Breathing efficiency and endurance performance

The comments and discussion in response to my recent post on breathing confirm that others have found it helpful to pay attention to breathing while running.  The evidence suggests that training that focuses on breathing is likely to improve both ventilation and running performance.  However, like many aspects of running, the question of how to maximise performance is complex.  There are four major practical questions:

1)      what type of training is beneficial;

2)      what aspects of running performance improve most;

3)      at what stage in a training program is breathing training most  likely to be beneficial;

4)      how great an improvement can be anticipated?

While a growing number of studies demonstrate that breathing training works in practice, the application of the findings from scientific studies carried out under standardized conditions, in a way that fits the circumstances of an individual athlete requires interpretation based an understanding of the underlying mechanisms.  Therefore it is helpful to start with a brief review of the various mechanisms by which breathing training might lead to improved performance.

The relaxation response

Improved breathing might be expected to improve running performance via several different mechanisms.  There is evidence that merely focussing on deep diaphragmatic breathing produces relaxation of unnecessary muscle tension, especially the tension in muscles of neck and shoulders which produces restrictive hunching of the shoulders.  It is also likely that the rhythmic contraction and relaxation of the abdominal muscles associated with diaphragmatic breathing minimizes the tense static contraction that results in painful cramp of abdominal muscles.   Herbert Benson, a cardiologist from Harvard and pioneer in the investigation of mind-body interactions, showed that mental centring – focussing of conscious awareness  of a location a few cm behind the navel resulted in a 5% reduction in energy  consumption at a fixed power output on an exercise bike (Benson 1975) .  Benson’s studies were performed at low power levels (corresponding to heart rate around 100 bpm) and are of little direct relevance to competitive athletes.  Nonetheless, it is very likely that the release of unnecessary shoulder tension and the maintenance of good trunk posture combats exhaustion during long races – marathons or ultramarathons – in which efficient husbanding of energy resources is crucial.

It is plausible that minimising static tension in the trunk muscles reduces the production of  toxic metabolites that prompt that the putative ‘central governor’  to impose restrictions on muscular effort during a long race.  My own experience indicates that conscious focus on diaphragmatic breathing in the later stages of a marathon can have a re-vitalising effect.   We will return in a later section to a discussion of how diaphragmatic breathing might also play a crucial role optimising the dynamic range of diaphragm muscle fibre length, thereby preventing a vicious cycle of respiratory muscle failure when the athlete is approaching a state of whole body exhaustion.

Respiratory muscle fatigue

At first sight, it appears the respiratory muscles have adequate capacity to move air at a rate exceeding that required for running.  Typically, maintaining a pace of 4 min/Km requires approximately 3 litres of oxygen per minute.   15 litres of air contains 3 litres of oxygen.  If we assume that it is inefficient to attempt to extract more than 20% of the oxygen from alveolar air during a single inspiration, maintaining a pace of 4 min/Km would require a ventilatory rate of about 75 litres of air per minute.   Peak ventilation rate for a man is typically 180-200 litres per minute.   It is clear that peak ventilation rate is more than adequate to provide the oxygen required to maintain a pace of 4 min/Km.  However breathing at anywhere near peak ventilatory rate is quite exhausting.   The diaphragm becomes exhausted quite quickly and takes a substantial time to recover.

Using a technique that assessed strength of diaphragmatic contractions by stimulating the phrenic nerve, which drives the diaphragm, after exercise, Johnson and colleagues (1996) from the Mayo Clinic in Minnesota  showed that  endurance exercise at above 85% of VO2max lasting 8-10 min caused a 15-30% reduction in the maximum force that the diaphragm could produce .  It took more than an hour for the diaphragm to recover it full strength.   However, this loss of muscle power was apparently due at least in part to increased competition for blood flow from the locomotor muscles. Reducing the load on the respiratory muscles by breathing a mixture of oxygen and helium had little effect on the time to exhaustion at 85% of VO2 max, implying that at that work rate, the principle cause of exhaustion was competition from the locomotor muscles for blood flow.  However, at 95% VO2 max, changing the load on respiratory muscles by adjusting the composition of inspired air did effect time to exhaustion, indicating that the diaphragm itself had become exhausted.  Thus, at high power output, exhaustion of the diaphragm is a limiting factor.  It seems very likely that in 3000m or 5000m races respiratory muscle exhaustion might be a limiting factor, and training designed to improve respiratory muscle endurance might be expected to improve performance.

At lower power outputs typical of 10Km to marathon pace, the major contribution to diaphragmatic exhaustion will be competition for blood from the locomotor muscles.   The most important way to minimise fatigue at these paces is to increase overall aerobic capacity and the efficiency of the entire musculature, especially the large locomotor muscles.  When it is crucial to avoid any waste of energy, ensuring maximal efficiency of the respiratory muscles is also likley to be worthwhile.    Because muscles work most efficiently when they are slightly stretched, the efficiency of respiratory muscles will be maximised by ensuring that they are working under conditions that ensure a fairly extensive dynamic range.  This is more likely to be achieved by deep diaphragmatic breathing.

Inspiratory muscle training

Inspiratory muscles can be trained by exercises that entail breathing- in through a device that contains a valve that does not open until the inspiratory muscles has produced a pre-determined reduction in air pressure in the mouth.  Thus the muscles must exert greater than usual force. Typically a training program involves one or two sets of thirty such inspirations daily for a period of several weeks.  Several commercially produced inspiratory muscle training devices are available.  One that is readily available on-line and has been used in several scientific studies is the PowerBreathe.

The majority of published studies that have tested the value of respiratory muscle training have reported benefits.  Andrew Edwards from Universal College of Learning in New Zealand and colleagues from Sheffield and Leeds in the UK, carried out a study in which two groups of 8 previously  untrained men were assigned to either respiratory training consisting of 30 daily inhalations against maximum resistance using the PowerBreathe, or 30 daily inhalations against minimal resistance (Edwards et al, 2007).  In addition both groups underwent a four week cardiovascular training program consisting of three running sessions per week (5x1000m; 3x1600m or 20 min run).   Mean inspiratory power increased 15% in the group who inhaled against maximal resistance and only 8% in the control group.  Those training against maximal resistance improved 5000m run time by 4.3% whereas the control group only improved by 2.2%.  However the magnitude of the improvement in running performance was not significantly correlated with the increase in inspiratory muscle power (possibly a consequence of lack of statistical power in such a small study),

Subsequently, Tong and colleagues (2008) from Hong Kong reported that a 6 week program consisting of 3 sets of 30 inspirations against a respiratory load increased by 50% produced a 30% increase in inspiratory muscle power; a 16 % increase in shuttle run performance; an 11% reduction in the rate of increase of perceived breathlessness; and less lactate accumulation during the shuttle run.  These beneficial effects were not seen in a control group who exercised against a minimal resistance.  The same investigators had previously demonstrated that a warm-up consisting of inspiratory muscle exercises similar to those used in the training program  improved performance and decreased the rate of rise of subjective breathlessness.  However, the warm-up produced a lesser increase in inspiratory muscle power than the training program.   They concluded that both the training program and the warm up had a beneficial effect on performance, but the mechanisms were different, with the benefits of training attributable to improved muscle function, while the warm-up merely produced greater tolerance of breathing discomfort.  My own opinion is that the benefits of the warm-up were likley to have been due, at least in part, to increased depth of inspiration resulting from greater mental focus on breathing.

More recently Lomax and colleagues from University of Portsmouth (2011) examined the independent and combined effects of an inspiratory muscle warm-up and inspiratory muscle training performed using a PowerBreathe device, on intermittent running to exhaustion.   The 12 male participants undertook four intermittent running tests, two before and two after 4-week of training consisting of 30 breaths twice daily at either 50% (experimental group) or 15% (control group) maximal inspiratory mouth pressure,.  Tests 1 and 4 were preceded by a warm-up consisting of 2 × 30 breaths at 40% of maximum inspiratory mouth pressure.   In the pre-training period, the inspiratory warm up produced significant increases in both inspiratory muscle power and distance covered in the shuttle run, in both groups.  After training, in the experimental group inspiratory muscle power increased by 20%  in the session without respiratory  warm-up and 27% in the session with warm-up.  Distance covered increased by 12%  in the session without respiratory  warm-up and 15% in the session with warm-up.   The investigators concluded that inspiratory muscle training and inspiratory muscle warm-up both increase running distance independently, but the greatest increase occurs when they are combined.

Inadequate respiratory drive

Respiration is controlled by the respiratory centre in the brain stem.  Decreases in oxygen concentration and/or increases acidity in the blood stimulate nerve endings in the walls of the large blood vessels.  These nerve terminals send messages to the brain stem and prompt increased respiratory drive, which is transmitted to the diaphragm by the phrenic nerve and to the accessory muscles of respiration in the chest and abdomen via the somatic nerves.  When running at paces at or above lactate threshold, the accumulation of acidity in the blood results produces a strong respiratory drive.  At these paces, the major issue to consider is respiratory muscle fatigue as discussed in the previous sections.  However at slower paces, typical of an ultra-marathon or during recovery runs, it cannot be taken for granted that automatic respiratory drive will be strong enough to achieve optimal ventilation.

The classic studies by WS Fowler (1949) demonstrated that uneven alveolar filling occurs in healthy people at rest, suggesting that when demand is low, respiratory drive is inadeqaute to ensure complete filling of the alveoli.  If alveolar filling remains non-uniform when exercising in the low aerobic zone, oxygenation of the blood will be inefficient.   Adopting a breathing pattern that ensures uniform alveolar filling would be expected to improve the oxygenation of the blood.    I suspect that the rapidly-achieved reductions in heart rate at a given power output achieved by conscious focus in breathing pattern when exercising in the low aerobic zone, described in my post on 22nd February, were probably due to a voluntary increase in depth of inspiration.  It is plausible that the mechanism was more even alveolar filling resulting in more efficient oxygenation of blood .

It is also plausible that the rapid improvements reported by Alexander Streltsov (1992) achieved by performing four inspiratory efforts preceding each expiration can be attributed to increased depth of inspiration.  Streltsov reported that his breathing technique led to improvements across the aerobic range, in international class athletes.  However I remain unconvinced that the particular ‘fractional’ breathing pattern proposed by Streltsov is necessary to achieve the benefit.  I believe that the crucial feature is deep diaphragmatic breathing, and this might be achieved by various different respiratory patterns.

Respiratory rate

What is the ideal respiratory rate?  Each breath must move a volume of air that is far larger than the capacity of the bronchial tree, as the air remaining within the bronchial tree at the end of each expiration contains stale air from which oxygen has been removed and CO2 added, and this stale air will be drawn back into the alveoli at the beginning of inspiration.  Therefore rapid, shallow breathing will be inefficient.  On the other hand, the rate at which air moves into the lungs will be greatest during early inspiration and the rate of expulsion will be greatest early in expiration when the stretched lungs recoil.  Therefore very prolonged inspiration and expiration will also be inefficient.   I have tried a number of different patterns and rates of breathing.  In my experience it is usually best to breathe at the rate that is most comfortable.

During a progressive run, I start with a 4:4 rhythm (inspiration for four steps; expiration for four steps) during warm-up, and then move to a 3:3 rhythm when running in the low aerobic zone.  When I become aware of an impulse to snatch an occasional ‘early’ inspiration before the full duration of expiration (which usually happens in mid-aerobic zone) I shift up to a 2:2 rhythm, which I can sustain up to the  lactate threshold.   If I deliberately try to prolong the inspiration ( e.g. by holding onto a 3:3 rhythm when I would feel more comfortable at 2:2) I find that my heart rate goes up a few beats per minute without an increase in pace.  However if I make a premature transition from 3:3 to 2:2, I often find that both pace and heart rate increase but without noticeable increase in effort.   Thus, in general I do not find it helpful to attempt to sustain duration of inspiration longer than is comfortable.  However, if I want to increase pace, I do find it helpful to move up to 2:2 a little sooner than I feel the need.

With regard to Streltsov’s proposal to make four consecutive inspiratory efforts before each expiration, I found that it worked well enough at the low end of the aerobic zone, and I am prepared to accept that it might be a useful training strategy to encourage a deep inspiration, but I have not found it helpful for ‘routine’ running in the mid or upper aerobic zones.


Both my own experiences and the scientific literature provide fairly clear evidence that attention to respiration is potentially beneficial. Both novices and experienced runners can benefit.  At low and modest paces, simply focussing on diaphragmatic breathing appears to produce more efficient running, possibly due in part to the relaxation response described by Benson, and in part to the greater efficiency associated with a large dynamic range of respiratory muscle contraction. For optimum performance at higher paces, such as 3000m or 5000m pace, it can helpful to perform inspiratory muscle training to improve respiratory muscle performance.  Specific exercises using a device such as Powerbreathe that increases the load on the breathing muscles can help increase the power and endurance of these muscles.    Furthermore, a respiratory warm-up can increase both respiratory muscle power and running performance


Benson H (1975) The relaxation response, New York:Morrow

Edwards AM, Wells C, Butterly R  (2007) Concurrent inspiratory muscle and cardiovascular training differentially improves both perceptions of effort and 5000-m running performance compared to cardiovascular training alone.  Br J Sports Med

Fowler WS (1949) Lung Function Studies. III. Uneven Pulmonary Ventilation in Normal Subjects and in Patients with Pulmonary Disease J Appl Physiol 2:283-299.

Johnson BD, Aaron EA, Babcock MA, Dempsey JA. (1996) Respiratory muscle fatigue during exercise: implications for performance. Med Sci Sports Exerc. 28(9):1129-37.

Lomax M, Grant I, Corbett J. (2011) Inspiratory muscle warm-up and inspiratory muscle training: Separate and combined effects on intermittent running to exhaustion. J Sports Sci. 22:1-7.

Streltsov A. (1992) The endurance reserves,

Tong TK, Fu FH, Chung PK, Eston R, Lu K, Quach B, Nie J, So R (2008) The effect of inspiratory muscle training on high-intensity, intermittent running performance to exhaustion. Appl Physiol Nutr Metab. 33(4):671-81.

13 Responses to “Breathing efficiency and endurance performance”

  1. Ewen Says:

    Thanks Canute. Some interesting results from those studies. The 16% improvement in shuttle run performance from Tong is relevant down here as sports testers use shuttle run endurance to help identify talented middle distance runners.

    I notice on the Powerbreathe site they only claim 7% faster recovery during repeated “sprints”. Even if in practise the improvement is not as much it could be worthwhile, even for the relaxation response you mention. A 5% improvement for 5k is the difference between 21 and 20 minutes.

    • canute1 Says:


      Thanks for your comment.

      I am looking forward to the day when you achieve that sub-20 min 5K target. If improved respiration contributes even a few percent to that achievement, I will be absolutely delighted

  2. Helen House Says:

    Thank you for your work on these two articles…always so clear.

    That last sentence

    ‘Furthermore, a respiratory warm-up can increase both respiratory muscle power and running performance’

    How else might you do a respiratory warm-up without a PowerBreathe device? Most of us start each run at a deliberately easy pace a general warm-up…that might be of value as a respiratory warm up as well? Can you think of any extra breathing exercise that might be worth adding to the warm-up.?

    • canute1 Says:

      Helen, Thanks for your comment. With regard to your question about respiratory warm-up, I has wondered whether a few minutes of deep breathing before training would help. However in my initial tests of this, I found that 5 minutes of slow forceful diaphragmatic breathing resulted in a substantial decrease in my peak expiratory flow rate (from 500 l/min to 340 l/min). I doubt that this was respiratory muscle exhaustion. I think that the forceful over-breathing had triggered bronchoconstriction. However I will try other strategies

  3. Ewen Says:

    Canute, I’ve purchased a PowerBreathe and will let you know how it goes. Initial thoughts are that it’s an excellent way to exercise the muscles required for inspiration.

  4. Rick Says:

    Hi Canute I asked Marius Bakken what he thought of the benefits of Diagrammic breathing and this is his reply to me.
    To your question : for well training athletes like yourself, from my honest opinion both as a runner but also from a medical perspective as a doctor, there is very little evidence that this will help.

    Problem is, the limiting factor for top trained runners is not in the lungs, but rather in the oxygen transport and diffusion capacity over the lung alverolaris to the blood stream and again over to the muscle cells, the enzyme activity to produce energy in the cells as well as the elastic abilities of the muscles + the overall running economy. Breating one way or the other (within reason) won’t help much with either of this.

    Now, for poorly/moderately trained it may have a benefit, but I’m not too sure. I’m generally open to research and have read what has been done with diagrammic breathing but I have a hard time being convinced based on the pure physiological aspect of it.

    However, if you want to find out for sure, the way to do this is to run back to back lactate acid tests on the treadmill to find if the threshold is altered using this. If you do this, I’d love to hear the results !

    I wish you all the best,

    kind regards, Marius

    Rick responds;
    Canute I do not have access to a treadmill or lactate testing, do you know of anyone having undertaken these tests using both normal and digramic breathing\ and what were the results ?

    • canute1 Says:

      Thanks for posting Marius Bakken’s reply. I take his opinion seriously, though there is not really adequate evidence to provide a definitive answer as to whether a well trained athlete would benefit from inspiratory training. The studies demonstrating performance benefits from inspiratory training described in this posting on 7th March had not tested well-trained athletes. However the study by Leddy described in my posting on 22 Feb was a study of experienced athletes, In that study respiratory training produced 50% increase time to exhaustion during treadmill running test at 80% of VO2 max on the day after training, an improvement that was maintained 1 week later. Furthermore 4 mile run time decreased by 4% and this improvement was maintained for a period of at least 3 months.
      I agree with Marius Bakken that (at least at first sight) the evidence indicates that ventilatory capacity is not likely the limiting factor in the process of delivering oxygen to the muscles. However the crucial issue that Marius does not mention is respiratory muscle fatigue. The study by Johnson and colleagues provides clear evidence that fatigue of respiratory muscles can occur after only about 10 minutes at high power output levels. While it might be that time to fatigue would be longer in very well trained athletes I think that Johnson’s evidence, together with the evidence provided by Leddy’s study suggests that experienced athletes might nonetheless benefit from respiratory muscle training.

      With regard to your query about studies that measured lactate, the study by Tong reported a significant reduction in lactate production during shuttle running after inspiratory muscle training- but I am not sure how well trained the athletes were prior to the respiratory training

  5. Scott Brown Says:

    Thanks for your recent comment Canute.

    I was very glad to have read it and touched by the sentiments.

    Nice mix of intelligence and sensitivity! Please keep in contact.

    • canute1 Says:

      I had left my comment on your blog because, like many others, my thoughts this weekend are with the Japanese and those living in Japan. In the past I have been an occasional visitor to your blog, partly because I have been amused by the banter between Ewen and yourself, and partly because I have been interested in your progress towards your marathon goal. I look forward to continuing to follow your progress, though perhaps the next few months might be a difficult period. Best wishes,

  6. steve Says:

    You have been gone for a while. I hope everything is OK with your health. I have really enjoyed following you Blog and reading the older information since finding you. Thanks for the education.

  7. canute1 Says:

    Steve, Thanks for your message. I had an accident. In retrospect it appears rather comical but at the time it was a bit worrying. I tripped and fell while leaping onto a moving canal narrow boat after the boat had emerged from a lock and I had closed the lock gate. I tore a lot of skin off my leg as I scraped it along a metal ridge on the gunwale. Then I fell in the canal and exposed an extensive mass of raw flesh to the rather mucky water – I think canal water is mainly dilute rat pee with added oil and grit. I had the leg attended to in hospital but a week later I developed extensive cellulitis of the leg accompanied by an overall feeling that I had nasty toxins coursing throughout my blood stream. Some antibiotics sorted out the infection, but it took quite while for the cuts on my leg to heal. I still have few technicolour splodges on my shin. However I have started running again though I have lost a lot of fitness, so it will probably be a long haul to get back into shape.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: