Thursday, September 26, 2019

Training intensity guidelines from HRV and NIRS data

Training intensity distribution continues to be a hot topic for both athletic and basic fitness exercise.  In order to train within these guidelines, one needs to have a handle on two basic concepts, the maximum lactate steady state and the first ventilatory threshold.  Both have been explored in detail in prior posts.  Over the past I have been asked by several athletes to help estimate their upper zone 1 limits (of a 3 zone model) and MLSS (beginning of zone 3).  This post will be a review and step by step guide as to how I do it and will show examples of cycling, running and skiing.

Many believe the first ventilatory threshold is the upper boundary of the zone of low intensity exercise, which should make up the bulk of ones endurance training.  As noted in my VO2 max study, the VT1 may not correspond exactly to the LT1, making reliance on the LT1 problematic. Whether or not VT1 and LT1 are different physiologic concepts, or just the same process, but presenting as different values due to methodology, is still unclear to me.  There are proponents to both theories.  Regardless, I would like to formally propose two "simple" procedures to obtain the upper limit of zone 1 (easy) exercise as well as an approximation of the MLSS, as the beginning of zone 3 (high intensity) training.  Exercise intensity in between (zone 2) should probably be limited, but can be defined as above VT1 and below MLSS.

For simplicity, I'm going to assume for this post that you have only a chest belt HRM and an O2 sensor, so no formal lactate measurements are needed.  

We will start at low intensity training and work our way up.....

Zone 1 upper limit estimation using DFA a1:
The DFA a1 is a measure of short term cardiac interbeat fractal complexity.  While at rest or during very light exertion, the DFA a1 will have values between .8 and 1.2.  According to the excellent review by Gronwald et al, as intensity rises, the DFA a1 will drop, passing the .5 area (uncorrelated white noise) during substantial efforts.  With very high intensity, values reach even lower levels.  However, for our purposes, we want to see at what power or heart rate DFA a1 begins to approach the white noise value of .5.  In my VO2 max study, VT1 occurred just before DFA a1 white noise values were reached.  In my experience, a DFA a1 value of .7 would represent a good limiting value for easy sessions.  This represents a correlated RR interval series, low physiologic stress and probably an intensity state below VT1.

Requirements: 
Polar H10 or similar, Hexoskin even better (less artifact).  Make sure the belt is snug and use conductive cream.

How to: 
In order to get an idea of the work rate associated with uncorrelated DFA a1, a progressive ramp of some type is needed.  This can be running, cycling or even skiing.  Having a measure of power is helpful in keeping the intensity as a constant.  If your power is fluctuating markedly, that would make interpretation difficult.  It is not necessary to reach exhaustion, going up to LT2 is more than enough.  Here is an example of a ramp that I did on an indoor trainer:
Each interval was 5 minutes long, the heart rate and DFA a1 was taken from the last 2 minutes of each section.  Watts were the average of the entire 5 minute block.  There was about a 30 watt rise per segment.  For a guide to using Kubios software in this regard see this, about half way down the page.

DFA a1 and heart rate:



DFA A1 and power:

The heart rate and power associated with a DFA a1 of .7 are 128 bpm and 175 watts.  According to the gas exchange test, my VT1 was at about 172 watts, close enough to the above.  
Why did I pick a DFA a1 of .7?  This seems to represent the point just prior to poorly correlated values and as an analogy, is the edge of the intensity cliff.  A small increase in effort above this leads to a white noise value of .5. Given cardiac drift, fatigue and hydration changes, a .7 value provides a safe buffer zone of stability.
As an example, notice that at just 25 watts or 7 bpm more, the DFA a1 becomes an uncorrelated .5, which is seen with cardiovascular stress.
Knowing this, the upper limit of zone 1 (easy) training would be at about 128 bpm/175 watts.

Another cyclist example:
A fellow cyclist asked me to do a similar calculation based on his data.  This was derived from an outdoor session where the ramps were not perfectly even (Kubios sample time was last 2 minutes of each interval).  Regardless, the data seemed to make sense so lets take a look:
  • The DFA a1 of .7 is reached at about 170 watts.  The value becomes uncorrelated (.5), about 35 watts higher.

Here is the DFA a1 vs heart rate:
  • The points don't line up as well but there is still reasonable fit with a heart rate of 143 as the upper zone of easy riding.  Causes of mediocre fit include RR artifacts which artificially raise complexity as well as uneven pacing during the interval.
  • Unfortunately, the ramp did not go above MLSS, so we don't have the same shape as my tracing (or the one below)

Skiing:
This was done by a friend who is a competitive cross country skier.  Ramp intensity was controlled by pace, heart rate and RPE.
  
Here is the DFA a1 vs heart rate relation:
Since he was using the Hexoskin and had virtually no artifacts, I used the last 60 seconds of data rather than 120 seconds.  It's been my observation that as long as artifacts are near zero, 1 minute time samples can be used.



 From the raw Kubios data below:
  • There is an excellent fit of DFA a1 to heart rate with the .7 limit at 130 bpm.
  • The DFA a1 becomes uncorrelated just 5 beats above this at 135 bpm. 

The following tracing is the above ramp plotted against a different session consisting of mainly zone 1 intensity, but provides a good example of what happens if a certain "threshold" is crossed:

  • The red markers are just different 2-4 minute sections of heart rate vs DFA a1.  This was recorded with the Hexoskin and artifacts were in the 0-3% range.   
  • This is a good example of using the DFA a1 as a guide to stay in zone one as well as showing how easy it is to drift out of that zone with just a minimal change in effort.
This is raw Kubios data:
  • Notice that if heart rate rises above 130 bpm, DFA a1 becomes uncorrelated.

    Take home points:
    • Doing a simple incremental ramp with 5 minute segments allows easy determination of the heart rate or power associated with a loss of RR interval interbeat fractal complexity.
    • Although there are limited recommendations in the literature, it should be logical to avoid lengthy training time at uncorrelated values.  Finding where your own heart rate or power reaches a value of .7 should provide enough of a buffer to avoid undo stress during a session dedicated to high volume but low intensity.
    • According to my gas exchange data, the .7 value was reached at the VT1.  At 30 watts above this it was already .5 which was the location of the LT1 by lactate testing.
    • Whether or not VT1 is fundamentally equivilant to LT1 or not, it is clear that they may not test out to the same power or heart rate figure.  To avoid the confusion about the VT1 vs LT1 significance, use of a very objective cardiac index seems a more reliable metric for low intensity zone demarcation.


    MLSS/LT2 estimation using muscle O2 desaturation pattern:
    The maximal lactate steady state can be estimated in many ways. Various methods such VT2, RCP, LT2, OBLA, critical power and even FTP are present in the literature.  Unfortunately, we are again faced with arguments and disputes as to which is "best" and if they are even representing the same metabolic process or not.  I am not going into that rabbit hole, but lets just say that they correspond to grossly similar work rates.  Performing constant rate exercise over this output power will result in a non steady state Ve, rising lactate levels and limitation of exercise time.
    Muscle O2 saturation kinetics have been used to explore this and appear to have a desaturation breakpoint corresponding to the rapid rise in lactate values.  Criticism to using this approach revolves around many legitimate reasons (skin thickness, depth of sensor penetration, etc), but the bottom line is that it appears to work.  Despite my dislike of the marketing hype (including false claims) from the manufacturers of these devices, I believe they have a role in providing information in high intensity zone demarcation.  After all, the original generation one BSX was only able to calculate the MLSS (based on a ramp) and did not provide muscle O2 data as an independent output.  So, in the spirit of the original BSX Insight, this post and the work of Murias et al. lets explore using muscle O2 sat for the MLSS.

    This procedure requires you to either use a Moxy or Humon Hex sensor (BSX is fine if you still have one).  If you do not have one already, get the Hex, it's much less costly and will provide good data.
    Placement of the sensor should be the same across sessions, so don't use the rectus femoris on one day and the vastus on another.  Stick to the same leg as well.  I favor the rectus femoris since it has a sharper desaturation at the MLSS than the VL.  However, the deltoid and costal locations are going to behave in a similar fashion (although for different reasons). 

    What we are going to look for is an O2 desaturation pattern that continuously downslopes over a 5 minute constant power interval.  Fluctuating power may make it difficult to decide on a saturation trend so keep the power or run speed steady.  Here is an example of a series of 5 minute intervals I did as part of a lactate test:
    Pay particular attention to the last 2 segments of 260 and 285 watts.  Notice that the rectus femoris saturation is relatively stable at 260w but continuously decreases at 285w.  Somewhere in between is the MLSS.  According to the LT2 calculations of that day it was about 270 watts.  
    Once you have an idea of the power zone where the downslope occurs, you can monitor it on a regular basis looking for fitness improvement or detraining.  In this case, you really only need to do two intervals, one about 10-15w below and another 10-15w above the last MLSS.

    Here is an example.  As part of a regular training ride outdoors, I did a five minute interval below and above the 270 watt area.  Notice that costal, deltoid and rectus femoris placement of sensors did not matter, the tracings are nearly the same.



    The final result here is that my MLSS is somewhere between those 2 power figures.

    To confirm this, a 5 1/2 minute interval at 275 watts does show relative stability of O2 sat past 3 minutes (270 watts would probably have been better):



    Other athlete examples.
    This was part of the outdoor cycling ramp done by the individual we were looking at above.
    Although this is less than ideal, it does provide a learning experience in the methods recommended. 
    Notice the saturation of the rectus femoris at 245 watts, it's steady or coming up.  The next increment at 268 watts does have a downslope pattern but it stabilizes at about 4 minutes into the section.  This would not be considered a valid marker of exceeding MLSS.  There should have been one more interval after this at 280 to 290w which probably would have exhibited the continuous downslope we are looking for.



    Skiing ramp:
    This was part of the test done by my friend the cross country skier.  The intensity modulation was a combination of heart rate, pace and "feel".  We will look at the heart rate where the rectus femoris downslope occurs.  We also have the treat of comparing the Moxy vs Hex in the same muscle group and person (but different legs).
    • Both the Moxy and Hex show very little desaturation at heart rates below 155 bpm.  According to some of the marketing claims of these devices they should help one train a low intensities for recovery.  That certainly does not seen to be the case here.
    • However, we see a definite continuous downslope of O2 saturation at a corresponding heart rate of 165 in both sensors.  This would indicate MLSS has been exceeded.  
    • Therefore the MLSS resides somewhere between a heart rate of 155 and 165. 
    • The tracings are not as smooth as constant bike power via meter, but are impressive given the limitations of ski monitoring.
    Time trials:
    Theoretically a self paced time trial should be associated with high, but steady state lactate and power levels at the MLSS/LT2/VT2.
    In a study looking at 30 minute cycling time trials, 13 triathletes had extreme variability in lactate levels, but kept output power remarkably constant.  Even though the lactate levels varied, they were all well above 4 mmol indicating near or above MLSS power.
    The conclusion of the study was 
    In summary, the cycling ITT30 was characterised by a fairly self selected
    constant energy expenditure (reflected by VO2 and
    blood lactate responses) equivalent to the energy demand associated
    to the second ventilatory threshold. In ITT events lasting
    30 min, self-selected pace could be a valuable metabolic index
    to determinate appropriate training and competition race. The
    primary advantage of the ITT30 test at self-selected intensity is
    the ease and objectivity with which endurance capacity can be
    determined
    In addition, observation of both Ve and heart rate showed continued rise throughout the 30 minutes:

    As part of the pack of info sent to me, the cross country skier also included a running time trial.  His results are very similar to the above with a steady rise in HR and Ve (via Hexoskin) to near max levels.


    • Heart rates were generally in the 155 to 165 range, with a steady rise throughout.
    • The rise in Ve was steady throughout and did parallel the heart rate.  
    • The values make sense if the time trail was performed near the MLSS. 


    Do these patterns make sense from the cycling TT observations?
    From the discussion:
    The continuous increase in VE measured over the ITT30
    (+ 23 l N min–1) is somewhat higher with other observations reported
    during cycling bouts at “maximal lactate steady state”
    but with lower percentage of sustainedVO2max compared with
    the present study. MacLellan and Cheung [23] measured a
    20 l N min–1 increase in V˙ E over 30 min while exercising from 71
    to 76% V˙ O2max and recently, Lajoie et al. [20] reported an increase
    of 15 l N min–1 over 60 min (from 71 to 79% V˙ O2max). In
    our study the continuous increase in VE was due mainly to a significant
    increase in Bf
    (r = 0.717, p < 0.01).
    And:
    In contrast with most laboratory conditions, ITT30 in
    the field shows a steady profile responses in HR
    [24]. The latter
    observation and our present findings allow us to hypothesise
    that thermoregulation could be probably the crucial factor leading
    to a cardiovascular drift in laboratory setting. During road ITT
    events £30 min, HR could be a more valuable metabolic index to
    determine appropriate training and competition pace compared
    to laboratory conditions. Considering % HRmax as a valuable index
    of exercise intensity [12], the present results indicated that the
    ITT30was performed at quite high exercise intensity levels (range
    of 87–96% HRmax).
    Therefore with a similar pattern, both heart rate and Ve rose in the running time trial consistant with the published study results.

    Conclusion:
    Power levels derived from both self seleted individual time trails as well as MLSS determined by gas exchange (and presumably equivilant measures) are very close.
    But from the standpoint of proper handling of training loads and stress, performing 30 minute time trials on a regular basis does not make sense.
    Getting surrogate information through O2 saturation kinetics would perferable.


    VO2 max power
    Although it is not possible to measure VO2 max directly from power or heart rate,one can can a general idea of the rise or fall of this metric.  It's been shown by Keir and others that a maximum 5 minute constant power interval will usually allow an individual to reach their VO2 max.  Obviously, the interval needs to be maximal, but if you were a few percent below max heart rate, that's probably good enough.  Augmenting this approach, David Poole has proposed a validity stage for VO2 testing - a constant power interval of 5 minutes duration to be done 20 minutes after the incremental ramp of the VO2 max session.


    A reasonable approach for an athlete interested in following VO2 max power would be doing a regular 4 to 5 minute constant power interval with longitudinal tracking of the heart rate.  For example, if I were to do intervals of 330 watts for 5 minutes on a regular basis and noted a gradual reduction in average heart rate, that would imply that VO2 max is increasing (at the same ambient temperature, humidity, time of day, hydration, altitude etc).  This is what the Garmin Firstbeat formula attempts to do but their implementation just does not work.

    Here is an example of what to look for.
    I recently took about 10 days off from riding due to travel and a respiratory illness.  After about 2 weeks I felt fine and decided to do a 5 minute interval.  I then compared this to a 5 minute interval I did a year before.  The 2018 session was after a good 3 months of sustained training with no travel or illness.  Ambient temperature as about the same at 75 F.
    This is the plot:



    • During the post detraining interval (date 2019) heart rate was faster to rise and reached higher levels than the session in 2018 which was at peak fitness.
    • Comparisons such as this should be done with outside conditions matched (heat, humidity) as well as duration of exercise session.  This comparison was of identical cycling routes.  For example, do not compare segments occurring at 30 minutes vs 90 minutes into an exercise session. 
    • This is good evidence that my VO2 max dropped a bit from lack of training.  Repeat testing in 6-8 weeks will be interesting to see.

    A comment on using muscle O2 for low intensity guidance.
    There is only minimal locomotor muscle O2 desaturation at low intensities in either my tracings, the cyclist or the ski athlete.  Given the day to day and site to site fluctuation in sensor values, using them for low intensity insight is misleading at best.  However both Humon and Moxy persist in claiming these devices can help athletes in their recovery efforts.
    Here are some web screenshots:

    Humon:



    Moxy:


    • Not only are there no studies supporting this, but as we have seen above, they simply don't manifest enough dynamic range at low intensity zones. 
    • I understand these companies want to be profitable but putting out misleading statements like the above is not the way to do it.  I'm looking forward to the day that we see an honest approach.  

    Update 10/23



    Summary:

    • Two key physiologic metrics are able to be obtained non invasively with simple equipment, namely VT1/LT1 and MLSS.  
    • Ramp type exercise efforts of at least 5 minutes in length are needed.
    • Whether or not VT1 is metabolically equivilant to LT1 is arguable, but corresponding power levels of each may vary significantly.  If you do have official testing data of each fogure, use the lower value as the upper limit of zone 1 training.  As a surrogate marker for this breakpoint, DFA a1 complexity loss may be helpful.
    • DFA a1 HRV values below .7 indicate the approach of uncorrelated cardiac interbeat fractal complexity.  By the time a value of .5 (white noise) is reached, the low intensity exercise threshold has already been passed. 
    • Muscle O2 desaturation patterns can provide information as to the location of the MLSS in relation to interval power or heart rate.  Direct locomotor muscle placement as well as respiratory muscle (costal) or even non involved sites (deltoid) seem to all have similar behaviors.  Care should be taken to maintaining interval length of at least 5 minutes to demonstrate the transition from stable to continued downsloping of the saturation tracing.  It is possible that during very long intervals there is a plateau in desaturation to a personal nadir.  
    • Once these zone markers are know it is recommended to spend the vast bulk of endurance exercise below the VT1/LT1/DFA a1 cutoffs.  The exact amount of relative training time here is unclear but many would say 85% or more should be in zone 1.
    • MLSS knowledge can be used as an intensity training guide for zone 3 intervals as well as a regular benchmark of fitness, race pace.
    • A constant maximal power interval of 5 minutes should achieve VO2 max in most subjects.  Although measurement of VO2 can't be measured directly, comparison of the heart rates associated with these intervals can be used a relative indicator of VO2 max fitness.  The exercise interval conditions must be matched for temperature, humidity, time of day, time into entire session, caffeine use and hydration.
    • Muscle O2 sensors do not have sufficient dynamic range to be useful for low intensity zone demarcation.


    See also:
    VT1 correlation to HRV indexes - revisited  
    Does the FTP relate to the MLSS - Yes, No, Maybe? 

    Measurement of Hemoglobin saturation breakpoints - use as a fitness monitoring tool 

    Smart trainer usage in physiologic testing and interval training 

    DFA a1 vs intensity metrics via ramp vs constant power intervals  


    DFA a1 decline with intensity, effect of elevated skin temperature 

    Friday, September 20, 2019

    Why is the Garmin Firstbeat VO2 estimate inaccurate, can we do better?


    As the final post in the "VO2 max estimation" series, I would like to take a stab as to why the Garmin-Firstbeat prediction is, at least in my case, inaccurate.  Since I don't have access to a large number of paired gas exchange versus Garmin watch VO2 metrics I can only speak to my experience.  
    To start with, as discussed before, the Firstbeat algorithm takes advantage of the near linear relationship between VO2 and heart rate.  This was explored previously but at the time I did not have my personal VO2 data, so let's take a look now and see if that holds.  Here is the gas exchange VO2 vs heart rate (last 30 sec average) from my ramp study:

      
    • The fit line of regression is quite good, with only a +- 2 BPM deviation from the observed points.  I did not use the last point from the final stage since the data was suspect (see prior post).
    • The VO2 to heart rate relationship seems very clear from this ramp study.

    Since we can't measure VO2 on the road, does the VO2 to power curve seem close enough to use Watts as a surrogate marker?

    • The correlation is very good with a solid relationship of power in watts to VO2.

    Finally, does power still match up with heart rate?  It should if VO2 = Power.

    • There is an excellent fit of heart rate to power, except for the last point which was excluded.

    As another example of the close relation of heart rate to power, here is a graph of an indoor ramp done with 5 minute stages to above the MLSS, but not to exhaustion.  




    • This plot is as close as the VO2 max ramp with the rate within 1-2 BPM of the fit line.


    So if VO2 were substituted for watts, we would have a very nice, near linear association of heart rate and VO2.  

    So whats the problem then?  Well at least 4 things come to mind.
    • The heart rate may be different if the stages were longer.
    • Ambient temperature/humidity will affect heart rate to variable degree.
    • Maximum heart rate needs to be reached for a true VO2 max.  Maximum rate prediction equations may or may not be valid for the given individual.
    • We are still stuck without an accurate VO2 to power equation.  In another discussion, the various formulas for watts/run speed to VO2 conversion were reviewed.  Perhaps the best is the Storer formula, but this was derived from doing 1 minute continuous cycling stages, 30w incremental ramps to failure, which Garmin does not utilize.  Even if the Garmin formula correcting for temp, humidity, exercise time was perfect, the critical missing link is what is the actual VO2 for a given max power.  

    An example of field conditions.
    I recently did a series of 5 minute intervals on the road (80F and humid) and wanted to see how they matched up with the 5 minute indoor ramp from above.  The power values were 180, 240, 300w in red below:

    The three points in red are plotted along the indoor ramp.  Except for the 300 watt value, the other two are quite a bit off.  

    This translated into a fair amount of watt differential:


    The reason I show this is to put the "Performance condition" metric that Garmin displays in perspective.  My guess is that they are simply using a predefined heart rate to power formula, and then comparing your current heart rate to that.  As noted above, even a mild temperature, humidity change can have a major effect, especially at the lower power figures.  

    Is there a better surrogate than heart rate for VO2?
    Although the relationship of heart rate to VO2 is superb in a lab setting or even at home on a trainer, in the real world it is not.  Do we have another choice for a physiologic surrogate?
    Historically, the plot of ventilation (Ve) to VO2 is also very solid.  However, instead of a nice linear relation, it's curvilinear.  

    Gastinger et al looked at Ve and heart rate relationships to VO2 and felt that Ve provided a better correlation to VO2.


     

    Notice that the Ve response does deviate off a simple linear regression (red line I drew), as opposed to the HR which is flat (and would have been even more pronounced as curvilinear if they were testing at higher work rates):


    Let's examine the Ve to VO2 graph from my test:
    This is a plot leaving out the last stage with it's questionable value.

    This was based on the following third order equation:

    Here is the Ve to power plot for the 5 minute stage ramp:
    On this one I ran a third order fit as well as a linear regression at below LT1 and above it.  They are actually pretty close but the curvilinear trace seems better.

    Now let's place some other cycling session data points on the graph.  The following are two different outdoor sessions with comparison to the indoor ramp study.  The first was done in warm humid weather and the second at about room temp but outside.

    Warm-humid:
       Equal ambient temperature:

      • There is a mild difference in watt equivalence between indoor and warm outdoor efforts (as expected).
      • The comparison of indoor to outside at equal temperature is closer especially at the higher ranges.
      • Any formula needs to take conditions as well as hydration into account.

      Summary:
      • Although in a lab or controlled setting VO2, heart rate, Ve and power are all well correlated, in actuality, there can be major differences to the indoor benchmark values out on the road.  
      • Although Ve to HR may be a better choice for relative VO2 assessment, there is no accurate conversion formula to derive O2 usage from Ve. 
      • The Garmin Firstbeat methods do have some solid science behind them but the final product almost has to be flawed.  Not only does the heart rate relationship markedly change with conditions, it will also be affected by elapsed time and hydration.  These type of changes my be more or less per individual making blanket correction difficult.  
      • Caution should be used with the concept of "performance condition" as well as recommendations of your Garmin gear as to how to train.  In perfectly matched conditions, cycling at higher power with lower HR certainly implies an improvement in fitness, but conditions are seldom matched.
      • However, the biggest Achilles heel of the Firstbeat formula must be the conversion of power to VO2.  Somewhere in their calculations, above and beyond heart rate as a measure of the fraction of VO2 max reached, there must be some internal watts/run speed transformation to oxygen usage.  My personal guess is that this is the main reason the estimation fails.  Remember, I used an accurate HRM/power meter during my VO2 max testing and the Garmin estimate was off over 11%. 
      • Until an accurate formula that converts watts/run speed to VO2 is available (probably never) no prediction algorithm can be precise for VO2 max.  
      • As a better method of measuring your VO2 max "status", surrogate tests such as 4 or 5 minute steady state max power/speed or the Wingate 60 are reasonable alternatives for current VO2 max status.

      See also: