This article begins with an overview that attempts to roughly answer a few questions. They are
– How long can I take off of training without losing muscle/strength?
– How much training is required to maintain muscle/strength?
– If I do lose muscle/strength, how quickly will I regain it?
I then quickly cover a couple of the mechanisms by which the phenomenon of “muscle memory” (rapid regains after detraining) might occur, and finish with some practical recommendations for reintroducing training if you have had time off or significant reductions in training load for a prolonged period of time.
For reference, the pictures above were taken over the course of 2 months of backpacking, in which I did 1-2 gym sessions per week, with some longer gaps in between. I’d say I maintained my physique well. Whilst this article is inspired largely by the impacts of COVID-19 on many of our training, the lessons in it are applicable in any time where your training has to take a back seat.
I’ve pulled out some literature in order to demonstrate a few things conceptually, but by the far the best resource on the topic of detraining/retraining that I’ve come across in reading for this is by Adam Tzur of Sci-Fit. You can find it here – I’d highly recommend reading it for a more comprehensive look at the state of the evidence.
The most basic concept that we need to grasp when considering retention vs loss of any adaptation is the “use it or lose it principle”, or “reversibility”. That is to say that adaptations arise as a result of stressors, and without continued application of those stressors, they fade away with time. Unfortunately (if you dislike training), you can’t just get super jacked and stop lifting at ALL and never lose any muscle. However, as we’ll see, you can get by with a greatly reduced stimulus without much loss.
With that in mind, under periods of detraining we expect to see eventual erosion of the muscle and strength adaptations that we gained during prior training. The question is over what timescale do these losses kick in, and at what rate, and the answer will depend partly on the nature of the adaptation itself.
First, let’s consider total breaks from training. 10-12 weeks of detraining is sufficient to see significant size and strength losses in individuals following cessation of a strength training regime (1,2). In a study by Staron, 6 women subjected to a 20 week training regime detrained for 30-32 weeks lost 13%, 29% and 32% of their end-of-program strength on the squat, leg extension and leg press respectively, although strength remained above their pre-training levels, suggesting some persistence of adaptation even at that length of time. Interestingly, limited losses of fiber cross-sectional area were observed in this study during the detraining period, which is surprising to me, as measures of muscle mass in other research do show decline. In a study by Bickel, for instance, during 32 weeks of detraining type I and type II muscle hypertrophy were reversed to pre-training levels after only 16 weeks of detraining.
McMaster performed a systematic review on the development, retention and decay rates of strength and power in elite rugby union, league and American football players and found notable losses in strength with prolonged detraining. When training ceased for 7.2 +/- 5.8 weeks, strength losses averaged 14.5 +/- 14.3%. Far larger losses were observed with longer durations of detraining, and trivial losses in the short-term, which is entirely logical.
From that we can take it that if you’re going to do absolutely no training for months on end, you will lose some gains. The rate at which you lose gains is like to accelerate over time, although you may still be at above pre-training levels of either strength or muscle size after even half a year of studiously avoiding exercise.
But what about short durations? Short of an enforced break from training, most people interested in this type of article aren’t going to take more than a couple of weeks off at a time. In short, things look pretty good.
In McMaster’s review, highly trained athletes (those with 3 or more years of resistance training under their belts, currently participating in elite sport) experienced trivial strength losses in up to 3 weeks of detraining. Even after 4 weeks, strength performance may be maintained, albeit with reductions in eccentric force and sport-specific power (stuff that is important for injury prevention, change of direction etc, but marginally less so for just being jacked). A 2020 study by Gavanda on 16 year old males also found that 3 weeks of detraining did not affect muscle thickness or strength gained by 12 weeks of prior training.
In a study by Hwang, 18-20yr old men performed 4 weeks of resistance training, 2 weeks of detraining, and then a further 4 weeks of resistance training. The 2 week break did not appear to reduce lower body strength. This is consistent with another study by Ogasawara, in which participants trained for a period of 24 weeks, either continuously or including two 3 week breaks. At the end of the program, both groups had made similar gains in muscle fiber CSA and strength, suggesting that periodic training breaks don’t appear to matter too much.
An important observation – triceps CSA decreased during the detraining periods in the periodic training group in Ogasawara’s study, but rapidly rebounded when training resumed. This may be attributable to changes in glycogen/water storage within the muscles. This relates to our lived experience – nearly everybody feels and looks a little flatter after very short-term detraining, but this doesn’t mean that true tissue losses have occurred.
To exactly answer the question of when and whether we lose muscle during a short training break is actually very difficult. I’d highly recommend reading that section of the Sci-Fit article that I shared earlier, as it renders most of what I’d have to say redundant. The long and short of it is that the observed short-term losses of lean body mass in these studies may be mostly glycogen and water. That doesn’t preclude some loss in contractile tissue on top, but it is likely somewhere between negligible and non-existent in the first few weeks. Considering that strength appears to be maintained over this time-scale, and that in studies such as the Ogasawara one, short training breaks don’t seem to massively impair growth over the longer-term, I find it hard to believe that your muscles will fall off in 2-3 weeks.
I mentioned that the timescale of losses of different adaptations may change, and that is true, although that doesn’t make giving very exact recommendations about training much easier. Hypertrophy is a morphological adaptation – if you’re interested in getting super jacked, you’re really just asking to build tissue. Maintenance of hypertrophy entails maintaining that tissue, and that’s it. Other adaptations, such as strength, are more multifaceted. Strength performance relies both on the amount of underlying tissue (muscle mass), other morphological adaptations (such as to tendons, and changes to muscle pennation angles etc) and neurological adaptations. Each of these characteristics may decay at different rates.
Highly technical tasks likely decay quicker than less technical tasks – in your lived experience, you might find that your squat groove is affected more by a training break than your general strength on the leg press. Tasks requiring very specific adaptations and high levels of coordination (such as sports specific power as mentioned in Mujika) also likely decay quicker than large scale tissue changes.
In practice, what this means is that depending on the duration of a detraining period, or the manner of maintenance training that you do, you may wholly preserve certain adaptations, whilst others fall by the wayside. You may entirely maintain your muscle mass, whilst still losing some top-end squat performance due to a lack of technical practice and exposure to sufficiently heavy weights. For strength athletes who cannot train with heavy loads for a prolonged time period, investing in maintaining muscle mass is almost certainly your best bet. The technical and neural adaptations required to regain your top end strength are more rapidly regained than tissue, and less easily preserved.
Now, if from all of the above, you concluded that you can take a holiday from training here and there, and your muscles won’t disappear, but that not training for multiple months on end is a bad idea, 5 stars. Most of us could probably have intuited that.
The more important question tends to be how we can maintain our gains when our normal training routine is heavily impacted, and to answer that question I am going to split things into answers directed at strength, and then hypertrophy.
It is likely that one exposure per week to heavy-ish weights, especially when taken near-failure, is sufficient to maintain general strength qualities. However, that comes with a number of caveats, which I’ll explore.
Firstly, to justify my recommendation, the evidence from McMaster on athletes suggests that in the initial 2-3 weeks of detraining, strength losses are trivial, whilst Hwang didn’t report strength losses after 2 weeks of detraining.
In the aforementioned study by Bickel, however, we get a slightly different picture. In this study, subjects performed 3 sets of 8-12 reps on 3 leg exercises (9 sets/d) on 3 occasions per week for a 16 week training block. After this, subjects either detrained completely, reduced to 1 training day per week, or reduced to 1 training day per week with only 1 set performed per exercise (ie, either performing 1/3 or 1/9 of their previous volume) with both maintenance prescriptions appearing to effectively maintain strength. However, in the lower volume maintenance group, muscle losses began to be observed after the 16th week of detraining, which may mean that over an even longer time-scale it would prove insufficient to fully maintain strength adaptations.
There were also age-dependent differences in responses observed, with young trainees maintaining (and even gaining) muscle during the detraining period, and the older adults not. This again leads me to think that absolutely minimal training doses might be best avoided where possible, as certain populations likely require higher training doses just to maintain.
It’s important to note that maintaining the improvements in a leg press 1RM garnered by training in the 8-12 range is a far cry from improving squat, bench or deadlift 1RM numbers for a powerlifter, however. A recent systematic review and meta analysis found that a single set of 6-12 repetitions with loads ranging from approx. 70-85% of 1RM on 2-3 occasions per week, when taken near failure, can produce suboptimal yet significant improvements in squat and bench press 1RM strength in resistance-trained men. However, whilst the participants in this research were “trained”, their 1RM squats and benches were about 1.7x bodyweight and 1.1x bodyweight respectively, still far from elite strength athletes.
For preservation of absolute peak performance, exposure to near-maximal loads is almost certainly required, especially amongst the more advanced. Consider that between peaks, even when baseline training performance is improving, many more advanced strength athletes would still struggle to beat their best competition lifts without a brief acclimation or peaking period. However, as a prescription for general strength maintenance, a low-ish volume of high-effort work is likely sufficient.
Despite the recommendation of performing a single set on 2-3 occasions/wk, I don’t believe it necessary to have greater than one weekly exposure per lift/movement pattern for strength maintenance. A 2018 meta-analysis on the role of frequency in developing strength found that when volume was equated, no significant effect of training frequency was found. That is to say that performing 3 sets on one day, or one set on 3 days is likely not going to make a large difference to your strength maintenance outcomes. However, one or the other may be more convenient or appealing for you. On a similar line of thinking, that strength doesn’t appear to decay over 2-3 weeks of detraining (per McMaster) and that 3 sets per exercise performed once per week (per Bickel) was sufficient for maintenance suggests to me that maximizing training frequency is not likely important for pure maintenance of strength. During overloading training periods, arranging training across a week such that you can do more of it, perform and recover well, may be prudent. However, if that’s your current situation, then you don’t need to worry about not losing your gains.
I mentioned caveats, so to (re)iterate them – without access to near-maximal loads (>85%) maintaining absolute top-end strength will be very difficult. People who are older, and most likely those who are more highly trained, may require higher training doses to maintain strength in the long-term. A single exposure per week is likely sufficient provided you perform some appreciable volume of work as compared to your normal training doses, otherwise multiple small exposures may be necessary.
For hypertrophy, the picture is similarly rosy, really. Per the Bickel study, total volume by 1/3, performed at 1/3 the frequency of prior training, was sufficient to promote continued hypertrophy for the first 16 weeks of the 32wk detraining period amongst young men. Conceptually, as with the strength literature, this checks out. It’s well-established that whilst higher volumes promote greater hypertrophy, at least to a point, very modest volumes can still promote size gains. Similarly to strength, higher frequencies appear advantageous for hypertrophy only in as far as they allow more volume to be performed, meaning that until session volumes are high enough that the marginal returns on work performed are low or negative (estimates I discuss here, here and here, and expanded on here and here by Krieger, sit around 6-12 sets per workout), it’s reasonably unimportant how you distribute your work.
Whilst hypertrophy on a per-set basis appears similar across a range of intensities from approximately 40% upwards (1, 2), at the lower end of this spectrum training to, or very close to, failure appears more important. Where inability to perform specific exercises or use loads sufficient to maintain strength might limit you there, it oughtn’t limit you maintaining muscle size. However, the freedom to manipulate exercises to maximise demands on target muscles can also help you move up the relative intensity continuum a little. Consider using exercises that isolate muscles more effectively, especially unilaterally, to ensure that you truly do perform high-effort sets rather than pooping out from the burn of doing sets of 40+.
As mentioned earlier, in the Bickel study it appeared that dropping volume to 1/9th of prior training volume was eventually insufficient to maintain gains. It is probably therefore prudent to aim for at least one exposure including multiple near-failure sets per week. However, as I discuss in this article, per-set your training may be less stimulative at home, and perhaps slightly higher volumes are warranted again.
To sum up hypertrophy – similarly to strength, one exposure to multiple near-failure sets per week is likely enough to maintain muscle size (albeit perhaps not muscle fullness if you are accustomed to much higher volumes than that). However, if you are limited in your ability to stress muscles through full ranges of motion or train at higher relative intensities, it might be prudent to perform more sets per training day, and/or include a second, lower-volume exposure each week for each muscle group. That is to say that you can probably get away with doing just a few very hard sets per week, but if your circumstances and motivation permit it, doing a few more is not a bad idea. A rock-solid maintenance plan might include one day of 6-8 hard sets for a muscle, with another 3-4 hard sets performed at the other end of the week.
Retraining and muscle memory
Let’s say that you didn’t do too well on the muscle/strength maintenance front and lost a significant amount. How long does it take to get your gains back?
I’m only aware of a little research looking at this, but what there is seems reasonably positive. The previously-mentioned study by Staron followed 6 women who performed 20wks of lower limb training. They then detrained for 30-32 weeks, before recommencing training. After a 1 week preconditioning phase, and 6 weeks of retraining their 1RMs were similar to the end of the prior 20 week training intervention.
Even in quite extreme circumstances, retraining happens much more quickly then making gains initially. In a study by Campbell young men had one leg suspended (this involves locking the knee and ankle joints, placing the leg in a sling, and propping up the opposite foot such that the suspended leg takes no load) for 3 weeks. This led to losses in joint torque and quadriceps size that were completely reversed within 3 weeks of commencing a retraining program of leg extensions on 3 occasions/wk.
The interesting question is why this phenomenon of muscle memory occurs. There are two good answers that I’m aware of, and both likely occur.
The first is that greater numbers of myonuclei acquired during training persist with detraining and enable more rapid regrowth. For a quick primer – muscle cells are really big. Mine especially are enormous. Cells are made up of organelles (little structures that do specific jobs), one of which, the nucleus contains genetic material (DNA). When we need to make proteins (the molecules that do stuff in our cells) we transcribe that genetic information from the DNA and synthesise them from amino acids using the plans encoded in our DNA. Because muscle cells are really big (mine especially), we have multiple nuclei spread along each muscle cell to ensure that all of the cell regulation and protein synthesis that needs doing can be done.
Related to this, then, is the concept of a limited myonuclear domain, which is to say that a single nucleus can only oversee those functions for a given amount of cell volume. That has been proposed as a cap on muscle growth and only modest growth appears possible with their addition. Luckily, muscle damage from training appears to aid in the addition of myonuclei by recruiting them from satellite cells (1), so when we train we don’t just max out the capacity our current nuclei give us, we raise the ceiling. Evidence from rodent models suggests that hypertrophy from overload exercise is actually preceded by the addition of myonuclei, and that even under conditions as severe as denervation of a given muscle, they are retained in detraining.
Human evidence mirrors this, with myonuclei preserved for 15 years or more, also appearing to be recruited from satellite cells prior to growth occurring, with higher numbers of myonuclei facilitating faster growth.
That leads me to the muscle memory part. Because myonuclei appear to hang around even where muscle fibers shrink during detraining, when we recommence training all of the relevant cellular machinery is there and raring to go to transcribe genetic material and synthesise new muscle proteins.
The second explanation is more complicated physiologically and has to do with the DNA itself. Whilst we think of our DNA as just a string of code, and changes to our genes as being solely to do with that code itself, a whole host of adaptations can occur that alter gene expression without altering the genetic code itself.
These are called epigenetic changes, and there is evidence that resistance training can induce epigenetic changes that contribute to the phenomenon of muscle memory. In a paper by Seaborne 8 males (average age 27.6 years old) completed 7 weeks of 3d/wk resistance training, 7 weeks of unloading, and 7 weeks of retraining. Muscle biopsies were taken at baseline, after an initial acute session, after the first 7 weeks of training, after the 7 weeks of unloading, and after the 7 weeks of reloading. Without discussing the training results in depth, they followed a similar pattern to what I’ve already discussed.
Now, the sort-of-complicated bit. Methylation of DNA strands is an epigenetic modification. It involves whacking a carbon and 3 hydrogens onto a DNA molecule. This doesn’t alter the genetic sequence itself, but it does alter the ease with which the cellular machinery that binds, reads and transcribes DNA binds with that segment of it. Methylation typically represses gene expression. Hypomethylation involves the removal of those same carbons and hydrogens, and is like taking the brakes off gene expression – stuff that was previously less likely to be read and transcribed is now done so more readily.
The researchers found that a large number of sites were hypomethylated following loading as compared to baseline, and different gene clusters followed different temporal patterns when they looked at DNA methylation/hypomethylation patterns and subsequent gene expression. Two particular clusters are of interest to us.
One demonstrated exactly what you would expect of muscle memory. Genes were hypomethylated after loading (ie their activity was increased) and this hypomethylation persisted during unloading and subsequent reloading. This suggests that some epigenetic modifications to DNA induced by overloading resistance training persist and facilitate retraining. In some instances hypomethylation occurred to an even greater degree during the reloading phase.
A second cluster was methylated (repressed) at baseline and hypomethylated (increased activity) after loading. Upon unloading, the genes reverted to a methylated state, and were then hypomethylated again with reloading. When the authors looked at gene expression, the degree of enhancement of these genes was greater during reloading than the initial loading phase, suggesting that despite the more rapid reversion to a methylated state during unloading, adaptations allowing for the increased expression of these genes remained.
What does this all mean? Well, these explanations aren’t mutually exclusive. The concept of acquiring and keeping additional myonuclei facilitates greater rates of protein synthesis. However, the DNA housed within those nuclei may also have undergone epigenetic modifications making them more fit for producing the proteins that make you jacked, and also contributing to an increased training response upon retraining. If you’ve had some time off, there is evidence supporting the phenomenon of muscle memory from both an applied and physiological level that suggests that you’ll regain your gains far more quickly than you made them initially.
How to retrain
So how DO we retrain? In the study by Staron reviewed above, subjects underwent a light familiarization week, and then went straight back into training as normal. However, I think it’s prudent to start with conservative loads and volumes, and gradually work your way back to full training.
Firstly, high levels of muscle damage can impede muscle growth (1). Novel exercises and loading patterns are predictors of muscle damage, and so being awfully gung-ho in jumping back into training may cause you more pain than the associated gain. Instead, start with conservative volumes and RPEs, and slowly increase training difficulty. By the time you reach truly challenging loads again, you’ll have reacquired some of the protection against disruption of the repeated bout effect, and be in a better place to endure and benefit from it.
Whilst I didn’t mention it under muscle memory, it may be that breaks from training resensitise anabolic pathways such that you will grow more from less (there is a summary here). I’m not incredibly confident in this hypothesis – it’s just as likely that you are more sensitive because you are re-growing rather than growing anew. However, if you do commence training with conservative volumes, you may inadvertently discover that you didn’t need to be doing as much training as you previously did, or you may progress from conservative training enough to surpass where you previously were before you have to push training difficulty up to keep gaining, and if that occurs, that’s a great thing too.
Secondly, large spikes in acute to chronic workloads appear to predict injury risk (1, 2). For a detailed discussion of this, listen to this episode of my podcast with Jamie Smith of Melbourne Strength Culture. In essence, however, a gradual reintroduction of training stress reduces the chances of sustaining an injury. If you have come off of a long period of not training, the last thing that you want is to then be injured and forced to either abstain from or heavily modify your training unnecessarily.
So, what does a prudent return to training look like for a hypertrophy trainee coming off of a complete break? Here’s an example that you could apply for any given bodypart.
|Training week||Day 1||Day 2|
|Week 1||2 sets of 8 to 12 with 3-4 RIR||2 sets of 6 to 10 with 3-4 RIR|
|Week 2||3 sets of 8-12 with 3-4 RIR||3 sets of 6-10 with 3-4 RIR|
|Week 3||4 sets of 8 to 12 with 2-3 RIR||3 sets of 6-10 with 3-4 RIR|
|Week 4||5 sets of 8 to 12 with 2 RIR||3 sets of 6 to 10 with 3 RIR|
(2 exercises on day 1)
|4 sets of 5-8 with 2 RIR|
2 sets of 10-15 with 2 RIR
|3 sets of 6 to 10 with 2 RIR|
(2 exercises on day 1 and 2)
|4 sets of 5 to 8 with 2 RIR|
3 sets of 10-15 with 1-2 RIR
|3 sets of 6 to 10 with 2 RIR|
2 sets of 8 to 12 with 2 RIR
Of course, this is just an example, but it lands you somewhere sensible after 6 weeks.
For strength trainees, my advice is in concept very similar. You may not even need to progress volumes of your highest intensity work much at all, and simply let intensity creep up whilst doing some accessory work for hypertrophy. If you have effectively maintained muscle size, you’ll rapidly reacclimate to lifting heavy loads, and your performance will likely increase markedly for quite some time before you need to push volume up higher again to drive adaptation.
And that’s a wrap. To sum up
– Our training adaptations are reversible. If we detrain completely for long periods, we can expect that they’ll go away at accelerating rates. However, short breaks are totally fine.
– For both hypertrophy and strength, lowish volumes and frequencies are sufficient to maintain the majority of our adaptations. For strength trainees, without accessing truly heavy weights we can expect to lose some top-end strength, and for hypertrophy trainees it may be prudent to do higher than the minimum necessary volumes if you are training with limited loads/equipment, as your exercises are likely to be a little less stimulative.
– Upon retraining we seem to regain losses very quickly. This is facilitated by a number of physiological processes.
– If you’ve had some time off, it is probably in your best interests to reintroduce training slowly. It might actually enhance your gains in the short and long term, but it will nearly certainly reduce your risk of injury.
If you LIKED this article, join my mailing list – it’s FREE and I can send you stuff directly to your inbox. You can also follow me on Instagram, where I frequently post training and diet analysis and advice on my stories.
I also have a podcast, Weakly Weights, available on iTunes and Podbean, where I discuss training for powerlifting in-depth. You can join our mailing list, too, for free sample programs when we discuss them.
Finally, I have now made home workout templates available for sale. They’re designed to be used with any combination of bodyweight, bands, kettlebells/dumbbells and weighted backpacks. They individualise volume prescriptions and offer feedback on when to progress your training to get the most out of limited equipment. You can find them here.