One of the abstracts featured yesterday examined weather conditions and concussion risk in college football. The APS article summarizes the study and its negative findings nicely, although the authors still feel that there may be a link under less ideal situations that NCAA games.
I was more interested in the contention on which they based their hypothesis, namely that a 2% loss of body fluid results in a 10% loss of cerebrospinal fluid (CSF) which plays a role in cushioning the brain from concussion.
In pediatrics, we see kids who have vomited and pooped out 2% of their body weight all the time. That kind of change in CSF volume should lead to screaming headaches, one would think, at least based on those of the folks who get a CSF leak following a spinal tap.
I dropped by the poster and got this reference:
Dickson JM, et al. Int J Sports Med 2005: 26: 481-485
My first peeve: they keep referring to their athlete subjects as "dehydrated." As a nephrologist (and a bit pedantic on this point), I can tell you these guys forced to exercise in a hot environment sweated out a mixture of salt and water. Normal sweat is hypotonic to body fluids, but sweating is not the pure "water" loss implied by the word "dehydration." I admit that the non-nephrologist world uses the term to imply extracellular volume depletion (EVD), but I will stick to the more technically correct term today.
So 6 subjects underwent MRI, then exercised for 90 minutes, losing 2.6% of their body mass by the end of the work-out. They then went back into the scanner. A number of urine and blood measurements were taken as well, although those measurements seem to be a mess. One subject went from a serum sodium of 136 to 201, which is higher than any serum sodium I have ever seen. His serum osmolality did rise, but only from 295 to 310.
These measurements make no sense.
Anyway, lets go back to the main point. They took pre- and post-EVD brain images and quantitated brain and ventricular volumes by point counting. They found that ventricular volume decreased by 10% on average, and that the change in ventricular volume correlated highly (r2=0.87) with the decrease in body mass. Let's look at their graph and see why this is crazy:
On the X axis we have the percentage of EVD, while the Y features the change in ventricular volume from baseline. That Y axis runs from -20% to +40%. So milder (2.1-2.2%) reductions in body weight showed decreased ventricular volumes while bigger changes in body weight showed increases?
I am not a neurophysiologist (thank any deity you care to here), but I have a hard time making these findings work.
Also, a subsequent study showed little or no changes in ventricular volume with EVD using more sensitive MRI techniques (Kempton MJ, et al. Human Brain Mapping 2009: 30:291-298).
Finally, I am unclear on how those ventricles protect from concussion. Ventricles are pockets of fluid within the brain that are easy to measure. Sure they may play a role in transmission of force, especially for contrecoup portions of the head injury. I would focus more on the layer of CSF that surrounds the brain. Of course, that is really hard to measure if it isn't enlarged.
Going back to the abstract, the authors found that most head injuries occur in moderate to warm temperatures. Of course, most college football games are played at those temperatures; however, even looking at the rate of concussions failed to show a relationship.
The bottom line: we can't blame the weather for head injuries at this time.