Do ice baths help or hurt your training?

Over the last few years, there's been lots of debate about the pros and cons of ice baths (and, more generally, of any technique that promises to speed up your post-workout recovery). The microscopic muscle damage and accompanying inflammation incurred by training, the thinking goes, is actually the stimulus that tells your body to adapt and get stronger. So if you use an ice bath to suppress that inflammation, you're actually suppressing your training responses.

This debate was kicked off way back in 2006 by a study by Motoi Yamane and his colleagues at Chukyo University in Japan, which found that strength and endurance gains were attenuated in subjects who took ice baths after training. There were some unusual aspects of the study, though – they tested handgrip strength in untrained subjects with a relatively light load, and the ice baths were as cold as 5°C (i.e. really cold), for as long as 40 minutes in total (i.e. really long). So there were some understandable questions about how applicable the results were to real-world sports situations.

Despite all the debate, there hasn't been much actual research testing this assumption. Probably the most interesting and real-world-applicable study came from Shona Halson and her colleagues at the Australian Institute of Sport last year. They looked at training adaptations in elite cyclists during three weeks of intense training with or without ice baths, and didn't see any major differences. If anything, the ice bath group did better. With that in mind, my general feeling has been that the training-suppression effects may exist in theory, but are unlikely to matter in practice.

Now, however, Yamane (now at Aichi Mizuho College) and his colleagues have published another study, in the International Journal of Sports Medicine, and it's an interesting one. They're basically following up on their previous work, but with some different parameters including a heavier load to induce greater training adaptations – and the results are quite striking.

This time, the protocol involved wrist strengthening, three times a week for six weeks. The training sessions were five sets of eight reps of wrist flexion, starting at 70% of one-rep max and moving up to 80% after three weeks. After the workouts, half of the 14 subjects cooled their arms in a 10°C ice bath for 20 minutes.

All of this is fairly impressive. There are also some other measurements; for example, wrist-flexor endurance (the number of reps they could do at 35 per cent of max) increased by 40 per cent in the control group, but only around 28% in the ice bath group. Blood measurements of interleukin-6 (a marker of inflammation) and vascular endothelial growth factor (a marker of blood-vessel adaptation) weren't significantly different, which means it's still not clear what explains the differences between the two groups.

So does this settle the debate and prove that ice baths are 'bad'? There are still some features of the study that make it hard to interpret. Wrist-flexor exercise is a fairly uncommon task, so there's no guarantee that more conventional forms of exercise involving bigger muscle groups would respond in the same way. The ice bath duration of 20 minutes is still quite a bit longer than most athletes use in practice (I think 10 minutes is pretty standard).

Perhaps most importantly, the balance between the advantages of quicker acute recovery versus the negatives of suppressed adaptation might be quite different in untrained volunteers compared to well-trained athletes. The untrained volunteers are getting the low-hanging fruit of quick adaptations without having to train too hard. Trained athletes are already well adapted, and have to train much harder to see any improvements; for them, ice baths might still pay off.

Still, the new study affirms that the original results from Yamane's group weren't just a fluke. I suspect (and hope) that this latest study will be followed by others that test more conventional exercise protocols with more typical ice bath recipes. At this point, it's clear that it's not entirely crazy to wonder whether ice baths might be counterproductive in some contexts.