Resistance training is well-established as a primary exercise-based strategy to enhance muscle mass in humans. The manipulation of program variables is believed to be a key factor in optimizing muscular gains (1,14). Variables such as volume, load, and frequency have been well explored in the literature (28–30). One variable that has not received as much attention is the set endpoint, operationally defined as the point at which a set of repetitions is terminated (34). The intensity of effort expended during a set is best estimated by how close an individual comes to reaching muscular failure, which can be defined as the point where the activated muscles are incapable of completing another complete repetition without assistance (22,25). Training to muscular failure has been promoted since the 1940s when DeLorme, a physician in the U.S. Army at the time, published a series of articles advocating the use of this method in resistance exercise (15). Although training to failure has long been used in resistance training programs, particularly by bodybuilders, the number of well-controlled studies that have explored this topic is small. Initially, Rooney et al. (21) indicated a benefit for training to failure (versus not training to failure) for gains in strength; however, these results were not corroborated by others (8). Recognizing the equivocal body of evidence, Davies et al. (4) recently conducted a meta-analysis for strength gains and concluded that similar increases in strength might be attained with both muscle failure and nonfailure training. However, this meta-analysis (as well as the majority of original studies) only focused on gains in strength; the effects of training to failure on muscle hypertrophy have been less explored.
Some researchers claim that training to failure is necessary to maximize muscle growth (2,7,38). This contention is at least in part based on the underlying belief that training to failure elicits full motor unit recruitment, which is considered an essential component for increases in muscle size (37). However, the applicability of this claim may be load-specific. High-threshold motor units are recruited almost immediately when lifting very heavy loads as high levels of force are required from the onset of the exercise (5). This is in contrast to low-load training, where recruitment of the larger motor units is delayed because a high level of force is not initially needed to lift the weight; as the set becomes more fatiguing, higher threshold motor units are then recruited to maintain force output. This physiological response has been demonstrated in studies showing that fatiguing concentric contractions produce a corresponding increase in surface electromyography activity during low-load training (27,32), but the effect diminishes with the use of progressively heavier loads. It therefore could be hypothesized that the need to train to failure becomes increasingly less relevant when training with high intensities of load.
It also has been hypothesized that training to failure augments muscular growth by increasing metabolic stress. There is some evidence that the buildup of metabolites—byproducts from anaerobic energy production—enhances the anabolic response to resistance training (26,35), although this claim remains speculative (36). Research shows that continuing a set to the point of volitional fatigue heightens energy demands, thereby resulting in a greater metabolite accumulation (10). This seemingly supports a beneficial metabolic effect of training to the point of failure. However, it is not clear whether the additional metabolic stress produced during an “all-out” set leads to a meaningfully greater accretion of muscle proteins compared with a set stopped short of failure. It is conceivable that a threshold exists for metabolic stress beyond which no further beneficial effects are realized. The purpose of this column is to (a) review the current literature as to the effects of failure training on hypertrophic adaptations and (b) draw evidence-based conclusions as to how the strategy should be used in practice, both in terms of integration with other resistance training variables and avoiding burnout, for those seeking maximize muscle growth.
OVERVIEW OF THE LITERATURE
There is surprisingly little longitudinal research investigating the topic of training to failure. An often-cited study in support of training performed to failure compared muscle growth in recreationally trained men performing a multiple set protocol of 10 reps with 60 seconds of rest between sets, whereby one group trained to failure and the other did not (11). Exercises consisted of the lat pull-down, shoulder press, and leg extension, with 3–5 sets performed per exercise. Results showed that the group training to failure gained significantly more muscle over the course of the 12-week study than the group that did not train to muscle failure (+13 versus +4%). Although on the surface, these findings may seem intriguing, the caveat here was that 1 group performed all sets continuously to failure while the group that did not train to failure took a 30-second break at the midpoint of each set. This protocol does not replicate traditional nonfailure training regimens where sets are stopped at a given number of repetitions from fatigue; thus, it has limited ecological validity.
In a study by Schott et al. (31), one group trained using a low-fatigue training protocol that included 4 sets of 10 contractions (each contraction lasted 3 seconds followed by 2 seconds of rest) with 2 minutes of rest between sets while another group trained with high levels of fatigue induced by performing 4 sets of 30-second contractions with 1 minute of between-set rest. Only the latter group experienced significant increases in muscle size after 14 weeks of training, adding further support for the importance of fatigue for gains in muscle mass. However, as in the study by Goto et al. (11), the protocol did not mirror resistance exercise performed in the practical context because it included only isometric muscle actions. Furthermore, the protocol in the low fatigue group included inter-repetition rest which, again, is not often used in traditional resistance exercise.
Sampson and Groeller (22) compared the effects of exercising with muscle failure using a more traditional resistance training protocol. Twenty-eight untrained young men performed 4 sets of arm curls at 85% 1 repetition maximum (1RM). Subjects were randomized either to perform sets to failure using a 2-second concentric and 2-second eccentric action or to stop approximately 2 reps short of failure while using either rapid shortening (explosive concentric and 2-second eccentric action) or stretch-shortening (explosive movement on both concentric and eccentric components). After 12 weeks, the average gain in biceps muscle cross-sectional area was ∼11% for all subjects combined, with no significant differences noted between groups. A key point to keep in mind here is that subjects trained with heavy loads equating to a 6RM. It therefore can be hypothesized that training to failure becomes less important when using heavy weights, which is consistent with the previously mentioned research on muscle activation. The study did have a potential confounding issue: the nonfailure groups actually performed a single set to failure at the end of each week to determine the load for the subsequent week of training. Whether this had a significant effect on the results is not clear.
Martorelli et al. (16) randomly assigned 89 active women to 1 of 3 groups: (a) a group that performed 3 sets of repetitions to failure at 70% 1RM, (b) a group that performed 4 sets of 7 repetitions not to failure, but with volume equalized to the failure condition, and (c) a group that performed 3 sets of 7 repetitions not to failure. Training consisted of free-weight biceps curls performed 2 days per week for 10 weeks. The authors observed significant main effects for time and for the interaction between the groups. However, the authors did not perform further post hoc analyses to determine where the differences between the groups occurred. The relative changes substantially favored the group training to muscle failure (17.5% versus 8.5% in the group not training to failure, with equal volume). The magnitude of differences between groups raises the possibility of a type II error, whereby statistically significant differences went undetected. The group that performed repetitions not to failure without matching volume to the 2 other groups did not significantly increase muscle thickness.
Even if there indeed was an advantage favoring the group that trained to failure in the study by Martorelli et al., it should be noted that the included participants were young women, and therefore, the results cannot be generalized to older adults. Older adults might experience slower postexercise recovery and may warrant a different approach to their program design (6). da Silva et al. (3) essentially used the same study design as Martorelli et al. (16) while including older men (66 ± 5 years) as participants. In this study, the group training to failure and the group that did not train to failure with equalized volume experienced similar increases in quadriceps muscle thickness; no significant pre-to-post changes occurred in the group that did not train to failure and that performed less volume than the 2 other training groups. These results may indicate that age (and the age-related recovery from exercise) is an important factor to consider when prescribing resistance exercise performed to muscle failure. Furthermore, the study by da Silva et al. (3) indicates that training to muscle failure may not be needed for increases in muscle size in older adults. Finally, the totality of findings also suggests that volume load may be an important variable when considering the relevance of training to failure.
Most recently, Nóbrega et al. (18) allocated 32 untrained men to perform 3 sets of leg extension exercise at either a high load (80% 1RM) or low load (30% 1RM), twice per week. The study used a within-subject design whereby each lower limb was randomized to execute these conditions either to failure or terminated at the point at which participants voluntarily interrupted training. After 12 weeks, increases in muscle cross-sectional area were statistically similar between conditions. It should be noted that the differences in volume load between failure and nonfailure training were rather slight in both the high-load conditions (training to failure: 26,694 kg; training not to failure: 26,042 kg) and in the low-load conditions (training to failure: 21,114 kg; training not to failure: 20,643 kg), suggesting that the nonfailure condition performed sets at close to full fatigue (likely only 1 to 2 repetitions shy of failure). These results, therefore, may indicate that training close to muscle failure may be similarly effective for increasing muscle size as training that includes reaching actual muscle failure.
A primary issue when attempting to draw evidence-based conclusions on the topic is qualifying the alternative endpoint to failure. Specifically, if failure is not the chosen option, then at what point should a set be terminated? One option may be to use the “repetitions in reserve” (RIR) scale proposed by Zourdos et al. (39), whereby an RIR of 0 equates to training to failure, an RIR of 1 equates to stopping 1 repetition short of failure, an RIR of 2 equates to stopping 2 repetitions short of failure, etc. However, as indicated by Hackett et al. (12), individuals may underestimate the number of repetitions to failure during the earlier sets and the accuracy of prediction may increase with subsequent sets. Other researchers also show that resistance training experience may increase the ability to accurately predict repetitions to failure (33), and therefore, when using this scale, a period of familiarization should be incorporated in the study design to keep the comparison between the groups valid.
A potential issue with continuous training to failure is that it may increase the potential for overtraining and psychological burnout (9). This hypothesis was supported by a study from Izquierdo et al. (13), who randomized members of the Spanish Basque ball team to perform 3 sets of 8 resistance exercises targeting the body's major muscle groups either to failure or non-failure using 70–80% of 1RM. Training was performed twice a week for 16 weeks. Results showed that training to failure blunted resting levels of anabolic hormones (Insulin‐like growth factor-1 and testosterone), an outcome indicative of nonfunctional overreaching (9). Thus, if failure training is used in a program, it seems prudent to do so judiciously. There is no research on the topic, but 1 strategy would be to limit its use to the last set of an exercise. For example, if muscle failure is incorporated in the first set of a given exercise, it is likely that the performance (concerning the total number repetitions) would be hindered on subsequent sets (24). By limiting the use of a muscle failure only on the last set of a given exercise, we may ensure that sufficient volume is achieved, although the effects of failure training on volume remain somewhat equivocal (24). As with all resistance training variables, failure training could also be periodized, so that it is used more extensively in a short training block (i.e., the peaking phase of a mass-building program) and less so during other training cycles.
Based on the limited available evidence, it would seem that stopping several repetitions short of failure when training with moderately heavy loads (6–12RM) does not seem to compromise hypertrophy, at least when training volume is equated. The paucity of research with light loads makes it difficult to extrapolate a specific recommendation, but a logical case (from a motor unit recruitment perspective) can be made that RIR values would need to be in the range from 0 to 2 to fully stimulate the highest threshold motor units, and thus maximize hypertrophic adaptations. More research is needed to strengthen evidence-based conclusions.
Other more advanced methods, such as the assessment of velocity loss also require attention in this context (23). In 1 study, Pareja-Blanco et al. (19) randomized resistance-trained men to 1 of 2 protocols that differed only in the amount of repetition velocity loss allowed in each set: 20 versus 40% of velocity loss. The 40% velocity loss group trained in close proximity to muscle failure while the 20% velocity loss group performed approximately half of the maximum number of repetitions. After the 8-week training intervention, greater hypertrophy in the vastus lateralis and intermedius was observed in the 40% velocity loss group. These findings suggest that training to failure (or very close to it) might be indeed of importance for maximizing increases in muscle size. One limitation here is that the groups were not matched for total volume in terms of the total number of performed repetitions as the 40% velocity loss group also performed more total repetitions. This may be relevant given the linear dose-response relationship between volume and muscle hypertrophy (28). It would be interesting for future studies to use a similar protocol while adding more sets to the small velocity loss group to equate the volume load comparison.
An important limitation with the current body of literature is that most studies to date have been conducted in untrained subjects. A case can be made that as a lifter gains more training experience, there is an increasing need to challenge the neuromuscular system with higher levels of effort. Support for this hypothesis can be inferred in the results of the study by Pareja-Blanco et al. (19), but additional research is needed to draw stronger conclusions.
Ultimately, training to failure also needs to be considered in the context of the whole resistance training program (Figure). One variable that is likely important to take into account here is training frequency. In a recent study, training to failure in each set using a 3 × 10-repetition protocol (compared to training without reaching failure using a 6 × 5-repetition protocol) slowed down the recovery up to 24–48 hours after exercise (17). This attenuated postexercise recovery will not likely be of practical importance if the training program is performed with a low weekly training frequency per muscle group (e.g., training each muscle group once per week). However, when training with a higher training frequency (e.g., training a muscle group +4 times week), training to failure should likely be used sparingly to allow a better neuromuscular condition before the subsequent training session. Slower rates of recovery may be more pronounced when exercising with a higher number of repetitions, which is another component that needs to be taken into account in program design (20).
It also is important to consider the specific exercises performed. Multijoint movements, particularly those performed using free weights and of a structural nature, are substantially more taxing on the neuromuscular system than single-joint exercises. It therefore seems pragmatic to limit the use of training to failure in exercises such as squats, deadlifts, presses, and rows. Alternatively, failure can be used more liberally when performing single-joint exercises because they are much less physically and mentally demanding.
1. American College of Sports Medicine. American College of Sports Medicine position stand. Progression models in resistance training for healthy adults. Med Sci Sports Exerc 41: 687–708, 2009.
2. Burd NA, West DW, Moore DR, Atherton PJ, Staples AW, Prior T, Tang JE, Rennie MJ, Baker SK, Phillips SM. Enhanced amino acid sensitivity of myofibrillar protein synthesis persists for up to 24 h after resistance exercise in young men. J Nutr 141: 568–573, 2011.
3. da Silva LXN, Teodoro JL, Menger E, Lopez P, Grazioli R, Farinha J, Moraes K, Bottaro M, Pinto RS, Izquierdo M, Cadore EL. Repetitions to failure versus not to failure during concurrent training in healthy elderly men: A randomized clinical trial. Exp Gerontol 108: 18–27, 2018.
4. Davies T, Orr R, Halaki M, Hackett D. Effect of training leading to repetition failure on muscular strength: A systematic review and meta-analysis. Sports Med 46: 487–502, 2016.
5. Duchateau J, Semmler JG, Enoka RM. Training adaptations in the behavior of human motor units. J Appl Physiol 101: 1766–1775, 2006.
6. Fell J, Williams D. The effect of aging on skeletal-muscle recovery from exercise: Possible implications for aging athletes. J Aging Phys Act 16: 97–115, 2008.
7. Fisher J, Steele J, Smith D. Evidence-based resistance training recommendations for muscular hypertrophy. Medicina Sportiva 7: 16–26, 2013.
8. Folland JP, Irish CS, Roberts JC, Tarr JE, Jones DA. Fatigue is not a necessary stimulus for strength gains during resistance training. Br J Sports Med 36: 370–373, 2002.
9. Fry AC, Kraemer WJ. Resistance exercise overtraining and overreaching. Neuroendocrine responses. Sports Med 23: 106–129, 1997.
10. Gorostiaga EM, Navarro-Amézqueta I, Calbet JA, Sánchez-Medina L, Cusso R, Guerrero M, Granados C, González-Izal M, Ibáñez J, Izquierdo M. Blood ammonia and lactate as markers of muscle metabolites during leg press exercise. J Strength Cond Res 28: 2775–2785, 2014.
11. Goto K, Ishii N, Kizuka T, Takamatsu K. The impact of metabolic stress on hormonal responses and muscular adaptations. Med Sci Sports Exerc 37: 955–963, 2005.
12. Hackett DA, Johnson NA, Halaki M, Chow CM. A novel scale to assess resistance-exercise effort. J Sports Sci 30: 1405–1413, 2012.
13. Izquierdo M, Ibañez J, González-Badillo JJ, Häkkinen K, Ratamess NA, Kraemer WJ, French DN, Eslava J, Altadill A, Asiain X, Gorostiaga EM. Differential effects of strength training leading to failure versus not to failure on hormonal responses, strength, and muscle power gains. J Appl Physiol 100: 1647–1656, 2006.
14. Kraemer WJ, Ratamess NA. Fundamentals of resistance training: Progression and exercise prescription. Med Sci Sports Exerc 36: 674–688, 2004.
15. Kraemer WJ, Ratamess NA, Flanagan SD, Shurley JP, Todd JS, Todd TC. Understanding the science of resistance training: An evolutionary perspective. Sports Med 47: 2415–2435, 2017.
16. Martorelli S, Cadore EL, Izquierdo M, Celes R, Martorelli A, Cleto VA, Alvarenga JG, Bottaro M. Strength training with repetitions to failure does not provide additional strength and muscle hypertrophy gains in young women. Eur J Transl Myol 27: 6339, 2017.
17. Moran-Navarro R, Perez CE, Mora-Rodriguez R, de la Cruz-Sanchez E, Gonzalez-Badillo JJ, Sanchez-Medina L, Pallares JG. Time course of recovery following resistance training leading or not to failure. Eur J Appl Physiol 117: 2387–2399, 2017.
18. Nóbrega SR, Ugrinowitsch C, Pintanel L, Barcelos C, Libardi CA. Effect of resistance training to muscle failure vs. volitional interruption at high- and low-intensities on muscle mass and strength. J Strength Cond Res 32: 162–169, 2018.
19. Pareja-Blanco F, Rodriguez-Rosell D, Sanchez-Medina L, Sanchis-Moysi J, Dorado C, Mora-Custodio R, Yanez-Garcia JM, Morales-Alamo D, Perez-Suarez I, Calbet JAL, Gonzalez-Badillo JJ. Effects of velocity loss during resistance training on athletic performance, strength gains and muscle adaptations. Scand J Med Sci Sports 27: 724–735, 2017.
20. Pareja-Blanco F, Rodriguez-Rosell D, Aagaard P, Sanchez-Medina L, Ribas-Serna J, Mora-Custodio R, Otero-Esquina C, Yanez-Garcia JM, Gonzalez-Badillo JJ. Time course of recovery from resistance exercise with different set configurations. J Strength Cond Res 2018 [Epub ahead of print].
21. Rooney KJ, Herbert RD, Balnave RJ. Fatigue contributes to the strength training stimulus. Med Sci Sports Exerc 26: 1160–1164, 1994.
22. Sampson JA, Groeller H. Is repetition failure critical for the development of muscle hypertrophy and strength? Scand J Med Sci Sports 26: 375–383, 2016.
23. Sanchez-Medina L, Gonzalez-Badillo JJ. Velocity loss as an indicator of neuromuscular fatigue during resistance training. Med Sci Sports Exerc 43: 1725–1734, 2011.
24. Santos WDND, Vieira CA, Bottaro M, Nunes VA, Ramirez-Campillo R, Steele J, Fisher JP, Gentil P. Resistance training performed to failure or not to failure results in similar total volume, but with different fatigue and discomfort levels. J Strength Cond Res 2019 [Epub ahead of print].
25. Schoenfeld BJ. The mechanisms of muscle hypertrophy and their application to resistance training. J Strength Cond Res 24: 2857–2872, 2010.
26. Schoenfeld BJ. Potential mechanisms for a role of metabolic stress in hypertrophic adaptations to resistance training. Sports Med 43: 179–194, 2013.
27. Schoenfeld BJ, Contreras B, Willardson JM, Fontana F, Tiryaki-Sonmez G. Muscle activation during low- versus high-load resistance training in well-trained men. Eur J Appl Physiol 114: 2491–2497, 2014.
28. Schoenfeld BJ, Ogborn D, Krieger JW. Dose-response relationship between weekly resistance training volume and increases in muscle mass: A systematic review and meta-analysis. J Sports Sci 35: 1073–1082, 2017.
29. Schoenfeld BJ, Grgic J, Ogborn D, Krieger JW. Strength and hypertrophy adaptations between low- vs. High-load resistance training: A systematic review and meta-analysis. J Strength Cond Res 31: 3508–3523, 2017.
30. Schoenfeld BJ, Grgic J, Krieger J. How many times per week should a muscle be trained to maximize muscle hypertrophy? A systematic review and meta-analysis of studies examining the effects of resistance training frequency. J Sports Sci 2018 [Epub ahead of print].
31. Schott J, McCully K, Rutherford OM. The role of metabolites in strength training. II. Short versus long isometric contractions. Eur J Appl Physiol Occup Physiol 71: 337–341, 1995.
32. Spiering BA, Kraemer WJ, Anderson JM, Armstrong LE, Nindl BC, Volek JS, Maresh CM. Resistance exercise biology: Manipulation of resistance exercise programme variables determines the responses of cellular and molecular signalling pathways. Sports Med 38: 527–540, 2008.
33. Steele J, Endres A, Fisher J, Gentil P, Giessing J. Ability to predict repetitions to momentary failure is not perfectly accurate, though improves with resistance training experience. PeerJ 5: e4105, 2017.
34. Steele J, Fisher J, Giessing J, Gentil P. Clarity in reporting terminology and definitions of set endpoints in resistance training. Muscle Nerve 56: 368–374, 2017.
35. Takada S, Okita K, Suga T, Omokawa M, Kadoguchi T, Sato T, Takahashi M, Yokota T, Hirabayashi K, Morita N, Horiuchi M, Kinugawa S, Tsutsui H. Low-intensity exercise can increase muscle mass and strength proportionally to enhanced metabolic stress under ischemic conditions. J Appl Physiol (1985) 113: 199–205, 2012.
36. Wackerhage H, Schoenfeld BJ, Hamilton DL, Lehti M, Hulmi JJ. Stimuli and sensors that initiate skeletal muscle hypertrophy following resistance exercise. J Appl Physiol (1985) 126: 30–43, 2019.
37. Wernbom M, Augustsson J, Thomeé R. The influence of frequency, intensity, volume and mode of strength training on whole muscle cross-sectional area in humans. Sports Med 37: 225–264, 2007.
38. Willardson JM, Norton L, Wilson G. Training to failure and beyond in mainstream resistance exercise programs. Strength Cond J 32: 21–29, 2010.
39. Zourdos MC, Klemp A, Dolan C, Quiles JM, Schau KA, Jo E, Helms E, Esgro B, Duncan S, Garcia Merino S, Blanco R. Novel resistance training-specific rating of perceived exertion scale measuring repetitions in reserve. J Strength Cond Res 30: 267–275, 2016.