Detailed Topic Index

Wednesday, January 30, 2019

Best metrics for tracking cold weather exercise - HR, Watts, Vent, O2 sat?

The choice of which physiologic metric is the "best" indicator of current effort or overall fitness is an interesting subject of debate.  For instance, one athlete may say that peak heart rate is a great index of intensity as well as state of conditioning.  In fact, in support of this, some fitness trackers use heart rate as such to calculate VO2 max or lactate thresholds. On the other hand, cyclists know quite well the value of power (watts), in demarcating training zones as well as comparing month to month overall physical conditioning status.  Another small group of enthusiasts like to use muscle O2 saturation change of the leg muscles for similar purposes.  However, which parameter would be most useful while doing an intense interval in the cold is another matter entirely.  A ride session in the recent cold wave has brought new insights into which metric is best (for me) and it's none of the above!

Some comments on using heart rate and/or power for fitness status and zone training.  Under ordinary circumstances these are very helpful parameters for effort tracking.  However as we saw in the post on cold weather exercise, the max heart rate may be several beats down, submax heart rate slightly higher and power (watts) diminished from multiple causes.  Even so, we should get some ballpark idea of our status.  However, what if both power and heart rate were reduced on a 3 minute VO2 peak interval?  This could be due to carbohydrate depletion, high RPE with central fatigue, over reaching or muscle strength loss.  Remember that over reaching generally is associated with lower power and heart rate vs fitness loss which is usually associated with a maintained max heart rate but lower power.  Exercise in the cold definitely alters these dynamics making conclusions difficult if not impossible.
Is there an available metric we can look at to say, "yes that interval was the best I could do under the circumstances from the cardiovascular standpoint"?  In other words, can we track some numeric value to help decide if we reached maximal intensity given the constraints of cold related changes.  I recently had a ride session on a very cold day, reviewed all the sensor tracings and will make a recommendation.

The following was from a difficult day in the cold, but rewarding from the physiologic data and understanding viewpoint.

The course and time of day were the same as in prior posts - my usual loop containing the 3 minute VO2 peak (I certainly did not hit max on this day) as well as the Wingate 60 second all out burst.  The ambient temperature was 40 degrees F but the wind chill temp was probably about 30 degrees.  

A word about wind chill
We have all heard of wind chill factors and such.  The wind does not actually make the item (person) colder but since the wind is taking away heat (by virtue of air to surface heat exchange), a certain amount of energy is lost.
Here is the conventional wind chill chart:
Riding at 20 mph in 40 degree F temp yields a "feels like" of 30 degrees.

But even more important is the heat loss in watts (a unit of work, just like in bike power):
Since adult males are about 2 meters squared, that means I'm losing about 2000 watts riding in that scenario.  But since I'm wearing clothing it's much less, but you get the idea.
The reason I'm bringing this up is that most lab hypothermia tests are not in a 20+ mph wind so direct comparison is difficult.  Clearly, on the road bike riding (or skiing) will introduce significant cooling and energy loss depending on your state of dress, wind exposure and surface area.

Back to my 3 min VO2 peak
I was careful to have plenty of carbs during the warmup as well as keeping power low to conserve stored energy for the intervals.  Subjectively I was feeling very well pre interval. At the finish I was quite winded and pretty exhausted (at the end of the 3 minutes).  I pedaled as hard as I could, and it was my best possible effort.
Here is the power tracing:
Power in red, Yellow is costal O2 sat, Blue is Moov HR
The first minute was a moderate fast start at 370 watts followed by only 260 watts for the next 2 minutes, with final average of 300 watts.  

This was way less than the usual 350-360 watts. 
Here is what it should have looked like on a warmer day at a similar average power (7/23/18).  Incidentally, this was not anywhere near a max effort, but I wanted to compare monitored values:
With Heart rate and Ventilation (7/23):

Comparing the two -
I was not surprised that VO2 peak was reduced in the cold after writing the previous post on cold weather exercise.  However, two metrics were odd.  The maximum heart rate was about 20 BPM lower than peak (and 10 BPM lower than the same interval power comparison in warm weather) and the costal O2 was 7% (was over 30% in the summer interval). 
If the poor power result in the cold was related to muscle power loss, fatigue, substrate deficiency (glucose) I would have expected low average power, lower heart rate but not a severely desaturated costal O2. That costal O2 desaturation is only seen with true maximal efforts, high work of breathing and cardiac output redistribution.  Under ordinary conditions, the costal O2 at a 300 watt average would have not dropped below 25-35%.

For instance here is another ride in very cold temperature early in the season (39F).  I was unable to even do a 3 min 300+ watt effort.  This is 250 watts for 5 minutes and costal O2 is clearly not dropping:

Here the limitation to a max effort may have been more related to glucose depletion or severe muscle weakness.  There does not appear to be any sign of cardiac output redistribution or high work of breathing.

Back to the 3 minute 300 watt interval at 40F:
The hexoskin data confirmed the Moov heart rate and showed near peak ventilation rates of 200-220 L/min:

So it looked like I certainly was performing at my maximal best effort for that set of conditions from the costal saturation drop and ventilation standpoints.  The only major discrepancy was the heart rate drop of 20 BPM. Since the cardiac output is stroke volume x heart rate, the constrained max heart rate significantly reduced the output and therefore the VO2 peak.  Less cardiac output also means that output redistribution will occur earlier, at a lower power than usually seen (which is what seems to have happened).

Another site measured was the calf using the Humon sensor.  Oxygen extraction was a bit less than usual:
Here is the 3 min x 300 watts at 40 degrees:
360 watts x 3 min at 60 degree temp:
Both pre interval saturation's were in the low 40s, but the higher wattage session dropped the calf a much greater amount (12 vs 25%), despite both being at my volitional maximum.  According to the literature, cold related systemic vasoconstiction should have reduced flow, enhancing extraction in the cold tracing.  Some reasons this did not occur could be related to less muscular effort or cold induced Hb curve shifting. 

Regardless, I don't think observation of the locomotor muscles will be helpful in deciding if the interval power was at our maximum capability under the given conditions.  There are many conflicting factors potentially leading to more or less desaturation.  In addition, if you are not able to fully desaturate in the cold, observation of this metric can't indicate if you have reached your full potential that day.

Wingate 60
Later on that ride, I did an all out Wingate 60 sec interval.  
Once again, the average was considerably reduced (about 60 watts), heart rate max about 12 BPM below usual, but costal O2 desaturation profoundly low at 3%.

Hexoskin data:
Peak ventilation was near max, especially at end interval.  Subjectively, it was a 10/10 perceived effort.  For the first time in doing that route over the years, I considered stopping to rest.

Calf O2 data:
Slightly better desaturation than on the 3 minute effort but it had warmed up a few degrees.
From a practical basis, not very helpful in deciding if I had any extra "gas in the tank", or if this was the best I could do.

So far the major physiologic discrepancy from normal temperature riding appears to be a markedly reduced heart rate (with a commensurate reduction in cardiac output).  Is there any literature addressing this?  We saw in the earlier post that max heart rate is only off by a few beats per minute.  However, as stated before, depending on how one sets up the test and hypothermic conditions, results could be quite different.  For example,
the conditions in this study, were the following:
Very different from a summer athlete wearing inadequate clothing.  One could argue that these subjects were simply doing their routine "thing" and did not undergo any novel hypothermic stress.

An older study done in 1979 used the following protocol:
Different core (esophageal) and muscle temperatures
were induced by swimming in cold water (l3-15°C) or
by submaximal bicycle ergometer exercise. The rate of
work (both in swimming and cycling) corresponded to
40-60s of each individual’s maximal oxygen uptake. The
duration of these exercise periods was 15-25 min. In three
of the subjects mean skin temperatures of 27 and 31°C
were obtained by setting the room temperature to either
+5 or +2O”C. After achieving a predetermined value for
body temperature 
Then the testing was done: 
the subjects moved to a bicycle ergometer
constructed for combined arm and leg exercise
(2). On this ergometer the subjects exercised at a rate of
work, which at control conditions (“normal” temperatures)
exhausted them within 5-8 min
. The rate of work
for each individual case was the same in
It then follows that power selected was a near VO2 peak  since this would be the max power over 3-4 minutes whereas they chose 5-8 min.  Coincidentally, this is very similar to my 300 watt power range.

What they found was the following:
With progressive reduction in body temp, the heart rate curve was shifted lower and the time of work was reduced.  The baseline heart rate in cold subjects was also lower.

 As temp fell, both heart rate, oxygen uptake and max work time were affected.  

At the greatest temp disparity, heart rate max was near 30 BPM lower than usual.  VO2 was less but ventilation was not markedly different.

From the paper:
The heart rate at 1 minute was reduced in all cold subjects to a similar degree, but this progressively deviated from the norm as the temperature was dropped.

And finally:

Bottom line from the study:
  • Huge drops in heart rate during intense exertion are possible with sufficient hypothermia.
  • Commensurate reduction in VO2 is seen.
  • Only minimal changes occur in ventilatory rates with hypothermia.
  • Work time is reduced. 
Back to my data:
What about the Deltoid and Biceps?
I did have sensors on each with the following data for the 3 minute 300 watt interval (at 40F as noted above):
Interestingly there was less than usual desaturation of each compared to a warmer day.  Biceps dropped to the 15% but is capable of 2-3% nadirs on a normal day at max effort.  Deltoid went from 73 to 51%, also less than usual for a max effort.  Could this have been related to the cold induced shift in the O2 dissociation curve?  I can't say, but my Fenix 5 watch was reading 57 degrees wrist temperature instead if 80 to 90 degrees (usual during warmer rides).  The watch was under the clothing layers and jacket, so the limb temp drop was real and not trivial.

Where does this leave us in regards to intense exercise in the cold?
It appears that attempting to do a maximum effort in the cold will be limited by a major reduction in cardiac output (heart rate) - with the following additional summary points.
  • Major reduction in heart rate
  • Less than normal locomotor muscle desaturation
  • About the same (or slightly less) ventilation rates
  • Similar patterns of deltoid and biceps desaturation but not reaching as low a value as usual on normal temp conditions
  • Similar pattern and values of costal O2 desaturation possibly related to both the factors of cardiac output redistribution and high ventilation rate.

Looking at this another way, let's go through each individual metric and make a value judgement in relation to reaching VO2 peak, training zone intensity and fatigue:
  • Heart rate monitor - poor, given the likelihood for substantial suppression of maximum values.
  • Power - poor, given the expected reduction in normal peak values and reduced time to exhaustion.
  • Locomotor muscle O2 saturation - poor, values may be the same or show less desaturation related to interaction of O2-Hb curve shifts, reduced muscle power, systemic vasoconstriction.
  • Biceps/Deltoid muscle O2 saturation - poor, same as above.
  • Costal O2 saturation - appears to be valid and similar to normal temperature conditions.  Potential reasons - could be acting as both an index of cardiac output redistribution at high loads as well as an index of high work of breathing (high ventilation rates).  Significant desaturation at this site can be a sign of impending decompensation and effort intensity.
  • Ventilation rates - appears to still be a valid marker of intensity zones as well as VO2 peak.  There may be a slight reduction form usual max values.

Several conclusions can then be inferred from the above, when exercising in the cold:
  • Heart rate values (by themselves) will not be indicative of training load, or attainment of VO2 peak.  Using them as such can be very misleading.
  • Locomotor muscle O2 desaturations will not be very informative regarding power zones or loading.
  • Deltoid and Biceps O2 saturation pattern may have some utility with regards to recovery from intense loads.  Persistent biceps desaturation (even at somewhat higher than usual values) may indicate a less than optimal state for another intense bout.
  • Ventilation rates do offer a reasonable measure of attaining near peak potential for that given time frame/temperature zone.  They appear to track more closely than heart rate for a given range of relative intensity.
  • Costal O2 desaturation appears to be a valid indicator of peak potential effort by combining both cardiac output redistribution factors as well as the effect of high ventilation rates (work of breathing increase).  Both the curve pattern (downward without plateau) and absolute values (near historic lows) can be used to help judge whether a given effort was at or near physiologic max.  Conversely, lack of costal O2 desaturation provides evidence that the effort was unlikely to have taxed the cardiovascular system to a significant degree.
  • Finally, during an intense interval, the presence of a subnormal heart rate in the face of normally high ventilation rates and/or significant costal O2 desaturation may indicate cold induced effects resulting in the impairment of VO2 peak, and the myriad issues of hypothermic exercise.

No comments:

Post a Comment