After a tough workout, my hand automatically reaches for ibuprofen. It's almost a reflex \u2013 my shoulder aches, my quad aches, I swallow a pill, and forget about it. A recent study on bioRxiv shows that this isn't a neutral habit for muscle recovery. And that a penny's worth of aspirin does the opposite in the same circumstances.<\/p>\n\n\n\n
Researchers from Canada and the US (Lu et al., April 2026) compared aspirin and non-aspirin NSAIDs \u2014 indomethacin, celecoxib, SC-236 \u2014 at two levels.<\/p>\n\n\n\n
In vitro<\/strong> myoblast precursor cells (C2C12), which would have fused into muscle fibres. Indomethacin and celecoxib at usual therapeutic concentrations notably inhibited myogenesis \u2013 fibre formation. Aspirin only inhibited it at unrealistically high concentrations.<\/p>\n\n\n\n In vivo<\/strong> Mice were subjected to acute muscle injury induced by barium chloride and then treated with either 3 mg\/kg\/day indomethacin or aspirin. Indomethacin:<\/p>\n\n\n\n Aspirin in the same dosage:<\/p>\n\n\n\n A separate experiment on transgenic mice lacking 12\/15-LOX (the enzyme that synthesises lipoxins and resolvins) showed that such animals exhibit an excessive and prolonged inflammatory response, confirming that the pro-resolving pathway is indeed important.<\/p>\n\n\n\n All NSAIDs block COX-2, the enzyme that produces inflammatory prostaglandins. This is what they're taught in med school. But aspirin does this in a special way.<\/p>\n\n\n\n Aspirin acetylates<\/strong> COX-2 irreversibly. Acetylated COX-2 does not die; it switches to a different job. Instead of prostaglandins, it starts to produce precursors for a class of molecules called aspirin-triggered specialised pro-resolving mediators (AT-SPMs) \u2013 aspirin-triggered lipoxins and resolvins.<\/p>\n\n\n\n These aren't \u201canti-inflammatory\u201d molecules in the usual sense. They are molecules Resolution<\/strong> Inflammation - they don't suppress it, they actively end it and initiate the recovery phase:<\/p>\n\n\n\n Other NSAIDs \u2013 ibuprofen, indomethacin, celecoxib \u2013 simply block COX-2 completely. Inflammation is suppressed, but along with it, the resolution pathway is also suppressed. The body has fewer signals to finish the job.<\/p>\n\n\n\n This study is on mice. A dose of 3 mg\/kg for a 70 kg human is approximately 210 mg of aspirin, which is half a standard tablet. This is below the typical analgesic dose (500 mg) and significantly above the cardiac dose (81 mg). It is too early to translate directly to humans.<\/p>\n\n\n\n But the direction is clear \u2014 and it aligns with what sports medicine already knows:<\/p>\n\n\n\n Ibuprofen after weight training is a bad idea.<\/strong> Roberts et al. (2017), Lilja et al. (2018) \u2014 separately showed that maximum doses of ibuprofen (1200 mg\/day) inhibit hypertrophy in humans. The same mechanism: blocking COX-2 = a weaker signal for adaptation.<\/p>\n\n\n\n Aspirin is another beast.<\/strong> Through acetylated COX-2, it doesn't switch off the signal. At low doses (\u2264200 mg), it may even help.<\/p>\n\n\n\n Paracetamol is neutral.<\/strong> It is not an NSAID, it does not block COX-2 in peripheral tissues. If an analgesic is needed on recovery days, paracetamol is the safest choice from an adaptation perspective.<\/p>\n\n\n\n This is not an argument against ibuprofen in general. In three situations, it remains useful:<\/p>\n\n\n\n What you shouldn't do is take ibuprofen preventatively \u201cso it doesn't hurt tomorrow\u201d or \u201cso you don't get DOMS\u201d. This reduces the adaptation signal without any real benefit.<\/p>\n\n\n\n A simple protocol based on what we know:<\/p>\n\n\n\n Again, research in mice. A direct clinical translation would require RCTs in humans, which likely nobody will fund (aspirin is an unpatented drug, so the pharma industry's interest is minimal). Therefore, we will likely be dealing with mechanistic confirmation from animal models plus indirect data from people who have taken aspirin for other reasons.<\/p>\n\n\n\n Aspirin has its side effects: gastrointestinal risks, anticoagulation. Before you take it regularly, you need to weigh this up. This particularly applies to people taking cardio-aspirin \u2013 extra doses are not necessary.<\/p>\n\n\n\n The most practical conclusion: The default post-workout painkiller is not ibuprofen<\/strong>. Better nothing. If anything, paracetamol.<\/p>\n\n\n\n\n
\n
Why do two similar NSAIDs work differently<\/h2>\n\n\n\n
\n
What does this imply for training?<\/h2>\n\n\n\n
When NSAIDs are actually needed<\/h2>\n\n\n\n
\n
What to do today<\/h2>\n\n\n\n
\n
Restrictions<\/h2>\n\n\n\n
\n\n\n\n