The Role of Intensity and Duration in Endurance Training

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/233855836

Intervals, Thresholds, and Long Slow Distance: the Role of Intensity and Duration in Endurance Training

Article · January 2009

CITATIONS READS 180 26,599

2 authors:

Stephen Seiler Espen Tønnessen Universitetet i Agder Kristiania University College

175 PUBLICATIONS 4,680 CITATIONS 81 PUBLICATIONS 2,288 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Connecting physiological and perceptual responses during HIIT View project

Prospective longitudinal study investigating the degree and the impact of stressors related to athletic training, School and everyday life on health and performance of upper secondary school athletes View project

All content following this page was uploaded by Espen Tønnessen on 21 May 2014.

The user has requested enhancement of the downloaded file. SPORTSCIENCE · sportsci.org Perspectives / Training Intervals, Thresholds, and Long Slow Distance: the Role of Intensity and Duration in Endurance Training Stephen Seiler1 and Espen Tønnessen2 Sportscience 13, 32-53, 2009 (sportsci.org/2009/ss.htm) 1 University of Agder, Faculty of Health and Sport, Kristiansand 4604, Norway. Email. 2 Norwegian Olympic and Paralympic Committee National Training Center, Oslo, Norway. Email. Reviewers: Iñigo Mujika, Araba Sport Clinic, Vitoria, Spain; Stephen Ingham, English Institute of Sport, Loughborough Uni- versity, Leicestershire, LE11 3TU, UK.

Endurance training involves manipulation of intensity, duration, and frequency of training sessions. The relative impact of short, high-intensity training versus longer, slower distance training has been studied and debated for decades among athletes, coaches, and scientists. Currently, the popularity pendulum has swung towards high-intensity interval training. Many fitness experts, as well as some scientists, now argue that brief, high-intensity interval work is the only form of training necessary for performance optimization. Research on the impact of interval and continuous training with untrained to moderately trained subjects does not support the current interval craze, but the evidence does suggest that short intense training bouts and longer continuous exercise sessions should both be a part of effective endurance training. Elite endurance athletes perform 80 % or more of their training at intensities clearly below their lactate threshold and use high-intensity training surprisingly sparingly. Studies involving intensification of training in already well-trained athletes have shown equivocal results at best. The available evidence suggests that combining large volumes of low-intensity training with careful use of high-intensity interval training throughout the annual training cycle is the best-practice model for development of endurance performance. KEYWORDS: lactate threshold, maximal oxygen uptake, VO2max, periodization. Reprint pdf · Reprint doc · Reviewer's Commentary

Interval Training: a Long History ...... 33 Exercise Intensity Zones...... 35 Training Plans and Cellular Signaling...... 36 Training Intensities of Elite Endurance Athletes ...... 38 Units for Training Intensity ...... 41 The 80-20 Rule for Intensity...... 42 Training Volume of Elite Athletes...... 43 Intensified-Training Studies ...... 44 Intensity for Recreational Athletes ...... 46 Case Studies of Training Manipulation ...... 47 Case 1–From Soccer Pro to Elite Cyclist ...... 47 Case 2–From Modern Pentathlete to Runner ...... 48 Valid Comparisons of Training Interventions ...... 49 Conclusions...... 49 References...... 50

The evening before the start of the 2009 defended his dissertation. The other had played European College of Sport Science Congress in the adversarial role of “førsteopponent.” Tøn- Oslo, the two of us were sitting at a doctoral nessen’s research on the talent development dissertation defense dinner that is part of the process included extensive empirical analyses time honored tradition of the “doctoral dispu- of the training characteristics of selected world tas” in Scandinavia. One of us was the relieved champion female endurance athletes. His career disputant (Tønnessen) who had successfully case-study series systematized training diary

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 33 logs of over 15,000 training sessions from three broader question of training intensity distribu- World and/or Olympic champions in three tion in competitive endurance athletes. sports: distance running, cross-country skiing, Interval Training: a Long History and orienteering. Common for all three cham- International running coach Peter Thompson pions was that over their long, successful ca- wrote in Athletics Weekly that clear references reers, about 85 % of their training sessions were to “repetition training” were seen already by the performed as continuous efforts at low to mod- early 1900s (Thompson, 2005). Nobel Prize erate intensity (blood lactate ≤2 mM). Among winning physiologist AV Hill incorporated the 40 guests sat coaches, scientists, and former intermittent exercise into his studies of exercis- athletes who had been directly or indirectly ing humans already in the 1920s (Hill et al., involved in winning more endurance sport 1924a; Hill et al., 1924b). About this time, Olympic gold medals and world championships Swede Gosta Holmer introduced Fartlek to than we could count. One guest, Dag Kaas, had distance running (fart= speed and lek= play in coached 12 individual world champions in four Swedish). The specific term interval training is different sports. In his toast to the candidate he attributed to German coach Waldemer Ger- remarked, ”My experience as a coach tells me schler. Influenced by work physiologist Hans that to become world champion in endurance Reindell in the late 1930s, he was convinced disciplines, you have to train SMART, AND you that alternating periods of hard work and recov- have to train a LOT. One without the other is ery was an effective adaptive stimulus for the insufficient.” heart. They apparently adopted the term be- So what is smart endurance training? The cause they both believed that it was the recov- question is timely: research and popular interest ery interval that was vital to the training effect. in interval training for fitness, rehabilitation, Since then, the terms intermittent exercise, and performance has skyrocketed in recent repetition training, and interval training have years on the back of new research studies and all been used to describe a broad range of train- even more marketing by various players in the ing prescriptions involving alternating work and health and fitness industry. Some recent inves- rest periods (Daniels and Scardina, 1984). In tigations on untrained or moderately trained the 1960s, Swedish physiologists, led by Per subjects have suggested that 2-8 wk of 2-3 Åstrand, performed groundbreaking research times weekly intense interval training can in- demonstrating how manipulation of work dura- duce rapid and substantial metabolic and car- tion and rest duration could dramatically impact diovascular performance improvements physiological responses to intermittent exercise (Daussin et al., 2007; Helgerud et al., 2007; (Åstrand et al., 1960; Åstrand I, 1960; Christen- Talanian et al., 2007). Some popular media sen, 1960; Christensen et al., 1960). As Daniels articles have interpreted these findings to mean and Scardina (1984) concluded 25 years ago, that long, steady distance sessions are a waste their work laid the foundation for all interval of time. Whether well founded or not, this in- training research to follow. In their classic terpretation raises reasonable questions about chapter Physical Training in Textbook of Work the importance and quantity of high- (and low-) Physiology, Åstrand and Rodahl (1986) wrote, intensity training in the overall training process “it is an important but unsolved question which of the endurance athlete. Our goal with this type of training is most effective: to maintain a article is to discuss this issue in a way that inte- level representing 90 % of the maximal oxygen grates research and practice. uptake for 40 min, or to tax 100 % of the oxy- In view of the recent hype and the explosion gen uptake capacity for about 16 min.” (The in the number of studies investigating interval same chapter from the 4th edition, published in training in various health, rehabilitation, and 2003, can be read here.) This quote serves as performance settings, one could be forgiven for an appropriate background for defining high assuming that this training form was some intensity aerobic interval training (HIT) as we magic training pill scientists had devised com- will use it in this article: repeated bouts of exer- paratively recently. The reality is that athletes cise lasting ~1 to 8 min and eliciting an oxygen have been using interval training for at least 60 demand equal to ~90 to 100 % of VO2max, years. So, some discussion of interval training separated by rest periods of 1 to 5 min (Seiler research is in order before we address the and Sjursen, 2004; Seiler and Hetlelid, 2005).

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 34

Controlled studies comparing the physiological Poole and Gaesser (1985) published a cita- and performance impact of continuous training tion classic comparing 8 wk of 3 × weekly (CT) below the lactate turnpoint (typically 60- training of untrained subjects for either 55 min 75 % of VO2max for 30 min or more) and HIT at 50 %VO2max, 35 min at 75 %VO2max, or 10 began to emerge in the 1970s. Sample sizes × 2 min at 105 %VO2max with 2-min recover- were small and the results were mixed, with ies. They observed no differences in the magni- superior results for HIT (Henriksson and Reit- tude of the increase in either VO2max or power man, 1976; Wenger and Macnab, 1975), supe- at lactate threshold among the three groups. rior results for CT (Saltin et al., 1976), and little Their findings were corroborated by Bhambini difference (Cunningham et al., 1979; Eddy et and Singh (1985) in a study of similar design al., 1977; Gregory, 1979). Like most published published the same year. Gorostiaga et al. studies comparing the two types of training, the (1991) reported findings that challenged CT and HIT interventions compared in these McDougall and Sale's conclusions regarding studies were matched for total work (iso- the adaptive specificity of interval and continu- energetic). In the context of how athletes actu- ous training. They had untrained subjects exer- ally train and perceive training stress, this situa- cise for 30 min, three days a week either as CT tion is artificial, and one we will return to. at 50 % of the lowest power eliciting VO2max, McDougall and Sale (1981) published one of or as HIT, alternating 30 s at 100 % of power at the earliest reviews comparing the effects of VO2max and 30 s rest, such that total work was continuous and interval training, directed at matched. Directly counter to McDougall and coaches and athletes. They concluded that both Sales conclusions, they found HIT to induce forms of training were important, but for differ- greater changes in VO2max, while CT was ent reasons. Two physiological assumptions more effective in improving peripheral oxida- that are now largely disproven influenced their tive capacity and the lactate profile. At the be- interpretation. First, they concluded that HIT ginning of the 1990s, the available data did not was superior for inducing peripheral changes, support a consensus regarding the relative effi- because the higher work intensity induced a cacy of CT vs HIT in inducing peripheral or greater degree of skeletal muscle hypoxia. We central changes related to endurance perform- now know that in healthy subjects, increased ance. lactate accumulation in the blood during exer- Twenty years on, research continues regard- cise need not be due to increased muscle hy- ing the extent to which VO2max, fractional poxia (Gladden, 2004). Second, they concluded utilization of VO2max, and work effi- that since stroke volume already plateaus at 40- ciency/economy are differentially impacted by 50 %VO2max, higher exercise intensities would CT and HIT in healthy, initially untrained indi- not enhance ventricular filling. We now know viduals. Study results continue to be mixed, that stroke volume continues to rise at higher with some studies showing no differences in intensities, perhaps even to VO2max, in well peripheral and central adaptations to CT vs HIT trained athletes (Gledhill et al., 1994; Zhou et (Berger et al., 2006; Edge et al., 2006; Overend al., 2001). Assuming a stroke volume plateau at et al., 1992) and others greater improvements low exercise intensity, they concluded that the with HIT (Daussin et al., 2008a; Daussin et al., benefit of exercise on cardiac performance was 2008b; Helgerud et al., 2007). When differ- derived via stimulation of high cardiac contrac- ences are seen, they lean in the direction that tility, which they argued peaked at about 75 continuous work at sub-maximal intensities %VO2max. Thus, moderate-intensity continu- promotes greater peripheral adaptations and ous exercise over longer durations and therefore HIT promotes greater central adaptations more heart beats was deemed most beneficial (Helgerud et al., 2007). for enhancing cardiac performance. While Controlled studies directly comparing CT newer research no longer supports their spe- and HIT in already well-trained subjects were cific conclusions, they did raise the important essentially absent from the literature until re- point that there are underlying characteristics of cently. However, a few single-group design the physiological response to HIT and CT that studies involving endurance athletes did emerge should help explain any differential impact on in the 1990s. Acevedo and Goldfarb (1989) adaptive responses. reported improved 10-km performance and

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 35 treadmill time to exhaustion at the same pace cation to two CT sessions, three HIT sessions up a 2 % grade in well-trained runners who and one LT session each week gave no addi- increased their training intensity to 90-95 tional adaptive benefit, but did increase subjec- %VO2max on three of their weekly training tive training stress and indicators of impending days. In these already well-trained athletes, overtraining (Billat et al., 1999). In fact, train- VO2max was unchanged after 8 wk of training ing intensification over periods of 2-8 wk with intensification, but a right shift in the blood frequent high-intensity bouts (3-4 sessions per lactate profile was observed. In 1996 -97, South week) is an effective means of temporarily African sport scientists published the results of compromising performance and inducing over- a single group intervention involving competi- reaching and possibly overtraining symptoms in tive cyclists (Lindsay et al., 1996; Weston et al., athletes (Halson and Jeukendrup, 2004). There 1997). They trained regionally competitive is likely an appropriate balance between high- cyclists who were specifically selected for and low-intensity training in the day-to-day study based on the criteria that they had not intensity distribution of the endurance athlete. undertaken any interval training in the 3-4 These findings bring us to two related ques- months prior to study initiation. When 15 % of tions: how do really good endurance athletes their normal training volume was replaced with actually train, and is there an optimal training 2 d.wk-1 interval training for 3-4 wk (six train- intensity distribution for long-term performance ing sessions of six 5-min high intensity work development? bouts), 40-km time trial performance, peak While arguments can be made that tradition, sustained power output (PPO), and time to resistance to change and even superstition may fatigue at 150 %PPO were all modestly im- negatively influence training methods of elite proved. Physiological measurements such as endurance athletes, sports history tells us that VO2max and lactate profile changes were not athletes are experimental and innovative. Ob- reported. Stepto and colleagues then addressed serving the training methods of the world's best the question of interval-training optimization in endurance athletes represent a more valid pic- a similar sample of non-interval trained, re- ture of “best practice” than we can develop gional cyclists (Stepto et al., 1999). They com- from short-term laboratory studies of untrained pared interval bouts ranging from 80 to 175 % or moderately trained subjects. In today’s per- of peak aerobic power (30 s to 8 min duration, formance environment, where promising ath- 6-32 min total work). Group sizes were small letes have essentially unlimited time to train, all (n=3-4), but the one group that consistently athletes train a lot and are highly motivated to improved endurance test performance (~3 %) optimize the training process. Training ideas had used 4-min intervals at 85 % PPO. These that sound good but don't work in practice will controlled training intensification studies essen- fade away. Given these conditions, we argue tially confirmed what athletes and coaches that any consistent pattern of training intensity seemed to have known for decades: some high- distribution emerging across sport disciplines is intensity interval training should be integrated likely to be a result of a successful self- into the training program for optimal perform- organization (evolution) towards a “population ance gains. These studies also seemed to trigger optimum.” High performance training is an a surge in interest in the role of HIT in athlete individualized process for sure, but by popula- performance development that has further tion optimum, we mean an approach to training grown in recent years. organization that results in most athletes staying If doing some HIT (1-2 bouts per week) healthy, making good progress, and performing gives a performance boost, is more even better? well in their most important events. Billat and colleagues explored this question in a Exercise Intensity Zones group of middle distance runners initially train- To describe intensity distribution in endur- ing six sessions per week of CT only. They ance athletes we have to first agree on an inten- found that a training intensification to four CT sity scale. There are different intensity zone sessions, one HIT session, and one lactate schemes to choose from. Most national sport threshold (LT) session resulted in improved governing bodies employ an intensity scale running speed at VO2max (but not VO2max based on ranges of heart rate relative to maxi- itself) and running economy. Further intensifi- mum and associated typical blood lactate con-

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 36 centration range. Research approaches vary, 2003) have employed the first and second venti- but a number of recent research studies have latory turnpoints to demarcate three intensity identified intensity zones based on ventilatory zones (Figure 1). The 5-zone scale in the table thresholds. Here we will examine an example above and the 3-zone scale below are reasona- of each of these scales. bly super-imposable in that intensity Zone 3 in Table 1 shows the intensity scale used by all the 5-zone system coincides well with Zone 2 endurance sports in Norway. A valid criticism in the 3-zone model. While defining five “aero- of such a scale is that it does not account for bic” intensity zones is likely to be informative individual variation in the relationship between in training practice, it is important to note that heart rate and blood lactate, or activity specific they are not based on clearly defined physio- variation, such as the tendency for maximal logical markers. Note also that 2-3 additional steady state concentrations for blood lactate to zones are typically defined to accommodate be higher in activities activating less muscle very high intensity sprint, anaerobic capacity, mass (Beneke and von Duvillard, 1996; Beneke and strength training. These zones are typically et al., 2001). defined as “anaerobic” Zones 6, 7, and 8.

Table 1: A typical five-zone scale to prescribe and moni- Training Plans and Cellular Signaling tor training of endurance athletes. Athletes do not train at the same intensity or Intensity VO2 Heart rate Lactate Duration for the same duration every day. These vari- zone (%max) (%max) (mmol.L-1) within zone ables are manipulated from day to day with the 1 45-65 55-75 0.8-1.5 1-6 h implicit goals to maximize physiological capac- 2 66-80 75-85 1.5-2.5 1-3 h ity over time, and stay healthy. Indeed, the 3 81-87 85-90 2.5-4 50-90 min former is quite dependent on the latter. Training 4 88-93 90-95 4-6 30-60 min frequency is also a critical variable manipulated by the athlete. This is particularly evident when 5 94-100 95-100 6-10 15-30 min comparing younger (often training 5-8 times The heart rate scale is slightly simplified compared to the per week) and more mature athletes at peak actual scale used by the Norwegian Olympic Federation, performance level (often training 10-13 ses- which is based primarily on decades of testing of cross- sions per week). Ramping up training frequency country skiers, biathletes, and rowers. (as opposed to training longer durations each Figure 1. Three intensity zones defined by physiological session) is responsible for most of the increase determination of the first and second ventilatory turnpoints in yearly training hours observed as teenage using ventilatory equivalents for O2 (VT1) and CO2 (VT2). athletes mature. Cycling might be an exception to this general rule, since cycling tradition dic- tates single daily sessions that often span 4-6 h among professionals. The ultimate targets of the training process are individual cells. Changes in rates of DNA transcription, RNA translation, and ultimately, synthesis of specific proteins or protein constellations are induced via a cascade of intracellular signals induced by the training bout. Molecular exercise biologists are unravel- ing how manipulation of intensity and duration of exercise specifically modifies intracellular signaling and resulting protein synthetic rates at the cellular or whole muscle/myocardial level (Ahmetov and Rogozkin, 2009; Hoppeler et al., 2007; Joseph et al., 2006; Marcuello et al., Several recent studies examining training in- 2005; McPhee et al., 2009; Yan, 2009). About tensity distribution (Esteve-Lanao et al., 2005; 85 % of all publications involving gene expres- Seiler and Kjerland, 2006; Zapico et al., 2007) sion and exercise are less than 10 y old, so we or performance intensity distribution in multi- do not yet know enough to relate results of day events (Lucia et al., 1999; Lucia et al., Western blots to the specific training of an

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 37 athlete. adaptations to training (Brigelius-Flohe, 2009; The signaling impact of a given exercise Gomez-Cabrera et al., 2008; Hansen et al., stress (intensity×duration) almost certainly 2005; Ristow et al., 2009; Yeo et al., 2008). It decays with training (Hoppeler et al., 2007; seems fair to conclude that while we suspect Nordsborg et al., 2003). For example, AMP important differences exist, we are not yet able activated protein kinase α2 (AMPK) activity to relate specific training variables (e.g., 60 min jumps 9-fold above resting levels after 120 min vs 120 min at 70 %VO2max) to differences in of cycling at 66 %VO2max in untrained sub- cell signaling in a detailed way. Our view of the jects. However, after only 10 training sessions, adaptive process remains limited to a larger almost no increase in AMPK is seen after the scale. We can still identify some potential sig- same exercise bout (McConell et al., 2005). naling factors that are associated with increased Manipulating exercise intensity and duration exercise intensity over a given duration (Table also impacts the systemic stress responses asso- 2) or increased exercise duration at a given sub- ciated with training. Making this connection is maximal intensity (Table 3). Some of these are further complicated by recent findings suggest- potentially adaptive and others maladaptive. ing that muscle glycogen depletion can enhance There is likely substantial overlapping of ef- and antioxidant supplementation can inhibit fects between extending exercise duration and

Table 2. Key physiological changes associated with an increase in exercise intensity from 70 %VO2max to ≥90 %VO2max for a given exercise duration. Induced change Possible signal Possible positive effect Possible negative effect Increased diastolic Increased myofiber Increased maximal stroke ?? filling and end- stretch/load (Catalucci et al., volume, compensatory ven- diastolic volume 2008; Frank et al., 2008; Pel- tricular wall thickening liccia et al., 1999; Sheikh et al., 2008)a Increased heart Increased rate pressure prod- None likely given superior None likely given superior rate and intraven- uct and myocardial metabolic oxidative capacity of cardiac oxidative capacity of cardiac tricular systolic load (see below) muscle muscle pressure Increased number Increased metabolic activity in Enhanced whole muscle fat Premature fatigue and in- of active muscle faster motor units (transduced oxidation/ right shift in lactate adequate stimulus of low fibers (motor via Cai and high energy phos- turnpoint threshold motor units? units) phate concentration shifts? (Diaz and Moraes, 2008; Hol- loszy, 2008; Ojuka, 2004) Expanded active Local mechanical and meta- A mixture of angiogenesis of ?? vascular bed via bolic signals (Laughlin and arteries, capillaries and veins motor unit activa- Roseguini, 2008) and altered control of vascular tion resistance (Laughlin and Roseguini, 2008) Increased glyco- Decreased intracellular pH Enhanced buffer capacity Premature fatigue at motor lytic rate within (Edge et al., 2006; Weston et unit level and reduced stimu- active fibers al., 1997) lus for oxidative enzyme synthesis Increased sympa- Cell exposure to increased ? Acutely delayed recovery of thetic activation epinephrine and norepineph- ANS (Seiler et al., 2007); rine concentration in blood Chronic down-regulation of (concentration×time) α- and β- adrenergic receptor sensitivity if repeated exces- sively (Fry et al., 2006; Leh- mann et al., 1997) aIf cardiomyocyte stretch induces intracellular signals leading to ventricular hypertrophy, then it is perhaps relevant that the myocardium may be stretched most in the moments of transition from work to recovery when heart rate drops and venous return remains transiently high.

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 38 increasing exercise intensity. time, a successful athlete will presumably or- It may be a hard pill to swallow for some ex- ganize their training in a way that maximizes ercise physiologists, but athletes and coaches do adaptive benefit for a given perceived stress not need to know much exercise physiology to load. That is, we can assume that highly suc- train effectively. They do have to be sensitive to cessful athletes integrate this feedback experi- how training manipulations impact athlete ence over time to maximize training benefit and health, daily training tolerance, and perform- minimize risk of negative outcomes such as ance, and to make effective adjustments. Over illness, injury, stagnation, or overtraining.

Table 3. Key physiological changes associated with increasing exercise duration at a submaximal exercise intensity of 60-70 %VO2max from 45 min to 120 min. Induced change Possible signal Possible positive effect Possible negative effect Increased number Increased stimulus for myelina- Improved technical stability, Technically maladaptive if of movement tion of active motor nerve movement economy race intensity motor pattern repetitions pathways (Fields, 2006; Ishi- were very different? bashi et al., 2006) Increased activa- Increased metabolic activity in Enhanced whole muscle fat ?? tion of fast motor faster motor units (transduced oxidation/ right shift in lactate units due to motor via Cai and high energy phos- turnpoint unit fatigue (Kamo, phate concentration shifts? 2002) (Diaz and Moraes, 2008; Holloszy, 2008; Ojuka, 2004) Enhanced glyco- ?? May amplify signal for syn- Potential accumulation of gen depletion thesis of specific oxidative fatigue if dietary CHO is enzymes (Chakravarthy and insufficient. Booth, 2004; Hansen et al., 2005) Increased relative Large increase in plasma free May amplify signal for mito- ?? fat oxidation fatty acid concentration chondrial biogenesis (Holloszy, 2008)

Training Intensities of Elite Endurance Athletes Empirical descriptions of the actual distribu- ing characteristics of 13 national class male, tion of training intensity in well-trained athletes New Zealand runners with favorite distances have only recently emerged in the literature. ranging from 1500 m to the marathon. They The first time one of us (Seiler) gave a lecture used heart rate data collected during training on the topic was in 1999, and there were few and related it to results from standardized hard data to present, but a fair share of anecdote treadmill determinations of heart rate and run- and informed surmise. Carl Foster, Jack Daniels ning speed at 4-mM blood lactate concentration and Seiler published a book chapter the same (misnamed anaerobic threshold at the time). year, “Perspectives on Correct Approaches to Over a data collection period of 6-8 wk corre- Training” that synthesized what we knew then sponding to the preparation phase, these ath- (read chapter here via Google books). At that letes reported that only 4 % of all training ses- time, much of the discussion and research re- sions were interval workouts or races. For the lated to the endurance training process focused remaining training sessions, average heart rate on factors associated with overtraining (a train- was only 77 % of their heart rate at 4-mM ing control disaster), with little focus on what blood lactate. This heart rate translates to per- characterized “successful training.” The empiri- haps 60-65 % of VO2max. The authors con- cal foundation for describing successful training cluded that while their physiological test results intensity distribution is stronger 10 years later. were similar to previous studies of well trained Robinson et al. (1991) published what was runners, the training intensity of these runners according to the authors “the first attempt to was perhaps lower than optimal, based on pre- quantify training intensity by use of objective, vailing recommendations to perform most train- longitudinal training data.” They studied training at or around the lactate/anaerobic threshold.

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 39

In one of the first rigorous quantifications of ance times in both long and short races were training intensity distribution reported, Mujika highly negatively correlated with total training et al. (1995) quantified the training intensity time in Zone 1. They found no significant cor- distribution of national and international class relation between the amount of high-intensity swimmers over an entire season based on five training and race performance. blood-lactate concentration zones. Despite spe- Rowers compete over a 2000-m distance re- cializing in 100-m and 200-m events requiring quiring 6-7 min. Steinacker et al. (1998) ~60 to 120 s, these athletes swam 77 % of the reported that extensive endurance training (60- 1150 km completed during a season at an inten- to 120-min sessions at <2 mM blood lactate) sity below 2 mM lactate. The intensity distribu- dominated the training volume of German, tion of 400- and 1500-m swim specialists was Danish, Dutch, and Norwegian elite rowers. not reported, but was likely even more Rowing at higher intensities was performed ~4- weighted towards high-volume, low-intensity 10 % of the total rowed time. The data also swimming. suggested that German rowers preparing for the Billat et al. (2001) performed physiological world championships performed essentially no testing and collected data from training diaries rowing at threshold intensity, but instead of French and Portuguese marathoners. They trained either below 2 mM blood lactate or at classified training intensity in terms of three intensities in the 6-12 mM range. speeds: marathon, 10–km, and 3–km. During Seiler collaborated with long time national the 12 wk preceding an Olympic trials mara- team rower, coach, and talent development thon, the athletes in this study ran 78 % of their coordinator Åke Fiskerstrand to examine his- training kilometers at below marathon speed, torical developments in training organization only 4 % at marathon race speed (likely to be among international medal winning rowers near VT1), and 18 % at 10–km or 3–km speed from Norway (Fiskerstrand and Seiler, 2004). (likely to be > VT2). This distribution of train- Using questionnaire data, athlete training dia- ing intensity was identical in high-level (<2 h ries, and physiological testing records, they 16 min for males and <2 h 38 min for females) quantified training intensity distribution in 27 and top-class athletes (<2 h 11 min and <2 h 32 athletes who had won world or Olympic medals min). But the top-class athletes ran more total in the 1970s to 1990s. They documented that kilometers and proportionally more distance at over the three decades: training volume had or above 10–km speed. increased about 20 % and become more domi- Kenyan runners are often mythologized for nated by low-intensity volume; the monthly the high intensity of their training. It is there- hours of high-intensity training had dropped by fore interesting that with data from another one-third; very high intensity overspeed sprint study by Billat et al. (2003), we calculated that training had declined dramatically in favor of elite male and female Kenyan 5- and 10-km longer interval training at 85-95 %VO2max; runners ran ~85 % of their weekly training and the number of altitude camps attended by kilometers below lactate-threshold speed. the athletes increased dramatically. Over this The first study on runners to quantify train- 30-y timeline, VO2max and rowing ergometer ing intensity using three intensity zones was performance improved by ~10 % with no that of Esteve-Lanao et al. (2005). They fol- change in average height or body mass. Most lowed the training of eight regional- and na- of the changes occurred between the 1970s and tional-class Spanish distance runners over a six- 1980s, coinciding with major adjustments in month period broken into eight, 3-wk mesocy- training intensity. cles. Heart rate was measured for every training Most recently, Gullich et al. (2009) de- session to calculate the time spent in each heart- scribed the training of world class junior rowers rate zone defined by treadmill testing. All told, from Germany during a 37-wk period culminat- they quantified over 1000 heart-rate recordings. ing in national championships and qualification On average these athletes ran 70 km.wk-1 dur- races for the world championships (online ing the six-month period, with 71 % of running ahead of print here). These were very talented time in Zone 1, 21 % in Zone 2, and 8 % in junior rowers, with 27 of 36 athletes winning Zone 3. Mean training intensity was 64 medals in the junior world championships that %VO2max. They also reported that perform- followed the study period. Remarkably, 95 % of

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 40 their rowing training was performed below 2 larization observed might have been due to mM blood lactate, based on daily heart-rate better management of intensity (keeping hard monitoring and rowing ergometer threshold training hard and easy training easy) among the determinations performed at the beginning of most successful athletes. This polarization the season. This heavy dominance of extensive might protect against overstress. endurance training persisted across mesocycles. Professional road cyclists are known for per- However, the relatively small volume of Zone 2 forming very high training volumes, up to and Zone 3 work shifted towards higher intensi- 35,000 km.y-1. Zapico and colleagues (2007) ties from the basic preparation phase to the used the 3-intensity zone model to track train- competition phase. That is, the intensity distri- ing characteristics from November to June in a bution became more polarized. It is important group of elite Spanish under-23 riders. In addi- to point out that time-in-zone allocation based tion, physiological testing was performed at on heart-rate cut-offs (the kind of analysis per- season start and at the end of the winter and formed by software from heart watch manufac- spring mesocycles. There was an increase in turers) underestimates the time spent perform- total training volume and a four-fold increase in ing high-intensity exercise and the impact of Zone 3 training between the winter and spring that work on the stress load of an exercise ses- mesocycles (Figure 2), but there was no further sion (Seiler and Kjerland, 2006). Although the improvement in power at VT1, VT2 or at outcomes are biased by this problem, there was VO2max between the end of the winter and still a clear shift in the intensity distribution spring mesocycles (Figure 3), despite the train- towards large volumes of low- to moderate- ing intensification. Anecdotally, this finding is intensity training. We also evaluated retrospec- not unusual, despite the fact that athletes feel tively whether there were any differences in fitter. It may be that VT2 and VO2max determi- junior training characteristics between a sub- nation using traditional methods can miss an group of rowers who went on to win interna- important increase in the duration that can be tional medals as seniors within three years (14 maintained at the associated workloads. of 36 athletes) and the remainder of the sample, Individual and team pursuit athletes in cy- who all continued competing at the national cling compete over about 4 min. The event level. The only physical or training characteris- appeals to sport scientists because the perform- tic that distinguished the most successful row- ance situation is highly controlled and amena- ers from their peers was a tendency to distribute ble to accurate modeling of the variables on their training in a more polarized fashion; that both sides of the power balance equation. is, they performed significantly more rowing at Schumacher and Mueller (2002) demonstrated very low aerobic intensities and at the highest the validity of this approach in predicting “gold intensities. We concluded that the greater po- medal standards” for physiological testing and

Figure 2. Cycling intensity and volume of elite Spanish U23 cyclists training in the period November to June. Data redrawn from Zapico et al. (Zapico et al., 2007). 300 Zone 1 Zone 2 Zone 3

250 21 5 57 200 42 Training 150 duration (h) 100 182 164 50

0 Winter mesocycle Spring mesocycle (Nov‐Feb) (Mar‐Jun) Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 41

Figure 3. Response to periodization of training intensity and volume in elite Spanish U23 cyclists. Physiological test results from tests performed before starting the winter mesocycle (Test 1), at the end of the winter mesocycle (Test 2), and at the end of the Spring mesocycle (Test 3). Data redrawn from Zapico et al. (2007).

500 VO2 max 450 VT2 400 VT1 350 (watts)

300 250

Power 200 150 Test 1Test 2 Test 3 power output in track cycling. However, less ners, we used heart-rate monitoring to quantify obvious from the title was the detailed descrip- all endurance sessions and determined three tion of the training program followed by the aerobic intensity zones based on ventilatory German cyclists monitored in the study, ulti- turn points. We also recorded the athletes' rat- mately earning a gold medal in Sydney in ing of perceived exertion (RPE) using the world-record time. These athletes trained to methods of Foster et al. (1996; 1998; 2001a) for maintain 670 W in the lead position and ~450 all training bouts. Finally, we collected blood W when following using a training program lactate during one training week to relate heart dominated by continuous low to moderate in- rate and perceived exertion measurements to tensity cycling on the roads (29-35,000 km.y-1). blood lactate values. In the 200 d preceding the Olympics, the ath- When comparing the three different intensity letes performed “low-intensity, high-mileage” quantification methods, we addressed the issue training at 50-60 % of VO2max on ~140 d. of how training intensity is best quantified. Stage races took up another ~40 d. Specific Heart-rate monitoring is clearly appealing. We track cycling at near competition intensities was can save heart rate data, download entire work- performed on less than 20 d between March and outs to analysis software, and quantify the time September. In the ~110 d preceding the Olym- heart rate falls within specific pre-defined in- pic final, high-intensity interval track training tensity zones. Using this “time-in-zone” ap- was performed on only 6 d. proach, we found that 91 % of all training time was spent at a heart rate below VT1 intensity, Units for Training Intensity Cross country skiers have rather legendary ~6 % between VT1 and VT2, and only 2.6 % of status in exercise physiology circles for their all 15-s heart rate registrations were performed aerobic capacity and endurance capacity in above VT2. We then quantified intensity by arms and legs. Seiler et al. (2006) studied 12 allocating each training session to one of the competitive to nationally elite male 17–y old three zones based on the goal of the training skiers from a special skiing high school in the and heart rate analysis. We called this the “session-goal approach”. For low-intensity continu- region. The mean VO2max for the group was 72 ml.kg-1min-1. They were guided by coaches ous bouts, we used average heart rate for the with national team coaching experience and entire bout. For bouts designed to be threshold were trained along similar lines to the seniors, training we averaged heart rate during the but with substantially lower volumes of train- threshold-training periods. For high-intensity ing. Like Esteve-Lanao (2005) did with run- interval-training sessions, we based intensity on

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 42 the average peak heart rate for each interval ful endurance athletes. About 80 % of training bout. Using this approach, intensity distribution sessions are performed predominantly at inten- derived from heart rate responses closely sities under the first ventilatory turn point, or a matched the session RPE (Figure 4), training blood-lactate concentration ≤2mM. The remain- diary distribution based on workout description, ing ~20 % of sessions are distributed between and blood-lactate measurements. The agree- training at or near the traditional lactate thresh- ment between the session-by-session heart-rate old (Zone 2), and training at intensities in the quantification and session RPE-based assign- 90-100 %VO2max range, generally as interval ment of intensity was 92 %. In their training training (Zone 3). An elite athlete training 10- diaries, athletes recorded 30-41 training ses- 12 times per week is therefore likely to dedicate sions in 32 d and described 75% of their train- 1-3 sessions weekly to training at intensities at ing bouts as low intensity continuous, 5% as or above the maximum lactate steady state. This threshold workouts, and 17% as intervals. rule of thumb coincides well with training studies demonstrating the efficacy of adding two Figure 4. Comparison of training intensity distribution in interval sessions per week to a training program well trained junior cross-country skiers using traditional (Billat et al., 1999; Lindsay et al., 1996; Weston heart-rate (HR) time-in-zone, session goal HR analysis, et al., 1997). Seiler and Kjerland (2006) have and session rating of perceived exertion (RPE). Time-in- previously gone so far as say that the optimal zone data represents total distribution of training time for intensity distribution approximated a “polarized all athletes combined. Data redrawn from Seiler and distribution” with 75-80 % of training sessions Kjerland (2006). in Zone 1, 5 % in Zone 2, and 15-20 % in Zone 100 Time-in-Zone HR 3. However, there is considerable variation in s how athletes competing in different sports and 90 Session goal HR 80 event durations distribute their training intensity Session RPE 70 within Zones 2 and 3. 60 Why has this training pattern emerged? We 50 do not have sufficient research to answer this 40 question, but we can make some reasonable 30 guesses. One group of factors would involve 20 the potential for this distribution to best stimu- 10

Percent of training session training of Percent late the constellation of training adaptations 0 required for maximal endurance performance. Zone 1 Zone 2 Zone 3 For example, large volumes of training at low intensity might be optimal for maximizing pe-

We have also recently observed the same ripheral adaptations, while relatively small time-in-zone mismatch when quantifying inten- volumes of high intensity training fulfill the sity distribution in soccer training (unpublished need for optimizing signaling for enhanced data). It seems clear that typical software-based cardiac function and buffer capacity. Techni- heart-rate analysis methods overestimate the cally, lots of low intensity training may be ef- amount of time spent training at low intensity fective by allowing lots of repetitions to engrain and underestimate the time spent at very high correct motor patterns. On the other side of the workloads compared to athlete perception of adaptation-stress equation is the stress induced effort. We think this mismatch is important, by training. Athletes may migrate towards a because the unit of stress perceived and re- strategy where longer duration is substituted for sponded to by the athlete is the stress of the higher intensity to reduce the stress reactions entire training session or perhaps training day, associated with training and facilitate rapid not minutes in any given heart-rate zone. recovery from frequent training (Seiler et al., 2007). Interestingly, Foster and colleagues The 80-20 Rule for Intensity reported a very similar intensity distribution by In spite of differences in the methods for professional cyclists during the 3 wk and 80+ quantifying training intensity, all of the above racing hours of the grand tours, such as the studies show remarkable consistency in the Tour de France. Perhaps this distribution repre- training distribution pattern selected by success- sents a form of pacing that emerges over the

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 43 months of elite training (Foster et al., 2005). than about 15 hours in a week. At an average -1 ”Low intensity”–between 50 %VO2max and running speed of 15 km.h , that would be at just under the first lactate turnpoint–represents most 225 km. Former world record holder in a wide intensity range in endurance athletes. the 5 km, 10 km, and marathon, Ingrid There is probably considerable individual varia- Kristiansen trained 550 h.y-1 when she was tion in where within this range athletes accumu- running (Espen Tønnessen, unpublished data). late most of their low-intensity training volume. At a younger age, when she competed in the Technique considerations may play in: athletes Olympics for Norway as a cross country skier, have to train at a high enough intensity to allow she actually trained 150 more h.y-1. Bente Skari, correct technique. For example, Norwegian one of the most successful female cross country Olympic flat-water kayak gold medalist Eric skiers ever, recorded peak annual training loads Verås Larsen explained that the reason most of of 800 h.y-1 (Espen Tønnessen, unpublished his Zone 1 continuous endurance training data). Annual training volume measured in tended to be closer to his lactate threshold than hours is around 1000 among world class row- normally observed was that he could not paddle ers. For example, Olaf Tufte recorded 1100 with competition technique at lower intensities training hours in 2004, when he took his first (Verås Larsen, personal communication). gold medal in the single scull event (Aasen, These qualifiers aside, we conclude that a large 2008). His monthly training volume for that fraction of the training within this zone is being year is shown in Figure 5. Of these hours, about performed at ~60-65 %VO2max, We note that 92 % were endurance training, with the remain- this intensity is about the intensity associated der being primarily strength training. An Olym- with maximal fat utilization in trained subjects pic champion swimmer like Michael Phelps (Achten and Jeukendrup, 2003), but it is unclear may record even higher annual training vol- why optimizing fat utilization would be impor- umes, perhaps as much as 1300 h (a reasonable tant for athletes competing over 3-15 min. guess based on training of other swimming medalists). Training Volume of Elite Athletes Obviously, training intensity distribution and The Kenyan marathoner, Italian cyclist, training volume together will determine the Norwegian rower and American swimmer are impact of training. Elite athletes train a lot, but all at the top of their sport, but when we meas- to be more specific requires some common ure their training volume in hours, they seem metric for comparing athletes in different quite different, with international success being sports. Runners and cyclists count kilometers, achieved with a two-fold or larger range in swimmers count thousands of meters, and row- hours per year (Figure 6). What can explain this ers and cross-country skiers count training difference? One explanation is that the muscle, hours. With a few reasonable assumptions, we tendon, and joint loading stress of the different can convert these numbers to annual training movements vary dramatically. Running im- hours. This physiological metric is appropriate, poses severe ballistic loading stress that is not since the body is sensitive to stress duration. present in cycling or swimming. There seems to Training volume increases with age in high- be a strong inverse relationship between toler- level performers, mostly through increased ated training volume and degree of eccentric or training frequency in sports like running and ballistic stress of the sport. Swimming, rowing, cross-country skiing, but also through increases and cross-country skiing are all highly technical in average session duration, particularly in cy- events with movement patterns that do not draw cling. A talented teenage cyclist training five on the genetically pre-programmed motor days a week might accumulate 10-15 h.wk-1. A pathways of running. Thus high volumes of professional cyclist from Italy performing a training may be as important for technical mas- 1000-km training week will likely be on the tery as for physiological adaptation in these bike between 25 and 30 h. disciplines. Rowers and speed skaters do less Cycling 30-35,000 kilometers a year at, say, movement-specific training than most other ~35 km.h-1 with occasional sessions of strength athletes, but they accumulate substantial addi- training, will add up to ~1000 h.y-1. An elite tional hours of strength training and other forms male marathoner would likely never run more of endurance training.

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 44

Figure 5. Annual training intensity distribution and volume of an Olympic champion rower. Data below are for two-time gold medalist Olaf Tufte in the training season 2003- 2004. The Olympic competition was held in August. Data redrawn from Aasen (2008). Training zones are as described in Table 1. 140 Training duration (h) 120 100 80 60 40 20 0 Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug

Rowing, Zones 1-2 Rowing, Zones 3-5 Strength training

Figure 6. Representative peak annual training volumes for champion athletes from different sports. Ballistic and eccentric loading differences, demands on technical entrainment, and non-specific training volume may all contribute to the differences. Annual Training Volume (h) 1400

1200 1000

800 600

400 200

0 Distance Orienteer Cross-country Rower Cyclist Swimmer runner skier

Intensified-Training Studies Is the “80-20” training intensity distribution Evertsen et al. (1997; 1999; 2001) published observed for successful athletes really optimal, the first of three papers from a study involving or would a redistribution of training intensity training intensification in 20 well-trained junior towards more threshold and high intensity in- cross-country skiers competing at the national terval training and less long slow distance train- or international level. All of the subjects had ing stimulate better gains and higher perform- trained and competed regularly for 4-5 years. In ance? Proponents of large volumes of interval the two months before study initiation, 84 % of training might invoke the famous pareto training was carried out at 60-70 %VO2max, principle and propose that in keeping with this with the remainder at 80-90 %VO2max. They “rule” of effects vs causes, these athletes are were then randomized to a moderate-intensity achieving 80 % of their adaptive gains with 20 (MOD) or a high-intensity training group % of their training and wasting valuable train- (HIGH). MOD maintained essentially the same ing energy. In the last 10 y, several studies have training-intensity distribution they had used been published addressing this question. previously, but training volume was increased

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 45

-1 from 10 to 16 h.wk . HIGH reversed their base- tensified program in Year 2 (VO2max, lactate line intensity distribution so that 83 % of train- threshold, race points). The positive responders ing time was performed at 80-90 %VO2max, from Year 1 showed a similar development in with only 17 % performed as low-intensity Year 2 as in Year 1. training. This group trained 12 h.wk-1. The It is interesting in this context to point out training intervention lasted five months. Inten- that many elite athletes now extend the perio- sity control was achieved using heart-rate moni- dization of their training to a 4-y Olympic cy- toring and blood-lactate sampling. cle. The first year after an Olympics is a “re- Despite 60 % more training volume in MOD covery season”, followed by a building season, and perhaps 400 % more training at lactate then a season of very high training volume, threshold or above in HIGH, physiological and culminating with the Olympic season, where performance changes were modest in both training volume is reduced and competition groups of already well-trained athletes. Find- specificity is emphasized more. Variation in ings from the three papers are summarized in the pattern of training may be important for Table 4. maximal development, but these large scale rhythms of training have not been studied. Table 4. Summary of a 5-month training intensification Esteve-Lanao et al. (2007) randomized 12 study with well trained cross-country skiers (Evertsen et sub-elite distance runners to one of two training al., 1997; Evertsen et al., 1999; Evertsen et al., 2001). groups (Z1 and Z2) that were carefully moni- High Moderate tored for five months. They based their training intensity intensity intensity distribution on the 3-zone model de- (n=10) (n=10) scribed earlier and determined from treadmill VO2max ↔ ↔ testing. Based on time-in-zone heart-rate moni- Lactate-threshold speed ↑ 3 % ↔ toring, Z1 performed 81, 12, and 8 % of train- 20-min run at 9 % grade ↑ 3.8 % ↑ 1.9 % ing in Zones 1, 2, and 3 respectively. Z2 per- Fiber type ↔ ↔ formed more threshold training, with 67, 25, Enzyme activities and 8 % of training performed in the three re- spective zones. That is, Group Z2 performed MCT 1 transporter ↔ ↓ 12 % twice as much training at or near the lactate MCT 4 transporter ↔ ↔ threshold. In a personal communication, the Citrate synthase ↔ ↔ authors reported that in pilot efforts, they were Succinate dehydrogenase ↑ 6 % ↔ unable to achieve a substantial increase in the total time spent in Zone 3, as it was too hard for Gaskill et al. (1999) reported the results of a the athletes. Total training load was matched 2-y project involving 14 cross-country skiers between the groups. Improvement in a cross- training within the same club who were willing country time-trial performed before and after to have their training monitored and manipu- the five-month period revealed that the group lated. The design was interesting and practically that had performed more Zone 1 training relevant. During the first year, athletes all showed significantly greater race time im- trained similarly, averaging 660 training hours provement (-157 ± 13 vs -122 ± 7 s). with 16 % at lactate threshold or higher (nomi- Most recently, Ingham et al. (2008) were nal distribution of sessions). Physiological test able to randomize 18 experienced national results and race performances during the first standard male rowers from the UK into one of year were used to identify seven athletes who two training groups that were initially equiva- responded well to the training and seven who lent based on performance and physiological showed poor VO2max and lactate-threshold testing. All the rowers had completed a 25-d progression, and race results. In the second post-season training-free period just prior to year, the positive responders continued using baseline testing. One group performed “100 %” their established training program. The non- of all training at intensities below that eliciting responders performed a markedly intensified 75 %VO2max (LOW). The other group per- training program with a slight reduction in formed 70 % training at the same low intensi- training hours. The non-responders from Year 1 ties as well as 30 % of training at an intensity showed significant improvements with the in- 50 % of the way between power at lactate

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 46

Table 5. Summary of physiological and performance vealed that the actual distribution was: Polar- changes in well trained rowers training for 12 wk at either ized 65-21-14 % and ACSM 31-56-13 %. low intensity or mixed intensity (70 % low, 30 % high) Comparing the intended and achieved distri- (Ingham et al., 2008). butions highlights a typical training error com- Low Mixed mitted by recreational athletes. We can call it (n=9) (n=9) falling into a training intensity “black hole.” It 2000-m speed ↑ 2 % ↑ 1.4 % is hard to keep recreational people training 45- VO2max ↑ 11 % ↑ 10 % 60 min a day 3-5 days a week from accumulat- Power at 2-mM lactate ↑ 10 % ↑ 2 % ing a lot of training time at their lactate threshold. Training intended to be longer and slower Power at 4-mM lactate ↑ 14 % ↑ 5 % becomes too fast and shorter in duration, and Various VO2 kinetics ↔ ↔ interval training fails to reach the desired inten- threshold and power at VO2max (MIX). In sity. The result is that most training sessions practice, MIX performed high intensity training end up being performed at the same threshold on 3 d.wk-1. All training was performed on a intensity. Foster et al. (2001b) also found that rowing ergometer over the 12 wk. The two athletes tend to run harder on easy days and groups performed virtually identical volumes of easier on hard days, compared to coaches' train- training (~1140 km on the ergometer), with ±10 ing plans. Esteve Lanao did succeed in getting % individual variation allowed to accommodate two groups to distribute intensity very differ- for variation in athlete standard. Results of the ently. The group that trained more polarized, study are summarized in Table 5. with more training time at lower intensity, im- Sixteen of 18 subjects set new personal bests proved their 10-km performance significantly for the 2000-m ergometer test at the end of the more at 7 and 11 wk. So, recreational athletes study. The authors concluded that LOW and could also benefit from keeping low- and high- MIX training had similar positive effects on intensity sessions at the intended intensity. performance and maximal oxygen consump- Interval training can be performed effec- tion. LOW training appeared to induce a greater Table 6. Typical training sessions performed by highly right-shift in the blood-lactate profile during trained athletes in five intensity zones (Aasen, 2008). sub-maximal exercise, which did not translate to a significantly greater gain in performance. If VO2 Examples of training Manageable Zone (%max) sessions durationa MIX training enhanced or preserved anaerobic capacity more than LOW, this may have com- 1 45-65 Continuous bouts 60-360 min pensated for the observed differences in blood- 2 66-80 Continuous bouts 60-180 min lactate profile. 3 81-87 6 x 15 min, 2-min rec 50-90 min 2 x 25 min, 3-min rec Intensity for Recreational Athletes 5 x 10 min, 2-min rec Elite endurance athletes train 10-12 sessions 8 x 8 min, 2-min rec and 15-30 h each week. Is the pattern of 80 % LT 40-60 min below and 20 % above lactate threshold appro- 50 x 1 min, 20-s rec priate for recreational athletes training 4-5 4 88-93 10 x 6 min, 2-3-min rec 30-60 min times and 6-10 hours per week? There are 8 x 5 min, 3-min rec almost no published data addressing this ques- 15 x 3 min, 1-min rec tion. Recently Esteve-Lanao (personal commu- 40 x 1 min, 30-s rec nication) completed an interesting study on 10 x (5 x 40 s, 20-s rec), recreational runners comparing a program that 2- to 3-min breaks was designed to reproduce the polarized train- 30-40 min steady state ing of successful endurance athletes and com- 5 94-100 6 x 5 min, 3-4-min rec 24-30 min pare it with a program built around much more 6 x 4 min, 4-min rec threshold training in keeping with the ACSM 8 x 3 min, 2-min rec 5 x (5 x 1 min, 30-s rec), exercise guidelines. The intended intensity 2- to 3-min breaks distribution for the two groups was: Polarized a Warm-up and rest periods in interval bouts are not 77-3-20 % and ACSM 46-35-19 % for Zones 1, included. 2, and 3. However, heart-rate monitoring re- LT, lactate threshold (max steady state); rec, recoveries.

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 47 tively with numerous combinations of work level, he adopted a highly intensive training duration, rest duration, and intensity. We have regime for cycling that was focused on training found that when subjects self-select running just under or at his lactate threshold and near speed based on a standard prescription, 4-min VO2max; for example, 2-3 weekly training work duration and 2-min recovery duration sessions of 4-5 × 4 min at 95 %VO2max. combine to give the highest physiological re- Weekly training volume did not exceed 10 h. sponse and maintained speed (Seiler and After 2.5 years of this high-intensity, low- Sjursen, 2004; Seiler and Hetlelid, 2005). How- volume training, Fostervold initiated coopera- ever, perceptual and physiological response tion with the Norwegian Olympic Center and differences across the typical work and recov- his training program was radically reorganized. ery spectrum are fairly small and performance Weekly training volume was doubled from 8-10 enhancement differences are unclear at best. h to 18-20. Training volume in Zone 2 was Some researchers have proposed that specific reduced dramatically and replaced with a larger interval regimes (e.g., 4 × 4 min at 95 volume of training in Zone 1. Training in Zone %VO2max) may be superior for achieving 5 was replaced with Zones 3 and 4, such that adaptive gains (Helgerud et al., 2007; Wisloff et total training volume at intensities at or above al., 2007), but other research studies and our lactate threshold was roughly doubled without observations of athlete practice suggest that a overstressing the athlete. The typical effective variety of combinations of work and rest dura- duration of interval sessions increased from ~20 tion are effective for long-term development. min to ~ 60 min (for example 8 × 8 min at 85- Table 6 shows typical combinations of intensity 90 %HRmax with 2-min recoveries). The in- and effective duration used by elite endurance tensity zones were initially based on heart rate athletes for workouts in the different aerobic but later adjusted relative to lactate and power training zones described earlier. All the exam- output measurements made in the field. Table 7 ples are taken from the training diaries of elite shows the training intensity distribution and performers. The effective durations for the volume loading for the athlete during the season different zones are utilized by highly trained before and after the change in training to a athletes. For those without the same training high-volume program. Table 8 shows the out- base, similar workouts would be performed but come. with less total effective duration. The athlete responded well to the training load amplification and reorganization. During Case Studies of Training Manipulation the 2005 season, after 2.5 y performing a low- Case studies are the weakest form of scien- volume, high-intensity program, a season train- tific evidence. But, for coaches and high per- ing with higher volume and lower average in- formance athlete support teams, each elite ath- tensity resulted in marked physiological and lete is a case study. So, we present here two performance improvement. Although the ath- case studies that we think are instructive in lete’s training de-emphasized both training near demonstrating the potential physiological im- his lactate threshold intensity and training at pact of successfully manipulating training volume and intensity distribution variables at Table 7. Comparison of weekly training intensity distribu- the individual level. Both cases involve Norwe- tion and total volume in 2004 season and 2005 season – Case 1. gian athletes who were followed closely by one of the authors (Tønnessen). Both would be Intensity zone Season 2004 Season 2005 considered already highly trained prior to the (%HRmax) (h:min) (h:min) training reorganization. 5 (95-100 %) 0:45 (8.5 %) 0:05 (0.5 %) Case 1–From Soccer Pro to Elite Cyclist 4 (90-95 %) – 0:40 (4.0 %) Knut Anders Fostervold was a professional 3 (85-90 %) 0:30 (5.5 %) 1:00 (5.5 %) soccer player in the Norwegian elite league 2 (75-85 %) 3:05 (36 %) 1:00 (5.5 %) from 1994 to 2002. A knee injury ended his 1 (55-75 %) 4:20 (50 %) 15:20 (85 %) soccer career at age 30 and he decided to switch Weekly totala 8:40 18:05 to cycling. Knut had very high natural endur- Annual totala 420:00 850:00 ance capacity and had run 5 km in 17:24 at age 12. After 15 y of soccer training at the elite HRmax: maximum heart rate. aEstimates based on diaries for the first 18 wk.

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 48

Table 8. Physiological testing before and after training wegian national time-trial championships, sec- reorganization – Case 1. onds behind former world under-23 time trial 8 wk 18 wk Change champions and Tour de France stage winners Pre post post 0-18 wk Thor Hushovd and Kurt Asle Arvesen. His BW (kg) 84 81 84 0 % failure to perform even better, given his excep- tionally high VO2max, was attributed to poorer VO2max (ml⋅kg–1⋅min–1) 81 90 88 11 % cycling efficiency and aerodynamics and a –1 VO2max (L⋅min ) 6.8 7.3 7.3 7 % lower fractional utilization at lactate threshold LT power (W) 375 420 440 14 % compared to the best professionals with many LT power (W⋅kg-1) 4.5 5.2 5.2 15 % years of specific training. In 2006 and 2007 he represented Norway in the world championship near VO2max, both of these physiological an- time trial. His absolute VO2max in 2005 was chors improved markedly. equal to the highest ever measured in a Fostervold won a bronze medal in the Nor- Norwegian athlete. Table 9. Comparison of actual training composition Case 2–From Modern Pentathlete to Runner during a hard training week, Fall 2003 and Fall 2004 – Prior to 2003, Øystein Sylta was a military Case 2. pentathlete (European champion in 2003). In Day Fall 2003 Fall 2004 the Fall of 2003 he decided to focus on distance Mon 60-min run, Z1-2 S1: 50-min run, Z1 running and is now nationally competitive, with S2: 65-min run, Z1 personal bests for 3000-m steeplechase, 5000- Tues 7x1000 m, 90-s re- S 1: 45-min run, Z1 m, and 10000-m of 8:31, 14:04 and 29:12 recovery, Z4 S2: 12 x 5-min, 1-min spectively. His case is interesting due to the recovery, Z3 dramatic change in training volume and inten- Wed S 1: 40-min run, Z1 S 1: 45-min run, Z1 sity distribution he undertook from 2003 to S 2: 50-min run, Z1-2 + S 2: 75-min run, Z1 2004 and associated changes in physiological 45-min strength test results. Thur 17x300m, 52s, 40-s S1: 45-min run, Z1 Prior to 2003, Sylta trained using a high- recovery, Z5 S 2: 12 x 3-min, intensity, low-volume training structure. When 1-min rec, Z4 he agreed to try a new approach, emphasis was Fri 55min run, Z1 45-min run Z1 placed on increasing training volume with low- intensity sessions and changing his interval Sat S 1: 40-min run Z1 + S 1: 45-min run, Z1 30-min strength S 2: 60-min run, Z1 training. He either trained long slow distance S 2: 4 x 7-min intervals, or long intense interval sessions. However, his 2-min recovery, Z3 total training distance at intensities above his Sun 100-min run Z1 150-min run Z1 lactate threshold was reduced and redistributed. From 2002/2003 to 2003/2004 he increased his Interval sessions were preceded and ended with 15-20- min easy running both seasons. In both seasons, easy total running distance from 3,500 to 5,900 km. runs were concluded with 5-8 x 100 m strides. He also reduced his strength training from 100 Intensity zones (Z) are as shown in Table 7. annual hours to 50. Table 9 shows a typical hard training week in the Fall of 2003 and Fall of 2004, and Table 10 summarizes the running Table 10. Annual training volume and intensity distribution in 2003 and 2004 – Case 2. specific training. His physiological adaption to the first year of restructured training is docu- Intensity zone 2003 season 2004 season mented in Table 11. 5 (95-100 %) 3 % (8 h) 0,5 % (2 h) From 2003 to 2009, Sylta’s threshold run- 4 (90-95 %) 12 % (33 h) 2,5 % (13 h) ning speed increased from 16.9 to 19.5 km.h-1. 3 (85-90 %) 13 % (36 h) 10 % (50 h) From 2002 to 2009, his 10-km time improved 2 (75-85 %) 18 % (49 h) 4 % (20 h) from 31:44 to 29:12, and 3000-m steeplechase 1 (55-75 %) 54 % (149 h) 83 % (412 h) from 9:11 to 8:31. In the first five months of Total for yeara 275 h 497 h training reorganization, his 3000-m steeple result improved by 30 s. a100 h of strength training in 2003 and 50 h in 2004 are not included in the totals. Both these case studies demonstrate that even in already well trained athletes, meaning-

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 49

Table 11. Physiological testing before and after training ferent. To answer a question like, “is near reorganization – Case 2. VO2max interval training more effective for Sep 03 Feb 04 Change achieving performance gains in athletes than Body mass (kg) 74 71 -4 % training at the maximal lactate steady state?”, the matching of training bouts has to be realistic VO2max (ml⋅kg–1⋅min–1) 76 83 9 % from the perspective of perceived stress and VO2max (L⋅min–1) 5.6 5.9 5 % how athletes train. Future studies of training Lactate threshold (km.h-1) 16.9 17.7 5 % intensity effects on adaptation and performance should take this issue of ecological validity into ful improvements in physiological test results account. and performance may occur with appropriate training intensity and volume manipulation. Conclusions Both athletes showed clear improvements in Optimization of training methods is an area physiological testing despite reductions in HIT of great interest for scientists, athletes, and training. Both seemed to respond positively to fitness enthusiasts. One challenge for sport an increase in total training volume and specifi- scientists is to translate short-term training cally, more low-intensity volume. intervention study results to long-term performance development and fitness training organiza- Valid Comparisons of Training Interventions tion. Currently, there is great interest in high- Matching training programs based on total intensity, short-duration interval training pro- work or oxygen consumption seems sensible in grams. However, careful evaluation of both a laboratory. As we noted earlier, this has been available research and the training methods of the preferred method of matching when com- successful endurance athletes suggests that we paring the effects of continuous and interval should be cautious not to over-prescribe high- training in controlled studies. Unfortunately, it intensity interval training or exhort the advan- is not realistic from the view of athletes pursu- tages of intensity over duration. ing maximal performance. They do not com- Here are some conclusions that seem war- pare training sessions or adjust training time to ranted by the available data and experience intensity in this manner. A key issue here is the from observations of elite performers: non-linear impact of exercise intensity on the • There is reasonable evidence that an ~80:20 manageable accumulated duration of intermit- ratio of low to high intensity training (HIT) tent exercise. We have exemplified this in Ta- gives excellent long-term results among en- ble 12 by comparing some typical training ses- durance athletes training daily. sions from the training of elite athletes. • Low intensity (typically below 2 mM blood The point we want to make is that the ath- lactate), longer duration training is effective lete’s perception of the stress of performing 4 × in stimulating physiological adaptations and 15 min at 85 %VO2max is about the same as should not be viewed as wasted training time. that of performing 6 × 4 min at 95 %VO2max, • Over a broad range, increases in total training even though total work performed is very dif- volume correlate well with improvements in

Table 12. Typical duration and intensity combinations used in training sessions by elite endurance athletes. Durationa Intensity Total Training loadc (min) (%VO2max) VO2b (L) (RPE.min) Basic endurance 120 60 360 240-360 Threshold training (lactate ~3-4 mM) 60 (4x15) 85 293 375 90 % intervals (lactate ~5-7 mM) 40 (5x8) 90 218 375-425 VO2max intervals (lactate ~6-10 mM) 24 (6x4) 95 152 300-350 aWarm-up not included. bOxygen consumption calculations based on a male athlete with 5 L.min-1 VO2max and include 15 min warm up at 50 %VO2max for threshold and interval sessions. Examples are based on a manageable accumulated duration at different interval training intensities, and drawn from the training diaries of elite athletes. cSession rating of perceived exertion x duration (Foster et al., 1996; Seiler et al., 2007).

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 50

physiological variables and performance. Åstrand PO, Rodahl, K.R. (1986). Physical Training. • HIT should be a part of the training program In: Textbook of Work Physiology. McGraw-Hill: of all exercisers and endurance athletes. Singapore. 412-476 However, about two training sessions per Beneke R, von Duvillard SP (1996). Determination week using this modality seems to be suffi- of maximal lactate steady state response in selected sports events. Medicine and Science in cient for achieving performance gains with- Sports and Exercise 28, 241-246 out inducing excessive stress. Beneke R, Leithauser RM, Hutler M (2001). • The effects of HIT on physiology and per- Dependence of the maximal lactate steady state on formance are fairly rapid, but rapid plateau the motor pattern of exercise. British Journal of effects are seen as well. To avoid premature Sports Medicine 35, 192-196 stagnation and ensure long-term develop- Berger NJ, Tolfrey K, Williams AG, Jones AM ment, training volume should increase sys- (2006). Influence of continuous and interval tematically as well. training on oxygen uptake on-kinetics. Medicine • When already well-trained athletes markedly and Science in Sports and Exercise 38, 504-512 intensify training with more HIT over 12 to Bhambhani Y, Singh M (1985). The effects of three training intensities on VO2 max and VE/VO2 ~45 wk, the impact is equivocal. ratio. Canadian Journal of Applied Sport Sciences • In athletes with an established endurance 10, 44-51 base and tolerance for relatively high training Billat V, Lepretre PM, Heugas AM, Laurence MH, loads, intensification of training may yield Salim D, Koralsztein JP (2003). Training and small performance gains at acceptable risk. bioenergetic characteristics in elite male and • An established endurance base built from female Kenyan runners. Medicine and Science in reasonably high volumes of training may be Sports and Exercise 35, 297-304; discussion 305- an important precondition for tolerating and 296 responding well to a substantial increase in Billat VL, Flechet B, Petit B, Muriaux G, training intensity over the short term. Koralsztein JP (1999). Interval training at VO2max: effects on aerobic performance and • Elite athletes achieve periodization of train- overtraining markers. Medicine and Science in ing with reductions in total volume, and Sports and Exercise 31, 156-163 modest increases in volume of training above Billat VL, Demarle A, Slawinski J, Paiva M, the lactate threshold. An overall polarization Koralsztein JP (2001). Physical and training of training intensity characterizes the transi- characteristics of top-class marathon runners. tion from preparation to competition mesocy- Medicine and Science in Sports and Exercise 33, cles. The basic intensity distribution remains 2089-2097 similar throughout the year. Brigelius-Flohe R (2009). Commentary: oxidative stress reconsidered. Genes and Nutrition 4, 161- Reviewer's Commentary 163 Catalucci D, Latronico MV, Ellingsen O, Condorelli References G (2008). Physiological myocardial hypertrophy: Aasen S (2008). Utholdenhet- trening som gir how and why? Frontiers in Bioscience 13, 312- resultater. Akilles Forlag: Oslo, Norway 324 Acevedo EO, Goldfarb AH (1989). Increased Chakravarthy MV, Booth FW (2004). Eating, training intensity effects on plasma lactate, exercise, and "thrifty" genotypes: connecting the ventilatory threshold, and endurance. Medicine dots toward an evolutionary understanding of and Science in Sports and Exercise 21, 563-568 modern chronic diseases. Journal of Applied Achten J, Jeukendrup AE (2003). Maximal fat Physiology 96, 3-10 oxidation during exercise in trained men. Christensen EH (1960). [Interval work and interval International Journal of Sports Medicine 24, 603- training.]. Internationale Zeitschrift fur 608 Angewandte Physiologie Einschliesslich Ahmetov, II, Rogozkin VA (2009). Genes, Athlete Arbeitsphysiologie 18, 345-356 Status and Training - An Overview. Medicine and Christensen EH, Hedman R, Saltin B (1960). Sport Science 54, 43-71 Intermittant and continuous running. Acta Åstrand I, Åstrand PO, Christiensen EH, Hedman R Physiolologica Scandinavica 50, 269-286 (1960). Intermittent muscular work. Acta Cunningham DA, McCrimmon D, Vlach LF (1979). Physiolologica Scandinavica 48, 448-453 Cardiovascular response to interval and Åstrand I ÅP, Christiansen EH, Hedman R (1960). continuous training in women. European Journal Myohemoglobin as an oxygen store in man. Acta of Applied Physiology 41, 187-197 Physiolologica Scandinavica 48, 454-460 Daniels J, Scardina N (1984). Interval training and

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 51

performance. Sports Medicine 1, 327-334 Fiskerstrand A, Seiler KS (2004). Training and Daussin FN, Ponsot E, Dufour SP, Lonsdorfer-Wolf performance characteristics among Norwegian E, Doutreleau S, Geny B, Piquard F, Richard R international rowers 1970-2001. Scandinavian (2007). Improvement of VO2max by cardiac Journal of Medicine and Science in Sports 14, output and oxygen extraction adaptation during 303-310 intermittent versus continuous endurance training. Foster C, Daines E, Hector L, Snyder AC, Welsh R European Journal of Applied Physiology 101, (1996). Athletic performance in relation to 377-383 training load. Wisconsin Medical Journal 95, 370- Daussin FN, Zoll J, Dufour SP, Ponsot E, 374 Lonsdorfer-Wolf E, Doutreleau S, Mettauer B, Foster C (1998). Monitoring training in athletes with Piquard F, Geny B, Richard R (2008a). Effect of reference to overtraining syndrome. Medicine and interval versus continuous training on Science in Sports and Exercise 30, 1164-1168 cardiorespiratory and mitochondrial functions: Foster C, Florhaug JA, Franklin J, Gottschall L, relationship to aerobic performance improvements Hrovatin LA, Parker S, Doleshal P, Dodge C in sedentary subjects. American Journal of (2001a). A new approach to monitoring exercise Physiology 295, R264-272 training. Journal of Strength and Conditioning Daussin FN, Zoll J, Ponsot E, Dufour SP, Research 15, 109-115 Doutreleau S, Lonsdorfer E, Ventura-Clapier R, Foster C, Heiman KM, Esten PL, Brice G, Porcari J Mettauer B, Piquard F, Geny B, Richard R (2001b). Differences in perceptions of training by (2008b). Training at high exercise intensity coaches and athletes. South African Journal of promotes qualitative adaptations of mitochondrial Sports Medicine 8, 3-7 function in human skeletal muscle. Journal of Foster C, Hoyos J, Earnest C, Lucia A (2005). Applied Physiology 104, 1436-1441 Regulation of energy expenditure during Diaz F, Moraes CT (2008). Mitochondrial prolonged athletic competition. Medicine and biogenesis and turnover. Cell Calcium 44, 24-35 Science in Sports and Exercise 37, 670-675 Eddy DO, Sparks KL, Adelizi DA (1977). The Frank D, Kuhn C, Brors B, Hanselmann C, Ludde effects of continuous and interval training in M, Katus HA, Frey N (2008). Gene expression women and men. European Journal of Applied pattern in biomechanically stretched Physiology 37, 83-92 cardiomyocytes: evidence for a stretch-specific Edge J, Bishop D, Goodman C (2006). The effects gene program. Hypertension 51, 309-318 of training intensity on muscle buffer capacity in Fry AC, Schilling BK, Weiss LW, Chiu LZ (2006). females. European Journal of Applied Physiology beta2-Adrenergic receptor downregulation and 96, 97-105 performance decrements during high-intensity Esteve-Lanao J, San Juan AF, Earnest CP, Foster C, resistance exercise overtraining. Journal of Lucia A (2005). How do endurance runners Applied Physiology 101, 1664-1672 actually train? Relationship with competition Gaskill SE, Serfass RC, Bacharach DW, Kelly JM performance. Medicine and Science in Sports and (1999). Responses to training in cross-country Exercise 37, 496-504 skiers. Medicine and Science in Sports and Esteve-Lanao J, Foster C, Seiler S, Lucia A (2007). Exercise 31, 1211-1217 Impact of training intensity distribution on Gladden LB (2004). Lactate metabolism: a new performance in endurance athletes. Journal of paradigm for the third millennium. Journal of Strength and Conditioning Research 21, 943-949 Physiology 558, 5-30 Evertsen F, Medbo JI, Jebens E, Nicolaysen K Gledhill N, Cox D, Jamnik R (1994). Endurance (1997). Hard training for 5 mo increases Na(+)- athletes' stroke volume does not plateau: major K+ pump concentration in skeletal muscle of advantage is diastolic function. Medicine and cross-country skiers. American Journal of Science in Sports and Exercise 26, 1116-1121 Physiology 272, R1417-1424 Gomez-Cabrera MC, Domenech E, Romagnoli M, Evertsen F, Medbo JI, Jebens E, Gjovaag TF (1999). Arduini A, Borras C, Pallardo FV, Sastre J, Vina J Effect of training on the activity of five muscle (2008). Oral administration of vitamin C enzymes studied on elite cross-country skiers. decreases muscle mitochondrial biogenesis and Acta Physiologica Scandinavica 167, 247-257 hampers training-induced adaptations in Evertsen F, Medbo JI, Bonen A (2001). Effect of endurance performance. American Journal of training intensity on muscle lactate transporters Clinical Nutrition 87, 142-149 and lactate threshold of cross-country skiers. Acta Gorostiaga EM, Walter CB, Foster C, Hickson RC Physiologica Scandinavica 173, 195-205 (1991). Uniqueness of interval and continuous Fields RD (2006). Nerve impulses regulate training at the same maintained exercise intensity. myelination through purinergic signalling. European Journal of Applied Physiology 63, 101- Novartis Foundation Symposium 276, 148-158; 107 discussion 158-161, 233-147, 275-181 Gregory LW (1979). The development of aerobic

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 52

capacity: a comparison of continuous and interval level. European Journal of Applied Physiology 86, training. Research Quarterly 50, 199-206 375-381 Gullich A, Emrich E, Seiler S (2009). Training Laughlin MH, Roseguini B (2008). Mechanisms for methods and intensity distribution of young exercise training-induced increases in skeletal world-class rowers. International Journal of Sport muscle blood flow capacity: differences with Physiology and Performance 4, In Press interval sprint training versus aerobic endurance Halson SL, Jeukendrup AE (2004). Does training. Journal of Physiology and Pharmacology overtraining exist? An analysis of overreaching 59 Suppl 7, 71-88 and overtraining research. Sports Medicine 34, Lehmann MJ, Lormes W, Opitz-Gress A, Steinacker 967-981 JM, Netzer N, Foster C, Gastmann U (1997). Hansen AK, Fischer CP, Plomgaard P, Andersen JL, Training and overtraining: an overview and Saltin B, Pedersen BK (2005). Skeletal muscle experimental results in endurance sports. Journal adaptation: training twice every second day vs. of Sports Medicine and Physical Fitness 37, 7-17 training once daily. Journal of Applied Physiology Lindsay FH, Hawley JA, Myburgh KH, Schomer 98, 93-99 HH, Noakes TD, Dennis SC (1996). Improved Helgerud J, Hoydal K, Wang E, Karlsen T, Berg P, athletic performance in highly trained cyclists Bjerkaas M, Simonsen T, Helgesen C, Hjorth N, after interval training. Medicine and Science in Bach R, Hoff J (2007). Aerobic high-intensity Sports and Exercise 28, 1427-1434 intervals improve VO2max more than moderate Lucia A, Hoyos J, Carvajal A, Chicharro JL (1999). training. Medicine and Science in Sports and Heart rate response to professional road cycling: Exercise 39, 665-671 the Tour de France. International Journal of Sports Henriksson J, Reitman JS (1976). Quantitative Medicine 20, 167-172 measures of enzyme activities in type I and type II Lucia A, Hoyos J, Santalla A, Earnest C, Chicharro muscle fibres of man after training. Acta JL (2003). Tour de France versus Vuelta a Physiologica Scandinavica 97, 392-397 Espana: which is harder? Medicine and Science in Hill AV, Long CNH, Lupton H (1924a). Muscular Sports and Exercise 35, 872-878 exercise, lactic acid, and the supply and utilization MacDougall D, Sale D (1981). Continuous vs. of oxygen. Parts I-III. . Proceedings of the Royal interval training: a review for the athlete and the Society of London. Series B: Biological Sciences coach. Canadian Journal of Applied Sport 96, 438-475 Sciences 6, 93-97 Hill AV, Long CNH, Lupton H (1924b). Muscular Marcuello A, Gonzalez-Alonso J, Calbet JA, exercise, lactic acid, and the supply and utilization Damsgaard R, Lopez-Perez MJ, Diez-Sanchez C of oxygen. Parts IV-VI. Proceedings of the Royal (2005). Skeletal muscle mitochondrial DNA Society of London. Series B: Biological Sciences content in exercising humans. Journal of Applied 97, 84-138 Physiology 99, 1372-1377 Holloszy JO (2008). Regulation by exercise of McConell GK, Lee-Young RS, Chen ZP, Stepto skeletal muscle content of mitochondria and NK, Huynh NN, Stephens TJ, Canny BJ, Kemp GLUT4. Journal of Physiology and Pharmacology BE (2005). Short-term exercise training in humans 59 Suppl 7, 5-18 reduces AMPK signalling during prolonged Hoppeler H, Klossner S, Fluck M (2007). Gene exercise independent of muscle glycogen. Journal expression in working skeletal muscle. Advances of Physiology 568, 665-676 in Experimental Medicine and Biology 618, 245- McPhee JS, Williams AG, Stewart C, Baar K, 254 Schindler JP, Aldred S, Maffulli N, Sargeant AJ, Ingham SA, Carter H, Whyte GP, Doust JH (2008). Jones DA (2009). The training stimulus Physiological and performance effects of low- experienced by the leg muscles during cycling in versus mixed-intensity rowing training. Medicine humans. Experimental Physiology 94, 684-694 and Science in Sports and Exercise 40, 579-584 Mujika I, Chatard JC, Busso T, Geyssant A, Barale Ishibashi T, Dakin KA, Stevens B, Lee PR, Kozlov F, Lacoste L (1995). Effects of training on SV, Stewart CL, Fields RD (2006). Astrocytes performance in competitive swimming. Canadian promote myelination in response to electrical Journal of Applied Physiology 20, 395-406 impulses. Neuron 49, 823-832 Nordsborg N, Bangsbo J, Pilegaard H (2003). Effect Joseph AM, Pilegaard H, Litvintsev A, Leick L, of high-intensity training on exercise-induced Hood DA (2006). Control of gene expression and gene expression specific to ion homeostasis and mitochondrial biogenesis in the muscular metabolism. Journal of Applied Physiology 95, adaptation to endurance exercise. Essays in 1201-1206 Biochemistry 42, 13-29 Ojuka EO (2004). Role of calcium and AMP kinase Kamo M (2002). Discharge behavior of motor units in the regulation of mitochondrial biogenesis and in knee extensors during the initial stage of GLUT4 levels in muscle. Proceedings of the constant-force isometric contraction at low force Nutrition Society 63, 275-278

Sportscience 13, 32-53, 2009 Seiler & Tønnessen: Intensity and Duration in Endurance Training Page 53

Overend TJ, Paterson DH, Cunningham DA (1992). D (1998). Training of rowers before world The effect of interval and continuous training on championships. Medicine and Science in Sports the aerobic parameters. Canadian Journal of Sport and Exercise 30, 1158-1163 Sciences 17, 129-134 Stepto NK, Hawley JA, Dennis SC, Hopkins WG Pelliccia A, Culasso F, Di Paolo FM, Maron BJ (1999). Effects of different interval-training (1999). Physiologic left ventricular cavity programs on cycling time-trial performance. dilatation in elite athletes. Annals of Internal Medicine and Science in Sports and Exercise 31, Medicine 130, 23-31 736-741 Poole DC, Gaesser GA (1985). Response of Talanian JL, Galloway SD, Heigenhauser GJ, Bonen ventilatory and lactate thresholds to continuous A, Spriet LL (2007). Two weeks of high-intensity and interval training. Journal of Applied aerobic interval training increases the capacity for Physiology 58, 1115-1121 fat oxidation during exercise in women. Journal of Ristow M, Zarse K, Oberbach A, Kloting N, Applied Physiology 102, 1439-1447 Birringer M, Kiehntopf M, Stumvoll M, Kahn CR, Thompson P (2005). Break through the speed barrier Bluher M (2009). Antioxidants prevent health- with the "new interval training". Athletics Weekly promoting effects of physical exercise in humans. 59, 62-63 Proceedings of the National Academy of Sciences Wenger HA, Macnab RB (1975). Endurance of the United States of America 106, 8665-8670 training: the effects of intensity, total work, Robinson DM, Robinson SM, Hume PA, Hopkins duration and initial fitness. Journal of Sports WG (1991). Training intensity of elite male Medicine and Physical Fitness 15, 199-211 distance runners. Medicine and Science in Sports Weston AR, Myburgh KH, Lindsay FH, Dennis SC, and Exercise 23, 1078-1082 Noakes TD, Hawley JA (1997). Skeletal muscle Saltin B, Nazar K, Costill DL, Stein E, Jansson E, buffering capacity and endurance performance Essen B, Gollnick D (1976). The nature of the after high-intensity interval training by well- training response; peripheral and central trained cyclists. European Journal of Applied adaptations of one-legged exercise. Acta Physiology 75, 7-13 Physiologica Scandinavica 96, 289-305 Wisloff U, Stoylen A, Loennechen JP, Bruvold M, Schumacher YO, Mueller P (2002). The 4000-m Rognmo O, Haram PM, Tjonna AE, Helgerud J, team pursuit cycling world record: theoretical and Slordahl SA, Lee SJ, Videm V, Bye A, Smith GL, practical aspects. Medicine and Science in Sports Najjar SM, Ellingsen O, Skjaerpe T (2007). and Exercise 34, 1029-1036 Superior cardiovascular effect of aerobic interval Seiler KS, Kjerland GO (2006). Quantifying training training versus moderate continuous training in intensity distribution in elite endurance athletes: is heart failure patients: a randomized study. there evidence for an "optimal" distribution? Circulation 115, 3086-3094 Scandinavian Journal of Medicine and Science in Yan Z (2009). Exercise, PGC-1alpha, and metabolic Sports 16, 49-56 adaptation in skeletal muscle. Applied Physiology Seiler S, Sjursen JE (2004). Effect of work duration Nutrition and Metabolism 34, 424-427 on physiological and rating scale of perceived Yeo WK, Paton CD, Garnham AP, Burke LM, exertion responses during self-paced interval Carey AL, Hawley JA (2008). Skeletal muscle training. Scandinavian Journal of Medicine and adaptation and performance responses to once a Science in Sports 14, 318-325 day versus twice every second day endurance Seiler S, Hetlelid KJ (2005). The impact of rest training regimens. Journal of Applied Physiology duration on work intensity and RPE during 105, 1462-1470 interval training. Medicine and Science in Sports Zapico AG, Calderon FJ, Benito PJ, Gonzalez CB, and Exercise 37, 1601-1607 Parisi A, Pigozzi F, Di Salvo V (2007). Evolution Seiler S, Haugen O, Kuffel E (2007). Autonomic of physiological and haematological parameters recovery after exercise in trained athletes: with training load in elite male road cyclists: a intensity and duration effects. Medicine and longitudinal study. Journal of Sports Medicine and Science in Sports and Exercise 39, 1366-1373 Physical Fitness 47, 191-196 Sheikh F, Raskin A, Chu PH, Lange S, Zhou B, Conlee RK, Jensen R, Fellingham GW, Domenighetti AA, Zheng M, Liang X, Zhang T, George JD, Fisher AG (2001). Stroke volume Yajima T, Gu Y, Dalton ND, Mahata SK, Dorn does not plateau during graded exercise in elite GW, 2nd, Heller-Brown J, Peterson KL, Omens male distance runners. Medicine and Science in JH, McCulloch AD, Chen J (2008). An FHL1- Sports and Exercise 33, 1849-1854 containing complex within the cardiomyocyte Published Nov 2009 sarcomere mediates hypertrophic biomechanical ©2009 stress responses in mice. Journal of Clinical Investigation 118, 3870-3880 Steinacker JM, Lormes W, Lehmann M, Altenburg

Sportscience 13, 32-53, 2009

View publication stats