You've been training hard for the marathon, the bicycle race, the swim meet.  You've been pushing yourself.

And one day you wake up with pain. You've overdone it. This is the moment NSAIDs – nonsteroidal anti-inflammatory drugs – were made for.

With the Broad Street Run in the offing, a lot of people will be popping the pills. But should they? What are the risks?

To learn more, we spoke with Garret FitzGerald, professor of medicine and systems pharmacology and translational medicine – and an NSAID expert — at the University of Pennsylvania's Perelman School of Medicine.

He is also a veteran of 20 marathons. So, he said, "I can relate."

Let's review: What are NSAIDs and how do they work? 

Garret FitzGerald
Sabina Louise Pierce
Garret FitzGerald

Along with statins, they're among the most commonly consumed classes of drugs on the planet. Among NSAIDs, the most commonly used are ibuprofen, acetaminophen, celecoxib (available only by prescription) and naproxen.  The NSAID class also includes aspirin. However, this is usually consumed at low doses – less than 100 milligrams a day — that protect against second heart attacks or strokes, but it's not much use for the relief of pain. The other NSAIDs are given at doses that relieve pain and inflammation. They all act by depressing the formation of short-lived fats that act locally and are called prostaglandins.

So let's say you're training and you develop a chronic problem such as a bone spur or tendinitis.  Can an NSAID be helpful?

Yes,  it can. They're pretty effective as far as pain relief is concerned. That's obviously important more generally at the moment in the context of the opioid crisis.  NSAIDs are the only real alternative pain relievers that we've got. It's important that they're not addictive. But they are associated with side effects that, while they are infrequent, can be serious.

What side effect is the most worrisome? And what about a recent report linking ibuprofen with male infertility?

These NSAIDs can be associated with very serious gastrointestinal [GI] bleeds.  The incidence is rare, maybe 1 percent or 2 percent of people exposed. However, because so many people take NSAIDs, that can amount to a substantial public health problem. They may also cause cardiovascular problems that are similarly infrequent but potentially serious. In both cases, the longer you take an NSAID, and the higher the dose, the more likely a side effect.  So the general advice is to take the lowest dose necessary to relieve pain for the shortest period of time.

The potential for these side effects is more common as you get older. The odds go up by the decade, after age 50, although it is not a sharp curve.  Clearly, if you have preexisting GI or cardiovascular disease, the risk of that type of side effect goes up.

Short of those serious GI side effects, people often complain of less serious things: dyspepsia, abdominal discomfort. They are harder to define, but much more commonly reported.  The frequency seems also to be related to dose.

There are some other issues to consider for the weekend warrior. Might masking bony or tendinous pain with an NSAID and continuing to exercise set us up for further tissue damage? At a more fundamental level, might the NSAID interfere with the healing response? In rodents, NSAIDs may delay tendon healing when given early after a tear, yet improve things when given later in the healing process; there is a lot that we have yet to understand. As for humans, there is an absence of evidence on tendon healing, but that's not necessarily the evidence of absence.

[As for the infertility study} I thought the science was well-done and the paper was phrased in a very guarded way. It was an unexpected finding – that a regulatory process that controls sex hormone production was somewhat compromised by ibuprofen. Testosterone levels were not depressed, but compensatory mechanisms to sustain normal levels were called into play. What it means, if anything, in terms of risk to the procreative potential among young men is completely unknown. It is a provocative, carefully performed study, the clinical implications of which may amount to nothing, or something. It is just too early to say.

What about taking an NSAID during or after a marathon?

If you've run a marathon and you feel a bit achy and painy – which, let's face it, is a reasonable effect — the best approach is to take nothing, have a hot bath, relax, and feel good about yourself because you've finished the race. We shouldn't need to medicate the body's normal reaction to running 26 miles.

The more serious issue is if you have a preexisting problem, or one that occurs in the course of a marathon. Some time ago I had chronic knee pain when I was training before a New York City Marathon. Deciding to be smart, I ran the marathon packing some ibuprofen in my shorts.  At mile 16, my knee really hurt – even more than the rest of my body – so I took some ibuprofen.  At the end of the race, I felt a bit more terrible than usual. I went to the bathroom in the hotel and I had a big GI bleed.

The problem is, when you are racing, your stress hormones are turned on, as are many of your inflammatory signals.  This is likely to set you up for an adverse effect to an NSAID.  If I were running a marathon again, I would not be taking an NSAID at mile 16, or anywhere else along the course.

Are there any relevant differences between the most common NSAIDs? 

We do not have good comparative data on the relative effectiveness of celecoxib, ibuprofen and naproxen as pain medicines. On average, celecoxib seems more likely to cause cardiovascular problems, perhaps naproxen less so, in some people, at least. Ibuprofen may lie somewhere in between. However these events are rare and we have no information at the individual level; we can't predict if you are the one who will have a problem or respond well to the drug.

On the GI side, celecoxib is less likely to cause a problem, on average, naproxen a bit more, again with ibuprofen probably in between. However, we do not have really robust information and nothing at the individual level.

Acetaminophen is also an NSAID although, at the most commonly consumed daily dose (one gram), only about half as effective as the others. Thus, it is useful only for trivial pain but is less likely, at this dose, to cause GI or cardiovascular problems than the others. It's not much good for the patient with arthritis, but perhaps a good place to start for the occasional athlete. At very high doses, acetaminophen can damage the liver, perhaps by a different mechanism than suppressing prostaglandins.

[Editor's note: There has been debate about how acetaminophen works and how it should be classified. Dr. FitzGerald says it is an NSAID, based on the chemistry of how it reduces inflammation.]