute per year) and a decrease in muscle mass (Trappe et al., 1996). Running economy at a fixed running speed has also been reported to decrease with age in some studies (e.g., dos Anjos Souza et al., 2023), though not all (e.g., Quinn et al., 2011), with this decline partially attributed to reductions in tendon stiffness (Karamanidis and Arampatzis, 2006), which, in turn, decreases the mechanical efficiency of the muscle (Bohm et al., 2019). Finally, the metabolic rate at which the lactate threshold occurs has been reported to decrease (Wiswell et al., 2000) or remain unchanged with age (Allen et al., 1985). While master athletes experience notable declines in cardiovascular function and endurance capacity after periods of detraining, they may maintain certain physiological adaptations acquired through lifelong training, potentially mitigating the effects of detraining (Barbieri et al., 2024).

Endurance exercise training induces, among other adaptations, increases in stroke volume and muscle oxidative capacity, resulting in an increase in V ˙ O₂max. There is little data in the literature on how rapidly these parameters adapt when highly trained individuals, who have exercised regularly for years, stop training for prolonged periods. The review by Burtscher et al. (2022) based on fewer than ten studies, suggests an almost linear decrease in V ˙ O_2max of up to about 20% following training cessation periods ranging from a few days to 12 weeks. For example, Coyle et al. (1984) showed that in young (≈30 years old) highly endurance-trained athletes, the reduction in absolute V ˙ O_2max could reach 16% after 12 weeks of training cessation. Although master athletes (>40 years old) have increasingly participated in endurance events in recent decades (Lepers and Stapley, 2016), we still lack comprehensive knowledge about the process of their detraining and retraining. Nichols et al. (2020) observed that following a 4-week detraining period (due to a collarbone fracture), the V ˙ O_2max of a female master (49.5 years old) cyclist decreased by 26%. Interestingly, after just 6 weeks of retraining, her V ˙ O_2max returned to its preinjury value, while peak power output (PPO) was still 6% lower, and it took 11 weeks to reach the preinjury value.

Conducting studies to examine the effects of detraining and retraining in competitive, well-trained athletes is difficult, as these athletes are often reluctant to voluntarily stop their training for a long period (e.g., >4 weeks). Consequently, even case reports represent valuable information on changes in physical fitness after periods of detraining and retraining.

In a previous case report, we demonstrated that a 12-week detraining program could benefit a well-trained master athlete by enhancing their muscle metabolic phenotype and aerobic capacity following a retraining protocol (Zanou et al., 2024). At the muscle level, detraining was associated with reduced levels of mitochondrial biogenesis and oxidative phosphorylation (OXPHOS) proteins. Conversely, it led to an increase in the levels of the ryanodine receptor type 1 (RyR1) protein, myosin heavy chain fast-twitch (fast MHC), and Sarco/Endoplasmic reticulum Ca2+ ATPase type 1 (SERCA) protein. In contrast, retraining elevated not only mitochondrial proteins but also glycolytic proteins beyond baseline levels, likely improving skeletal muscle metabolism. In this companion article, we describe the effects of a 12-week detraining period followed by a retraining period on cardiorespiratory fitness, muscular strength, and body composition in the same highly trained male master triathlete.

Figure 1 shows that absolute V ˙ O_2max decreased by 6.8% for the cycling test and 8.5% for running test after the 12-week detraining period, while PPO for cycling and MAV for running decreased by 9.9% and 8.6%, respectively. When expressed relative to body mass, the reduction in V ˙ O_2max reached 9.1% for cycling and 10.9% for running, respectively (see Table 1). After 3 months of retraining, these variables improved considerably and were even slightly higher than baseline values. Indeed, compared to baseline values, absolute V ˙ O_2max was 2.4% higher for running and 5% higher for cycling, respectively. PPO increased by 3.5% after retraining, while MAV was similar to baseline. HR_max increased by 7-8 bpm after detraining and was slightly lower than baseline values after retraining (Table 1). V ˙ E_max decreased after detraining, particularly for running, and almost returned to baseline values after retraining.

Analysis at submaximal exercise intensities revealed that after 12 weeks of detraining, CE (at an intensity corresponding to 53% of baseline PPO) decreased by 6.2% and remained 2.6% below baseline after retraining (Table 1). Similarly, running economy (at an intensity corresponding to 68% of baseline MAV) was 22% higher after detraining and remained 17% higher after retraining compared to baseline. Heart rate and ventilation at these submaximal intensities were higher after detraining than baseline values for cycling and running. After retraining, these variables returned to baseline values for cycling but not for running.

Table 1 showed that, after detraining, maximal strength capacity of the knee extensor muscles decreased by an average of 8.2%, ranging from 2.7% to 14.9%, depending on the contraction mode. After retraining, maximal strength capacity remained slightly below baseline values (−3.2% on average). The fatigue index over 50 repeated dynamic contractions increased by 34.6% after detraining and remained higher by 7.7% after retraining compared to baseline values. After detraining, jumping performances decreased by an average of 9.2% (−5.2% for the CMJ and −13.1% for the SJ), respectively. After retraining, jumping performances remained slightly below baseline values (−3.6% on average).

Body fat percentage increased from 10.1% to 13.3% during detraining (Table 2). The participant gained 2.5 kg of fat and lost 2.2 kg of fat-free mass. The loss of fat-free mass was more pronounced in the trunk (−5.7%) than in the arms (−1%) and legs (−1.5%). After 12 weeks of retraining, the participant was lighter than at baseline, and his percentage of body fat had dropped to 8.7%. His body fat mass was 1.6 kg lower than the baseline value, and his free-fat mass was 1 kg lower. Bone mineral content remained unchanged during the detraining and retraining periods (Table 2).

This study reports on changes in cardiorespiratory fitness, body composition and knee extensor maximal strength capacity of a highly trained 54-year-old master triathlete following 12 weeks of detraining and retraining. At baseline, his V ˙ O_2max was 62 mL kg⁻¹.min⁻¹, placing him above the 95th percentile of the age-related reference values provided by the ACSM’s Guidelines for Exercise Testing and Prescription (2021), indicating his high level of fitness.

The decrease in V ˙ O_2max (9%–10%) after 12 weeks of detraining was quite similar when measured in cycling or running and is in line with the lowest values reported earlier in literature for a similar detraining period (Burtscher et al., 2022; Coyle et al., 1984). Because no measurement was made during the 3 months of detraining, the kinetics of the decline in V ˙ O_2max could not be assessed. The decline in maximal O₂ pulse - defined as V ˙ O₂max.HRmax⁻¹ by 8.8% for cycling and 14.3% for running suggests that maximal stroke volume and/or arteriovenous O₂ difference decreased (Hagberg et al., 1985). Previous studies have suggested that a rapid decline in VO_2max occurs during the first 2–3 weeks with a less marked subsequent decline (Coyle et al., 1984; Klausen et al., 1981), but this is not always the case (Henriksson and Reitman, 1977). The initial phase of decrease in V ˙ O_2max in the first 2–3 weeks is thought to be due to a decrease in stroke volume, while the subsequent decrease would be attributed to a decrease in maximal arterio-venous oxygen difference (Coyle et al., 1984), as supported by the companion paper showing a decrease in mitochondrial biogenesis and oxidative phosphorylation proteins (Zanou et al., 2024).

When the participant was 40, his V ˙ O_2max, PPO, and MAV were 71 mL kg⁻¹.min⁻¹, 425 W, and 20 km h⁻¹, respectively. At the age of 53.8 (i.e., at baseline), these values were reduced by 12%–13%, corresponding to a loss of 7%–8% per decade, which is in line with existing literature, (around 7% per decade up to the age of 60 (Lepers et al., 2020; Valenzuela et al., 2020). Therefore, the 3 months of detraining (i.e., ∼10% loss of V ˙ O_2max), was equivalent to aging by around 15 years, from a cardiorespiratory point of view.

After detraining, RE assessed at 12 km h⁻¹ was much more altered than V ˙ O_2max (22% vs. 10.9%), and it was still 17.7% higher after retraining compared to baseline, while V ˙ O_2max returned to its initial value. This observation suggests that following a detraining period, the return to initial values is much slower for RE than for V ˙ O_2max (Jones, 2016). However, this needs to be confirmed with a larger sample of participants. After retraining, maximal leg strength capacity remained slightly depressed (−3%), which may partly explain the changes in RE. Other factors, such as biomechanical factors, could also play a role. For example, previous studies have shown that vertical, leg and knee stiffness were negatively correlated with RE (Li et al., 2021; Liu et al., 2022). However, older runners tend to show a less favorable balance between stiffness and reactive strength, possibly affecting performance and increasing energy costs at submaximal speeds (Jaén-Carrilo et al., 2021). It is worth noting that the proportion of training time devoted to running during the retraining period was lower compared to the pre-detraining period (25% vs. 30%), and running speeds were lower, especially at the beginning of the retraining period. The lower volume of running training may contribute to the reduced running economy seen after 12 weeks of retraining. Conversely, cycling efficiency was less affected by detraining than running economy (6.2% vs. 22%). The smaller reduction in energy cost for cycling compared to running during detraining can be attributed to the distinct characteristics of each activity. Running involves stretch-shortening cycles with eccentric muscle actions, whereas cycling is a non-weight-bearing activity that primarily uses concentric muscle actions and reduced osteoarticular stress. This observation suggests that after a detraining period, it takes longer for a triathlete to regain optimal energy cost in running compared to cycling, although this hypothesis requires further validation. Additionally, the relative time spent on cycling during the retraining phase was greater than during the pre-detraining period (47% vs. 38%). This increased focus on cycling during retraining may explain why cycling efficiency was nearly restored to baseline levels.

After detraining, the decline in maximum strength capacity was more pronounced for the concentric contractions at high contraction speeds (11%–15% at 120°.s⁻¹ and 180°.s⁻¹) compared with those at lower speeds (60°.s⁻¹) or under isometric conditions (3%–4%). The ability to maintain a level of muscle torque over 50 dynamic contractions (estimated by the fatigue index) was much more diminished (35%) than the maximum strength capacity, suggesting that the subject was more fatigable after detraining. These results are consistent with the increase in fast MHC observed in the participant.⁶ The decline in strength capacity could be explained by the reduction in leg fat-free mass, which was reduced by 1.5% during the detraining period. The leg fat-free mass did not return to baseline after retraining, nor did strength capabilities completely return to baseline. It should be noted that total mass and body fat percentage were lower after retraining than at baseline, which contributed to the increased relative V ˙ O_2max.

One of the limitations of this study is that the lactate threshold was not assessed. Lactate threshold has been identified as one of the main determinants of endurance, in addition to V ˙ O_2max and efficiency (i.e., the oxygen cost to generate a given running speed or cycling power output), (Joyner and Coyle, 2008). It would also have been interesting to monitor the time course of detraining and retraining, as well as other variables such as diet and physical activity, using accelerometry. Moreover, parameters such as voluntary activation level and muscle architecture would provide valuable insights into neuromuscular adaptations. Except the 400-m swim test, the athlete did not perform any cycling or running performance tests, such as a time to exhaustion at a given intensity or a time trial. During the triathlon racing season, starting 6 months after the end of the detraining period, the athlete’s performance level was in line with his usual level. Further studies are needed to better understand physiological changes after controlled periods of detraining and retraining in elite athletes with a high training volume (>20 h per week). Finally, the amplitude of changes between baseline and detraining or retraining for the physiological parameters, such as V ˙ O_2max, reported in this case report, should be considered with caution, considering the reliability that is inherent to the measurement. Indeed, Hopkins et al. (2001) showed that the coefficient of variation for V ˙ O_2max and PPO could range from 1.5% to 5%.

This case report illustrates the changes in cardiorespiratory fitness following 12 weeks of detraining and retraining in a lifelong endurance athlete following his competitive season. After a retraining period of the same duration as the detraining one, the master athlete reached and even exceeded his baseline level of cardiorespiratory fitness, but his running economy and lean mass remained slightly depressed.