Archive for the ‘Injury’ Category

Polarized Training and Injury Prevention

December 29, 2016

Avoiding injury is one of the major goals of training for distance running.  On account of the impact forces experienced at footfall on every stride, runners are uniquely prone to injury.  However, effective strategies for preventing injury are elusive.  In recent years, advocates of techniques such as Pose have claimed that injuries are largely due to poor running technique, and have promised that the problem can be overcome by proper technique.  In particular, they have identified heel striking as a cardinal problem.  However, there is very little evidence to support this claim.   Others have advocated stretching during warm-up as a strategy to reduce risk of injury, though the evidence provides little support for this claim.  Advocates of barefoot running have proposed that running shoes are the problem, but again there is little evidence to support the claim that running barefoot or in minimalist shoes reduces risk of injury.  Conversely, the manufacturers of running shoes have placed blame on foot orientation problems such as over-pronation and claimed that motion control shoes can reduce this risk.  Yet again, the evidence is slight, though at least one study had found that over-pronation is associated with increased risk of injury.

It is likely that a wide variety of factors contribute to injury in different individuals.  Meta-analyses that pool the findings of many studies are only likely to identify risk factors that are common to many athletes.  Two risk factors emerge consistently: a history of previous injury; and a large weekly volume of training.   Lisa Callaghan has provided an up-to-date review of the evidence.

Prior injury

A history of previous injury might predispose to subsequent injury simply because the athlete has not corrected problems that contributed to the first injury.  It is also possible that unsatisfactory recovery from the previous injury plays a role.  Muscles, tendons and other connective tissues tear when subjected to force that exceed the limits of their resilience.  A cardinal factor in the resilience of connective tissues is the elongated spring-like structure of collagen fibres, making them resilient against forces acting along the direction of the fibre.  During the initial stages of repair following injury, collagen is laid down with random orientation providing a framework for tissue renovation, but full resilience requires remodelling such that the collagen fibres become aligned in the required direction.  Therefore effective recovery requires early mobilization to promote the laying down of appropriately aligned fibres, perhaps augmented by slow stretching.

Training volume

Observational studies report that training volumes of 65 Km (40 miles) or more per week are associated with higher rates of injury [Fields et al; van Ghent et al].   In part these observations might simply reflect the greater duration of exposure to risk of injury, though it is likely that fatigue plays an important role.   In particular fatigue impairs neuromuscular coordination increasing the likelihood of poor coordination between different types of muscles fibres within a muscle and poor coordination between muscles that act as agonists and antagonists, resulting in excessive local forces within tissues.

Polarized training

Simply limiting training volume is unlikely to be a satisfactory strategy for many runners, making it desirable to identify alternative strategies to reduce the damaging effects of fatigue.  As the forces exerted increase with increasing pace, it might be expected that injury risk would be greater at faster paces. However the observation by Van Middelkoop and colleagues that among marathon runners, those who do interval training have a lower risk of knee injury raises an intriguing question.  Could it be that interval training provides greater protection than training at   somewhat lesser paces in the vicinity of lactate threshold?   Interval training, in which short efforts at fast but sub-maximal pace are separated by recovery periods, tends to promote the development of neuromuscular coordination with relatively mild muscle fatigue.  As discussed in my recent post, interval training is likely to promote a favourable balance between anabolic and catabolic hormones, leading to strengthening of tissues. In contrast, running for a sustained period at threshold pace might produce fatigue with the associated risky deterioration of neuromuscular coordination during the session, and also tip the balance towards the catabolic effects of cortisol, promoting subsequent breakdown of tissues.

Even more speculatively, the viscoelastic character of the musculotendinous unit might result in a peak risk of damage to muscles and tendons at threshold paces. Viscoelastic materials offer strong resistance to brief sudden onset forces but less resistance to sustained forces.  Although force is greater at sprinting pace, time on stance decreases.  At speeds above LT there is actually a decrease in the impulse acting through the foot at each step because the increased force is more than compensated by  reduction in time on stance.  The product of forces x time on stance actually decreases, as illustrated in figure 1 based on data from the study by Weyand and colleagues.


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.

Thus, it is possible that the risk of tissues tearing is actually less at sub-maximal paces substantially above LT than in the vicinity of LT.   Nonetheless, it is crucial to prime the requisite neuromuscular coordination during the warm-up (for example by moderate intensity strides) and it is generally desirable to avoid absolutely maximal effort that taxes neuromuscular coordination to the limit, during training.

The other pole of polarised training is low intensity running.  This has the potential to build resilience of the muscles, tendons and the other connective tissues engaged during running by repeated application of moderate forces.   Provided training volume is built-up gradually and excessive fatigue is avoided, the risk of injury is low.


While the predisposing and precipitating factors causing running injuries remain controversial, consistent evidence indicates that a high weekly training volume increases the risk.  In contrast, the observation that interval training provides some protection suggests that polarised training might diminish the risk.  Observational evidence and also speculation based on principles of biomechanics and physiology suggest that high intensity sessions have the potential to build effective neuromuscular coordination, while low intensity training would be expected to enhance the resilience of muscles, tendons and other connective tissues with relatively little risk.  Nonetheless, as with any type of training, it is important to build up the training load gradually, and to warm up for each session in a manner the primes the requisite neuromuscular coordination.


September 6, 2010

My post on 4th September was motivated largely by Ewen’s recent question about my plans for running a half marathon this year.  My running has been disrupted by several health problems and now these have receded into the background, I am trying to build up training volume.  However, I am struggling against the re-emergence of several long standing musculo-skeletal problems.  Currently my most troublesome problem is metatarsalgia.

By way of introduction to the topic, I had devoted most of my post on 4th September to the rather tortuous history of Dathan Ritzenhein’s metatarsals. Despite a very bright career as a high school athlete and many subsequent major achievements, including a brief tenure as the 5,000m US record holder, and a bronze medal in the world half-marathon championships in 2009, his potential has frequently been hampered by metatarsal fractures.  Most dramatically, metatarsal pain led him to drop out halfway through the 10,000m at the Athens Olympics in 2004.

He is currently a member of the elite team of US distance runners participating in the Nike Oregon project.  Under the guidance of his coach Alberto Salazar he has made three major adjustments to deal with his metatarsal problems.  First, he moved from the beautiful but austere environment of Boulder Colorado, where the trails are rock-hard, to the soft moist terrain of Portland, Oregon.  Secondly, Nike’s head of biomechanics, George Valiant, deigned some shoe inserts which relieve the pressure on the downward protruding head of the third metatarsal of his right foot.  Finally, and in my opinion, most controversially, he has, abandoned heel-striking for something approaching a mid-foot landing.  I presented my reasons for questioning the wisdom of third of these changes in my post on 4th September.  Now it is time to describe the history of my own metatarsals.

The history of my metatarsals

Although by nature a forefoot runner, I have always had problems with my metatarsal heads.  Since birth, the second metatarsal head in both my right and left feet has protruded downwards.  In childhood , I used to wear out my shoes from the inside.  By the time I reached my teens, a few months after I obtained a new pair of shoes, a hole appeared in the insole as a result of abrasion by the callous on the underside of my forefoot.  Perhaps surprisingly, my feet scarcely suffered at all. I ran all of my marathons in the same pair of Onitsuka Tigers – the fore runner of today’s minimalist shoes.  Although I had to take special precautions to deal with my congenitally peculiar toes, I suffered no pain in the vicinity of the metatarsal heads.  I think that in those days my body’s ability to repair itself far outstripped the rate of tissue damage.  I simply developed thicker callous. However, that has changed as the fat pads between the metatarsal heads and the callous have disappeared with age, and my capacity for tissue regeneration has waned.

About a decade ago, I went shopping for a new pair of street shoes and was frustrated by the fact that every pair I tried caused pain in my forefoot.  I was puzzled as to why I had not had such a problem before.  Even the worn old shoes that I had worn into the shop were quite comfortable.  At first I failed to draw the obvious conclusion but later when I started to suffer serious metatarsalgia (pain beneath the metatarsal heads) while running, the answer came to me.  My old shoes were comfortable because the insole have been hollowed out by the abrasive action of my foot.

I therefore hollowed out the insole of my running shoes, in a similar manner to the way George Valiant created inserts for Ritz eight years later.  The pain diminished substantially, though unfortunately did not resolve entirely.  Perhaps I lacked George’s engineering skills but I suspect the main problem was that I had developed inflammation of the fascia on the underside of the forefoot.  Although the hollowed-out insole shifted the pressure away from the metatarsal head, stress was transferred to a different region of the fascia, which continued to tug on the inflamed area.  I think that is why orthotics rarely provide full relief from plantar fasciitis.

However as the inflammation settled, I found that I could run without pain. Nonetheless, as an extra precaution, I avoided running on hard surfaces as much as possible.  In addition I embarked on regular exercising of the intrinsic muscles of my feet – mainly variations on toe curling – with the expectation that strengthening these muscles would improve their ability to help distribute the load at foot strike thereby controlling the rise in in the tension in the fascia. Together these precautions proved quite effective. Since taking up running again three years ago, I have had minimal trouble from metatarsalgia.

The price of complacency

Eventually I became a little complacent.  I stopped doing the exercises to maintain strength in the intrinsic foot muscles and I became lackadaisical about hollowing out the insole when I obtained new shoes.  There were slowly evolving signs that all was not well: since the episode of acute arthritis in the early months of this year, I had been aware of an increase in forefoot pain, but in the setting of the various other aches and pains that afflicted me, it appeared trivial and I ignored it.  However, I suffered a rude awakening two weeks ago when I set out for my ill fated tempo run. Because I had been late home from work, I ran along a paved sidewalk rather than risk the uneven riverside path in the dark.  I am not sure what was the main culprit: lack of hollowed insole; the hard surface; the alteration of gait due to my knee problems; the effect of lingering systemic inflammation or the accumulation of stress due to my recent return to running.  Whatever the cause, the outcome was a sharp pain in the forefoot.  The metatarsalgia had returned with a vengeance.  The following morning I could scarcely bear to put my foot on the ground.  Hollowing-out the insole of my street shoes provided only slight relief. I was amazed at the ferocity of the sudden exacerbation.   I wondered whether or not it might be a stress fracture.  Focal tenderness of the second metatarsal head added weight to this possible diagnosis, though my experience of similar pain in the past indicated that it would be unwise to jump to a rapid conclusion.

An abrupt drop in high frequency Heart Rate Variability that morning confirmed that I was markedly stressed, no doubt mainly due to the widespread minor musculoskeletal trauma arising from my tempo run, to which the metatarsalgia was only one contributor.  As shown in the figure presented in my post of 30 August, HRV remained depressed for two days, but then returned to a healthy level, indicating that my recovery mechanisms had risen to the challenge and dealt with the systemic stress level.  The focal pain in my forefoot was also substantially reduced but nonetheless, still quite appreciable.  Even with a hollowed-out insole, I could not bear to take my weight on my forefoot while standing.  Running was unthinkable.

Becoming a heel-striker

Gradually the pain in my forefoot diminished and by the fourth morning after the tempo run I decided it was time to try to run.  However it was clear that landing on my forefoot was out of the question.  Circumstances dictated that I should become a heel-striker. To minimise the force at each foot fall, I adopted a quite high cadence and short stride.  To my delight, I found this high cadence, heel-striking style was actually less painful than walking.  After a few Km, the pain in my forefoot had disappeared entirely, a time course typical of pain due to chronically inflamed connective tissue, which tends to feel better once recently formed local adhesions have been remodelled and local blood flow has increased.  This is not the usual time course of pain from a fracture, which tends to increase as the distance run increases.  So I was relived to realise that stress fracture was unlikley.

I was feeling very relaxed at a pace between 5:30 and 6 min per Km.  There was no sign of the discomfort in my knee that had plagued me in recent weeks.  I had intended to run about 6Km. but was feeling so relaxed that I extended this to almost 14 Km.  Although I still felt comfortable, I stopped simply because this was several Km further than I had run in the preceding two months, and I feared that accumulating tiredness would increase the risk of further injury.  My average pace was 5:45 Km/min and average heart rate 120, confirming the previous evidence that my loss of aerobic fitness has not been severe.  Later in the day I suffered the expected aches in knee and forefoot, but on the whole, my body had coped well.

In subsequent days my forefoot has remained tender. When I stand-up from a sitting position pain from the patella-femoral joint and also from the point where the ITB rubs against the lateral femoral condyle confirms that I still have lingering irritation of tissues in these areas, but provided I run with a mild degree of heel-strike and a high cadence, both my forefoot and my knees are comfortable when running.

The future

What the next few weeks will bring remains uncertain, but I have growing optimism that conversion from forefoot striking to slight heel striking, at least during low-aerobic training runs, might be the key to building up to an adequate volume of training.  If I can achieve a reasonable volume of running, I would like to run a half-marathon before the end of the year. Currently, I have the Worksop Halloween half marathon pencilled into my diary.  However that is less than two months away, and in the intervening period I will be doing some travelling.  I am scheduled to deliver talks at conferences in both Germany and China in October.  Unfortunately attending conferences does not diminish the load of routine work, so October will be a busy month.  Therefore, it is far from clear that I will be able to get my legs adequately conditioned for a half marathon by the end of October, and it would be foolish to race if my legs are seriously ill-prepared.

Whatever happens with my own race preparations, I am of course looking forward eagerly to the outcome of Dathan Ritzhenhiem’s experiment with the transition from heel-striking to mid-foot striking.  I hope that on 7th November in New York he at least improves upon his previous best marathon time even if he does not win what promises to be a great race.  But even if his experiment has a successful outcome, I am dubious about Alberto Salazar’s belief that there is one ‘best way’ to run.  I am increasingly inclined to think that while there are indeed rational principles that govern running mechanics, each individual needs to discover how best to apply those principles to his or her own situation.  The heel-strike debate is probably one of the least important issues for most marathon runners, but for Ritz, I think that it is potentially an important issue, and that in abandoning heel-striking he is taking a risk.

Can Dathan Ritzenhein win the 2010 New York City Marathon?

September 4, 2010

After Meb Keflezighi’s victory in the New York City Marathon last year and his fifth place in Boston this year, he will start as one of the favourites this year, though it promises to be a great race.  Haile Gebreselassie will be making his New York debut, but he is in no other sense a debutante.  It will fascinating to see whether or not he still has the form that carried him to the world record in Berlin in 2008.  I understand that Tesfaye Jifar who set the course record almost a decade ago, will be back again this year.  Among the somewhat younger contenders in New York on 7th November will be Dathan Ritzenhein [1].  He made a rather disappointing New York debut in 2006 but is returning to New York after some strong performances on the track, and a bronze medal at the World Half-Marathon Championships in a time of 60:00 in Birmingham in 2009.

But really this blog post and the next will be about me almost as much as Dathan Ritzenhein, and the sub-title might well be ‘Will Canute be fit enough to run the Worksop Half marathon on Halloween?’  I am writing this in response to Ewen’s recent question about my prospects of running a half marathon this year in light of the fact that my year has been blighted by illness.  In my return to running two weeks ago, I struggled to maintain a pace of 5 min/Km during an attempted modest tempo run.   The reason for a rather far-fetched comparison of myself with one of  America’s leading  distance runners is that Ritz has also frequently been sidelined by injury, and if one digs a little deeper into the details, there are some interesting parallels, but also interesting differences in the way that we have responded to a similar problem.

My main problem this year has been an episode of arthritis that started in January and lingered for many months.  It started in my neck, and then spread to my knees, especially the left knee.  Although the acute inflammation settled several months ago, I have subsequently been plagued by a variety of irritating problems around the knee joint, especially  patello-femoral pain and also irritation of the iliotibial band.  I suspect that both of these problems can be attributed largely to a temporary  alteration of my gait to protect the femoro-tibial joint (the main load-bearing joint at the knee) during the period when the acute arthritis was resolving. However, I think the presence of acute systemic inflammation and/or my altered gait has also unsettled several of my other long-standing trouble spots, including my metatarsals.     At present my most frustrating problem is metatarsalgia.

The history of Ritz’s metatarsals

Dathan Ritzenhein  has suffered metatarsal  problems for years.  After a promising display of talent in high school athletics, culminating in a bronze medal at the IAAF World Junior Cross-Country Championships in 2001, he had went to college in Boulder, Colorado.  Following a successful freshman year, his sophomore year was blighted by two metatarsal stress fractures.   The next year he won the National Collegiate cross country championship but again suffered a stress fracture, and limped home in last place in the 10,000m trials for the 2004 Olympics.  Nonetheless due to various mishaps to the initially selected runners, he made the Olympic team, but dropped out halfway through the race in the Athens on account of pain from the stress fracture.   After the Olympics he left college athletics to become a professional and joined Brad Hudson’s coaching group in Boulder.

Boulder is a quirky university town set in awe-inspiring but austere landscape on the eastern slope of the Rockies.  I knew Boulder as it was in the days before Ritz attended college there, but I do not expect that the terrain has changed greatly in the past decade.   Within the city are many paved cycle paths, including the well known creek- side path, which at first sight appears an attractive running route,  but the concrete surface is very hard.  Extending up into the nearby foothills is a further network of unpaved trails but these are mostly hard earth and rock.    Being in the centre of the north American landmass, Boulder also happens to be more than a mile (1600m) above sea level.   It is not as high as towns such as Eldoret in the Rift Valley district of western Kenya, or the mountains near Addis Ababa in Ethiopia, where high country lying between 2000 and 3000m above sea level  has become a Mecca for athletes seeking the secrets of African distance runners.  However Boulder’s  combination of thin air, hard rocky ground and relatively few trees create an environment that is hard on the body of a serious  distance athlete.

In her account of  training in the mountains of Ethiopia,  Hilary Stellingwerff noted ‘Finally, on all my recovery runs, the Ethiopian athletes stressed the importance of running on soft ground in the forest to make sure you go slow enough to really recover. They don’t worry too much about their pace, but instead about “getting good oxygen” from the trees and “soft ground” for the body.’ [2]

I will return to the question of whether or not  the harshness of the environment makes an appreciable contribution to the risk of injury in a future post when I respond to Ewen’s other recent question about the value of monitoring Heart Rate Variability.  However I think it is plausible that the austere environment, and especially the hard trail surfaces of Boulder contributed to several of Ritz’s injuries and illnesses over the years.

Softer ground

In May 2009 Ritz left Brad Hudson to joint Alberto Salazar’s group at the Nike Oregon project in Portland.   Although I do not know Portland, I had lived for almost a decade in Vancouver, BC, and I am fairly familiar with the Pacific Northwest.  I find it hard to imagine anywhere in the world  that could be more congenial to the body and spirit of a distance runner than the moist and verdant Pacific Northwest.  Added to the idyllic natural surroundings is the high tech support provided by Nike, which includes a house with artificially thinner air.  The athletes can live and sleep in the rarefied atmosphere that encourages accumulation of red blood cells, yet avoid the stress of high altitude training by doing their rigorous training at normal atmospheric pressure.  However even in this runners’ paradise, Ritz continued to suffer injury.   So the hard surfaces of Boulder were not the only cause.

In an interview with Peter Gambaccini for the Racing News blog at Runner’s World in July of this year Salazar admitted  ‘Dathan continues to have some foot problems which he’s had for years. I had thought that just by keeping him on soft surfaces and making sure that he’s recovered that this would be taken care of.’ [3]

Shoe inserts

In an attempt to overcome the continuing problems Salazar and the Nike team have implemented two changes.  First they identified the fact that the head of Dathan’s third metatarsal on the right foot protrudes downwards.  To relieve the pressure,  Nike’s head  of biomechanics, George Valiant, produced a hollowed-out insert for his running shoe.  This produced an immediate relief  which I find understandable, because I had made a similar modification to the insoles of my own running shoes about 8 years ago, and , as I will describe in my next post, this provided a partial relief to my own problems.   In the interview reported in the Racing News blog  Peter Gambaccini also spoke to Dathan himself.  He reported ‘I feel really comfortable now. The inserts feel real good. There’s still a little bit of refining on them, but at this point, I feel like when I train daily now, it feels good and my body’s getting used to it.’

Changing  from heel-striking to mid-foot landing

Salazar’s other innovation was to encourage Dathan to change from heel-striking.  Alberto Salazar believes that there is a right way to run and that right way does not include heel-striking.   In the interview for the Racing News blog, Gambaccini asked about the change from heel striking and Dathan repliedI was definitely more of a heel-striker, so I’m definitely getting on to my midfoot more. I wouldn’t say I get all the way up to my toe. I think I’m more pretty much efficient for the marathon if I stay in more of a midfoot stance anyway. ……. Initially, the problem was we tried to focus solely on changing that without being strong enough to do it. We went back to trying to build it up from the strength side so it (the stride change) naturally took over instead of trying to think about it consciously

It is of interest to note that Ritz emphasized the necessity of building strength to support the transition from heel-strike towards the forefoot.  In that interview he did not give further details, but  I suspect he was referring largely to the greater calf strength required when running on the forefoot, though I wonder whether he was also referring to the necessity for greater strength of the intrinsic muscles of the foot, which are called upon to take a larger role in distributing the  forces of impact.  I seriously doubt Salazar’s wisdom in the decision to change from heel-striking for a runner with metatarsal problems, and will return to that issue when I focus on my own tentative approach to my metatarsal problems.

Is the heel-strike debate a red herring?

In my own speculation about running style (described in ‘Running: a dance with the devil’ in the side panel)  I have advocated forefoot landing, but I believe that even more important than forefoot landing in a high cadence and short time on stance.  A recent study by Heiderscheit and colleagues from Wisconsin [4] confirms that increasing cadence by 10% without making any conscious attempt to  change  other aspects of running style results in a substantial reduction in stress at knee and hip.   I continue to believe that if there is one change that is worth making to running style, it is increasing cadence, at least up to a rate in the range 180-200 steps per minute.  I think that above 200 there are diminishing benefits, except when sprinting.  But the nagging question remains: is it also worthwhile to change from heel-strike to forefoot strike.

There are three main arguments favouring a change.  First, it would be expected that landing on the forefoot will result in greater capture of the energy of impact as elastic energy in the muscles and tendons of the foot and calf, and that this energy might be recovered at lift-off from stance.  Secondly, the absorption of impact energy as elastic energy will prevent the sharp rise in ground reaction force immediately after foot-strike.  The jarring effect of this rise in force transmitted upwards through knee and hip might be expected to increase risk if musculo-skeletal injury, though there is little evidence supporting this.   Thirdly, from the evolutionary perspective, it is probable that the human frame evolved to facilitate barefoot running, and barefoot runners usually land on mid or forefoot.

However, the extensive anecdotal evidence of increased rate of calf injuries following transition to forefoot landing suggests that the injury risk associated with the transition is high unless the runner makes a determined effort to strengthen the muscles of foot and calf.   Studies such as the Capetown study of Pose [5] suggest that the transition can be associated with less stress at the knee, but the more recent study by Heiderscheit and colleagues [4] indicates that the reduced  stress on the knee with Pose style might be due at least in part to increased cadence. With regard to the evolutionary argument, it might well be that forefoot striking was best suited to the barefoot running on the African savannah 2 million years ago, but most of us now run on paved surfaces much of the time.  Furthermore the elegant longitudinal arch of the foot suggests to me that the human foot evolved to absorb and store impact energy efficiently when both forefoot and heel are grounded.

In principle heel -striking and forefoot striking are distinctly different, but in fact there is a continuum.  At one extreme, the entire force of impact is borne by the heel; at the other extreme the impact is taken entirely on the forefoot.  I consider that both of these extremes are likely to increase risk of injury.  In the middle of the range is mid-foot striking in which the initial impact is taken equally on forefoot and heel.  In this style, the impact forces within the foot are immediately distributed along the length of the longitudinal arch.  But of course, the runners’ stance is a dynamic event in which the peak vertical ground reaction force occurs around mid-stance, and perhaps that it the point at which it is most beneficial to have both forefoot and heel grounded.

If one is aiming to have both forefoot and heel grounded around midstance, the possibilities for ankle posture at foot-strike stance range from  plantar flexion to mild dorsiflexion, but I suspect that the factor that plays the greatest role in determining the softness of the landing is the degree of flexion of the knee.  As the knee flexes at impact, the quads, which are far bulkier than any muscles below the knee,  will absorb impact energy.  If the degree of tension in the quads is low, the landing will be soft and the risk of injury low, but the recovery of elastic energy will be relatively slow.  If higher tension is maintained in the quads, the leg will act like a stiff spring, retuning energy rapidly and promoting efficiency, at the price of somewhat greater initial rate of rise of the vertical ground reaction  force and possibly greater risk of injury.

I suspect that there is an inevitable trade-off between efficiency of energy recovery and risk of injury, determined largely by the amount of tension in the quads.  I also suspect that for a long distance runner, the orientation of the ankle matters relatively little provided it is within the moderate range that allows an equable dissipation of impact forces along the longitudinal arch by midstance.  If so, the heel-strike v fore-foot debate is largely irrelevant, unless the athlete has anatomical features that make a particular part of the foot more vulnerable.  For a runner with downward protruding metatarsal heads, I suspect that a mild degree of heel-strike might actually be preferable.

I have taken a particular interest in the way Dathan Ritzenhein has dealt with his problem because I have faced some similar issues.  By trial and error I had discovered some of the same strategies as Ritz, though in one potentially important respect I have taken a different path.  But this post is already long enough so I will defer the history of my own metatarsal problems to my next post.




[4] Heiderscheit, BC.; Chumanov, ES.; Michalski, MP.; Wille, CM.; Ryan, MB (2010) Effects of Step Rate Manipulation on Joint Mechanics during Running. Medicine & Science in Sports & Exercise: doi: 10.1249/MSS.0b013e3181ebedf4; Jun 23. [Epub ahead of print]

[5] Arendse RE Noakes TD, Azevedo LB, Romanov N, Schwellnus MP, Fletcher G.  (2004) Reduced Eccentric Loading of the Knee with the Pose Running Method. Medicine & Science in Sports & Exercise: Vol 36 pp 272-277

The sweet-spot on the crest of the inverted U

March 21, 2010

There is abundant evidence that a moderate amount of running increases health and life-expectancy.  However there is also some disconcerting evidence that excessive amount of running might damage health, impair performance and perhaps decrease life expectancy.  The main sources of evidence are:

1)      One of the most robust predictors of  running injuries is volume of training, as demonstrated in a rigorous review by  van Gent and colleagues (Br J Sports Med  41: 469-480; 2007)

2)      Athletes appear to be at increased risk of developing heart rhythm disturbances such as atrial fibrillation (which leads to increased risk of stroke) and perhaps even more dangerous rhythms such as ventricular tachycardia, as discussed in my recent posts.

3)      Some recent evidence suggests that large training volume is associated with increased stiffening of the walls of the arteries which creates a risk of elevation of blood pressure, and also a risk of heart attack.  Increased calcification of the arteries might arise from chronic inflammation.

4)      The data for physical deterioration with age in athletes is rather inconsistent, but some studies demonstrate that athletes suffer a more rapid deterioration of the VO2max  with age than sedentary individuals (Tanaka and Seals, J Appl Physiol 95: 2152–2162, 2003)

5)      Excessive training can produce an over-training syndrome which has some similarities to chronic fatigue syndrome.  This impairs athletic performance and tends to be associated with features such as weakened immune responses that jeopardize health.

These rather disconcerting observations must be set against the much larger body of evidence demonstrating beneficial effects of running on health.  Nonetheless, it appears that the relationship between health benefits and training volume might have an inverted U shape, such that the optimum health is achieved at some intermediate value of training volume.  Similarly, the relationship between performance and training volume also exhibits an inverted U shape.

Are the sweet spots fixed by our genes?

These observations raise several important questions.  First, is the sweet spot at which training volume produces maximum health benefits the same as the sweet spot that produces peak performance?  Secondly, is the location of either or both of these two sweet spots adjustable.  In other words, can we move the sweet spots to a higher training volume by appropriate training strategy, thereby promoting greater heath and greater peak performance.

The answer to the second question is almost certainly yes, a sensible training strategy that builds up training volume gradually allows us to cope with a higher training volume and hence to achieve a higher peak performance.   It is also probable that we can move the sweet spot for health benefit to a higher level.  To maximize our chance of success in moving either or both of the sweet spots to a high training volume, we need to understand something about the way in which training produces either benefit or harm.

The nature of the training effect

It is widely accepted that training produces stress on the body which results in damage and a short term loss of power but produces a adaptive response during subsequent recovery that makes us fitter than before.  There are several mechanism that produce this training effect.

First, oxidative metabolism (required to fuel muscle contraction) produces  reactive molecules known as reactive oxygen species (ROS)  that have the potential to damage the proteins and the DNA in the cells of our body by a process of oxidation.  However, the production of moderate quantities of ROS mobilizes the body’s defense against ROS and this can make the body’s tissues even stronger.  (Incidentally, the probable reason that antioxidant supplements are not generally beneficial is that such supplements might suppress the body’s own defense system.).

Secondly, mechanical damage to muscle fibres releases chemical messengers that mobilize satellite cells (precursors of muscle fibres) causing them to fuse with muscle fibres resulting in hypertrophy. 

In a more general sense, anything that causes tissue damage – either gross trauma or release of local toxins, is likely to produce an inflammatory response which promotes repair, but if there is inadequate opportunity for the repair process, the inflammation might become chronic and produce long lasting damage, as discussed in my post two weeks ago.

The nature of the training response provides support for several of the strategies that have been developed by coaches and athletes on the basis of trial and error, and offers guidance in how best to deploy these strategies.

Assessing adequate recovery 

The most important principle is that training must be followed by adequate recovery.  However, this principle leaves two questions unanswered.  First: how much recovery is needed.  The traditional answer is to listen to the body and indeed that is a good starting point.  However it is almost certain that we can do better by monitoring apsects of body physiology that reflect the amount of stress the body is experiencing.  Of these measurements, Heart Rate (HR) and Heart Rate Variability (HRV) are of the greatest practical utility for the amateur athlete.  However interpretation of both HR and HRV has several pitfalls and I am currently in the process of reviewing the evidence for the best way of dealing with these pitfalls.

Optimizing training strategy 

The second unanswered question is what program allows us to achieve the optimum balance between training and recovery. I do not think there is a single answer.  It depends on the individual and on their goals.  In particular, the question of focus on high volume or high intensity has no single answer.   When the main goal is increasing speed at the anaerobic threshold (probably the most important predicator of performance over distances from 5Km to the half marathon) either high volume, high intensity or a combination of both can be effective.  However this does not mean that it does not matter what program we follow. It means we need to be a little subtle in making our choices to give the best result.   When the goal of maintaining a moderately high level of performance over a period of many years  is added to the ‘wish list’ we need to be even more subtle in making our choices.

The major options are:

1)      periodization – an essential feature introduced in a systematic way by Arthur Lydiard.  While Lydiard’s observation of the need to divide the training program into base-building and peak periods appears very sound, the science that Lydiard introduced to justify the nature of the base-building period is questionable, and I am still dubious about the universal suitability of any simplistic conclusion such as the proposal that a large training volume is the most important feature of base building.

2)      Variety within each period of a periodized program.  In fact Lydiard did recommend variety and in particular, included some speed work and hill work even during base building.  I think the scientific evidence (which I reviewed in several of my posts in April 2009) suggesting that at least a moderate amount of work above the anaerobic threshold does not damage the aerobic base, provides us with even greater flexibility in planning the base-building phase than Lydiard might have proposed .

3)      Cross-training.  In light of the clear evidence that eccentric muscle contraction produces substantial damage to muscles,  I believe that cross training is a very valuable way of producing useful metabolic adaptations and increases in capillary density that improve aerobic capacity, while producing minimal eccentric damage to muscle.  Hence, I do a substantial amount of my aerobic training on the elliptical cross trainer.  There are several caveats that must be added.  First, elliptical  training and running both load the heart, so it is important to make sure that sparing the leg muscles does not result in undue load on the heart.  I use HR and HRV to monitor this. Secondly, it is important to ensure that the leg muscles have been adequately conditioned to cope with eccentric stress before racing, so a substantial amount of running is essential at all stages of the program and running should become the major component in race preparation.

The observation that elite athletes tend to have a limited duration of tenure at the top of the rankings, even after allowing for age, suggests that the strategies used by many elite athletes have not been optimal for ensuring longevity.   Maybe this is because it is impossible to make the two ‘sweet spots’ coincide.  Perhaps the optimum training load for achieving one’s peak performance is higher than the optimum training load for maximizing longevity.  I suspect this might be the case, but nonetheless, think it is worth while to do what we can to bring the two sweet spots as close together as possible, so that we can achiever our peak performance with only minimal damage to our potential longevity – longevity of both running performance and general health .  Perhaps the most important ingredient for achieving this is the judicious use of rest or low intensity training when the body is showing signs of stress.  This brings us back to the question of monitoring body stress, a topic to which I will return subsequently.

Sloppy snow and reactive oxygen

February 5, 2009

In the past eight years, snow in the east midlands has never remained on the ground for more than 24 hours, but this week we have had lingering snow for five days. It snowed again last night and continued in the morning. Unlike the initial flurries of crisp polar snow on Sunday, today’s precipitation was typical sloppy English snow, created as a result of a weakening stream of cool sub-arctic air from the east meeting warm moist maritime air from the southwest. When I set out for work the slope down the escarpment to the river was treacherous, so I decided to run to work rather than cycle. The riverside path was delightful as the fresh snow was largely undisturbed, but elsewhere was mush. Apart from free flowing traffic on the few major roads that had been gritted, the roads were in chaos. I am sure that running was preferable to any other form of transport today. The round journey to and from work is about 15 Km. I was pleased that there was no trace of discomfort in my hamstring

This week, in my continuing attempt to assemble the evidence about possible long term muscle damage from forms of training such as downhill running; plyometrics or simple long runs, that entail large amounts of eccentric muscle contraction, I have been looking into the mechanism of damage at the cellular level. There is very strong evidence, which I will review in greater detail in a later post, that sudden increases in amount of strenuous exercise cause damage to muscle via the generation of re-active forms of oxygen – various atomic and molecular forms of oxygen with an extra electron attached. These are generated by aerobic metabolism and cause damage within the muscle fibres. Furthermore, the generation of these reactive forms of oxygen is much greater in the elderly. At first sight, this appears to provide clear-cut support for the value of supplementary antioxidants such as vitamin C.

However, as with almost all processes in the human body, there are natural compensation processes. Training helps build up chemical processes that neutralise the reactive forms of oxygen. Inappropriate supplementation with anti-oxidants might at least in principle diminish this natural intrinsic protective process. On the other hand, in the elderly, the development of natural ability to counteract over-reactive forms of oxygen is more sluggish. So far, I have not managed to identify evidence establishing whether eccentric and concentric muscle contractions differ in their ability to promote protection.

So what can we conclude. As is often the case, the evidence is not adequate to allow definitive conclusions, but several guidelines appear sensible.

1) this evidence confirms that sudden increases in training load are more likely to result in long term damage, even when there is no overt injury.

2) slow build up of training load is likely to help build up of the ability to counteract oxidative damage.

3) plentiful natural anti-oxidants in the diet (eg vitamin C from citrus fruits etc) are likely to be beneficial. As an ‘oldie’, I am veering towards adding supplements, but want to look further into the possible danger of suppression of intrinsic defence mechanisms before consuming an amount in excess of that in the diet that humans have adapted to over our evolutionary history.

Polar snow

February 1, 2009

Ewen, as you imply, De Castella’s demanding 10 mile sessions including fairly fast down-hill running at Stromlo almost certainly contributed to his good performance in Boston, but it is intriguing to speculate that they might also have contributed to the fact that he is now ‘well and truly retired’ at 51. Of course there are many possible reasons why a former world record holder might choose to take it easy as middle age approaches.

However, it is disconcerting that some evidence indicates that elite athletes who stop training tend to deteriorate faster in middle age than sedentary individuals.   In a comparison of 64 sedentary men with 89 endurance-trained men, Pimentel and colleagues (Journal of Applied Physiology, volume 94, pp 2406-2413) found a more rapid decline in maximum oxygen uptake (VO2max) after age 50 in the endurance trained men. Not surprisingly the accelerated deterioration was associated with reduced training volume, though the causal mechanism was not established. In a 30 year follow-up study of men who had participated in the 1966 Dallas bed rest study in their youth, McGuire and colleagues (Circulation, 104, 1350-1357, 2001) found that the cardiovascular deterioration due to 3 decades of aging was less than that due to 3 weeks of bed-rest at age 20. Of special note with regard to the mechanism of deterioration, they found that the decrease in VO2max could be attributed mainly to decreased ability of muscle to extract oxygen from blood. In other words, the deterioration with age was largely due to deterioration within the muscles, though whether this deterioration was merely a loss of aerobic enzymes, or to the loss of fibres, and/or capillaries is unknown.

In the 4th edition of his book ‘Lore of Running’, Tim Noakes proposes that the springiness of muscles is significantly compromised by large numbers of runs over 21k, and he advocates that runners seeking a sustained quality running career should minimise eccentric muscular damage. So, I will continue to be cautious about forms of training that focus on eccentric contraction. Maybe the most important thing is allowing adequate recovery, especially when the legs start to show signs of cumulative fatigue over several consecutive days.

With regard to the recovery from my recent hamstring injury, my easy 7 Km run today went well despite the rather stripey weather. Fitful sunshine alternated with brief flurries of snow. Although the flurries were brief, the flakes were small and compact, typical of polar snow borne by a north-easterly airstream. Nonetheless it was good to be out of doors. After a gentle warm up, I gradually increased pace up to 5:30 min per Km for the 6th Km and was not aware of any discomfort in my hamstring. So I hope that after a week or two of gentle running I will be able to return to moderately intense efforts by mid February.

Downhill running, eccentric contractions and torn hamstrings

January 31, 2009

A week or two ago, Ewen’s comment about down-hill running led me to pose the question about the balance of risks and benefits of fast downhill running – and more generally, the risks of long term muscle damage from forms of training that focus on eccentric contraction. I am still looking into the scientific evidence on this question – but it is a tricky subject because the type of long term damage that I am concerned about might not emerge for years and it is virtually impossible to conduct rigorous comparative studies over that time scale. There is little doubt that at least some people who do a lot of very demanding training do eventually suffer long term damage (see Grobler et al, Br J Sports Med, 38,697-703). This seems to be a greater problem for runners than cyclists, but whether this is due to eccentric loads, is not easy to establish.

Even more important for the average runner, is whether a minor, but nonetheless appreciable degree of long-term damage occurs in the majority of runners, but this is even harder to establish. In a fairly recent review Brancaccio and colleagues point out that downhill running is associated with increased release of the muscle protein creatine kinase into the blood-stream and that this might be associated with significant damage of muscles. They conclude that in individuals with evidence of muscle damage, intense prolonged exercise may produce negative effects, as it does not induce the physiological muscle adaptations to physical training given the continuous loss of muscle proteins. (Br Med Bull. 2007;81-82:209-30). I am still working on assembling as good a collection of evidence as possible.

Meanwhile I have been grateful to receive interesting comments from several people based on their experiences. Thomas, who is preparing for Boston this year, commented on a friend who had injured himself with down hill running in preparation for Boston, with its notorious down-hills.   Jason pointed out that he has experienced a gradual reduction in the DOMS produced by downhill running after a  moderate amount of down hill running at speed incorporated within his overall training plan. I think that taken together, these comments confirm that downhill running at speed has some risk, but a period of moderate amounts of fast downhill training leads to adaptive changes so that it is no longer as damaging. This of course is the training effect that is the goal of most training programs, and makes sense. It suggests that someone preparing for Boston might benefit from a moderate amount of down hill running.

It still leaves open the question of very long term consequences – but even if very long term adverse consequences are possible, this is merely a specific example of the risk that we take with any demanding training – and it is probable that gradual adaptation is less damaging than a single major stressful event without adequate preparation. So on balance, if I were preparing for Boston, I think I would incorporate a moderate amount of downhill running at marathon pace or slightly faster, into my program. However, a least until I have a better understanding of the situation I am trying to minimize the amount of high demand eccentric work in my program, and in particular, I do only a very small amount of light plyometrics.

However, my experience last week when I went all-out to hold off a challenge in a 100 metres sprint late in the afternoon with inadequate warm-up despite rapidly dropping air temperature, an hour or so after a mile race, illustrated the fact that subjecting ones muscles to large eccentric loads without adequate preparation is almost certainly more damaging than a gradual build up of eccentric loading in a sensible training program. However, as I remarked last week, sometimes life is more fun when you throw caution to the winds and I do not regret the choice I made, despite the fact that my wings have been clipped for the time being.

In fact my injury is not too bad; my hamstring complains a bit if I inadvertently stretch it when bending down, but it feels OK when jogging a few metres. Tomorrow I will probably go for a short, slow paced run.

Dancing with the devil in 2008

December 31, 2008

Despite lingering constriction of the throat and sore ears, I decided it was time to recommence moderate intensity running today with three 1Km repeats at 5K pace (around 4:20 min/Km ). I did the first two repetitions in 4:20 and 4:21 at mean heart rate 142 and 143. Then during the third, the cold air upset my airways and I started to develop an asthma attack. I completed that repetition in 4:32 at HR 144, and jogged home. I had taken one puff of my long-acting inhaler early in the morning, but it appears that in this weather I should increase the dose. Nonetheless I was quite pleased that despite my recent bout of flu, my pace and heart rate in the first two repetitions were similar to the recordings in a 6x1Km session on 6th December.


This a picture of the Trent beside the path where I do 1Km repeats, taken in May of this year. Ignore the incorrect date setting on my camera.

I had started this year with the intention of developing a running style that minimised risk of injury, especially the risk of further damage to my right knee which has been ravaged by intermittent attacks of inflammatory arthritis since childhood. I have read fairly widely and been influenced by an number of authors, especially Gordon Pirie, and to a lesser extent by Nicholas Romanov, the inventor of the Pose technique. Some of the principles of Pose appear to violate the laws of physics, especially the laws of conservation of energy and momentum, but despite its shaky theoretical foundation, the experimental evidence confirms that Pose places less stress on the knee. The study by Arendse and colleagues from Tim Noakes laboratory in Cape Town demonstrated a 50% reduction in work done at the knee joint when running Pose style compared with heel striking. So I attempted to work out a rational basis for a running style including what I regarded as the good features advocated by Pirie and by Romanov, but with a rational foundation based on the laws of physics. The details are described in the articles in the side bar entitled ‘A dance with the devil’

Unfortunately I have had a somewhat disappointing year, mainly due to respiratory problems, and it is difficult to judge whether my running style has been a success. The good news is that my knee is coping fine. It has caused far less trouble this year than it did in my mid-fifties before I recommenced running; and also, less trouble than last year, when I had remained injury free for most of the year, but suffered a flare up of arthritis in late summer that caused me to descend stairs on all fours for a week or so. That was the final impetus to develop a safer running style.

While my knee has been virtually trouble free this year, I have suffered two minor injuries. In neither case was running style the main cause, though in each case, I think that my new style did exacerbate the problem.

One of the crucial features that I learned from Pirie and Romanov is the principle of lifting the foot from stance as quickly as possible. This ensures that much of the energy stored in the muscles, tendons and ligaments at footfall is recovered via elastic recoil. Because muscle and tendon is viscoelastic, recovery of energy is greatest when recoil is rapid. The time on stance should be around 100 milliseconds or less. If cadence is around 180 steps per minute, which experience indicates provides optimum efficiency, the duration of each stride is 333 milliseconds. Therefore, when time on stance is less than 100 milliseconds, the major part of the stride is spent airborne. The vertical ground reaction force that must balance the downwards impulse of gravity acts only for a minor fraction of the gait cycle, and hence this ground reaction force will be several times body weight. For cadence180 steps per minute and time on stance of 100 milliseconds, the average vertical ground reaction force will be three times body weight, and for shorter times on stance, it will be even greater.

Another, more controversial feature borrowed from Pirie and Romanov is that it is most efficient to land on the forefoot. If vertical reaction force is over three times body weight, this places a substantial load on the ball of the foot. In addition, a substantial tension develops in the Achilles tendon. If the knee is slightly flexed at the time, a large portion of this strain is taken by soleus, the deeper of the two calf muscles. Unlike the larger gastrocnemius muscle, soleus does not cross the knee. Hence when the knee is flexed, tension in gastrocnemius is low and soleus bears the main burden.

So how did the principles I borrowed from Pirie and Romanov contribute to my injuries? The main culprit for the first injury was a stone on the path. I landed with the stone beneath the head of my second metatarsal, and a force greater than three times my body weight was transmitted directly onto the metatararsal head and adjacent metatarso-phalangeal joint. Ouch! This part of my foot has always been vulnerable and I was hobbling for several weeks afterwards.

The other injury occurred when I made a fairly rapid increase in the intensity of my training. After several months of low intensity training, I did two speed sessions within a week. In the second session, while running fairly fast uphill over irregular ground, I suffered a minor tear of soleus. Undoubtedly the main culprit was increasing intensity too quickly. I had not conditioned soleus adequately for the task, so I do not think that I should lay too much blame on my change of running style. However, this injury emphasises the importance of conditioning soleus well before increasing training intensity. It also suggests that when running longer distances, it is best to allow the heel to touch the ground in mid stance to avoid the risk that repeated micro-tears will combine to produce a full-blown tear of the muscle.

Despite the occasional tribulations, I am still greatly enjoying running and looking forward to an even better year, next year.

Happy New Year and good running in 2009.

Running in the twilight

November 23, 2008

In the past few days a northerly air-stream has brought the temperature down, and the dawn light this morning revealed a slight dusting of snow in the roof tops – a rare occurrence in the English Midlands in recent years. The ground temperature was still too high to allow the snow to settle. Midmorning, there was another flurry of soggy snow flakes but then sun came out and the temperature rose a degree or two. I was busy doing various odd and ends about the house, and daylight was fading rapidly by the time I got ready to go for a run. I decided that it would be best to follow the fairly open path beside the River Trent to make the most of the last glimmer of sunset reflected off light cloud above and the water beside me, rather than risk tripping over tree roots in the evening gloom on my more usual woodland route.

The open river bank path exposed me to the wind which had swung to the west and strengthened. Any gain in temperature due to the replacement of Arctic air from the north with Atlantic air from the west was negated by the increased wind chill factor. I was wearing light-weight shorts but also a long sleeve top and gloves, so with the wind at my left shoulder on the long reach of the river as it flows northeast from Clifton village to Wilford, I felt comfortable but invigorated. It was tempting to increase the pace, but bearing in mind my recent muscle problems, I restrained myself to a gently fluctuating pace ranging from about 6min per Km to 5 min per Km (for a 100 metre or so at a time). There was no trace of discomfort from my right calf.

As I passed a clump of trees just before reaching the point where Fairham Brook joins the Trent, I was subjected to scrutiny by a predatory owl. He swooped down silently from behind me, but clearly recognized I was far too big a morsel for supper, and wheeled away over the river. At Fairham Brook I turned for home. Now the chilly wind was on my right cheek and I realised that I was starting to wheeze. I have not yet adopted the habit of using my salbutamol inhaler regularly when I run, but it was the exacerbation of my asthma by cold air last winter that led to the need for anti-asthma medication for the first time in my life. So today’s run was a valuable reminder that this winter I should use my inhaler before running whenever the temperature is low.

My distance today was only a little over 6Km and peak pace was no faster than my estimate of my current marathon pace, but it was a good session. There was no trace of pain in my calf; I learned a little more about managing my asthma. I celebrate the fact that as I approach the twilight years of late middle age, I can still enjoy running in the countryside. But I have a wistful awareness that the carefree days when I could run, play football or climb mountains without concern for my body are now well behind me. In those long-gone days I had many mild problems – congenitally twisted toes; inflammatory arthritis; mild asthma – but those problems were a trivial nuisance that never amounted to anything disabling. I am certainly glad that I never let concern for these minor infirmities become a reason to wrap myself in cotton wool. Even now I think that I am likely to remain healthiest by maintaining a fairly demanding training schedule. I just need to learn the difference between the incidental twinges that are part of growing old and a significant muscle tear such as I suffered a few weeks ago.

Running again

November 15, 2008

It was great to go for an easy paced 5 Km run today. Though a wintery sun struggled to penetrate the clouds and fine droplets of rain hung in the air, a variegated pattern of russet, brown and gold carpeted the woodland floor. In places the vivid brown of beech leaves dominated, elsewhere it was the dull brown of oak or the brilliant gold of maple. Under the stand of larch, a shower of fresh yellow needles added to the deep pile that has accumulated over the years creating my favourite running surface.

In light of my recent calf troubles, I ran with small steps, focusing on lifting my foot from the ground rather than actively pushing off to initiate each airborne phase. There was a scarcely perceptible ache in my right calf, concentrated in the vicinity of the muscle tissue damaged by the vicious nocturnal cramp I had suffered recently, in the week following a mild tear of my right soleus muscle arising for an injudicious increase training intensity a month ago. It appears that my lingering troubles are more closely related to the nocturnal cramp than the original injury. It is difficult to know whether the two events were even related. The primary precipitant for the nocturnal cramp was apparently dehydration, and the cramp affected both legs simultaneously. However, it is possible that irritability of the previously injured muscle contributed to greater damage on the right. Whatever the cause, it is clear that I must take things easily in the next few weeks, but I am happy to be running again.