Friday, October 26, 2018

O2 desaturation during acidosis - the Bohr effect

A preface to the following observations and reason for this post:
I was out doing a training ride including a VO2 max interval (350w for 3 min) in cooler weather than usual.  About 2 min after the interval, I felt pretty rested and decided to do about 5 minutes at the MLSS (not rested enough for a repeat VO2 max/MAP, so chose a moderate power range).  Although I generally do not pay much attention to leg muscle O2, I was surprised that the saturation of the RF was at historic nadir values with the MLSS interval instead of it's usual mild desaturation at that work rate.  Since I never stack two such efforts close together (Florida heat index makes that difficult) this was a novel finding that I felt worthy of discussion.

One of the many ways the body adapts and compensates to best perform high intensity exercise is by altering the O2 hemoglobin dissociation curve.  Exercise above the lactate threshold will create an acidic environment shifting the curve, thereby releasing O2 from hemoglobin more readily at a given degree of O2 content.  In addition the temperature of the muscle, CO2, and 2,3 DPG will all contribute to O2 unloading in severe intensity zones.
For an excellent summary please read this review, and the following is an excerpt. 
The dissociation curve shift during acidosis has also been called the Bohr effect. 

In general circumstances the curve is as follows:

However, with intense exercise it becomes even more pronounced:
 With the commentary:

This post will address whether it is possible to demonstrate the O2 dissociation shift with NIRS, namely, the muscle O2 sensor.

While the Bohr effect is undoubtedly occurring while we exercise in the intense domains, actually seeing it is another matter.  There are multiple reasons in why this is so.  Other variables affecting O2 saturation are simultaneously occurring including change in arterial flow, micro-vascular circulation changes, external muscular compression, venous outflow obstruction and especially patterns of muscle fiber usage.  The last if of recent interest in physiology, recruitment patterns of locomotor muscles may play a huge role in observed O2 dynamics seen with NIRS.

On a practical basis, the muscle O2 desaturation curve may not be always be a continued drop in an all out maximal effort.  

Rectus Femoris 1 min max:
 Vastus Lateralis 1 min max:
  • In both examples, the O2 saturation actually increased despite a max effort and expected severe acidosis at the end.

However, the locomotor muscle O2 desaturation is generally in a downward direction with a ramp:
With innumerable examples in the literature and forum posts.

Initial Observation:
"Usual" VO2 max interval with a MLSS much later on:
The following tracing is my "usual" response of the RF to an initial VO2 max interval (350w x 3 min) followed by a MLSS (250w) interval:

An entire 3 hour session with an initial 4 min VO2 max, 1 min max then a 250 watt MLSS x 5 min at the end:

What about a very short rest between them?
This is what I saw when the separation between the intervals was very short (and the basis for this post):
The usual 3 min interval followed by a MLSS interval but separated by a short rest.  The "rest" was about 2.5 minutes of intermittent easy pedaling.
  • The RF O2 desaturation during the 250w MLSS interval reached (and maybe exceeded) the nadir of the Fast start 350 watt interval.  
  • As mentioned, the saturation of that muscle during the moderate continuous power of 250w is not this low.  
  • It did take several minutes for the saturation to drop to a low steady state level.  
  • I would not have been surprised if the O2 reached the nadir during the second interval if it was done at a higher power, but given practical aspects of fatigue, that would not be possible.  
  • One would expect the lactate to be very high post VO2 max effort and not to drop during the MLSS, did this create a lactate "clamp"......

Is it possible that the (very) high lactate, acidosis resulting from the initial high intensity bout lead to the enhanced unloading of O2 during the second interval?  The second interval was done at the MLSS, so presumably the high lactate that was currently present would have remained.  This power zone is associated with equal lactate production and removal, therefore a large excess should not be readily metabolized.  In other words, the lactate was "clamped" at a high level throughout the MLSS, however the cycling power needed to do so was at an achievable zone to continue riding.  

The respiratory data does show a slightly higher ventilation rate in the MLSS interval post short rest session vs long (there is artifact in some of the heart rate data since I wasn't sweating in the cooler weather):

Short, 2.5 min rest:

Longer rest (done several days before):
  • The minute ventilation was higher as expected in the 2+ minute rest session.  The usual value for ventilation during my MLSS is about 160.
  • Heart rate was similar in both trials. 
  • The above comparisons (short vs long "active rest") were done on different days.

I also decided to repeat the tests with sensors on both the Rectus Femoris and Vastus Lateralis and do both a short rest and longer rest MLSS all in one days session to better compare (ambient temp same, sensor location same)

First, a 3 min VO2 max at about 350w, 2 min active rest, them MLSS pace at 258w:

Very similar to the session done before, with the RF O2 at nadir levels during the last segment of the MLSS.

And a repeat MLSS done 20 minutes later:
  • The last minute average O2 sat did not reach the prior effort, 46% vs 42%.

To compare what I was capable of maximally dropping the RF O2 to, here is a 1 minute max (510w) of the same session:
  • The minimal RF O2 was about the same as in the 3 min VO2 max and short rested MLSS (all 41-42%)

The Vastus Lateralis (same intervals, just a different sensor):

  • Here the desaturation did not quite reach the same nadir after a short rest.
But it was still lower than the MLSS interval 20 minutes later:
  • The last 60sec average was 54% vs 52%.
  • The spread is relatively small but seems consistent.  I would not have trusted this if it was done on different days, or even after getting off the bike (sensor location change).

Maximum desaturation at peak effort:
  • The 50% value is the same as the nadir during the initial part of the VO2 max interval (same pattern as the RF).

Respiratory patterns:

  • No artifact in HR today (made sure plenty of moisture on the Hexoskin)
  • HR, ventilation at peak levels during the 350w x 3min - VO2 max achieved.
  • The ventilation post short rest during MLSS was much higher than usual.  Probably from respiratory compensation and attempted buffering of the acidosis.

MLSS after 20 min of easy riding:
  • The HR was close to the prior MLSS bout.
  • Ventilation was less than the prior MLSS (220 vs 188 L/min).
  • If the first (short rest) MLSS was done under significant acidosis (unlike the second), the reduced respiratory compensation would make sense.

A literature review was done but I could find little regarding NIRS observations and induced acidosis.

An interesting study did look at voluntary respiratory alkalosis induced by hyperventilation.  Although the authors were looking at VO2 kinetics, transitions between different priming exercise zones, the following was presented showing some change in the vastus lateralis saturation with respiratory alkalosis:


Another study looked at O2 extraction (as well as other issues) after a heavy priming exercise.  They did note an elevation of deoxy hemoglobin after the heavy priming exercise:

From the paper:
"A metabolic acidosis accompanies exercise
in the heavy-intensity domain, resulting in a rightward shift of
the oxyhemoglobin dissociation curve enabling a greater offloading
of O2 from Hb at a given PO2 and may account for the
greater muscle deoxygenation (HHb) during HVY 2."

Finally, there was a study looking at arm cycling vs leg cycling to induce a mild lactate elevation, then observing  vastus lateralis O2 saturation.
The rational was as follows:
"The purpose of the present study was to gain better insight
into the underlying mechanisms of the sigmoid pattern of
deoxy[Hb ? Mb] response to incremental exercise by
means of priming high-intensity exercise of different types,
i.e. arm and leg exercise. The Bohr effect will have a systemic
effect due to the establishment of a metabolic acidosis
(and an increase in core temperature), and will be elicited by
both the priming leg and arm exercise. However, during the
ramp exercise following the priming leg exercise, it can be
suggested that sequential recruitment of motor units consisting
of different fibre types (Henneman 1985) will be
disturbed as a consequence of muscle fatigue from the previous
high-intensity leg exercise (Burnley et al. 2002;
Krustrup et al. 2004; Layec et al. 2009). Priming high intensity
leg exercise has shown to increase the motor unit
recruitment in subsequent exercise (Layec et al. 2009) and
according to the size principle (Henneman 1985) the additional
muscle fibres that will be recruited are likely to be fast twitch
fibres with a different C(a-v)O2 profile (Ferreira et al.
The authors were looking at whether the enhanced O2 desaturation after a priming exercise was related to the Bohr effect vs muscle recruitment patterns (and hyperemia).  Since a higher usage of type 2 fibers could cause higher O2 extraction, the leg primed interval should behave differently than the arm primed interval if systemic acidosis was kept constant (arm and leg cycling creating the same acidosis).

The protocol:

Subjects VO2 max/MAP:
It is convenient that the MAP (max aerobic power, or power at VO2 max) is close to my personal value.

Interesting that the EMG was different in the leg primed ramp as opposed to the arm primed ramp.  The post leg exercise evidently lead to a greater degree of muscle recruitment, throughout the ramp session:

Lactate levels were similar between groups, but did drop to lower levels by the time the second ramps were to be done.  More comments about this to follow.

 The O2 saturation comparison:
  • The peak percent deoxy Hb was the same with each priming exercise (agrees with my tracings).
  • However the relative desaturation was higher at the lower work rates only in the priming leg exercise, not the priming arm exercise.
  • The differential was highest at lower work loads, 30% of peak, or about 1/3 of the MAP (120w in my case).
 From the paper:
Up to this point, the mechanisms underpinning this
sigmoid deoxy[Hb ? Mb] response have not been established.
In the present study it was investigated whether this
typical response is mediated by either systemic or local
mechanisms by applying either a priming high-intensity
arm or leg exercise. Measurements of lactate concentration
were used to quantify the metabolic acidosis which should
induce the Bohr effect, i.e. a rightward shift in the HbO2-
dissociation curve. The elevated lactate concentration
measured in the present study indicates that this metabolic
acidosis was probably present and of a similar magnitude at
the onset of the second ramp exercise for both LL and AL
It should be pointed out, however, that measurements of
pH would have provided a more accurate indication.

The results of the present study support the hypothesis
that the sigmoid increase in deoxy[Hb ? Mb] to ramp
exercise can mainly be explained at a peripheral level. It has been shown in rats that muscles consisting of different fibre types are characterized by a specific relationship
between O2 delivery and O2 extraction. More specifically,
the fast-twitch muscles are characterised by a greater
fractional O2 extraction compared to slow-twitch muscles
(Behnke et al. 2003; McDonough et al. 2005). In this
context, Ferreira et al. (2006) reported different profiles of
arteriovenous O2 difference (C(a-v)O2)as a function of
_VO2m between ‘slow-twitch’ and ‘fast-twitch’ rat muscles

The authors felt that the Bohr effect was not present and the findings were possibly due to recruitment changes.  In addition they also found the baseline O2 sat, before the second interval (post priming) was higher in the leg priming session (this was seen in my data as well-see above):
The reason for this enhancement was thought to be from hyperemia.

Finally their conclusion:
From the present study it can be concluded that priming
high-intensity leg exercise disturbs the traditionally
observed sigmoid pattern of deoxy[Hb ? Mb] in response
to ramp exercise. When the ramp exercise was preceded by
the leg exercise, deoxy[Hb ? Mb] increased immediately
from the onset of the ramp exercise. The observation that
the sigmoid pattern remained unchanged when the exercise
was preceded by the high-intensity arm exercise indicates

that the Bohr effect is probably not the mechanism behind
the sigmoid increase
. It is likely that the mechanisms relate
to the relative hyperaemia at the onset of the subsequent
ramp exercise and/or to the sequential recruitment of
muscle fibre types with different characteristics

Therefore, no Bohr effect.  The shift was felt to be related to hyperemia and/or recruitment changes.

But was this an adequate look for the Bohr effect?
Although the rest period was 3 minutes, the second exercise bout was also preceded by 3 minutes of 50 watt cycling.  Then a ramp was begun with increasing power of 25 watts per minute.  The net result of the 3 minute rest plus 3 minutes at 50w plus 3-5 minutes of ramping at low power equals about 10 minutes of time for lactate to be metabolized.  My point is that it is very possible that the local muscular lactate/pH was already near baseline by the time the ramp reached significant power.
This is not to take away from the other findings of EMG changes or a shift in the O2 extraction curve, which certainly can be attributed by priming exercise in the legs (and not the arms).  But, the protocol used here is much different than what I employed.   

  • The O2 hemoglobin dissociation curve is designed for optimal O2 unloading during exercise (Bohr effect).
  • Respiratory alkalosis will shift the curve the other way.
  • Attempts to demonstrate the Bohr effect in the literature by NIRS have been limited, and according to one study, not found.
  • However, the lack of positive results may relate to not achieving sufficient acidosis during the test interval.
  • From a theoretical standpoint, the Bohr effect should occur if substantial acidosis occurs.
  • According to the data presented here, there is an enhanced locomotor muscle desaturation while cycling at a (presumed) high lactate.
  • Whether the enhanced desaturation is related to acidosis or other factors (muscle recruitment patterns) is unclear.  However, the test efforts were done relatively close in time, making recruitment and hyperemia perhaps less likely.  Also it is certainly possible that the desaturation change is multi-factorial and related to all the above.
  • If one desires to train at lower O2 saturation levels at a submax power rate - a prior "priming" interval at maximal aerobic power, followed by a short "rest" and then followed by a prolonged MLSS interval seems promising.


No comments:

Post a Comment