What is the best way to increase lactate threshold?

There are five physiological variables that need to be trained in preparation for a good marathon:

  • VO2max – this measures the maximum rate at which oxygen can be delivered to tissues and hence the maximum rate at which muscles can generate energy.
  • Speed at VO2max.
  • Pace at lactate threshold as a proportion of pace at VO2max. For a well-trained marathoner, race pace is near to lactate threshold.
  • Ability to conserve glycogen so that glucose supply is not exhausted before 26.2 miles.
  • Resilience of leg muscles to sustain pounding for the duration of the marathon with only minimal loss of power.

These five variables are all trainable to at least some extent, though the first two are largely determined by genetic factors. These two variables set the ultimate limit on performance. The other three can be trained to the level where they no longer impose the limit. But nonetheless, the way in which any of them is trained is likely to affect the others, and hence the choice of training schedule must take account of all of the requirements.

As discussed in my previous post about the physiology of Paula Radcliffe, once you have dealt with any remediable defects of strength or form that impede your speed, if you want to push to the very edge of the limitations that your genes and/or the aging process have placed on VO2max and speed at VO2 max, the main focus of training should be on increasing pace at lactate threshold. Therefore in this post, I will address the question of how best to increase pace at lactate threshold, while minimising risk of injury and taking into account the need to ensure that none of the other four requirements are undermined.

Threshold training

The most obvious way to increase pace at lactate threshold is to do a lot of running near lactate threshold. This will encourage the development of the mechanism for transporting lactate out of muscles and for metabolising lactate in other tissues such as liver and heart, thereby not only conserving fuel but also minimising the accumulation of acidity. Thus lactate threshold will be pushed upwards to a faster pace. This is the approach that was employed by Paula Radcliffe, with striking success. I think it is highly likely that this is the approach can work well for many runners, at least in the short term, but there are dangers in this approach that limit its value.

The greatest of these dangers is undue accumulation of stress. This is likely to lead to sustained high levels of cortisol that damage tissues. Such damage not only decreases the ability to generate the power essential for achieving optimum speed at VO2max,but also decreases the resilience of muscles and increases risk of injury.   Furthermore, the impact forces at foot strike increase greatly with speed, so the direct physical trauma imposed on the legs is substantially greater during threshold training than low intensity training.  As shown in figure 1 (showing data reported by Peter Weyand and colleagues) the impulse transmitted through the leg (the product of average force  x time on stance) rises very rapidly as speed increases from low speed reaching a peak at typical tempo speeds and then actually decreases a little a higher speed due to decreased time on stance.  Since energy is consumed while force is sustained and muscle failure will occur when the required force can no longer be sustained, I suspect that impulse might be a better predictor of likelihood of damage than the magnitude of the force.  If so, tempo speeds are likely to be especially damaging.

Figure 1. Upper panel: the average vertical force (expressed relative to body weight) during stance as a function of running speed.  Lower panel: the vertical impulse (average force x duration of stance) transmitted through the leg during stance as a function of running speed.

Figure 1. Upper panel: the average vertical force (expressed relative to body weight) during stance as a function of running speed. Lower panel: the vertical impulse (average force x duration of stance) transmitted through the leg during stance as a function of running speed.

I believe that Paula Radcilffe achieved her phenomenal marathon record of 2:15:25 in 2003 not only because she did a lot of training at a pace near lactate threshold but also because she employed a strengthening program that minimised loss of power and provided some protection against injury. However, I also think that it is likely that in Paula’s case, her strategy did ultimately lead to injury that frustrated her hopes of Olympic gold. The question of whether or not she could have achieved her phenomenal run in London in 2003 without putting herself at risk of injury in the longer term remains unanswered and perhaps will remain unanswered until someone else breaks her record via a less stressful training program.

Low intensity training.

The alterative to relying on highly developed enzymes for metabolising lactate is minimising production of lactate. Lactate is produced when glucose is metabolised in the absence of a copious supply of oxygen. Fat metabolism generates energy via aerobic metabolism and hence is not able to meet needs when oxygen supply is seriously limited. However, provided there is some oxygen available, fat metabolism generates energy without the production of lactic acid because the pathway of fat metabolism leads directly into the Krebs cycle (as illustrated in my post of 5th Dec 2013). The rate at which energy is produced by fat metabolism is relatively slow, and therefore, for most athletes, fat metabolism is inadequate to meet the requirements in the upper aerobic zone. However, the capacity to generate energy by fat metabolism can be increased. Such an increase will not only help conserve glucose, but also minimise production of lactic acid when running in the upper aerobic zone.   Thus, increased ability to metabolise fat would be expected to raise lactate threshold such that a faster pace can be achieved at threshold. One way to promote the capacity to metabolise fat is to do a lot of running at slow speeds. This mobilises slow twitch fibres that preferentially utilise fat.

A large amount of slow running will also help develop the muscle resilience to cope with a long duration of running, though perhaps not the resilience required to maintain marathon pace for a long period. A schedule that consists entirely of slow running is unlikely to develop the neuromuscular coordination required to achieve a high speed at VO2 max, nor the coordination required to protect the muscles against damage at marathon pace. Furthermore, merely minimising the generation of lactate is not adequate for optimum performance, since once pace increases to the level where lactate does begin to accumulate, the accumulation will be rapid unless the ability to metabolise lactate is also well developed.. In addition, a large amount of slow running would also be expected to lead to sustained high cortisol levels unless the body is well adapted to long slow runs. So low intensity training alone is unlikely to be the answer.

Polarised training

We are faced with several competing demands: the need to raise lactate threshold without undue accumulation of stress, while also maintaining the neuromuscular coordination and power required to run fast. This suggests that some higher intensity training is required. The key question is whether it is possible to combine low intensity and high intensity training in a manner that achieves the advantages of both without each damaging the benefits produced by the other.

High intensity training (above lactate threshold) does actually enhance fat metabolism while increasing aerobic enzymes. Therefore, in itself high intensity training, at least in moderation, would not be expected to harm the benefits derived from low intensity training. High intensity training also enhances ability to metabolise lactate. In addition, high intensity training promotes release of anabolic hormones. However, the risk is rapid accumulation of stress. The greater impact forces at higher speeds increase the risk of physical trauma to muscles. Thus, high intensity training is potentially dangerous unless done judiciously.

On the other hand, excessive low intensity training might harm neuromuscular coordination required for faster running, but the contrast between the pictures of Ed Whitlock training in the Evergreen Cemetery and racing suggests that only a small proportion of higher intensity running is required to maintain the required neuromuscular coordination.

Thus, there is little reason for believing that a judicious combination low and high intensity training will be mutually antagonistic. The major issue to be addressed is avoidance of accumulated stress from both types of training. The accumulation of stress is probably best dealt with by gradual build up.

The question of how much high intensity training is required to develop adequate ability to metabolise lactate, or alternatively, whether at least some threshold training is required remains unanswered. The evidence from the training of elite athletes suggests that at least some threshold training should be included in the mix.

Cruise intervals

There is an alternative strategy for enhancing capacity to metabolise lactate: cruise intervals in which periods of running at or perhaps a little faster than lactate threshold pace alternate with recovery periods during which the lactate is cleared from the system. Jack Daniels advocated moderately long periods at tempo pace with short recovery to enable tempo pace to be maintained longer. It is likely that Zatopek’s legendary interval sessions were a variant of cruise intervals with the faster epochs appreciably faster than lactate threshold pace, as Ewen pointed out in his comments on my post about Zatopek in 2009. I find that I recover well from cruise intervals with moderately short effort epochs (e.g 6 minutes) a little above lactate threshold.


On balance, the evidence indicates that polarised training is best if one wants to achieve year on year development, or to slow the deterioration with age. But it remains unclear whether a strategy that produces year on year development will ultimately lead to one’s best possible performance. Alternatively, if one’s goal is to produce one’s best possible marathon without concern for longevity, might a large amount of threshold training be best? Is it better to flash with the brilliance of Paula Radcliffe in 2003 but burn like a meteorite, or is it best to glow with the unassuming brightness of Ed Whitlock, like Sirius in the night sky?

Perhaps Yoshihisa Hosaka will break Ed’s M70-75 record in a few years’ time providing evidence suggesting that Whitlock might have done better with more intense training. Perhaps some yet unknown female marathoner will eclipse Paula Radcliffe’s record after less stressful training. The future will answer these questions. But at least for the time being, my own evaluation of the evidence favours the polarised approach: a large amount of low intensity running to enhance fat metabolism thereby minimising the production of lactate in the upper aerobic zone, together with a small proportion of high intensity training and a similar proportion of threshold training, perhaps in the form of cruise intervals, to enhance lactate metabolism.


8 Responses to “What is the best way to increase lactate threshold?”

  1. Laurent Therond Says:

    Hello there,
    Just wanted to say I keep reading your posts with the greatest interest. I wish I had more time to contribute to your research efforts!

  2. Laurent Therond Says:

    In 1984, Dudley et al showed that slow twitch fibers were maximally (aerobically) trained at an intensity of about 83% of VO2max.
    Fast Oxidative Glycolytic fibers were maximally trained at anywhere from 83-116% of VO2max.
    Fast Glycolytic fibers weren’t significantly trained until 100% of VO2max, and weren’t maximally trained until the workload reached 116% of VO2max.

    In 1983, Harms and Hickson showed that doubling training volume at 60% of VO2max improved performance by 40-100%.

    Furthermore, Dudley et al. showed that 15 minutes at or above 100% of VO2max elicited almost identical adaptation in FOG fibers as 60-90 minutes at Lactate Threshold!

    • canute1 Says:


      Thanks for your comment. As I see it, Dudleys’ studies (of rats) demonstrate that maximal training of slow twitch fibres is achieved by moderate duration, moderately intense running, and fast twitch aerobic fibres by intense short sessions. If one’s only goal is to improve aerobic capacity, Dudley’s data indicate that relatively short, fast sessions are the most efficient approach. However the point of my post is that such approach is risky because physical trauma to muscle and also because of metabolic stress. Cortisol elevation is much higher at VO2/VO2max= 60% than at VO2/VO2max=40%. So for year-on-year development I think the polarised approach is better. However, this leaves open the question of whether it is best to aim for maximal development in a short period or slow development over a sustained period.

      With regard to the study by Harms and Hickson, they demonstrate (also in rats) that low intensity running is just as effective as faster running for developing slow twitch red fibres (in soleus) but faster running was more effective for developing fast twitch white fibres in vastus lateralis. Nonetheless, doubling the amount of relatively slow running improved running performance substantially.

      It is also necessary to consider possible differences between rats and humans. In human muscles, I believe there is a greater tendency for the major leg muscles to have a mixture of fibre types than in rats. Maybe this is because standing on two legs demands more activity in major muscles of the legs to maintain posture. Therefore, I think that rats are informative about the types of exercise that develop specific fibre types but perhaps less informative about the effects of exercise on running performance.

      Whatever the species differences in muscle fibre composition, If we assume that we can develop the ability to run at marathon pace using mainly slow twitch fibres, then the Harms and Hickson result indicates that low intensity running can provide adequate increase in oxidative enzymes.

  3. Robert Osfeidl Says:

    When I read Steve Magness’ book Running Science I was suprised with just how short the tempo segments that elites us are.

    I have often done 8+ mile tempo runs but typically do them at 20 to 60 seconds per mile slower than threshold. The elite’s seem to do more shorter sections at threshold pace, or use intervals where the pace alternates between a little faster and little slower than threshold.

    Perhaps keeping the mileage at really stressful paces lower is important to keeping overall balance of training. Which brings one back to the polarised approach to training 😉

    So perhaps I should be exchanging my planned 8 mile tempo run at 6:40 min/mile pace with a couple mile easy, then 4 miles at 6:20 min/mile then a couple miles easy.

    • canute1 Says:

      Thanks for your comment. I certainly think it is desirable to avoid long periods at threshold pace unless the muscles are well prepared for this challenge. Damage to muscles is much more likely after substantial degree of exhaustion, thereby leading to a high risk after long periods at threshold pace. Perhaps an additional advantage of cruise intervals is that the recovery epoch during cruise intervals would be expected to allow at least partial recovery of the muscle fibres. Nonetheless, in a marathon run near to threshold, one will inevitably face the need for muscles to resist damage despite exhaustion, so I think that a program including cruise intervals in which the duration of the effort epochs increases gradually over a period of several months might be the answer.

  4. Ewen Says:

    Thanks again Canute. Must say I’m looking forward to Paula’s incredible marathon record being surpassed by someone who chooses to “glow with the unassuming brightness of Ed Whitlock, like Sirius in the night sky” Fantastic description that!

    I’m wondering if the damaging impact forces at higher speeds can be limited in a number of ways: running on soft surfaces, such as dirt or the manicured grass of Stromlo; using the new breed of cushioned shoes such as the Hoka Clifton; doing some sessions on a low gravity treadmill such as the Alter-G?

    I like the idea of cruise intervals. I think Deek’s 400m interval session could be described as such – 8 x 400m where the 400s were 5k pace, with 200m ‘fast’ recoveries in 40-45 seconds or so.

    • canute1 Says:

      Thanks for you comment.

      I agree that running on softer surfaces is desirable. I think you are lucky to have Stromlo and also the grass track at Calwell. In fact Paula Radcliffe often recommended running on soft surfaces, though I do not know what proportion of her near-threshold training she did on roads.

      I am inclined to think that it is preferable to use well-padded shoes for most of the training on roads because I suspect that microscopic muscle damage is reduced by padding. If a runner makes an abrupt transition to less padded shoes while continuing with the same volume of training, he/she often reports DOMS which is probably a marker for microscopic damage. I am at present using well-padded shoes for training. However as I get nearer to racing a marathon, I will gradually increase the proportion of training in my racing shoes

      I agree that Deeks’ 8×400 session could be described as a cruise interval session. I have experimented with the duration of the effort epoch ranging from 1 minute to 10 minutes. I do not have any clear answer as to what duration is best. It seems to me that whatever length of effort epoch one chooses, the principle should be that the effort epoch is comfortably hard, such that an appreciable amount of lactate is being produced (indicated by an appreciable increase in respiratory effort; eg up to about 40-50 breaths per /minute for the first effort epoch) and the recovery should be sufficient for most of that lactate to be dissipated, but not all, so that successive effort epochs become a little harder. The session should stop before the peak respiratory rate reaches 60 breaths per minute

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: