As a cycling enthusiast and resident of Florida, winter is generally a great time to train. The miserable summer heat and humidity are gone, which usually translates to an optimal environment to push exertional boundaries. However, on occasion, the temperature drops below some sort of threshold where performance can dramatically drop. Coupled with riding gear best suited to high temperature, minimal time to cold adapt, interesting hits to exercise metrics may occur. In this post we will cover some basic concepts of cold induced physiology change as well as some of my personal observations with the Hexoskin and multiple muscle O2 sensor data. In addition, some general health related concerns and cold temp compensation strategies will be outlined.
Cold induced exercise has many areas of physiologic effect that all lead to a decline in overall physical performance.
Impairments occur along the entire neuromuscular chain from the muscle contraction apparatus, neurologic reflex arc as well as the cardiovascular system.
At the local level, upon cooling, muscular force may reduce through several biochemical pathways. The end result is less maximum force. In addition, shivering thermogenesis (involuntary muscle twitching) can lead to reduced efficiency of the contraction/relaxation cycle of a given muscle. Shivering thermogenesis can also lead to depletion of glucose reserves, potentially causing early "bonking". For example, if in warm/cool weather you are able to go 1 hour before carbohydrate replenishing, this time benchmark may be reduced depending on how cold you become. Besides using carbohydrates, shivering can also use precious O2 reserves, thereby diverting this away from the intended exercise activity.
In terms of the cardiovascular system, some very dangerous changes can occur which could induce cardiac ischemia, arrhythmia and death.
A prime mechanism for many of these issues is activation of the sympathetic nervous system resulting in elevation of resting heart rate, blood pressure, and peripheral vasoconstriction. There can also be parasympathetic stimulation from the "diving reflex". This dual confluence is felt to be a factor in the high risk of fatal arrhythmia in cold weather exercise. In addition, if there is impairment of coronary vasodilatation from occlusive disease (plaque), the subject will be at higher risk of poor coronary perfusion leading to myocardial infarction when cardiac blood flow demand rises with intense exertion (from higher BP, HR, systemic vasoconstriction - see B below):
Apparently stoke volume is not markedly affected due to enhanced preload pressure from venous return. As we will see below, VO2 peak and max heart rates can alter by variable degrees depending on the study, protocol, duration and temperature.
Isolated effects on Muscle:
Both peak force, fatigue resistance and muscular efficiency (O2 consumption per unit of work) decline in low temp exercise according to a study using a frog muscle model.
Efficiency, O2 consumption:
The bottom line is that cold muscle loses significant advantageous exercise properties that can markedly effect performance.
VO2 max and heart rate during a ramp treadmill:
This study looked at VO2 max and heart rate during a treadmill ramp test. They found a reduced VO2 max, shorter test duration and slight decline in peak HR at low temp.
In addition, at a submaximal exertion (50% VO2 max), the O2 consumption was higher at each step of cooler temps:
This is particularly concerning since much of a training/racing session will not be done at max output. Given the higher degree of O2 usage at moderate power levels, one would expect a commensurate elevation of fuel usage. However, although fuel use does increase, the proportion of carbohydrates burned seems to stay the same:
The SHIV group was subjected to cold and shivering conditions, the LOW vs MOD refer to either low or moderate exercise intensity.
Back to Heart rate:
Max heart rate in the cold is either about the same or less than neutral conditions. But what about at submax power levels? Here is a comparison at 50% VO2 max:
It appears that as the temperature drops, heart rate rises during steady state mid level power.
However, there are other studies that do not show much in the way of heart rate change with temperature reduction:
Although HR did not really change, VO2 did rise with temp drop as many other publications have shown.
- The end result is that you may exhaust your carbohydrate reserves rapidly during exercise in the cold.
What about cold and wet conditions:
Adding the element of rain to the already cold air temperature places even more stress on an already marginal compensation process. This was shown in the following study.
The subjects jogged at 70% VO2 max speed in 5 degree temps with and without rain. Results showed a higher VO2, higher nor-epinephrine, higher lactate, higher ventilation rate in the rainy conditions:
VO2 (O2 consumption):
Lactate, Norepi levels:
Therefore, if one is exercising in the cold and subsequently gets wet, an additional physiologic burden occurs that should be considered. There appears to further effects on O2 consumption, stress hormone levels, lactate and ventilation rates.
Multiple sites of change can occur at the level of the muscle as noted above. From a recent review:
As far as muscle O2 via NIRS, a recent paper looked at the effects of hypothermia with and without supplemental inspired O2. They did find performance decreases with temp drop that were corrected by the subjects breathing a higher percent of oxygen.
I circled the cold exposed group who had lower power and longer time for completion than the control and hyperoxia group.
The fact that correction of the performance decline was possible with a higher O2 breathing mixture was a new and novel finding. It does make one wonder if some of the effect of low temp on performance is related the change in the O2-Hb dissociation curve. As an adaptation to exercise and muscle warming, hemoglobin affinity for O2 decreases, allowing more O2 offload to the active muscle. Unfortunately, the reverse is also theoretically true, in cold conditions the Hb-O2 affinity rises, making tissue O2 offload decline. Whether this is actually occurring in the athletes is unclear. To make matters more complex, cold associated vasoconstriction may also play a role in tissue O2 extraction as well as absolute levels.
The study did look at muscle O2 levels in the VL, cerebral O2:
There was a small but significant drop in TOI (muscle O2) in the VL as well as cerebral O2 in the low temp group. The addition of inhaled O2 did correct both sites back to the normal range.
However, the authors did comment that measuring O2 at this site/depth of the muscle may not be totally representative of muscle O2 kinetics:
Lower ambient temperature leads to significant changes in both circulatory and local neuromuscular physiology that is generally detrimental to performance. The decline in power was corrected by a breathing a gas mixture higher in O2. This also corrected the changes seen in muscle and cerebral O2.
This tracing was done on my typical 45 mile ride (1 hr modulating power warm up then 3 min max aerobic power/VO2 peak power interval, then returning home). It was 39F degrees and I was not dressed warmly. In addition, it was the first real cold spell for the area so I was not cold adapted at all.
Here is my attempt at a 3 min MAP, VO2 max interval that should have been at 350 watts:
The power was barely above 300w for 45 seconds then I was unable to sustain even that. Average power over the 5 minutes was 240w, less than my lactate threshold power (under normal circumstances). I did not have a sensor on the legs.
The costal, biceps O2 declined minimally, which is expected since power was not very high.
Slightly higher ventilation rates were noted (usually about 140-150 for MLSS).
Unfortunately the Hexoskin wasn't picking up heart rate well, but this is data from my Garmin Fenix 5 watch:
A 3 hour ride in slight rain at 50F after several weeks of cold adaptation (with warmer clothing):
- Baseline pre interval heart rate was higher.
- Steady state interval HR was similar to normal temp sessions.
- Ventilation rate was above what it would have been with normal temp.
- Power was markedly down. Either the muscles were unable to develop the intended force, neuromuscular coordination was disrupted, or by that time I had run low on carbohydrate stores (bonking).
- Mild changes only in non locomotor muscle O2 (costal, biceps) indicating minimal cardiac output redistribution.
This time the 3 min power was in the usual range.
The VL O2 desat was less than usual, but this appears due to a lower initial baseline. The net drop in VL was about the same as in warmer conditions.
Costal O2 drop was similar to standard conditions.
Heart rate and ventilation:
- Ventilation was in the same range as usual.
- Heart rate was lower (generally near high 160s to 171).
Wingate 60 - max 1 min power:
The temp had warmed to about 54%
- Peak and average power was slightly less than usual (505 vs 520w).
- VL O2 desat was similar to the prior 3 min VO2 max interval, with baseline higher than the prior interval (temp rose several degrees by that time).
- Baseline VL saturation continued to be lower in the cold (usually about 70% as opposed to 60% here. However, other sessions in warmer weather have had baselines as low as 60%.
- Costal O2 desat very similar to 3 min session.
Heart rate and ventilation:
During exertion, the change in percent desaturation was about the same as warmer conditions. The presence of the lower baseline VL saturation could be related to cold induced vasoconstriction or simply sensor placement variation. Although cardiac output redistribution did not appear to be any more pronounced, it was noted in some of the above studies that overall VO2 (per unit of effort) was increased. So it is likely that the lower baseline O2 saturation is related to multiple mechanisms.
In conclusion, potential mechanisms for these changes include:
- Peak heart rate slightly lower than with normal temps (167 vs 171).
- Peak ventilation about the same as normal and prior MAP interval.
- Decrease in peak strength.
- Higher proportional O2 use (less efficient).
- Diminished muscle endurance.
- Increase usage of both fatty acids and carbohydrate per unit of power.
- Slight decrease in muscle O2 of locomotor muscles.
- Possible role of altered O2-Hb affinity, making O2 offload decrease.
- Higher systemic vascular resistance via vasoconstriction.
- Higher risk of arrhythmia, heart attack.
- Higher heart rates at submax power.
- Lower peak heart rate.
- Lower VO2 peak/max.
- Higher VO2 (usage of oxygen) at submax power.
- Higher ventilation rates at submax power.
A few days later the temp warmed up into the 70's. I repeated the 3 min MAP interval. The average power was 365, max HR 170, ventilation 230 indicating excellent return of VO2 peak. Desaturations of the costal, VL, RF and calf were similar to the wet/cool weather conditions. It seems that at least by my data, NIRS does not show major alteration in desaturation patterns in locomotor muscles.
My guess is that there could be two opposing effects that cancel each other out:
On one hand there should be relative vasoconstriction from sympathetic stimulation (leading to lower muscle O2 from decreasing flow) as well as low temperature shifting of the O2 dissociation curve (leading to higher overall muscle O2 from higher Hb-O2 affinity).
- Depending on the degree of hypothermia, acclimation and occurrence of rain, the amount of performance impairment can vary widely. At extreme conditions, it may not even be possible to sustain usual intensity levels. Expectations of exercise power, endurance may need to be curtailed.
- Strategies to compensate for activity in the cold include adequate clothing (including the legs), higher consumption of carbohydrates during the activity, adjustments to power/heart rate thresholds used to modulate training zones.
- Consideration of cardiac risk factors before embarking on cold weather exercise - age, family history, lipid levels, prior history of diabetes, vascular disease, arrhythmia, hypertension etc.
And finally - There does appear to be an inverse U shaped curve as far as overall performance - low at extreme cold, extreme heat, but optimal in the comfortable temperature range.
Cold weather exercise part 2 - what sensor is best to rely on....
Some time ago I reviewed the Humon Hex muscle O2 sensor. There were several major issues making it a very poor choice for interested athletes. Recently I purchased another unit (Black Friday sale) and decided to give it another chance. Perhaps the most glaring problem in the original one I used was the extraordinarily weak Ant+ signal making it almost not usable.
The tracings below are some preliminary results from just 2 rides, with more to follow. I will be looking at 2 major parameters, signal strength (drop outs) and O2 saturation patterns.
The following figures were from an interval of 360 watts for 90 seconds then 240 watts for 45 seconds. A BSX sensor was on my L costal and the Hex was on the R costal area. I am sure they were not over the exact corresponding mirror areas, but were close. In addition, the Hex still needs a calibration step (according to tech support this is used to adjust the led brightness for different skin pigmentation). I picked the costal area because it is a great non locomotor site to follow, and the Moxy, in most hands does not work there. If the Hex does, that would be a major advantage.
Humon Hex on R costal
Power in red, Yellow is O2
BSX on the L costal
Power in red, Yellow is O2
Red is power, purple is THb
Red is power, purple is THb
- Both signals are solid with no drop outs.
- The nadir is lower with the BSX (10 vs 30%), but the baseline and recovery are very similar (within 3%).
- The overall pattern is similar but there are subtle differences (the green arrows).
- At this point I do not know why the desaturation patterns were different.
Deltoid (no BSX data, this is Humon Hex only)
This was shown on the last post, but from the standpoint of signal strength, expected data I show it again. In addition, no light shielding was used. The BSX would not have provided data without shielding:
Red is power, yellow is O2
- As with the O2 data, there are no drop outs or loss of signal.
- The general pattern is similar, with the start of cycling power there is a drop in THb.
- However, the Humon Hex has a gradual drop at the interval start, whereas the BSX abruptly declines then stays level until power declines (green vs purple arrows).
Same ride as above, but I simply elevated my arm (lateral raise) holding an empty water bottle:
Red is power, yellow is O2, purple is total Hb
- Good signal strength. The watch was on my R wrist but the sensor on my L deltoid. The signal needed to pass through a fair amount of tissue. Lack of light shield was not an issue.
- Consistent, reproducible results. Both nadir values were similar with intervals reaching VO2 peak.
- As expected the total Hb drops (external muscle compression) along with the O2.
- No signal loss.
Measurement of Hemoglobin saturation breakpoints - use as a fitness monitoring tool
Smart trainer usage in physiologic testing and interval training
- The Humon Hex now has excellent signal strength. No drop outs were seen!
- The Deltoid tracing was internally consistent, similar baseline and nadir values seen with VO2 peak intervals. A simple lateral raise shows the expected drop on O2 and total Hb.
- An unexpected bonus is the tolerance for outside light. No other shielding is needed. This is not a trivial point. Veteran muscle O2 users know all to well the problems with getting a good clean tracing free from light contamination.
- Usage for monitoring Costal O2 and THb is possible. Fit and stability were fine. Signal was free of drop outs. The overall curve shapes were similar to the BSX but not exactly so. I will continue to explore the Humon Hex as a costal O2 monitor but it seems very promising.
- Cost is much lower than Moxy. I was able to get one for just over $200 during the sale.
An individual's VO2 max/peak power or running speed, has been used as a measure of aerobic fitness as well as predictor of race times. Besides using gas exchange data during a lab ramp test, other modalities can give a rough idea of the VO2 max. A previous post looked at the average power produced over a 3-5 minute interval as one surrogate. Recently I came across a paper from a few years ago that also gives an alternate index of VO2 max status from a different type of testing procedure. This method is quick, simple and at least in my experience, very reproducible. The procedure is a 60 second Wingate test, or an all out 60 second effort. First we will review a published study, then look at some of my personal results with emphasis on O2 saturation of legs, arms, and costal muscles. Since I do have the Hexoskin, we also have a super accurate EKG plus ventilation data.
A word about the Wingate test. In the lab the test is done on a load controlled bike with a fixed resistance based on the subjects weight. However, one can also do something similar in the field. My procedure is to use a moderately steep hill with a constant grade, approaching it at about the same speed, gearing, heart and ventilation rates. I always use the same hill. Although this departs from the strict lab definition, it is a more realistic setting as far as performance on ones own bike, position, as well as pedal resistance is concerned.
In addition, the Wingate 30 second test is usually thought of as a measure of both alactacid and lactacid potentials, not aerobic physiology. However, others in the field have shown that at longer interval times, an index of aerobic capacity is possible.
The study aim was to compare the peak VO2 responses of an incremental ramp test to both a Wingate 45 and Wingate 60 second effort.
There theory was:
It was hypothesized that: (a) an increase in the duration ofThere were 3 tests, an incremental, a Wingate 45 and a Wingate 60 second.
Some basic results were:
Interestingly this comment was made:
From my personal perspective, I must agree. There are times when I can barely stay on the bike after an all out 1 minute. Because of this I have never wanted to try an all out 3 minute test which has also been used as an index of critical power.
How did they fare in regards to VO2 max, max HR, ventilation rates?
the Wingate test to 60 s would elicit peak VO2
values that were equivalent to that of an incremental stepwise test to voluntary
fatigue, and (b) the central and peripheral factors that determine
the peak VO2 would be significantly related between the Win60
and incremental exercise tests.
- The Wingate 60 VO2 max was very close to the incremental test which did verify the initial hypothesis.
- Peak heart rate was higher in the incremental test than the Wingate 60.
- Ventilation rate was higher in the Wingate 60 than incremental test.
- Oxygen pulse (a measure of stroke volume) was the same in Wingate 60 and incremental.
NIRS measurements of the VL:
Unfortunately, the area of interest is buried in the first part of the graph. It would have been nice to see the fluctuations during the Wingate 60 but we do have these comments:
In regards to total Hb:
- The O2 sat declined more rapidly in the Wingate tests than incremental.
- Wingate O2 rapid decline occurred in the first 35 seconds, then leveled off or slightly increased. Notice the mention of different patterns.
- During active recovery, O2 saturation rose rapidly and did overshoot baseline before returning toward normal.
There are some practical and more esoteric points made by this study.
First and most importantly:
Therefore, we can use the Wingate 60 as an index of the VO2 peak which is perhaps a better term than the max since the index is represented by a best effort. In other words performance on a Wingate 60 after a 3 hour intense training ride will be quite different than after a 20 minute warmup. Peak is not necessarily a maximal possible result.
- During the incremental ramp, the THb gradually rose, then leveled off or dropped slightly.
- In contrast, both Wingate tests showed a decrease in THb during the first 30-40s, then rose but not to baseline.
- During active recovery the THb returned toward baseline with a plateau in passive recovery.
The wingate 60:incremental test peak values were
- VO2 96%
- Heart rate 92%
- O2 pulse 99%
- Lactate 99%
- The O2 desaturation was more pronounced with the incremental test.
- Several potential reasons were given including changes muscle recruitment patterns, the Bohr effect.
- The THb abrupt decrease was seen only in the Wingate tests only.
- This has been discussed in this blog before - the application of maximal force by the leg muscles resembles that of weight training, resulting in high intramuscular pressure and hence flow reduction.
This point was of great interest to me. It was explored by the following:
- A higher Ventilation rate occurred during the Wingate vs incremental test.
- There was a greater degree of ventilation per unit of O2 consumed in the Wingate 60.
- Mention was made of chest muscle O2 monitoring to follow respiratory function effects.
- Recommendation was made to follow both locomotor as well as indirect muscles of exercise.
Final points made:
The following two intervals were done on the same day, O2 measurements on the VL, RF and costal areas. The first is a 4 minute slightly fast start done at MAP, with the second test an all out Wingate 60s.
The HR, Ventilation:
- "A valid measure of the peak VO2 can be attained during a Win60 test"
- "A greater proportion of the peak VO2 during the Win60 test was due to oxygen utilization by the respiratory muscles when compared to the incremental test because of the higher VE and VE/VO2 ratio during the Win60 test.
- VL desaturation was greater with incremental ramp than Wingate test.
Costal (yellow) and RF (blue) O2, red is power:
VL (yellow) only with total Hb in purple:
Compared to the Wingate 60 (same day, no sensor position change):
Heart rate and Ventilation:
Costal (yellow), RF (blue) O2, red is power:
VL (yellow) )2, total Hb (purple), power is red:
Examples of other muscle groups:
Although not done on the same day as above, I have some non locomotor muscle data during both a MAP and Wingate 60. Each pair was done on the same day, with both max HR and ventilation very similar (data not shown).
- The O2 desaturation of each muscle location was near identical.
- The pattern of decline and gradual up slope was similar in the VL, RF in both the 4 minute MAP and Wingate 60 trials.
- The pattern of straight downward slope was the same in the costal tracings.
- Heart rate and ventilation at the interval end was very similar.
MAP 3 minutes
Costal O2 (yellow), Biceps O2 (blue), power in red:
Costal O2 (yellow), Biceps O2 (blue), power in red:
The MAP 3min, Wingate 60 are on the same graph, max HR, ventilation rates were equivilant:
- Both Costal O2, biceps O2 have similar desaturation patterns and values in the MAP and Wingate 60 intervals.
- The biceps desaturation is severe and lasts well into the recovery portion.
- The pattern and degree of desaturation in the MAP, Wingate 60 intervals were virtually identical.
- Incidentally, the sensor employed here was the Humon Hex. It performed very well and an updated review is in the works.
- The maximal/peak aerobic power (MAP) also referred to as VO2 max/peak can be reached either by a conventional ramp or a maximal all out 60 second effort.
- The data presented here was the result of a much different protocol than the literature reviewed. The published study compared a ramp with a Wingate 60, whereas I am comparing a relatively steady 3-4 minute interval done at near peak aerobic power.
- According to my results (subject number=1), the peak/maximal aerobic power appears attainable with either the Wingate 60 or a 3-4 minute near max steady state effort.
- The steady state MAP interval yielded similar ventilation rates to the Wingate 60 as opposed to the ramp testing. Whether this correlates with higher percent of VO2 peak used by respiratory muscles as seen in the ramp comparison was not tested.
- However, the costal O2 nadir was remarkably similar in each test condition, perhaps indicating similar degrees of proportional VO2 usage for the respiratory muscles.
- The peak HR, ventilation rates were similar in both scenarios.
- The nadir O2 for the muscle groups tested were similar in both the Wingate 60 and MAP intervals.
- Patterns of O2 desaturation were similar in the Wingate 60 and MAP tests.
- The Wingate 60 appears to be a reasonable method for fitness monitoring, VO2 max power, and muscle desaturation nadir measurements. It's short nature makes it a more convenient modality than longer testing efforts.
VO2 max related posts