Note: This article is from a previous issue of MASS, my monthly research review with Eric Helms and Mike Zourdos. Each issue of MASS contains 9 pieces of content. At least 6 (typically 7) are articles that each break down the findings and application of a recent study, 2 are videos covering broader topics, and we occasionally have guest posts or special features for the last article.
This article is a review of a 2017 study from Androulakis-Korakakis et al. titled The effects of exercise modality during additional ‘high-intensity interval training’ upon aerobic fitness and strength in powerlifting and strongman athletes.
- HIIT cycling and HIIT lifting (squats and deadlifts for high reps with short rest periods) seem to have similar effects on strength.
- HIIT cycling likely improves aerobic conditioning more than HIIT lifting does.
- HIIT cycling would probably be the more judicious choice for coaches and athletes, as it’s likely safer and it allows for greater flexibility when structuring a training week.
Cardiovascular training can have a host of benefits. The benefits that get the most press are health-related, for good reason: cardiorespiratory fitness is a strong predictor of all-cause mortality (2). However, cardiorespiratory fitness may have some benefits for strength athletes as well. Higher levels of cardiorespiratory fitness may influence how well you can recover between sets, and thus how much volume you can tolerate in a training session.
Unfortunately, it’s almost universally true that strength athletes don’t particularly like doing cardio. If you chose a sport that revolves around lifting heavy objects exactly one time, then it makes sense that cardio may not click with you. As such, many people wonder if they can get the benefits of cardiovascular training by simply training with high reps, or training with short rest periods. You’re going to be winded after a high rep set of squats or deadlifts, so high-rep training must be a good cardiovascular conditioning tool, the thinking goes.
It’s also been proposed that using high-rep work as your cardiovascular training may help mitigate the interference effect – the fact that people generally gain less muscle mass and strength when doing both strength and aerobic training versus doing exclusively strength training (3).
This study set out to test these ideas. Over 8 weeks, a group of competitive powerlifters and strongmen did two sessions of high intensity interval training (HIIT) per week, using either squats and deadlifts with 60% of 1RM or cycle sprints as the exercise stimulus. Predicted VO2max (a measure of cardiorespiratory conditioning) and predicted 1RM knee extension improved in both groups. Strength increases were similar between groups, but the group performing cycle sprints improved their predicted VO2max more, indicating that cycle sprints are likely a more effective conditioning tool than high-rep squats and deadlifts.
Purpose and Research Questions
There were two primary research questions:
- Would high-rep squats and deadlifts mitigate the interference effect relative to cycle sprints (or even provide an additional positive stimulus for strength adaptations), allowing for larger strength gains?
- Would high-rep squats and deadlifts allow for similar increases in cardiorespiratory fitness relative to cycle sprints?
The authors hypothesized that both forms of HIIT would cause similar aerobic and strength adaptations.
Subjects and Methods
The subjects were 16 males with at least 2 years of training experience, ranging from 21 to 28 years old. All were competitive strength athletes, with a mix of powerlifters and strongmen included. They were randomized into two groups of eight subjects apiece. Two people in each group reported performing some form of aerobic training in addition to their strength work, while the other six subjects in each group didn’t do any additional aerobic training. All athletes reported that they were in a hypertrophy or general strength phase, and they reported training 2-5 times per week with 60-85% 1RM loads to a 7-8.5 RPE. Since they reported RPE with a decimal, that means they were likely using a reps in reserve-based RPE scale, so a 7-8.5 RPE would mean they generally pushed each set until they had 1.5-3 reps left in the tank.
The YMCA 3-minute step test was used to estimate VO2max. The test works by having the participants do 12-inch step-ups for three minutes, with 24 step-ups per minute at a pace dictated by a metronome. After the test, heart rate is measured for one minute. VO2max is then predicted by using the equation (for men): 111.33 – (0.42 x heart rate in beats per minute). This prediction has been shown to correlate well with measured VO2max (r = 0.83) in the general population (4).
Strength was assessed via predicted 1RM knee extension. Knee extensions were chosen instead of squats or deadlifts since the participants had experience with the exercise, but none were using it in their training at the time. The participants’ strength training wasn’t controlled, so the researchers wanted to ensure that the strength assessment wasn’t overly affected by the participants’ different training programs (i.e. strength in the squat may change considerably in 8 weeks if switching from low to high intensities or high to low intensities due to specificity, but actual knee extension strength wouldn’t fluctuate as much). All participants could do at least two reps with the entire weight stack of the knee extension machine, so predicted 1RM based on Dohoney’s formula (5) was used as a strength measure instead of 1RM. The equation used was 82.07 + (0.76 x weight) + (5.66 x reps), and its predicted 1RMs have shown to correlate well (r = 0.82) with actual 1RM knee extension strength.
One group was labeled Aerobic Mode (AM), and the other was labeled Strength Mode (SM).
The AM group performed HIIT twice per week using cycle sprints. Each session started with a 5-minute warm-up at a 5-6 RPE on a modified 0-10 Borg scale (with 0 representing no effort, and 10 representing maximal effort), followed by 7 rounds of 30-second sprints at 8-9 RPE interspersed with 90 seconds of pedaling at 5-6 RPE. The final sprint was followed by a 5-minute cooldown at 4-5 RPE.
The SM group performed HIIT twice per week using squats on one day and deadlifts on the other day. They were allowed to use whatever stance width and bar position they wanted for squats as long as they were capable of squatting to legal powerlifting depth (the anterior surface of the thigh at the hip passed below the top of the knee), and they were allowed to use whichever deadlift stance they preferred. They warmed up to 60% of 1RM, and did 7 sets of as many reps as it took to reach an 8-9 RPE (i.e. until they had 1-2 reps left in the tank), interspersed with 90 seconds of passive rest. Each set took 16-30 seconds on average, and the participants performed 8-15 reps per set with a fast rep cadence (about one second for both the eccentric and concentric). After the seventh set, they walked on a treadmill at a 4-5 RPE pace for 5 minutes to cool down.
As previously mentioned, strength training wasn’t controlled. The researchers ran into a common problem when attempting to study competitive athletes: The athletes wouldn’t agree to take part in the study if it meant that the researchers were in control of their strength training for two months. As such, this study loses a quite a bit of face validity since a major variable wasn’t controlled, but it gains a little ecological validity, as it examined the effects of HIIT added to actual “real-world” training programs. It would have been better if the researcher persisted with recruitment until they could get a batch of lifters who were willing to follow a standardized strength training protocol as well, but as it is, this is the study we’re working with.
Pre-training, the AM group was significantly lighter (86±7kg versus 100±kg), and the SM group had a significantly higher predicted VO2max (71.1±2.7 mL⋅kg-1⋅min-1 versus 66.5±3.2 mL⋅kg-1⋅min-1). There were no other significant pre-training differences.
Predicted 1RM knee extension strength increased to a similar degree in both groups: 7.2±3.7kg for the AM group, and 6.6±3.5kg for the SM group. Both of these corresponded to moderate effect sizes of 0.62 for the AM group, and 0.57 for the SM group. The between-group effect size was trivial: 0.17.
Both groups also had increases in predicted VO2max: 5.3±2 mL⋅kg-1⋅min–1 for the AM group, and 2.8±1.4 mL⋅kg-1⋅min-1 for the SM group, corresponding to large within-group effect sizes of 1.79 and 0.95, respectively. There was a significant difference between groups (p<0.05), with VO2max increasing more in the AM group, corresponding to a large between-group effect size of 1.45 (6).
There were a few things I really liked about this study. The first is that it was “real world” science. It’s the type of study any coach or personal trainer could run on their clients, as it was carried out entirely in a normal gym with readily-available equipment: a barbell, a box for step ups, and a knee extension machine. It was also ecologically valid, meaning it looked like the type of training people would do in the real world. When most strength athletes add HIIT into their training, they don’t proactively make huge changes to the rest of their program to accommodate it (perhaps unwisely) in my experience – they just add it on top of their other training. With that in mind, I really like the decision to measure strength with the knee extension machine instead of squats or deadlifts. It’s an exercise that’s easy to perform, one that wouldn’t be unduly influenced by the lifters’ varied training programs (since none of them included knee extensions in their training), and something that is obviously a valid measure of quad strength (which one would expect to be affected in some way by HIIT squats or cycle sprints).
However, there were also several shortcomings. The biggest shortcoming was, obviously, that the resistance training wasn’t controlled. The other notable shortcoming was that the main outcomes (knee extension strength and VO2max) were predictions, not measurements. I don’t think this is a huge issue with knee extension strength, as the predicted changes in 1RM were simply based on a rep-max equation; even if the predicted 1RMs weren’t accurate, you do still know that the participants could do more reps with the same weight, indicating increased strength, increased strength endurance, or both. However, using prediction equations pretty clearly caused issues with assessing aerobic fitness in this population.
To explain why this is a problem, you first need to understand a little bit about aerobic physiology. Your body uses the oxygen you breathe in the final step of aerobic metabolism. When you’re exercising harder, oxygen needs increase. As such, measuring the total amount of oxygen your body can use, relative to your body size, tells you how good of shape you’re in by telling you how much energy your body can produce in a given period of time to power your exercise. That measurement is called VO2max, and it’s usually expressed in milliliters of oxygen per kilogram of body mass per minute (mL⋅kg-1⋅min-1). Though VO2max is one of the core measurements in exercise science focusing on aerobic exercise, it’s not something you run across very often in the strength training literature, so let me give you some context for these measurements.
I think you see the problem here. Pre-training, both groups had predicted VO2maxes in the high 60s or low 70s, and both groups had predicted VO2maxes in the 70s post-training. Quite frankly, there’s just no way that’s accurate. Previous research (7) indicates that strength athletes generally have measured VO2maxes in the neighborhood of 50mL⋅kg-1⋅min-1. Either the researchers in this study found a group of powerlifters and strongmen with a level of aerobic fitness that would make most marathon runners jealous (even though 75% of the participants reported doing no dedicated cardiovascular training), or the regression equation used to predict VO2max based on heart rate after a 3-minute step test doesn’t work well for strength athletes. I think the latter option is the more likely one. It would be interesting for future research to investigate why the step test over-predicts VO2max for strength athletes – this may give further insights into the mechanisms by which strength training improves exercise economy in endurance athletes (8).
One final point about the VO2max predictions before moving on: Even though they clearly weren’t accurate (valid) for this population, it’s likely they were still sufficiently precise (reliable) to tell us something about change in aerobic fitness. Accuracy tells you how close a measurement is to the true value it’s supposed to measure (known as validity in research). Precision tells you how closely measurements cluster and whether they track reliably with changes in the true value they’re supposed to measure (known as reliability in research). For example, if you weigh 80kg and your scale gives you readings between 79.7kg and 80.3kg, it’s accurate but not incredibly precise. If you weigh 80kg and it says you weigh 76kg on the dot every time, and if you gain 2kg it says you weigh 78kg on the dot every time, its accuracy is lower, but its precision is higher; the actual weights are wrong, but the change in weight is correct and the measurement doesn’t fluctuate. With good measurement tools, accuracy and precision can both be very high; that’s the goal in science. However, even if the accuracy of a measurement leaves something to be desired (as it does in this case), as long as the measurement is reasonably precise, you still have a good idea about change. Previous research shows that the step test is reliable (i.e. precise; 4).
With all of that in mind, here are the tentative takeaways from this study:
- Both HIIT approaches likely affected strength similarly. Since strength training wasn’t controlled, it’s possible that the reported strength gains were influenced by average differences in training programs between groups. However, testing strength using an exercise that wasn’t included in any of the lifters’ programs probably mitigated that issue. Since there wasn’t a group that only performed strength training for comparison, it’s impossible to say whether the two HIIT protocols helped with strength development, hindered strength development, or had no effect on strength development. However, we can say that whatever effect the HIIT training had on strength, it was likely similar for both groups.
While HIIT cycling and HIIT squats and deadlifts can both increase aerobic fitness in strength athletes, HIIT cycling is likely more effective at doing so. I’m hedging this takeaway a little bit for four reasons:
- It’s possible that the reliability (precision) of VO2max predictions using the step test is also bad in strength athletes; after all, the accuracy of the predictions was likely terrible in this study, while the same prediction equation has been shown to be quite accurate in research on the general population.
- It’s possible that one group’s strength training was structured in such a way as to allow for more vigorous HIIT training. For example, if one group systematically had hard squat workouts before one of the HIIT training sessions, they may not have been able to put adequate effort into their HIIT sessions. However, I think this is unlikely. Smaller increases in predicted VO2max were seen in the SM group, and I really don’t think those subjects would be crushing a high volume squat session knowing they had high rep HIIT squats or deadlifts the next day.
- It’s possible that the larger increase in predicted VO2max in the AM group was due to novelty; obviously the participants had experience with squats and deadlifts, but weren’t as experienced with cycling. However, I think that’s an unlikely explanation due to the way VO2 was predicted. “Training for the test” can increase measured VO2 (i.e. cycling and running may cause the same “true” increase in VO2max, but the measured increase may be larger if your training revolves around running instead of cycling, and you measure VO2 on the treadmill), but the exercise used for testing in this study was step-ups, which weren’t trained by either of the groups. I’m not aware of any literature showing that novel exercises inherently increase VO2max independent of metabolic demands (i.e. a marathon runner probably won’t experience a boost in VO2max by switching to swimming or cycling for training).
- Predicted VO2max was higher pre-training in the SM group anyways, so it’s possible that they experienced smaller gains in aerobic fitness because they simply didn’t have as much room to improve. A different statistical test that accounts for pre-training group differences, such an ANCOVA, would have been more appropriate for analyzing the group differences here instead of the t-tests that were used. However, given the magnitude of difference between groups (the AM group’s predicted VO2max increased almost twice as much as the SM group’s did; p=0.012 in the t-test and ES=1.45), it’s still very likely that HIIT cycling was more effective at improving aerobic fitness.
Taken in totality, and viewing these data through the lens of an athlete and coach, this study suggests that strength athletes would likely do better by adding HIIT cycling to their training rather than high rep, short rest squats and deadlifts if the goal is to improve cardiovascular fitness. Not only did the AM group have larger increases in predicted VO2max, but they also had similar strength increases. There was a measurable advantage for HIIT cycling, but not HIIT squats and deadlifts.
On a more practical level, if you put yourself in the athletes’ shoes here, the HIIT cycling affords you much more flexibility in your training than the HIIT squats and deadlifts. HIIT cycling is hard, and the first time you do it, you may be a little sore the next day, but you’ll adapt to it quickly. It may hinder recovery from squatting or deadlifting a bit, but the impact will likely be manageable (especially if you’re not either very lean or in a big calorie deficit). However, high rep squats or deadlifts with 60% of 1RM and short rest between sets is always going to be brutally hard. This study didn’t collect data on how the athletes needed to alter their strength training to accommodate the added HIIT, but I’m very confident that the SM group would have (or at least should have) made larger adjustments to their squat and deadlift training in the days following their HIIT workouts to give themselves adequate time to recover, likely reducing the amount of heavier training they could manage. In essence, getting your HIIT work from squats or deadlifts gives you smaller aerobic benefits and necessitates a lower average weekly training intensity, decreasing the amount of highly specific strength work you can manage.
Safety also needs to be considered. Fatigue when cycling is perfectly safe. Specifically chasing metabolic fatigue during high rep squats and deadlifts, on the other hand, increases the risk of technique breakdown and putting your body in compromised positions that could increase injury risk.
Future studies should replicate this study, but with tighter controls (putting all participants on a standardized strength training program), more accurate measurements (VO2max via spirometry, and knee extension strength using a machine that could tolerate higher loads, or simply 1RM squat and deadlift strength, which would be valid if the strength training was controlled), and a control group only performing strength training. Furthermore, it would be interesting to see other weighted exercises tested that are more commonly used for conditioning (i.e. kettlebell swings or sled pushes).
Application and Takeaways
If you’re a strength athlete adding HIIT to your program, this study indicates that cycling may be a more judicious choice than high rep, short rest squats and deadlifts. The additional squats and deadlifts didn’t provide an advantage for strength development, and the HIIT cycling led to larger increases in aerobic fitness. Moreover, from a practical perspective, it’s easier to modify a training program to accommodate HIIT cycling, and HIIT cycling is likely the safer option as well.
Get 9 articles like this every month
If you enjoyed this article, consider subscribing to the MASS research review. We publish 9 research review articles – just like this one – every month. You can get an 80-page free issue (Value: $29) or go ahead and subscribe.
- Androulakis-Korakakis P, Langdown L, Lewis A, Fisher J, Gentil P, Paoli A, Steele J. The effects of exercise modality during additional ‘high-intensity interval training’ upon aerobic fitness and strength in powerlifting and strongman athletes. J Strength Cond Res. 2017 Apr 21. doi: 10.1519/JSC.0000000000001809.
- Kodama S, Saito K, Tanaka S, Maki M, Yachi Y, Asumi M, Sugawara A, Totsuka K, Shimano H, Ohashi Y, Yamada N, Sone H. Cardiorespiratory fitness as a quantitative predictor of all-cause mortality and cardiovascular events in healthy men and women: a meta-analysis. JAMA. 2009 May 20;301(19):2024-35. doi: 10.1001/jama.2009.681.
- Wilson JM, Marin PJ, Rhea MR, Wilson SM, Loenneke JP, Anderson JC. Concurrent training: a meta-analysis examining interference of aerobic and resistance exercises. J Strength Cond Res. 2012 Aug;26(8):2293-307. doi: 10.1519/JSC.0b013e31823a3e2d.
- Beutner F, Ubrich R, Zachariae S, Engel C, Sandri M, Teren A, Gielen S. Validation of a brief step-test protocol for estimation of peak oxygen uptake. Eur J Prev Cardiol. 2015 Apr;22(4):503-12. doi: 10.1177/2047487314533216.
- Dohoney P, Chomiak J, Lemire D, Kovacs C. Prediction of one repetition maximum (1-RM) strength from a 4-6 RM and 7-10 RM submaximal strength test in healthy young adult males. Journal Of Exercise Physiology Online. 2002 Aug;5(3):54-59.
- If you read the full text of the study, you’ll notice that the effect sizes reported in this write-up don’t match the effect sizes reported in the study. There’s a simple reason for that: they didn’t use standard effect size calculations. They divided the average increases by the standard deviations of the increases instead of dividing the average increases by the pre-training standard deviations for their given measures. The effect sizes they reported tell you about the variability of the change, not the magnitude of the change relative to its variability. Fortunately, the authors reported enough data to that I could do the correct calculations myself.
- Häkkinen K, Alén M, Komi PV. Neuromuscular, anaerobic, and aerobic performance characteristics of elite power athletes. Eur J Appl Physiol Occup Physiol. 1984;53(2):97-105.
- Støren O, Helgerud J, Støa EM, Hoff J. Maximal strength training improves running economy in distance runners. Med Sci Sports Exerc. 2008 Jun;40(6):1087-92. doi: 10.1249/MSS.0b013e318168da2f.