THANK YOU FOR VISITING SWEATSCIENCE.COM!
As of September 2017, new Sweat Science columns are being published at www.outsideonline.com/sweatscience. Check out my bestselling new book on the science of endurance, ENDURE: Mind, Body, and the Curiously Elastic Limits of Human Performance, published in February 2018 with a foreword by Malcolm Gladwell.
- Alex Hutchinson (@sweatscience)
***
Interesting new study in the September issue of Journal of Sports Sciences that Amby Burfoot recently tweeted about, which appears to show that running on grass causes more hemolysis — the rupture of red blood cells — than running on asphalt.
It’s been known for a long time that prolonged running causes hemolysis. In fact, most forms of prolonged exercise can cause some hemolysis, simply because blood is being pumped around so vigorously and exposed to high oxygen fluxes. But running is particularly susceptible because of the footstrike: the cells get squashed as they pass through the capillaries of the foot at the moment of impact. Is this a problem? Well, excessive hemolysis can play a role in iron insufficiency — but this is generally only a problem if there are other exacerbating factors like overtraining. In this case, hemolysis is mainly interesting not because it’s a serious problem, but because it can tell us something about how different surfaces affect footstrike impacts.
The new study, by researchers in India, is very simple. Ten runners ran for an hour at 60-70% max heart rate on grass, while another 10 did the same on asphalt. Blood samples were taken before and after. The researchers expected to find more hemolysis from the harder asphalt surface, but they instead found the opposite. So what’s going on? Although the grass was superficially smooth and even, they speculate that its underlying unevenness affected the runners’ strides:
Running on uneven and inconsistent surfaces like the beach or grass can cause more injuries because each step creates varying pressures and forces in the feet, ankles, knees and hips as runners most constantly adjust to the surface. These natural surfaces also tend to slope and create a dangerous off-centre force on the ankles and feet while running. Even though the grass surface appeared to deform relatively more than asphalt, it was assumed that the uneven nature could have led to inappropriate pressure distribution and impact forces on the foot, which could have resulted in an increased haemolysis in these runners.
As Burfoot points out, this is reminiscent of Benno Nigg’s ideas: whether a surface is hard or soft, your leg automatically adjusts to cushion the impact. But on surfaces where you’re unable to correctly predict exactly how your foot will land — i.e. grass — that automatic adjustment can’t take place, and that’s when strong uncompensated forces shoot up your legs.
So it all fits together, right? Well, I have one caveat. Here’s the data from the new study for the two key measures of hemolysis (unconjugated bilirubin and lactate dehydrogenase, for those keeping score at home):
As you can see, the pre-run differences between the two groups (which were assigned randomly) are greater than the change from pre- to post-run! So statistical analysis may suggest that the two groups responded differently, but for now I’d treat this finding very cautiously. This study really should have been conducted as a randomized crossover trial, so that each runner was measured under both conditions. Bottom line: great idea for an interesting study, but until the results are replicated somewhere else I’d take the conclusions with a grain of salt.
I am certainly no expert in statistics, but I wonder if the statistical basis for their conclusion is sound.
As I gather from the abstract and the graphs, the they performed a ranked tests on each group to see if:
1 Group A differed significantly before and after
2 Group B differed significantly before and after
They found that, for both markers, the statistic for group B was more significant than that of group A.
Perhaps I am wrong, but it I don’t think you can conclude from that hence there is a significant difference between group A and group B. The statistics tell us only if there is a significant within-group-difference. Moreover, the data are ranked, so they represent directional information. So, rougly put, the statistic tells us if the biggest differences in a group are mainly increases or mainly decreases. Comparison of the absolute value of the difference of group A and B is thus impossible.
Of course I see that the LDH shows no significant difference in A and a significant difference in B, so if LDH is a strong marker, this provides at least some clue. Again though, the ranking makes it difficult to say anything about the size of the difference between group A and group B. In theory it is possible that all members group B had a tiny change in their LDH, but all in the same direction, while group A had huge changes in every possible direction.
For clarity: I can see the different increase of UBR and LBH between groups A and B. What I was trying to say is that we cannot say on the basis of the statitstics whether the difference between A and B is significant.
@RH: I agree with your doubts. The more I read of the paper, the less I understood the basis for their claim that the differences between the two groups were significant. Here’s the relevant passage from the discussion:
“There were no significant differences (pre-post difference between groups) in the degree of haemolysis incurred between the two surfaces (unconjugated bilirubin: P=0.27, lactate dehydrogenase:P=0.20, haemoglobin: P=0.20, serum ferritin: P=0.63). However, athletes who ran on grass demonstrated an increase in the haematological parameters (unconjugated bilirubin: P<0.01, lactate dehydrogenase: P<0.05) when compared to athletes who ran on asphalt (unconjugated bilirubin: P<0.05, lactate dehydrogenase: P<0.241), indicating an increased haemolysis on grass (Figures 1 and 2)."
Figures 1 and 2 are the ones I included in the blog entry above. I simply don't understand what that passage means -- it seems to me to be making two contradictory statements. In the end, I decided it wasn't really worth trying to understand. From looking at the data (and number of subjects and the study design etc.), it was clear to me that, even if there's some theoretical statistical basis for concluding that these results are significant, the data simply doesn't pass the smell test. It's not necessarily wrong, but it doesn't convince me that it's necessarily right, no matter what p-values they claim.
An interesting post anyway. I really did’nt know I was smashing my blood to bits. I always thought it was the asphalt that suffers.
In a clash between human and asphalt, always bet on the asphalt!
(But for the record, I should reemphasize: this hemolysis is an inevitable part of running, and for that matter occurs with most sports. It’s generally only a problem for people whose iron status is already borderline. In this case, it’s most interesting as a way of probing the forces associated with footstrike on different surfaces.)
@RH We are a group of students analyzing this article in a subject called Biomedical Themes (second year of medical school, University of Bergen) and we have the exact same doubts. The entire study seems a bit hastily done – I would certainly not call this definite knowledge until it has been verified in a bigger study.