Immediately after stumbling across the finish line of the 2019 Ultra-Trail du Mont-Blanc, the legendarily punishing mountain-ultra-trail event, 56 runners hobbled over to the National Ski and Mountaineering School in the French resort town of Chamonix. Waiting there was a team of scientists with a roomful of lab equipment to measure the precise physiological toll of their exertions. The problem: one of the tests required them to run on a treadmill for four minutes while their breathing was measured. Many of the super-fit ultrarunners couldn’t stay on the treadmill for that long, so the scientists altered the protocol on the fly and shortened the treadmill run to three minutes.
Running ultras is hard; so is studying them. Can you really get volunteers to run on a laboratory treadmill for 24 hours? Well, yes, you can—and Guillaume Millet, a researcher at the Université Jean Monnet Saint-Etienne in France (and himself an accomplished ultrarunner) has been there, done that, and published the paper. But you get more plentiful and realistic data by studying ultrarunners in the wild. So Millet and his colleagues recently published the sixth in a series of studies from that 2019 UTMB. Here are some of the insights from those studies about what it takes to run through the mountains for hours on end, and how your body responds to the challenge.
Long and Short
This year, about 10,000 runners will participate in seven different UTMB races over the course of a week in late August. The runners studied in 2019 ran one of two “short” distances (25 and 34 miles) or three “long” ones (62, 90, and 105 miles). The distance is just part of the story: the longest race, for example, also includes almost 33,000 feet of climbing and descending. The men’s winning time in 2019 was over 20 hours.
The range of distances provided a perfect opportunity to test how leg muscle fatigue responds to different durations of running. It’s pretty obvious that your quadriceps and calves will be weaker right after you run 100 miles. But it’s less clear where that fatigue originates. Is it your muscles? Your brain? The signaling pathway along your spinal cord that sends messages from the brain to the spinal cord? Millet and his colleagues put the runners through a series of neuromuscular strength tests before and after the races. There was a test of voluntary strength, but they also used magnetic stimulation of the brain and electric stimulation of the nerves to elicit involuntary muscle contractions, in order to tease out exactly where strength was lost.
After the long races, voluntary quad strength decreased by 38 percent, compared to just 27 percent in the short races. Part of the fatigue arose in the brain: even though the subjects were trying to push as hard as possible, the outgoing signal from their brains was smaller. The muscles themselves were also weaker: for a given level of electric stimulation, they produced less force. (The spinal cord played only a minor role.) In the quads, the difference between short and long races was explained by more muscle fatigue, rather than more brain fatigue.
Surprisingly, though, calf strength declined by 28 percent after both the short and long races: in this case, the extra distance didn’t seem to make a difference. When you compare these results to previous ultrarunning studies, a somewhat confusing picture emerges. Beyond a certain point—about 15 hours of racing, the data suggests—longer races don’t seem to make your muscles more tired. That may be because you can go faster in shorter races, and intensity is a key cause of fatigue—particularly if you’re hammering down quad-busting mountainsides. The scientific picture remains murky, but if you happen to run one of these races, you might want to adopt “After 15 hours, it won’t get any worse!” as an encouraging mantra.
Men and Women
Ultrarunning is one of the rare sports where top women sometimes beat top men—a feat that always prompts discussion about the physiological differences between sexes, and whether women have ultra-friendly characteristics that help them overcome the edge in muscle strength and red blood cell count that men get from testosterone. That’s a long and involved debate, but one of the hypotheses is that women’s muscles fatigue more slowly than men’s. It does appear that women have, on average, a higher proportion of endurance-linked slow-twitch muscle fibers, and better blood flow to feed those fibers.
In the UTMB data, women did indeed seem to show less muscle fatigue after the race. Here, for example, is the individual (dashed lines) and average (solid lines) data for quadriceps strength in men (blue) and women (red), before (PRE) and after (POST) the races:
The men were stronger before the race and stronger after the race—which makes sense because they had to propel bigger and heavier bodies through the mountains—but they had a bigger strength decline. This fits with previous research showing better muscular endurance in women.
There’s a twist, though. The researchers also asked each runner to rate their “competitive intentions” on a scale of 0 to 10, with 0 corresponding to “I tried to do the best time possible” and 10 corresponding to “Fun mode: my only goal was to finish the race.” Here’s what those scores looked like in the short and long races:
In this case, the men seemed to be more focused on their time, particularly in the short races—which, it turns out, is where the differences in muscle fatigue were most pronounced. This opens a different can of worms regarding potential sex differences in competitiveness. On one hand, this idea seems irrelevant to the question of why top women can compete with top men in ultra races, because the women who win races are clearly not in “fun mode.” On the other hand, women have repeatedly been shown to pace themselves better in endurance events, an observation that may be linked to overly competitive (or, to use the technical term, “stupid”) early pacing by men.
Millet’s new data can’t answer these questions, but it adds to the evidence that patterns of fatigue tend to be different in men and women. The elephant in the room, though, is participation rates. Only 257 of the 2,543 starters in 2019 were women. Until the numbers are more even, it’s risky to draw any general conclusions about sex differences.
Flat and Hilly
There have been numerous attempts to figure out which physiological traits predict how you’ll do in an ultra trail race. For ordinary road marathons, the three key parameters are VO2 max (the size of your aerobic engine), lactate threshold (which roughly tells you how much of your engine capacity you can use over a long period of time), and running economy (the efficiency of the engine). But those three factors are less useful in trail ultras: a study I wrote about a few years ago found that standard lab tests had decent predictive ability over 50K, less value over 80K, and no use at all over 160K.
Two of the things that make trail ultras so different are (as the name suggests) the terrain and the distance. It’s one thing to measure running economy on a treadmill in the lab. But how much does your running economy change when you’re climbing a steep hill? Or when your legs are rubberized by 20 hours of running? Millet and his colleagues explored both those questions: they tested running economy on a level treadmill, and also at an uphill gradient of 15 percent, before and immediately after the races.
Here again there was a counterintuitive result: running economy got worse (meaning the runners had to spend more energy to cover a given distance) after the short races, but not the long races. Previous research has shown that both intensity and duration of exercise can hurt running economy, but there seems to be a threshold where if you’re going slow enough, your running economy won’t suffer no matter how long you’re out there. In fact, an earlier study found that running economy actually improved after the 200-mile Tor des Géants race, perhaps because an ordeal that brutal trims any wasted motion from your stride.
As for the effect of slope, previous research has found that the most efficient runners on level ground aren’t necessarily the most efficient going uphill: running up mountains is a unique and specific skill. But the new data found that post-race changes in efficiency on level ground were strongly correlated with changes in uphill efficiency, which suggests that the underlying cause—mostly likely stride-altering fatigue in the muscles, rather than changes in your metabolism—affects your stride whatever the terrain.
For better or worse, none of this makes UTMB any easier. Millet even co-wrote a whole book called How to Succeed at UTMB (the English translation is unfortunately out of print), collecting the accumulated scientific research and practical wisdom from runners and coaches who specialize in mountain trail ultras. It’s a hefty read, and drives home the point that, from a physiological point of view, these races are not simply extra-long marathons. “It’s more complicated,” Millet told me at a conference a few years ago. “That’s probably why I like it so much: it’s more interesting.”
For more Sweat Science, join me on Twitter and Facebook, sign up for the email newsletter, and check out my book Endure: Mind, Body, and the Curiously Elastic Limits of Human Performance.