sports medicine Volume 39 Issue 6 June 2009

Sports Med 2009; 39 (6): 423-438 0112-1642/09/0006-0423/$49.95/0

LEADING ARTICLE

ª 2009 Adis Data Information BV. All rights reserved.

Steps Per Day The Road to Senior Health? Yukitoshi Aoyagi1 and Roy J. Shephard2 1 Exercise Sciences Research Group, Tokyo Metropolitan Institute of Gerontology, Itabashi, Tokyo, Japan 2 Faculty of Physical Education and Health, University of Toronto, Toronto, Ontario, Canada

Abstract

In older adults, as in younger individuals, habitual moderate-intensity physical activity is associated with a reduced risk of various chronic health conditions, including certain types of cardiovascular and musculoskeletal disease and certain forms of cancer. However, the pattern of physical activity associated with such benefits remains unclear. One problem is that most investigators have examined patterns of physical activity using either subjective questionnaires or accelerometer or pedometer measurements limited to a single week, despite clear evidence of both the unreliability/invalidity of questionnaires and seasonal changes in activity patterns. Since 2000, we have thus conducted an interdisciplinary study examining the habitual physical activity and health of elderly people living in a mediumsized Japanese town (the Nakanojo Study). In about one-tenth of some 5000 available subjects aged ‡65 years, physical activity has already been assessed continuously for 24 h/day for >8 years using a specially adapted pedometer/ accelerometer. This device has a storage capacity of 36 days and can distinguish >10 intensities of physical activity (expressed in metabolic equivalents [METs]). Data have to date been summarized as daily step counts and daily durations of activity of 3 METs, averaged over a 1-year period. This article provides a detailed overview of both factors influencing habitual physical activity and relationships between such activity and health in an elderly population. To date, analyses have been cross-sectional in type. Substantial associations have been noted between the overall health of participants and both the daily duration of effort undertaken at an intensity of >3 METs and the daily step count. In men, the extent of health is associated more closely with the daily duration of activity of >3 METs than with the daily step count, whereas in women, the association is closer for the step count than for the duration of activity >3 METs. In both sexes, the threshold amount of physical activity associated with better health is greater for physical than for mental benefits: >8000 versus >4000 steps/day and/or >20 versus >5 min/day at an intensity >3 METs, respectively. In other words, better physical health is seen in those spending at least 20 min/day in moderate walking (at a pace of around 1.4 m/s [5 km/h]) and a further >60 min of light activity per day. In contrast, better mental health is associated with much smaller amounts of deliberate physical activity.

Aoyagi & Shephard

424

The daily step count and the daily durations of activity of 3 METs are all influenced by meteorological factors, particularly precipitation and mean ambient temperature. Activity decreases exponentially to about 4000 steps/day as precipitation increases. Excluding the influence of rainfall, the daily step count peaks at a mean outdoor temperature of around 17C; above and especially below such readings, physical activity decreases as a quadratic function of temperature. Seasonal changes in microclimate should thus be considered when designing interventions intended to increase the habitual physical activity of elderly people. The observed associations between physical activity and health outcomes point to a need for longitudinal analyses; these should examine potential causal interpretations of the current findings and elucidate possible additional mediating variables.

1. The Nakanojo Study in an International Perspective In older adults, as in younger individuals, habitual moderate-intensity physical activity is associated with a reduced risk of various chronic health conditions, including certain types of cardiovascular and musculoskeletal disease and certain forms of cancer.[1-5] However, the details of this association (intensity and total amount of activity) remain unclear.[6] One problem is that most of those investigating the associations have used questionnaires asking about the frequency and/or duration of a given type of physical activity during a typical recent week.[7-17] Such subjective responses necessarily provide only limited estimates of the volume and intensity of physical activity undertaken, particularly in older adults, many of whom have difficulties in recalling recent events and/or have some loss of cognitive function.[18] Pedometers and accelerometers have been employed more recently to assess physical activity patterns with greater accuracy.[19-37] Such measurements are objective, but unfortunately, data collection has usually covered no more than a single week, despite clear evidence of seasonal changes in activity, particularly among elderly people.[38-40] Recent observations[39] show that the numbers of days of recording needed to obtain reliable estimates of an individual’s habitual physical activity over an entire year are substantially fewer for random or seasonally distributed data than for consecutive sampling. However, ª 2009 Adis Data Information BV. All rights reserved.

even with random or seasonal data collection, the necessary observation period exceeds common practice. In an older adult with no formal employment, >10 random or seasonally selected days are needed to estimate a year’s activity with >90% reliability.[39] At least in short-term studies (£1 week), a further problem may arise from the individual’s reaction to wearing a recording device. Thus, Clemes et al.[22] noted that if people realized that they had been fitted with a pedometer, they walked some 13 000 additional steps in the first week of observation. Since 2000, we have been conducting an interdisciplinary study on the habitual physical activity and health of elderly people that addresses some of these problems (table I).[38-46] The test site is the medium-sized Japanese town of Nakanojo, located about 150 km northwest of Tokyo. National census data show a population of 17 491 (8501 men and 8990 women); 29.1% of these are aged ‡65 years (25.7% of men and 32.3% of women). Our subjects include all willing residents ‡65 years of age except those who are severely demented or bedridden (a total of almost 5000 participants). All have completed a simple conventional questionnaire[45] once a year, and in about one-tenth of the sample, physical activity has been assessed continuously 24 hours per day for >8 years using a specially adapted uniaxial pedometer/accelerometer (modified Kenz Lifecorder, Suzuken Co., Ltd, Nagoya, Aichi, Japan),[38-40] as recommended by Janz.[47] Technical details of this monitoring Sports Med 2009; 39 (6)

Steps Per Day for Senior Health?

425

Table I. Details of components of the Nakanojo Study cited in this report Research

Sample

Reference

total no. (male/female)

age range (mean – SD) [y]

Meteorology and habitual physical activity of the elderly (selected environmental factors [day length, mean ambient temperature, duration of bright sunshine, mean wind speed, mean relative humidity and precipitation] vs daily step count measured by pedometer/accelerometer over a 450-day period)

41 (20/21)

65–78 (71 – 4)

38

The minimum number of observation days needed to obtain reliable estimates of the annual habitual physical activity of an elderly individual (consecutive vs random or seasonal sampling of daily step count, based on observed intraindividual variations in pedometer/accelerometer data)

81 (37/44)

65–83 (71 – 4)

39

Sex, age, season and habitual physical activity of the elderly (comparisons of 1 y and/or 1 mo pedometer/accelerometer measurements [average daily step count and the daily durations of physical activity 3 METs] between men and women, between those aged 3 METs] vs calcaneal QUS parameters for bone density [SOS], structure [TI], and stiffness [OSI] and vs risks of osteoporosis and fracture)

172 (76/96)

65–83 (73 – 4)

42

Habitual physical activity and signs of metabolic syndrome in the elderly (1 y pedometer/ accelerometer measurements [average daily step count and the daily duration of physical activity >3 METs] vs the presence or absence of five diagnostic markers [BMI, TG, HDL-C, BP, and BG])

220 (91/129)

65–84 (72 – 4)

43

Habitual physical activity and HR-QOL in the elderly (1 y pedometer/accelerometer measurements [average daily step count and the daily duration of physical activity >3 METs] vs overall SF-36 score and its eight constituent dimensions [PF, RP, BP, GH, VT, SF, RE, and MH])

181 (73/108)

65–85 (73 – 6)

44

A new self-reported recall questionnaire for assessing four domains of habitual physical 3084 (1398/1686) activity (transportation, exercise/sports, housework, and labour) common among older Japanese (the first vs the second administrations of the PAQ-EJ; 1 mo pedometer/ accelerometer measurements [average daily step count and the daily durations of physical activity 3 METs] vs the corresponding scores for the PAQ-EJ; comparisons of PAQ-EJ scores between men and women, between independent and dependent individuals, and among those aged 65–74, 75–84 and 85–99 y)

65–99 (75 – 7)

45

Habitual physical activity and psychosocial status of the elderly (1 y pedometer/ accelerometer measurements [average daily step count and the daily duration of physical activity >3 METs] vs mood state [HADS], cognitive function [MMSE], and symptoms of anxiety, depression, and dementia)

65–85 (72 – 4)

46

184 (83/101)

BG = blood glucose; BMI = body mass index; BP = blood pressure[43] or bodily pain[44]; GH = general health; HADS = hospital anxiety and depression scale; HDL-C = high-density lipoprotein cholesterol; HR-QOL = health-related quality of life; METs = metabolic equivalents; MH = mental health; MMSE = mini-mental state examination; OSI = osteosonic index; PAQ-EJ = physical activity questionnaire for elderly . Japanese; %HRmax. = percentage of maximal heart rate; %HRreserve = percentage of heart rate reserve; %VO2max = percentage of maximal oxygen intake; %VO2reserve = percentage of oxygen intake reserve; PF = physical functioning; QUS = quantitative ultrasonic; RE = role limitations because of emotional problem; RP = role limitations because of physical health; RPE = rating of perceived exertion; SD = standard deviation; SF = social functioning; SF-36. = medical outcomes study 36-item Short-Form Health Survey; SOS = speed of sound; TG = triglycerides; TI = transmission index; VO2max = maximal oxygen intake; VT = vitality and energy.

device are widely available.[48-54] Motion sensors used by other investigators have differing sensitivity thresholds, filtering devices and/or ª 2009 Adis Data Information BV. All rights reserved.

attachments,[55-58] but of >10 monitors tested to date, our design of pedometer/accelerometer offers the most consistently accurate estimates of Sports Med 2009; 39 (6)

Aoyagi & Shephard

426

both step count (intramodel reliability 0.998; absolute accuracy –8 years of observation. To date, physical activity patterns have been summarized as the number of steps taken per day and the daily durations of activity in three intensity categories expressed in metabolic equivalents (6 metabolic equivalents [METs]). The primary aims of the Nakanojo Study so far have thus been to: (i) determine the overall patterns of daily physical activity that are most closely associated with good health and the absence of disease in the elderly; and (ii) identify personal, social and environmental factors that are important in allowing an amount of physical activity likely to maintain physical condition and delay the aging process. Secondary research objectives have included improving the effectiveness of interventions designed to promote habitual physical activity through the development of techniques for: (i) valid and reliable pedometer/accelerometer monitoring of such activity; and (ii) electronic evaluation and feedback of information about personal activity in the light of the various criteria noted above. Based on this extensive experience, the present article provides an overview of associations between patterns of habitual physical activity and the physical, psychosocial, mental and metabolic components of health in the elderly. It also explores possible interactions between step count and the amounts of light and moderately vigorous physical activity, and it examines psychological and meteorological factors that modulate the intensity and total volume of physical activity undertaken. Relevant evidence reviewed from other laboratories concerns mainly objective accelerometer and pedometer assessments of physical activity. 2. Characteristics of Habitual Physical Activity in the Elderly In characterizing habitual physical activity, it is important to recognize that the relative ª 2009 Adis Data Information BV. All rights reserved.

intensity of any absolute rate of working is age dependent.[59] Thus, the intensity of activities undertaken by an elderly person over a typical day (figure 1) can be divided into three categories: low (6 METs).[1,2,4] Most national and international physical activity guidelines and position statements[1-5] recommend that the elderly participate in regular physical activity as a means to prevent disease, promote health and especially to delay functional loss. The moderate intensity commonly recommended for such individuals corresponds to 50% of the person’s oxygen intake reserve (50–60% of maximal oxygen intake . [VO2max]) or 50% of the heart rate reserve (60–70% of maximal heart rate).. [1,2,4] There are significant correlations between VO2max and both maximal (r > 0.80) and preferred gait speeds (r > 0.65) in the elderly, irrespective of sex.[41] Thus, 60% of the maximal walking velocity and/or 110–115% of the preferred walking velocity are appropriate exercise recommendations for an elderly person.[41] In our laboratory, we calculate these two velocities from the time that an individual takes to move as fast as possible and at a usual comfortable pace from the 3 m to the 8 m mark on a flat 11 m walkway. Pedometer cut points corresponding to the threshold of moderate-intensity walking are roughly 100 steps/ min in both men and women.[60] To date, we have examined associations between pedometer/accelerometer measurements of habitual physical activity and several aspects of physical and psychosocial health in older people (table I). In our first analysis, physical activity patterns recorded over an entire year were summarized as the average step count per day and the daily durations of low- (3 METs) intensity activity. The daily step count (x) and the daily duration of physical activity >3 METs (y) were significantly correlated with each other in both men and women (for pooled male and female data, y = [1.96 · 10-7]x2 + [1.16 · 10-3]x; r2 = 0.93; figure 2). The form of the relationship was such that almost no physical activity >3 METs was recorded in subjects who took 3 METs; from 12 000 to 18 000 steps/day, each 1000 steps/day added another 7.5 min/day of >3 METs; and beyond 18 000 steps/day, each 1000 steps/day added another 10 min/day of >3 METs. In other words, those study participants who took 10 000 steps/day averaged >30 min/day at >3 METs. Tudor-Locke and Bassett[61] proposed classifying pedometer determinations of physical activity in healthy younger adults according to the following schemes: (i) a ‘sedentary’ lifestyle (3 METs (figure 2). The cut points are such that individuals falling in Q1 take 2000 to 3 METs, but in women, the decrease was in activity of 3 METs than with the daily step count, whereas in women, the closer association is with step count.[42,44] In the elderly, large fractions of the daily step count reflect minor movements (at an intensity 3 METs (figure 2). However, there remains an urgent need for longitudinal research to determine whether activity influences mood or whether the converse is the case. Among those with a clinical diagnosis of depression (n = 8, corresponding to 4.3% of our sample),[46] all except one male were taking 3 METs. It seems likely that the depressed individuals rarely went outdoors. A 450-day analysis of 41 participants in the Nakanojo Study[38] demonstrated an exponential decrease in physical activity from 6600 to 4000 steps/day as precipitation increased, and from this we inferred that a count 3 METs, and it is this type of activity that is associated with mental health. We also saw significant associations between daily step count, the daily duration of physical activity >3 METs and the overall HR-QOL as assessed by the medical outcomes study 36-Item Short-Form Health Survey (SF-36) scale (potential range of score 0–100)[68,69] [figure 2]. After co-varying for age, the overall HR-QOL in both men and ª 2009 Adis Data Information BV. All rights reserved.

Aoyagi & Shephard

women was substantially higher (>10 units) in the second through fourth quartiles (Q2–Q4) than in the first quartile (Q1) of physical activity, whether classified by step count or the duration of activity of >3 METs.[44] The threshold volume of habitual physical activity associated with many aspects of better physical health, ranging from higher levels of mucosal immune function such as salivary secretory IgA[26] to freedom from musculoskeletal diseases such as sarcopenia and osteoporosis,[42] is at least 7000–8000 steps/day and/or at least 15–20 min/day at an intensity >3 METs in both men and women (figure 2). This seems significantly greater than that associated with better psychological health. Among our subjects who met the minimum criteria for better physical health, all except a few females either showed no evidence of osteoporosis or exceeded the diagnostic T-score of -2.5 associated with an increased risk of fractures.[42] Part of any benefit to bone may come from ultraviolet exposure rather than from physical activity per se. The US National Institutes of Health[70] have recommended at least 15 min/day of exposure to sunlight in order to meet body vitamin D requirements and thus facilitate calcium absorption. Our observations[42] suggested associations between a larger daily volume of exercise at an intensity >3 METs (which we believe is indicative of outdoor activity in our sample) and better bone health (as indicated by higher calcaneal quantitative ultrasonic [QUS] scores). In contrast, all subjects who would be categorized as sedentary (those taking 3 METs (figure 2). The estimated risk of fractures in the lowest two quartiles (Q1 and Q2) was several times higher than that in Sports Med 2009; 39 (6)

Steps Per Day for Senior Health?

the top quartile (Q4), although the risk for this last group did not differ significantly from that for the second highest quartile (Q3).[42] Further work is needed to explore how much of the lower predicted risk of fractures in more active individuals is attributable to sunlight exposure and how much is related to the mechanical effects of exercise. The evidence of higher immune function seen in individuals taking >7000 steps/ day was substantiated by Shimizu et al.,[26] who applied a similar methodology to 284 healthy Japanese adults of comparable age (71 – 5 years; table I) over 14 days. In terms of pedometer counts, the salivary concentration and secretion rate of IgA were significantly higher for Q3 than for Q1, although salivary flow rates showed no interquartile differences.[26] Our observations[43] suggest that the threshold volume of habitual physical activity associated with absence of the metabolic syndrome may be even greater (figure 2). The five diagnostic criteria[71,72] that we adopted were: (i) a body mass index ‡25 kg/m2; (ii) a fasting serum triglyceride ‡1.7 mmol/L (‡150 mg/dL); (iii) a fasting serum high-density lipoprotein cholesterol 8000 steps/day and/or >20 min/day at an intensity >3 METs.[43] Few subjects who met such activity levels showed three or more of the selected metabolic risk factors while taking any prescribed medications.[43] These various observations seem in general keeping with current health guidelines for older adults as proposed by Health Canada and the US Centers for Disease Control and Prevention,[73] and the American College of Sports Medicine and the American Heart Association.[4] All of these groups recommend that in order to prevent disease and promote health, the elderly should ª 2009 Adis Data Information BV. All rights reserved.

431

undertake moderate-intensity aerobic activity totalling a minimum of 30 min on 5 days each week, with some recommendations offering the alternative of vigorous-intensity aerobic activity for a minimum of 20 min on at least 3 days per week. However, our observations also undeline recent WHO comments[74] that physical activity recommendations that have focused on other aspects of physical and psychosocial health may need to be increased substantially if the obesity epidemic is to be controlled and the spread of the metabolic syndrome countered. The greater levels of activity recommended by the WHO should also help to reduce the functional impairment and disability commonly found among older individuals with the metabolic syndrome.[75] Several small-scale interventional studies,[76-79] mostly on somewhat younger people, have demonstrated that at least in such individuals additional health benefits can accrue if habitual physical activity is increased further to 10 000–20 000 steps/day over a period of 6–24 weeks. In 30 hypertensive male industrial workers, specific benefits included a lowering of both systolic and diastolic blood pressure, an increased capacity for endurance exercise . (VO2max), and a reduced sympathetic nerve activity.[76] In 15 postmenopausal women with hypertension, both body mass and systolic blood pressure were reduced.[77] In nine middle-aged individuals with type 2 diabetes mellitus, waist girth and systolic blood pressure were decreased,[78] and in 14 hospitalized obese patients with type 2 diabetes, body mass was reduced and insulin sensitivity increased.[79] Although the immediate success of such interventions seems impressive, these findings are unlikely to be replicated in ‘real-world’ situations, where it is difficult to sustain high levels of physical activity. Given that many of the acute benefits are lost within a few weeks of ceasing exercise,[80] it is probably important that elderly people are encouraged to engage in modest levels of voluntary physical activity that they are likely to maintain, rather than focus on a short-term regimen of intensive and closely supervised training.[3,81,82] Applying exponential regression models to our year-long data,[42] the QUS parameter for Sports Med 2009; 39 (6)

432

the calcaneus appeared to maintain a positive association with physical activity up to a volume of 8000–10 000 steps/day (and/or 20–30 min/day at an intensity >3 METs). The OSI was only slightly greater in those who exceeded this step count;[42] we thus suspect that activity exceeding this limit is likely to add little to bone health in an elderly person. The possibility of a ceiling response is echoed in a study by Kitagawa et al.;[25] based on 7 days’ observation of 143 Japanese women aged 71 – 6 years, they reported a mean count of 8401 – 3404 steps/day, with a quadratic attenuation of increments in the calcaneal QUS parameter above 12 000 steps/day. Nevertheless, their estimates of both the mean level of activity and the turning point are at least 2000 steps/day greater than what we observed in Japanese women of similar age (table I; 6288 – 2556 and at most 10 000 steps/day, respectively). Possibly, the discrepancy reflects the brief unblinded observation period in the study of Kitagawa and associates[25] (see the 1-week study of Clemes et al.,[22] in which there was a reactive increase of pedometer counts [from 9541 – 3186 to 11 385 – 3763 steps/day] when subjects knew that their activity patterns were being recorded). Our current data[39] show that >37 consecutive days of observation are required to achieve >90% reliability when estimating the annual habitual physical activity of a female who no longer has formal employment. It is possible that Kitagawa et al.[25] would have reached similar conclusions to ours given a longer period of observation. Certainly, the existence and level of any such ceiling need further exploration. In summary, accurate year-long observations of habitual physical activity suggest that better overall health is seen in elderly individuals who take an average of >8000 steps/day and/or spend >20 min/day at an intensity >3 METs (figure 2). Moreover, the extent of physical and psychosocial health seems greater in older individuals who undertake a larger proportion of their daily activity at an intensity >3 METs (i.e. those who are on or above the dotted line in figure 2). Thus, it would be useful to test the causal nature of the inference that among those who currently take a baseline of 5 or >10 min/day of physical activity >3 METs would lead to practical and clinically significant health advantages (figure 2). 4. Factors Affecting Habitual Physical Activity in the Elderly Consistent participation in moderate levels of physical activity appears necessary to optimize health.[1-5] Various authors have identified the day of the week[31,39,83,84] and the month or season of the year[38-40,83-91] as periodic factors that commonly influence the physical activity of free-living humans. The periods of 2.3, 3.5 and 7.0 days observed in power spectrum analyses of our individual step count data appear unrelated to sex, age or exercise habit,[39] but they may be modulated substantially by symptoms of anxiety and depression. The data suggest that in those with symptoms of anxiety and/or depression, intraindividual variations within a 1-week period are less marked than in the remainder of our sample, whereas the variations over an approximately 3-month period are greater than in other subjects. The implication may be that a depressed person has less interest in and/or less concern for things, and in consequence his or her lifestyle becomes more monotonous. However, this tendency may vary with the season (particularly in those with seasonal affective disorder)[92] and other short-term fluctuations of mood state. Variability in habitual physical activity is linked not only to common endogenous factors such as mood state, but also to exogenous influences, particularly precipitation, day length, and the range of ambient temperatures.[38-40,83-91] Public health recommendations[86] have noted that a short day length and extremes of ambient temperature are potential barriers to healthenhancing outdoor physical activity. Such environmental factors contribute to commonly observed seasonal variations in habitual physical activity. Several cross-sectional[90,91] and longitudinal studies[83-89] of younger adults from both Europe and North America have reported that in most climates, a person’s physical activity peaks in summer and reaches its nadir in winter. Sports Med 2009; 39 (6)

Steps Per Day for Senior Health?

Looking at a large US population, the Centers for Disease Control and Prevention[86] reported that leisure-time physical activity was increased during the summer and reduced during the winter months; this trend seemed true for both sexes and most age and racial/ethnic groups. Similarly, Pivarnik et al.[89] demonstrated that the average weekly leisure-time energy expenditure of 2843 Michigan adults was 15–20% higher during the spring and summer months than at other seasons. Applying a 24-hour physical activity recall technique to 580 Massachusetts residents five times per year, Matthews et al.[88] demonstrated seasonal changes in total physical activity of 1.4 MET h/day (507 kJ/day [121 kcal/day]) in men and 1.0 MET h/day (293 kJ/day [70 kcal/day]) in women. Physical activity peaked in July, and variations seemed to be associated with environmental conditions, specifically precipitation (measured as the number of rainy days per month), mean ambient temperature, and number of hours of daylight before 8am and after 5pm.[88] However, detailed study of relationships between environmental factors and habitual physical activity has been hampered by limitations in the available data sets, particularly the frequency of collection and the reliability/validity of the data.[93] The full extent of differences in physical activity between the most and the least active months can be detected by accurate year-long pedometer/accelerometer measurements,[38,40] but may be missed by infrequent subjective questionnaire estimates of physical activity.[85,88-90] Thus, the magnitude of the seasonal differences (15–20%) reported by Pivarnik et al.[89] is only a half of that (30–40%) detected by our full-year observations.[40] We examined relationships between the daily physical activity of free-living elderly people and selected environmental factors (day length, mean ambient temperature, duration of bright sunshine, mean wind speed, mean relative humidity and precipitation).[38] Habitual physical activity decreased exponentially from 6600 to 4000 steps/ day with increasing precipitation.[38] We suggested that a count winter), day of the week (weekday > weekend), type of day (workday vs non-workday), and participation in sport/exercise (day with > day without sport/exercise). Our continuous yearlong pedometer/accelerometer assessments of 95 seniors,[40] likewise, showed clear seasonal variations in the month-averaged daily step count and the daily durations of physical activity 3 METs, irrespective of sex or age. Our selected measurements of habitual physical activity peaked in spring and/or autumn and reached their nadir in the winter months.[40] Many of the acute health benefits of physical activity are lost within a few weeks of ceasing exercise.[80] It is therefore possible that environmental factors could cause seasonal variations in the immediate risk of various chronic health conditions. The incidence of and mortality from coronary heart disease do indeed exhibit winter peaks in countries both north and south of the equator,[108] although this is probably due more to the direct effects of cold on blood pressure and the effects of snow-shovelling than to short-term reductions in habitual physical activity. Likewise, unusually hot spells are associated with an increase in deaths,[109] in part due to the direct circulatory effects of heat strain, but possibly influenced also by a seasonal decrease in the volume of moderate physical activity. In summer, when our daily step count approximated the average for the year, the proportion of activity 3 METs,[40] probably a manifestation of behavioural thermoregulation as described above.[97,98] Behavioural thermoregulation might also explain our findings[40] that physical activity 3 METs peaked in November, when the temperature was decreasing. Such observations suggest that in order to optimize health-promotional initiatives and reduce both direct and indirect health risks, due account should be taken of seasonal changes in the microclimate of elderly people. In many parts of the world, there is a need to encourage indoor (air-conditioned or climate-controlled) physical activity, such as mall walking, during extremes of summer and winter weather in order to maintain participation in at least moderate levels of physical activity. In areas where indoor facilities are not readily available, the time of day when exercise is taken can be changed, or a less stressful seasonally appropriate climate-adjusted target chosen. For instance, in a region where mean ambient temperatures range from -5C to 30C, older adults might attain a yearly average of 8000 steps/day and 20 min/day of physical activity >3 METs (figure 2) by taking 9000 steps/day and spending 25 min/day at an intensity >3 METs in the spring, 8000 steps/day and 15 min/day at >3 METs in the summer, 9000 steps/day and 30 min/day at >3 METs in the autumn, and 6000 steps/day and 10 min/day at >3 METs in the winter. 5. Conclusions Current cross-sectional data indicate that the overall health of older people is associated with both the year-averaged daily duration of physical activity undertaken at an intensity >3 METs and the year-averaged daily step count. In men, the extent of health seems associated more closely with the daily duration of activity >3 METs than with the daily step count, whereas in women, the closer association is with step count. In both sexes, the threshold amount of physical activity associated with better physical health is >8000 steps/day and/or >20 min/day at an intensity Sports Med 2009; 39 (6)

Steps Per Day for Senior Health?

>3 METs. The corresponding figures for better mental health are >4000 steps/day and/or >5 min/day at >3 METs. The physical health data can be equated with a daily total of at least 20 min of moderate activity (at a pace of some 1.4 m/s [5 km/h]) and at least 60 min of light activity. In contrast, good mental health seems associated with quite limited amounts of deliberate physical activity (although it remains to be assessed how far poor mental health is responsible for a lack of physical activity, rather than the converse). Plainly, the observed associations between physical activity and health outcomes also need examining longitudinally in order to test causal inferences and elucidate mediating variables. Both the intensity and the total amount of physical activity are influenced by meteorological factors, particularly mean ambient temperature and the extent of precipitation. In our study, habitual physical activity decreased exponentially to approximately 4000 steps/day with significant precipitation. Excluding the influence of rainfall, the daily step count peaked at a mean outdoor temperature of around 17C; above and especially below this temperature, physical activity decreased in a quadratic fashion. Seasonal changes in microclimate need to be taken into account when designing interventions to increase the physical activity of elderly people throughout the entire year. Acknowledgements This article focuses particularly on data from an interdisciplinary study on the habitual physical activity and health of elderly people living in Nakanojo, Gunma, Japan (the Nakanojo Study). The Nakanojo Study was supported in part by grants (Grant-in-Aid for Encouragement of Young Scientists: 12770037 and Grant-in-Aid for Scientific Research [C]: 15500503, [C]: 17500493, and [B]: 19300235) from the Japan Society for the Promotion of Science. The authors gratefully acknowledge the expert technical assistance of the research and nursing staffs of the Tokyo Metropolitan Institute of Gerontology (particularly Mr Hyuntae Park, Mr Sungjin Park, Dr Fumiharu Togo, Dr Akitomo Yasunaga and Mr Eiji Watanabe), The University of Tokyo (especially Dr Kazuhiro Yoshiuchi), and the Nakanojo Public Health Center. We would also like to thank the subjects whose participation made the Nakanojo Study possible. No sources of funding were used to assist in the preparation of this manuscript. The authors have no conflicts of interest that are directly relevant to the content of this manuscript.

ª 2009 Adis Data Information BV. All rights reserved.

435

References 1. American College of Sports Medicine Position Stand. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Med Sci Sports Exerc 1998 Jun; 30 (6): 975-91 2. American College of Sports Medicine Position Stand. Exercise and physical activity for older adults. Med Sci Sports Exerc 1998 Jun; 30 (6): 992-1008 3. Aoyagi Y, Shephard RJ. Aging and muscle function. Sports Med 1992 Dec; 14 (6): 376-96 4. Nelson ME, Rejeski WJ, Blair SN, et al. Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc 2007 Aug; 39 (8): 1435-45 5. Shephard RJ. Aging, physical activity, and health. Champaign (IL): Human Kinetics, 1997 6. Blair SN, Haskell WL. Objectively measured physical activity and mortality in older adults. JAMA 2006 Jul 12; 296 (2): 216-8 7. Brahm H, Mallmin H, Michaelsson K, et al. Relationships between bone mass measurements and lifetime physical activity in a Swedish population. Calcif Tissue Int 1998 May; 62 (5): 400-12 8. Coupland CA, Grainge MJ, Cliffe SJ, et al. Occupational activity and bone mineral density in postmenopausal women in England. Osteoporos Int 2000 May; 11 (4): 310-5 9. Evans EM, Ross KM, Heinrichs KL, et al. Ultrasound of the calcaneus and bone mineral density differs in older black and white women but is not impacted by current physical activity. Osteoporos Int 2005 Dec; 16 (12): 1755-60 10. Greendale GA, Barrett-Connor E, Edelstein S, et al. Lifetime leisure exercise and osteoporosis: the Rancho Bernardo study. Am J Epidemiol 1995 May 15; 141 (10): 951-9 11. Kakiyama T, Matsuda M, Koseki S. Effect of physical activity on the distensibility of the aortic wall in healthy males. Angiology 1998 Sep; 49 (9): 749-57 12. Laaksonen DE, Lakka HM, Salonen JT, et al. Low levels of leisure-time physical activity and cardiorespiratory fitness predict development of the metabolic syndrome. Diabetes Care 2002 Sep; 25 (9): 1612-8 13. Lee C, Russell A. Effects of physical activity on emotional well-being among older Australian women: crosssectional and longitudinal analyses. J Psychosom Res 2003 Feb; 54 (2): 155-60 14. Medras M, Slowinska-Lisowska M, Jozkow P. Impact of recreational physical activity on bone mineral density in middle-aged men. Aging Male 2005 Sep-Dec; 8 (3-4): 162-5 15. Nguyen TV, Center JR, Eisman JA. Osteoporosis in elderly men and women: effects of dietary calcium, physical activity, and body mass index. J Bone Miner Res 2000 Feb; 15 (2): 322-31 16. Pescatello LS, Murphy DM, Anderson D, et al. Daily physical movement and bone mineral density among a mixed racial cohort of women. Med Sci Sports Exerc 2002 Dec; 34 (12): 1966-70 17. Washburn RA, McAuley E, Katula J, et al. The physical activity scale for the elderly (PASE): evidence for validity. J Clin Epidemiol 1999 Jul; 52 (7): 643-51

Sports Med 2009; 39 (6)

436

18. Stewart AL, Mills KM, King AC, et al. CHAMPS physical activity questionnaire for older adults: outcomes for interventions. Med Sci Sports Exerc 2001 Jul; 33 (7): 1126-41 19. Bennett GG, Wolin KY, Puleo E, et al. Pedometerdetermined physical activity among multiethnic lowincome housing residents. Med Sci Sports Exerc 2006 Apr; 38 (4): 768-73 20. Cavanaugh JT, Coleman KL, Gaines JM, et al. Using step activity monitoring to characterize ambulatory activity in community-dwelling older adults. J Am Geriatr Soc 2007 Jan; 55 (1): 120-4 21. Chan CB, Spangler E, Valcour J, et al. Cross-sectional relationship of pedometer-determined ambulatory activity to indicators of health. Obes Res 2003 Dec; 11 (12): 1563-70 22. Clemes SA, Matchett N, Wane SL. Reactivity: an issue for short-term pedometer studies? Br J Sports Med 2008 Jan; 42 (1): 68-70 23. Dwyer T, Hosmer D, Hosmer T, et al. The inverse relationship between number of steps per day and obesity in a population-based sample: the AusDiab study. Int J Obes 2007 May; 31 (5): 797-804 24. Gerdhem P, Dencker M, Ringsberg K, et al. Accelerometermeasured daily physical activity among octogenerians: results and associations to other indices of physical performance and bone density. Eur J Appl Physiol 2008 Jan; 102 (2): 173-80 25. Kitagawa J, Omasu F, Nakahara Y. Effect of daily walking steps on ultrasound parameters of the calcaneus in elderly Japanese women. Osteoporos Int 2003 May; 14 (3): 219-24 26. Shimizu K, Kimura F, Akimoto T, et al. Effect of freeliving daily physical activity on salivary secretory IgA in elderly. Med Sci Sports Exerc 2007 Apr; 39 (4): 593-8 27. Strycker LA, Duncan SC, Chaumeton NR, et al. Reliability of pedometer data in samples of youth and older women. Int J Behav Nutr Phys Act 2007 Feb 17; 4: 4 28. Sugawara J, Otsuki T, Tanabe T, et al. Physical activity duration, intensity, and arterial stiffening in postmenopausal women. Am J Hypertens 2006 Oct; 19 (10): 1032-6 29. Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 2008 Jan; 40 (1): 181-8 30. Trost SG, McIver KL, Pate RR. Conducting accelerometerbased activity assessments in field-based research. Med Sci Sports Exerc 2005 Nov; 37 (11 Suppl.): 531S-43S 31. Tudor-Locke C, Burkett L, Reis JP, et al. How many days of pedometer monitoring predict weekly physical activity in adults? Prev Med 2005 Mar; 40 (3): 293-8 32. Tudor-Locke C, Bassett Jr DR, Rutherford WJ, et al. BMIreferenced cut points for pedometer-determined steps per day in adults. J Phys Act Health 2008 Jan; 5 (1 Suppl.): 126S-39S 33. Walsh JT, Charlesworth A, Andrews R, et al. Relation of daily activity levels in patients with chronic heart failure to long-term prognosis. Am J Cardiol 1997 May 15; 79 (10): 1364-9 34. Ward DS, Evenson KR, Vaughn A, et al. Accelerometer use in physical activity: best practices and research recommendations. Med Sci Sports Exerc 2005 Nov; 37 (11 Suppl.): 582S-8S

ª 2009 Adis Data Information BV. All rights reserved.

Aoyagi & Shephard

35. Westerterp KR. Pattern and intensity of physical activity. Nature 2001 Mar 29; 410 (6828): 539 36. Wyatt HR, Peters JC, Reed GW, et al. A Colorado statewide survey of walking and its relation to excessive weight. Med Sci Sports Exerc 2005 May; 37 (5): 724-30 37. Yoshioka M, Ayabe M, Yahiro T, et al. Long-period accelerometer monitoring shows the role of physical activity in overweight and obesity. Int J Obes 2005 May; 29 (5): 502-8 38. Togo F, Watanabe E, Park H, et al. Meteorology and the physical activity of the elderly: the Nakanojo Study. Int J Biometeorol 2005 Nov; 50 (2): 83-9 39. Togo F, Watanabe E, Park H, et al. How many days of pedometer use predict the annual activity of the elderly reliably? Med Sci Sports Exerc 2008 Jun; 40 (6): 1058-64 40. Yasunaga A, Togo F, Watanabe E, et al. Sex, age, season, and habitual physical activity of older Japanese: the Nakanojo Study. J Aging Phys Act 2008 Jan; 16 (1): 3-13 41. Aoyagi Y, Togo F, Matsuki S, et al. Walking velocity measured over 5 m as a basis of exercise prescription for the elderly: preliminary data from the Nakanojo Study. Eur J Appl Physiol 2004 Oct; 93 (1-2): 217-23 42. Park H, Togo F, Watanabe E, et al. Relationship of bone health to yearlong physical activity in older Japanese adults: cross-sectional data from the Nakanojo Study. Osteoporos Int 2007 Mar; 18 (3): 285-93 43. Park S, Park H, Togo F, et al. Yearlong physical activity and metabolic syndrome in older Japanese adults: crosssectional data from the Nakanojo Study. J Gerontol A Biol Sci Med Sci 2008 Oct; 63 (10): 1119-23 44. Yasunaga A, Togo F, Watanabe E, et al. Yearlong physical activity and health-related quality of life in older Japanese adults: the Nakanojo Study. J Aging Phys Act 2006 Jul; 14 (3): 288-301 45. Yasunaga A, Park H, Watanabe E, et al. Development and evaluation of the physical activity questionnaire for elderly Japanese: the Nakanojo Study. J Aging Phys Act 2007 Oct; 15 (4): 398-411 46. Yoshiuchi K, Nakahara R, Kumano H, et al. Yearlong physical activity and depressive symptoms in older Japanese adults: cross-sectional data from the Nakanojo Study. Am J Geriatr Psychiatry 2006 Jul; 14 (7): 621-4 47. Janz KF. Physical activity in epidemiology: moving from questionnaire to objective measurement. Br J Sports Med 2006 Mar; 40 (3): 191-2 48. Crouter SE, Schneider PL, Karabulut M, et al. Validity of 10 electronic pedometers for measuring steps, distance, and energy cost. Med Sci Sports Exerc 2003 Aug; 35 (8): 1455-60 49. Kumahara H, Schutz Y, Ayabe M, et al. The use of uniaxial accelerometry for the assessment of physicalactivity-related energy expenditure: a validation study against whole-body indirect calorimetry. Br J Nutr 2004 Feb; 91 (2): 235-43 50. Kumahara H, Ishii K, Tanaka H. Physical activity monitoring for health management: practical techniques and methodological issues. Int J Sport Health Sci 2006; 4: 380-93 51. Rafamantanantsoa HH, Ebine N, Yoshioka M, et al. Validation of three alternative methods to measure total

Sports Med 2009; 39 (6)

Steps Per Day for Senior Health?

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

energy expenditure against the doubly labeled water method for older Japanese men. J Nutr Sci Vitaminol 2002 Dec; 48 (6): 517-23 Schneider PL, Crouter SE, Lukajic O, et al. Accuracy and reliability of 10 pedometers for measuring steps over a 400-m walk. Med Sci Sports Exerc 2003 Oct; 35 (10): 1779-84 Schneider PL, Crouter SE, Bassett DR. Pedometer measures of free-living physical activity: comparison of 13 models. Med Sci Sports Exerc 2004 Feb; 36 (2): 331-5 Warms C. Physical activity measurement in persons with chronic and disabling conditions: methods, strategies, and issues. Fam Community Health 2006 Jan-Mar; 29 (1 Suppl.): 78S-88S Le Masurier GC, Tudor-Locke C. Comparison of pedometer and accelerometer accuracy under controlled conditions. Med Sci Sports Exerc 2003 May; 35 (5): 867-71 Le Masurier GC, Lee SM, Tudor-Locke C. Motion sensor accuracy under controlled and free-living conditions. Med Sci Sports Exerc 2004 May; 36 (5): 905-10 McClain JJ, Sisson SB, Washington TL, et al. Comparison of Kenz Lifecorder EX and ActiGraph accelerometers in 10-year-old children. Med Sci Sports Exerc 2007 Apr; 39 (4): 630-8 Tudor-Locke C, Ainsworth BE, Thompson RW, et al. Comparison of pedometer and accelerometer measures of free-living physical activity. Med Sci Sports Exerc 2002 Dec; 34 (12): 2045-51 Bouchard C, Shephard RJ. The model and key concepts. In: Bouchard C, Shephard RJ, Stephens T, editors. Physical activity, fitness and health. Champaign (IL): Human Kinetics, 1994: 77-97 Tudor-Locke C, Sisson SB, Collova T, et al. Pedometerdetermined step count guidelines for classifying walking intensity in a young ostensibly healthy population. Can J Appl Physiol 2005 Dec; 30 (6): 666-76 Tudor-Locke C, Bassett Jr DR. How many steps/day are enough? Preliminary pedometer indices for public health. Sports Med 2004 Jan; 34 (1): 1-8 Hill JO, Wyatt HR, Reed GW, et al. Obesity and the environment: where do we go from here? Science 2003 Feb 7; 299 (5608): 853-5 Bravata DM, Smith-Spangler C, Sundaram V, et al. Using pedometers to increase physical activity and improve health: a systematic review. JAMA 2007 Nov 21; 298 (19): 2296-304 Caspersen CJ, Pereira MA, Curran KM. Changes in physical activity patterns in the United States, by sex and cross-sectional age. Med Sci Sports Exerc 2000 Sep; 32 (9): 1601-9 Martinez-Gonzalez MA, Varo JJ, Santos JL, et al. Prevalence of physical activity during leisure time in the European Union. Med Sci Sports Exerc 2001 Jul; 33 (7): 1142-6 Weiss DR, O’Loughlin JL, Platt RW, et al. Five-year predictors of physical activity decline among adults in lowincome communities: a prospective study. Int J Behav Nutr Phys Act 2007 Jan 18; 4: 2 Japan Ministry of Health, Labour and Welfare. The national nutrition survey in Japan 2001 [in Japanese]. Tokyo: Daiichishuppan, 2003

ª 2009 Adis Data Information BV. All rights reserved.

437

68. Fukuhara S, Suzukamo Y. Manual of SF-36v2 Japanese version [in Japanese]. Kyoto: Institute for Health Outcomes & Process Evaluation Research, 2004 69. Ware JE, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): I, conceptual framework and item selection. Med Care 1992 Jun; 30 (6): 473-83 70. National Institutes of Health Osteoporosis and Related Bone Diseases, National Research Center. Bone health and osteoporosis: a guide for Asian women aged 50 and older. Bethesda (MD): National Institutes of Health, 2005 71. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001 May 16; 285 (19): 2486-97 72. World Health Organization Western Pacific Region, International Association for the Study of Obesity, International Obesity Task Force. The Asia-Pacific perspective: redefining obesity and its treatment. Sydney (NSW): Health Communications Australia Pty Limited, 2000 73. Shephard RJ. Whistler 2001: a Health Canada/CDC conference on ‘‘Communicating physical activity and health messages: science into practice’’. Am J Prev Med 2002 Oct; 23 (3): 221-5 74. Erlichman J, Kerbey AL, James WP. Physical activity and its impact on health outcomes: paper 2, prevention of unhealthy weight gain and obesity by physical activity – an analysis of the evidence. Obes Rev 2002 Nov; 3 (4): 273-87 75. Blaum CS, West NA, Haan MN. Is the metabolic syndrome, with or without diabetes, associated with progressive disability in older Mexican Americans? J Gerontol A Biol Sci Med Sci 2007 Jul; 62 (7): 766-73 76. Iwane M, Arita M, Tomimoto S, et al. Walking 10 000 steps/day or more reduces blood pressure and sympathetic nerve activity in mild essential hypertension. Hypertens Res 2000 Nov; 23 (6): 573-80 77. Moreau KL, Degarmo R, Langley J, et al. Increasing daily walking lowers blood pressure in postmenopausal women. Med Sci Sports Exerc 2001 Nov; 33 (11): 1825-31 78. Tudor-Locke CE, Myers AM, Bell RC, et al. Preliminary outcome evaluation of the First Step Program: a daily physical activity intervention for individuals with type 2 diabetes. Patient Educ Couns 2002 May; 47 (1): 23-8 79. Yamanouchi K, Shinozaki T, Chikada K, et al. Daily walking combined with diet therapy is a useful means for obese NIDDM patients not only to reduce body weight but also to improve insulin sensitivity. Diabetes Care 1995 Jun; 18 (6): 775-8 80. McArdle W, Katch F, Katch V. Exercise physiology: energy, nutrition, and human performance. 3rd rev. ed. Philadelphia (PA): Lea & Febiger, 1991 81. Aoyagi Y, Katsuta S. Relationship between the starting age of training and physical fitness in old age. Can J Sport Sci 1990 Mar; 15 (1): 65-71 82. Aoyagi Y, Katsuta S. The starting age of training and its effect on reduction in physical performance capability with aging. In: Kaneko M, editor. Fitness for the aged, disabled, and industrial worker. International series on

Sports Med 2009; 39 (6)

Aoyagi & Shephard

438

83.

84.

85.

86.

87.

88.

89.

90. 91.

92. 93.

94.

95.

96.

sports sciences, 20. Champaign (IL): Human Kinetics, 1990: 118-24 Chan CB, Ryan DA, Tudor-Locke C. Relationship between objective measures of physical activity and weather: a longitudinal study. Int J Behav Nutr Phys Act 2006 Aug 7; 3: 21 Tudor-Locke C, Bassett DR, Swartz AM, et al. A preliminary study of one year of pedometer self-monitoring. Ann Behav Med 2004 Dec; 28 (3): 158-62 Bergstralh EJ, Sinaki M, Offord KP, et al. Effect of season on physical activity score, back extensor muscle strength, and lumbar bone mineral density. J Bone Miner Res 1990 Apr; 5 (4): 371-7 Centers for Disease Control and Prevention. Monthly estimates of leisure-time physical inactivity – United States, 1994. MMWR Morb Mortal Wkly Rep 1997 May 9; 46 (18): 393-7 Haggarty P, McNeill G, Manneh MK, et al. The influence of exercise on the energy requirements of adult males in the UK. Br J Nutr 1994 Dec; 72 (6): 799-813 Matthews CE, Freedson PS, Hebert JR, et al. Seasonal variation in household, occupational, and leisure time physical activity: longitudinal analyses from the seasonal variation of blood cholesterol study. Am J Epidemiol 2001 Jan 15; 153 (2): 172-83 Pivarnik JM, Reeves MJ, Rafferty AP. Seasonal variation in adult leisure-time physical activity. Med Sci Sports Exerc 2003 Jun; 35 (6): 1004-8 Uitenbroek DG. Seasonal variation in leisure time physical activity. Med Sci Sports Exerc 1993 Jun; 25 (6): 755-60 US Department of Health and Human Services. Patterns and trends in physical activity. In: Physical activity and health: a report of the Surgeon General. Atlanta (GA): US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, 1996: 173-208 Lurie SJ, Gawinski B, Pierce D, et al. Seasonal affective disorder. Am Fam Physician 2006 Nov 1; 74 (9): 1521-4 Trost SG, Owen N, Bauman AE, et al. Correlates of adults’ participation in physical activity: review and update. Med Sci Sports Exerc 2002 Dec; 34 (12): 1996-2001 Czeisler CA, Dumont M, Duffy JF, et al. Association of sleep-wake habits in older people with changes in output of circadian pacemaker. Lancet 1992 Oct 17; 340 (8825): 933-6 Duffy JF, Dijk DJ, Klerman EB, et al. Later endogenous circadian temperature nadir relative to an earlier wake time in older people. Am J Physiol 1998 Nov; 275 (5 Pt 2): R1478-87 Togo F, Aizawa S, Arai J, et al. Influence on human sleep patterns of lowering and delaying the minimum core body temperature by slow changes in the thermal environment. Sleep 2007 Jun 1; 30 (6): 797-802

ª 2009 Adis Data Information BV. All rights reserved.

97. Aoyagi Y. Endurance training, heat acclimation, and protective clothing: the thermophysiology of exercising in a hot climate [dissertation]. Toronto (ON): University of Toronto, 1996 98. Aoyagi Y, McLellan TM, Shephard RJ. Interactions of physical training and heat acclimation: the thermophysiology of exercising in a hot climate. Sports Med 1997 Mar; 23 (3): 173-210 99. Aoyagi Y, McLellan TM, Shephard RJ. Effects of training and acclimation on heat tolerance in exercising men wearing protective clothing. Eur J Appl Physiol Occup Physiol 1994 Mar; 68 (3): 234-45 100. Aoyagi Y, McLellan TM, Shephard RJ. Effects of 6 versus 12 days of heat acclimation on heat tolerance in lightly exercising men wearing protective clothing. Eur J Appl Physiol Occup Physiol 1995 Mar; 71 (2-3): 187-96 101. Aoyagi Y, McLellan TM, Shephard RJ. Determination of body heat storage in clothing: calorimetry versus thermometry. Eur J Appl Physiol Occup Physiol 1995 Mar; 71 (2-3): 197-206 102. Aoyagi Y, McLellan TM, Shephard RJ. Determination of body heat storage: how to select the weighting of rectal and skin temperatures for clothed subjects. Int Arch Occup Environ Health 1996 Jun; 68 (5): 325-36 103. Aoyagi Y, McLellan TM, Shephard RJ. Residual analysis in the determination of factors affecting the estimates of body heat storage in clothed subjects. Eur J Appl Physiol Occup Physiol 1996 May; 73 (3-4): 287-98 104. Aoyagi Y, McLellan TM, Shephard RJ. Effects of endurance training and heat acclimation on psychological strain in exercising men wearing protective clothing. Ergonomics 1998 Mar; 41 (3): 328-57 105. McLellan TM, Aoyagi Y. Heat strain in protective clothing following hot-wet or hot-dry heat acclimation. Can J Appl Physiol 1996 Apr; 21 (2): 90-108 106. Gordon CJ. Relationship between preferred ambient temperature and autonomic thermoregulatory function in rat. Am J Physiol 1987 Jun; 252 (6 Pt 2): R1130-7 107. Bruce DG, Devine A, Prince RL. Recreational physical activity levels in healthy older women: the importance of fear of falling. J Am Geriatr Soc 2002 Jan; 50 (1): 84-9 108. Pell JP, Cobbe SM. Seasonal variations in coronary heart disease. QJM 1999 Dec; 92 (12): 689-96 109. Qiu D, Tanihata T, Aoyama H, et al. Relationship between a high mortality rate and extreme heat during the summer of 1999 in Hokkaido Prefecture, Japan. J Epidemiol 2002 May; 12 (3): 254-7

Correspondence: Dr Yukitoshi Aoyagi, Exercise Sciences Research Group, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakaecho, Itabashi-ku, Tokyo 173-0015, Japan. E-mail: [email protected]

Sports Med 2009; 39 (6)

Sports Med 2009; 39 (6): 439-468 0112-1642/09/0006-0439/$49.95/0

REVIEW ARTICLE

ª 2009 Adis Data Information BV. All rights reserved.

Exercise and Bone Mass in Adults Amelia Guadalupe-Grau, Teresa Fuentes, Borja Guerra and Jose A.L. Calbet Department of Physical Education, University of Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Canary Islands, Spain

Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Experiments with Animals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Studies with Humans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Cross-Sectional Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Young Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Young Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Longitudinal Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Young Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Young Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3 Premenopausal Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4 Middle-Aged Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Aging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Postmenopausal Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Older Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Practical Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

439 442 443 443 443 447 447 448 452 453 454 454 455 460 461 462

There is a substantial body of evidence indicating that exercise prior to the pubertal growth spurt stimulates bone growth and skeletal muscle hypertrophy to a greater degree than observed during growth in non-physically active children. Bone mass can be increased by some exercise programmes in adults and the elderly, and attenuate the losses in bone mass associated with aging. This review provides an overview of cross-sectional and longitudinal studies performed to date involving training and bone measurements. Crosssectional studies show in general that exercise modalities requiring high forces and/or generating high impacts have the greatest osteogenic potential. Several training methods have been used to improve bone mineral density (BMD) and content in prospective studies. Not all exercise modalities have shown positive effects on bone mass. For example, unloaded exercise such as swimming has no impact on bone mass, while walking or running has limited positive effects. It is not clear which training method is superior for bone stimulation in adults, although scientific evidence points to a combination of high-impact (i.e. jumping) and weight-lifting exercises. Exercise involving high impacts, even a relatively small amount, appears to be the most efficient for enhancing bone mass, except in postmenopausal women. Several types of resistance exercise have been tested also with positive results, especially when the

Guadalupe-Grau et al.

440

intensity of the exercise is high and the speed of movement elevated. A handful of other studies have reported little or no effect on bone density. However, these results may be partially attributable to the study design, intensity and duration of the exercise protocol, and the bone density measurement techniques used. Studies performed in older adults show only mild increases, maintenance or just attenuation of BMD losses in postmenopausal women, but net changes in BMD relative to control subjects who are losing bone mass are beneficial in decreasing fracture risk. Older men have been less studied than women, and although it seems that men may respond better than their female counterparts, the experimental evidence for a dimorphism based on sex in the osteogenic response to exercise in the elderly is weak. A randomized longitudinal study of the effects of exercise on bone mass in elderly men and women is still lacking. It remains to be determined if elderly females need a different exercise protocol compared with men of similar age. Impact and resistance exercise should be advocated for the prevention of osteoporosis. For those with osteoporosis, weight-bearing exercise in general, and resistance exercise in particular, as tolerated, along with exercise targeted to improve balance, mobility and posture, should be recommended to reduce the likelihood of falling and its associated morbidity and mortality. Additional randomized controlled trials are needed to determine the most efficient training loads depending on age, sex, current bone mass and training history for improvement of bone mass.

The most important function of bone tissue is to withstand and transmit forces without breaking. The strength of bone depends on the amount of tissue, its material composition and how bone material is organized microarchitecturally and geometrically (shape and size).[1,2] As summarized by Seeman and Delmas,[3] optimal bone tissue characteristics are defined by optimal levels of stiffness, flexibility and lightness. To efficiently withstand and transmit loads, bone must be stiff and able to resist deformation. However, it cannot be too stiff – i.e. unable to absorb some energy by shortening and widening when compressed, and by lengthening and narrowing when submitted to traction – otherwise the energy imposed during loading will be released by structural failure. Conversely, bone cannot be too flexible, because on loading it could easily deform beyond its peak strain, and fracture.[3] Bone must also have the ability to continually adapt to changes in physiological and mechanical environment. The mechanical properties of bone are determined by two major factors: the characteristics ª 2009 Adis Data Information BV. All rights reserved.

of the collagen matrix and the degree of mineralization, i.e. the amount of calcium hydroxyapatite crystals deposited on and between the collagen fibres. Bone strength is primarily determined by tissue mass and stiffness. While stiffness is mainly determined by the mineral phase,[4-7] the collagen matrix contributes primarily to bone toughness resilience (i.e. the ability to absorb energy without breaking).[8-10] Increasing bone mineral density (BMD) results in greater stiffness but lower flexibility.[11] Collagen, of which about 95% is type I collagen, comprises about 80% of the total protein in bone.[12] Collagen fibres are packed together by the formation of inter- and intramolecular crosslinks. Mature crosslinks such as pyridinoline (PYD) and deoxypyridinoline (DPD) reach a maximum concentration between 15 and 40 years of age, and their concentrations are lower in trabecular bone than in cortical bone.[13] If there are too many crosslinks, the ability to absorb energy diminishes, i.e. the bone becomes more brittle. Likewise, without the collagen matrix the bone becomes less elastic and more brittle.[14] In Sports Med 2009; 39 (6)

Exercise and Bone Mass in Adults

humans it has been shown that the compressive biomechanical ultimate strength of bone is correlated, independently of BMD, with the ratio PYD/DPD, but not with PYD, DPD or pyrrole separately.[15] Non-fibrillar organic matrix acts as the ‘glue’ that holds the mineralized fibrils together.[16] Bone strength also depends on the orientation of osteons (and thus collagen fibres) within the cortical bone.[17] Longitudinal fibres are found in regions supporting tensile loads, while transverse fibres predominate in regions under compressive loading.[2,18] Part of the bone plasticity in response to loading depends on its capacity to reorient its collagen fibres. For example, it has been reported in dogs that a 10% reduction in vertebral BMD elicited by a strenuous progressive running programme (up to 40 km/day) for 1 year did not change the bone mechanical properties.[19] These dogs, compared with their sedentary counterparts, showed reorganization of the collagen fibres in a more parallel manner without changes in the concentration of crosslinks, suggesting that collagen reorganization during exercise may contribute to the maintenance of bone strength despite decreased mineral density.[19] Bone mechanical properties are modified depending on loading, such that bone strength is enhanced or reduced in response to either increased or reduced mechanical loading.[3,20-22] The adaptive response is very complex and depends on the characteristics of loading history, but also on systemic and local factors, which include neuroendocrine, endocrine and paracrine changes in metabolites, cytokines, growth factors, hormones, vitamins and minerals.[23-27] Excellent reviews have been published recently on the molecular mechanisms that mediate adaptive responses of bone tissue to changes in loading, and the interested reader is referred to them.[24-26,28,29] The main signals for bone adaptation to mechanical loading are the rate and magnitude of strain, which should reach minimal levels or threshold to elicit structural modifications in bone.[30-33] To enhance bone mass or BMD in non-physically active humans, bone tissue must be submitted to mechanical strains above those experienced by daily living activities.[31,34] ª 2009 Adis Data Information BV. All rights reserved.

441

Although experimental evidence indicates that mechanical loads must be great to augment bone mass, to induce bone strains sufficient to cause microdamage and stimulate bone formation through the repair of damaged tissue,[35,36] the intensity of loading is not the only stimulus for bone accretion, as demonstrated by Rubin et al.[37,38] These authors demonstrated that high frequency vibration (20–50 Hz) of very low magnitude (1 hour a day had a reduced risk of hip fracture, but the beneficial effect was lost if the activity level was reduced.[150] In the Study of Osteoporotic Fracture, a longitudinal study including 9704 women >65 years of age and followed for about 8 years, women in the highest quintile of current activity level had a 42% lower hip fracture risk than the least active quintile of women, and self-reported walking time was associated with a 30% reduction in hip fracture risk during a 4.1-year follow-up.[43] Studies evaluating the question of whether bone mass is maintained after a reduction or cessation of exercise show contrasting results, irrespective of the level of BMD found in retired athletes. To evaluate the hypothesis that exercise during growth reduces the clinical problem of fragility fractures, it would be needed to demonstrate that retired athletes have fewer fractures than controls. Wyshak et al.[151] compared a large cohort (n = 10 796) of former female college athletes with sedentary controls aged 21–80 years. The number of former athletes with fractures after retirement was no different than among the controls. Among women aged ‡60 years, who were fracture-free up to the age of 40 years, the rate of any fracture at age ‡40 years was 29% for former athletes compared with 32% for non-athletes, a nonsignificant difference. Nordstro¨m et al.[41] measured BMD in two cohorts; the first comprised 65 young male ice hockey players, 73 young soccer players (two high-impact sports) and 61 age-matched controls. Measures were taken again after 5 years; at that time, 55 athletes had retired from their active sports career. The second cohort comprised 400 former soccer and ice hockey players and 800 age- and sex-matched controls. At baseline, Sports Med 2009; 39 (6)

Exercise and Bone Mass in Adults

all active groups had higher BMD values at whole body, femoral neck, lumbar spine and arms compared with controls; after 5 years the young retired athletes still had a 4–8% higher BMD than controls, whereas young athletes increased the difference in BMD compared with the controls at femoral neck and arms. These results suggest that higher BMD persists until several decades after retirement. Furthermore, retired athletes had fewer fractures than controls. Therefore, it seems that exercise during childhood and adolescence may be associated with lower risk of sustaining fragility fractures during old age in men,[94] but in women these beneficial results only persist if exercise practice is maintained. The effects of exercise protocols on bone density have also been reported in older populations; a review of recent longitudinal studies is provided in tables IV and V for women and men, respectively. 2.4 Postmenopausal Women

Nelson et al.[152] completed a 1-year randomized, controlled trial of high-intensity resistance training in postmenopausal women. The results of the study demonstrated that women in a 2 days/ week resistance training programme gained an average of 1% in BMD of the femoral neck and lumbar spine whereas the control group lost 2.5% and 1.8% at these sites, respectively. In addition, the resistance-trained women tended to maintain WBBMC of the skeleton whereas the women in the control group had a 1.2% decline in WBBMC. Also, the resistance-trained women had 35–76% increase in strength, 14% improvement in dynamic balance, and a 1.2 kg increase in total body lean mass and a 27% increase in physical activity unrelated to the intervention, whereas the control group showed declines in all of these parameters. In agreement with these results, Kerr et al.[153] reported that postmenopausal bone mass can be significantly increased by a strength regimen that uses high loads and a low number of repetitions (3 · 8 RM) but not by an endurance regimen that uses low loads and a high number of repetitions (3 · 20 RM). In 1–7 years, postmenopausal women, following ª 2009 Adis Data Information BV. All rights reserved.

455

9 months of strength training with intermediate loads (2 · 10–15 RM), lumbar spine BMD was enhanced by 1.6%. In this study, each subject performed one set of 10–12 RM (increasing progressively) for upper body training and one set of 10–15 RM for lower body training. In contrast, the women from the control group experienced a 3.6% decline in lumbar spine BMD.[159] Altogether, these studies show that the peak load is more important than the number of loading cycles in increasing bone mass in postmenopausal women. In late postmenopausal women (aged 60–72 years), 9 months of endurance training (mostly
running at 60–70% of V O2max, three to four times a week, for 35–50 min/session) either alone or in combination with hormone replacement therapy (HRT) resulted in significant increases in lumbar spine and femoral neck.[154] Exercise and HRT resulted in independent and additive effects on the BMD of the lumbar spine and Ward’s triangle, and a synergistic effect on whole body BMD. These effects were accompanied by a reduction in serum osteocalcin levels, indicating that increases in BMD in response to HRT and to exercise + HRT were due to decreased bone turnover.[154] The lack of change in serum osteocalcin and IGF-I in response to exercise alone suggests that the increases in BMD were due to decreased bone resorption and not to increased formation.[154] Other studies have reported just the maintenance of BMD in postmenopausal women with resistance training.[155,162] There is evidence that postmenopausal women respond differently to a resistance training programme than do premenopausal women.[163,164] Bassey et al.[163] studied the effects of a vertical jumping exercise regimen on BMD using randomized controlled trials in both pre- and postmenopausal women, the latter stratified for HRT. The exercise consisted of 50 vertical jumps on 6 days/week of mean height 8.5 cm, which produced mean ground reactions of 3.0 times bodyweight in the young women and 4.0 times in the older women. In the premenopausal women, the exercise resulted in a significant increase of 2.8% in femoral BMD after 5 months. In the postmenopausal women, there Sports Med 2009; 39 (6)

Study

Subjects n

M/F

age (y)

Training

Frequency

Exercises

Protocol time (mo)

456

ª 2009 Adis Data Information BV. All rights reserved.

Table IV. Effects of training protocols on bone tissue adaptations in young men, middle-aged men and age-specific sex comparisons: longitudinal studies Training intensity

Other

Bone measurement site (s)

Results

Young men EXM = 18 EXS = 19 C = 19

M

18–30

RT

5/wk 2–4 sets · 4–12 rep

Military press, bench press, seated chest fly, seated triceps extension, seated lateral pull down, seated wide grip row, seated reverse fly, seated biceps curl, abdominals, inclined leg press, 2-leg knee extension, 2-leg hamstring curl, seated calf raise

3

80% 1 RM

500 mL M/S in EXM; 500 mL S/S in EXS; 500 mL C/S in C

WB

NC

Ballard et al.[138] (2006)

YMEX = 13 YWEX = 12 CM = 12 CW = 11

M/F

20–22

RT + ET

5/wk 3 sets · 12failure rep

Bench press, inclined bench press, shoulder press, lat pulldown; cable rows, arm curl and extensions, hip sled, squats, calf raises + aerobics

6

70% 1 RM; 70%
VO2max

EX: 42 g P/S C: 70 g C/S

T, WB, A, L

› T vBMD in YMEX and YWEX › ABMC in YMEX and YWEX

Ryan et al.[139] (2004)

YWEX = 8 OWEX = 11 YMEX = 13 OMEX = 10

M/F

Y = 20–29 O = 65–74

RT

3/wk 2 sets, failure

Leg press, chest press, leg curl, lat pulldown; leg extension, military press, seated row, triceps pulldown, abdominal crunch, biceps curl, sit ups

6

12–15 RM

WB, SP, FN, WT, GT

› FNBMD in ESP

Fujimura et al.[136] (1997)

EX = 8 C=7

M

23–31

RT

3/wk 2–3 sets · 10 rep

Leg extension, leg curl, bench press, sit up, back extension, arm curl, wrist curl, half squat leg lunge, lateral pull down, back press

4

60–80% 1 RM

WB, FN, SP, R

NC

Continued next page

Guadalupe-Grau et al.

Sports Med 2009; 39 (6)

Hartman et al.[137] (2007)

ª 2009 Adis Data Information BV. All rights reserved.

457

1RM = 1 repetition maximum; A = arms; BMD = bone mineral density; C = control subjects; CM = male control subjects; C/S = supplemental carbohydrate; CW = female control subjects; ESP = entire study population; ET = endurance training; EX = exercising subjects; EXM = exercise + supplemental milk products; EXS = exercise + supplemental soy products; FN = femoral neck; GT = greater trochanter; L = whole leg; lat = latissimus; M/F = male/female; M//S = supplemental milk; N = number of subjects; NC = no changes; O = old; OMEX = older men exercising subjects; OWEX = older women exercising subjects; P/S = supplemental protein; R = radius; rep = repetitions; RT = resistance training; SP = lumbar
spine; S/S = supplemental soy; T = tibia; vBMD = volumetric BMD; V O2max = maximum oxygen consumption; WB = whole body; WT = Ward’s triangle; Y = young; YMEX = young men exercising subjects; YWEX = young women exercising subjects; › indicates significant increase p < 0.05; fl indicates significant decrease p < 0.05.

› 2.0% SP in EX › 3.8% FN in EX SP, FN 5–15 RM 4 Chest press, overhead press, lat pull, upper back row, leg press, leg extension 3/wk 1–2 sets · 15 rep RT 54–61 EX = 11 C=7 Menkes et al.[143] (1993)

M

EX = 70 C = 70 Huuskonen et al.[144] (2001)

Middle-aged men

Study

Table IV. Contd

Subjects n

M

M/F

50–60

age (y)

ET

Training

3–5/wk 30–60 min

Frequency

Exercises

Brisk walking

48

Protocol time (mo)

40–60%
VO2max

Training intensity

Other

SP, FN, GT, WT

Bone measurement site (s)

Results

fl SP, FN, GT, WT BMD in EX

Exercise and Bone Mass in Adults

was no significant difference between the exercise and control groups after 12 months (total n = 123) nor after 18 months (total n = 38). HRT status did not affect this outcome, at least up to 12 months. Sugiyama et al.[164] studied a group of Japanese female volunteers aged around 50 years divided into premenopausal women with a regular menstruation cycle and postmenopausal women within 5 years since menopause. About half of the subjects in each group chose to be non-exercisers. The remainder followed a 6-month training programme consisting of rope skipping (100 jumps/day, with an interval of 2–3 days). In total, they completed 10 days per month, 60 days during the study period. Among the premenopausal women, the BMD in the exercise group increased significantly compared with the controls for total hip (+1.6%) and femoral neck (2.4%), but changes at the whole body and lumbar spine levels were not significant. In contrast, there were no significant differences at any measurement sites among the postmenopausal women. Interestingly, in the premenopausal exercise group, the baseline value of urinary g-carboxyglutamate (Gla) residues (an indirect measure of osteocalcin carboxylation) was inversely correlated (r = -0.62) with the change in whole body BMD. The latter could indicate that bone gain induced by high impact exercise could become greater in proportion to the degree of deterioration in bone material properties.[164] Therefore, although optimum training strategies are still under discussion, it is generally acknowledged that the training should be population specific. Stengel et al.[155] tested the hypothesis that power training was more effective than conventional strength training for maintaining BMD at lumbar spine and hip. Forty-two postmenopausal women performed a 12-month training programme; the only difference between the two groups was the velocity at which movements were performed. The training protocol specified a 4-second concentric, 4-second eccentric sequence in the resistance training group, and a concentric fast/explosive 4-second eccentric sequence in the power training group. In addition, all women performed gymnastics and home training sessions. Women involved Sports Med 2009; 39 (6)

Study

Subjects n

Training

458

ª 2009 Adis Data Information BV. All rights reserved.

Table V. Effects of training protocols on bone tissue adaptations in older women, men and age-specific sex comparisons: longitudinal studies Frequency

Exercises

Protocol Training Other intensity time (mo)

Results Bone measurement site (s)

2/wk 3 set · 8 rep

5 weight-lifting exercises

12

WB, SP, FN

› › fl fl

M/F age (y)

Older women EX = 20 C = 19

Kerr et al.[153] (1996)

REX EEX C

Kohrt et al.[154] (1995)

REX C

F

RT

3/wk

Weight-bearing exercises

12

Stengel et al.[155] (2005)

PEX = 21 REX = 21

F

54–60 PT RT

4/wk

2 weight-lifting sessions 1 gymnastics session 1 home training session

12

70–90% 1 RM

1.500 mg Ca/S, 500 Vit-D/S

SP, WH, FN, T, IT

NC FNBMD in PEX NC SPBMD in PEX fl 0.9% SPBMD in REX fl 1.2% WHBMD in REX

Chien et al.[156] (2000)

EX = 22 C = 21

F

48–65 ET + HIT

3/wk 50 min

Treadmill walking + stepping exercise

6

70–85%
VO2max

Osteopenic subjects HRTh

WB, SP, FN

› 6.8% FNBMD in EX

Kohrt et al.[157] (1997)

GREX JREX C

F

60–74 ET RT + ET

3/wk GREX: 30–45 min JREX: 2–3 sets · 8–12 rep 15–20 min

GREX: walking, jogging, stair climbing JREX: overhead press, biceps curl; triceps extension, leg press, leg extension, leg flexion, bench press, squats

11

GREX: 60–85 MHR JREX: 8–12 RM, 60–85 MHR

WB, SP, FN, GT, W

› 2.0% WBBMD in GREX › 1.6% WBBMD in JREX › 1.8% SPBMD in GREX › 1.5% SPBMD in JREX › 6.1% GTBMD in GREX › 5.1% GTBMD in GREX

Verschueren et al.[158] (2004)

VEX: 25 EX: 22 C: 23

F

60–70 VEX: VEX, EX: VEX and EX: RT + WBV 3/wk leg extension, EX: RT 1–3 leg press sets · 10–15 rep

VEX, EX: 20–8 RM

WB, F, SP

› 0.9% FBMD in VEX NC in EX and C

F

50–70 RT

RT ET

80% 1 RM

1% SPBMD in EX 1% FNBMD in EX 1.8% SPBMD in C 2.5% FNBMD in C

12

6

HRTh

Continued next page

Guadalupe-Grau et al.

Sports Med 2009; 39 (6)

Nelson et al.[152] (1994)

Study

Pruitt et al.[159] (1992)

Subjects n EX: 17 C: 10

Training

Frequency

Exercises

M/F age (y) F

52–56 RT

3/wk Biceps curl, lat 1 set · 10–15 pulldown, bench press, rep wrist roller, leg press, leg ab/adduction, leg curl, leg extension, trunk extension, hip extension, lateral flexion

Protocol Training Other time intensity (mo)

Bone Results measurement site (s)

9

10–15 RM

SP, FN

› 1.6% SPBMD in EX fl 3.6% SPBMD in C

WH, FN, GT, SP

› 1.9% SPBMD HMEX › 1.3% GTBMD HMEX › 2.0% GTBMD HWEX

WB, SP, FN

› 2.8% FNBMD in EX

Hypertensive WB, SP, FN, subjects IT, WH

› 1.7% SPBMD in MRT

Older men Maddalozzo et al.[160] (2000)

MMEX = 12 M/F 50–60 MRT HRT HMEX = 12 MWEX = 9 HWEX = 9

MRT: 3/wk 3 sets · 10–13 rep HRT: 3/wk 3 sets · 2–10 rep

M: Leg press, leg extension hamstring curls, arm curl, triceps press, chest press, Pec deck, shoulder press, side lateral raise, lat pulldown, seated row, abdominal crunch, calf raise H: free weight back squat, deadlift, biceps curls, sit ups, triceps extension, chest press, incline shoulder press, high lat pull down, leg curl, gripper, calf raise.

6

MRT: 40–60% 1 RM HRT: 70–90% 1 RM

Ryan et al.[161] (1994)

EX = 21 C = 16

M

3/wk 2 sets · 15 rep

Leg press, chest press, leg curl, lat pull down, leg extension, military press, adductor, abductor, upper back, triceps, lower back, abdominals, biceps curl

4

5 RM

Stewart et al.[145] (2005)

MRT = 26 WRT = 31 C = 58

M/F 55–75 RT + ET

Bench press, shoulder press, 3/wk 2 sets · 10–15 seated mid-rowing, lat pulldown, leg extension, rep leg curl, leg press

6

51–71 RT

Dietary control

459

Sports Med 2009; 39 (6)

1RM = 1 repetition maximum; Ca/S = supplemental calcium; EEX = endurance training exercising subjects; ET = endurance training; EX = exercising subjects; F = femur; FN = femoral neck; GREX = ground reaction forces exercising subjects; GT = greater trochanter; H = high-intensity exercises; HIT = high-impact training; HMEX = men high-intensity exercising subjects; HRT = high-intensity resistance training; HRTh = hormone replacement therapy; HWEX = women high-intensity exercising subjects; IT = intertrochanteric subregion; JREX = joint reaction forces exercising subjects; lat = latissimus; M = moderate-intensity exercises; M/F = male/female; MHR = maximal heart rate; MMEX = men moderate-intensity exercising subjects; MRT = moderate resistance training; MWEX = women moderate-intensity exercising subjects; N = number of subjects; NC = no changes; pec = pectoralis; PEX = power exercising subjects; PT = power training; rep = repetition; REX = resistance training exercising subjects; RT = resistance training; SP = lumbar spine; T = tibia;
VEX = vibratory exercising subjects; Vit-D/S = supplemental vitamin D; V O2max = maximum oxygen consumption; W = wrist; WB = whole body; WBV = whole body vibration; WH = whole hip; › indicates significant increase p < 0.05; fl indicates significant decrease p < 0.05.

Exercise and Bone Mass in Adults

ª 2009 Adis Data Information BV. All rights reserved.

Table V. Contd

Guadalupe-Grau et al.

460

in the power training group maintained their BMD at lumbar spine and hip, whereas resistance training women experienced a significant decrease in lumbar spine and hip sites from baseline. These results indicate that to elicit a osteogenic response in older women, the strains and ground reaction forces required may be higher than those able to elicit a similar or even greater response in younger women. Palombaro[165] reviewed the effects of walkingonly programmes on BMD at various skeletal sites. This meta-analysis showed that walking interventions alone did not attenuate bone loss at the skeletal sites reported. Thus, other forms of exercise in addition to walking should be incorporated into training regimens for patients at risk for osteoporosis. Supporting this notion, Chien et al.[156] examined the efficacy of a 24-week aerobic plus high exercise programme for osteopenic postmenopausal women, and this appeared to be effective in offsetting the age-related decline of BMD, especially at the femoral neck, which showed a significant improvement of 6.8% in the exercise group. Kohrt et al.[157] applied two different training protocols to postmenopausal women at risk of osteoporosis. The first protocol consisted of exercises that stimulated the skeleton through ground-reaction forces (walking, jogging, stair climbing), while the second protocol included activities eliciting joint-reaction forces (weightlifting, rowing). The intensity was initially set at a low level (60% maximal heart rate, 12 RM) and progressed with training (to 85% maximal heart rate, 8 RM). After 11 months, BMD was increased at the whole body level and femoral neck in both groups, but the effects were greater in the ground reaction than the joint-reaction group.[157] These results could be explained by the combination of walking, jogging and stair climbing, which may generate ground reaction forces between 2.8–6 times bodyweight[166] in the ground reaction group, and the use of free weights in the resistance training combined with rowing in the joint reaction group. However, more randomized, controlled studies testing aerobic plus high-impact training in older adults are needed. ª 2009 Adis Data Information BV. All rights reserved.

Whole body vibration training in postmenopausal women has been shown to increase femoral neck BMD and balance more than walking.[167] Compared with resistance exercises progressing from low (20 RM) to high (8 RM) loading conditions, 6 months of static and dynamic knee extensor exercises on a vibration platform (35–40 Hz; 2.28–5.09 g) enhanced hip BMD by 0.9%.[158] In another study, whole body vibration inhibited bone loss in the spine and femur of postmenopausal women.[168] These authors performed a 1-year prospective, randomized, double-blind, placebo-controlled trial of 70 postmenopausal women who undertook brief periods (70% of 1 RM) with 3–4 sessions per week and 2–3 sets per session.[169] Although significant effects can be observed after 4–6 months in some locations, the efficacy of the training programme is greater when extended for ‡1 year. Combining strength training with aerobic exercise may also result in positive effects on BMD. Whole body vibration alone or in combination with exercise may help to increase or at least prevent BMD decline with aging in postmenopausal women. However, the gains in bone density and neuromuscular functions achieved by training are lost 5 years after cessation of training.[170] Continuous highintensity weight-loading physical activity is probably necessary to preserve bone density and neuromuscular function in older women. 2.5 Older Men

Older men have been much less studied than older women, possibly because of the lower osteoporotic fracture incidence in men.[171] One of Sports Med 2009; 39 (6)

Exercise and Bone Mass in Adults

these studies compared the effects of either a moderate (three sets of 10–13 repetitions at 40–60% of 1 RM) or high (three sets of 2–10 repetitions at 70–90% of 1 RM) intensity resistance training programme (with exercises involving all major muscle groups) in men and women aged 50–60 years.[160] Both older men and older women achieved significant increases in muscular strength and muscle mass regardless of intensity or training protocol.[160] The high-intensity training men experienced a significant increase in lumbar spine and greater trochanter BMD; however, women training with high-intensity increased greater trochanter BMD only slightly, maybe because these women were primarily early postmenopausal (within 36 months), a time during which there is accelerated bone loss of 2–6.5% per year. High-intensity free weight training was tolerated well by older adults and produced BMD changes in only 6 months. In older men, high-intensity training was more osteogenic at the lumbar spine than moderateintensity training. In agreement with Maddalozzo and Snow[160] Stewart et al.[145] reported no effect on BMD in men (55–75 years old) following 6 months of multistation machine at 50% of 1 RM followed by 45 minutes of aerobic training at 60–90% of their maximal heart rate. Nevertheless, this training programme resulted in other positive effects such as gains in lean mass, reduced fat mass and improved aerobic capacity.[160] Bone mass improvement has been observed in older men (mean age 61 years) after a relatively short training period.[161] In this study, femoral neck BMD was enhanced by 2.8% following 3 months of high-intensity training (5 RM; three times per week).[161] These results could be conflicting, because Frost[172] has argued that short-term increases in BMD measured by photon absorptiometry may reflect transient increases. In general, 1–3% BMD improvement in loaded bones can be achieved in old men with 6 months of strength training using heavy loads (above 70% of 1 RM, three times per week), while loads below 60% of 1 RM are unlikely to have a positive influence on bone mass.[173] ª 2009 Adis Data Information BV. All rights reserved.

461

3. Practical Recommendations Any prescription of exercise aiming to improve bone mass must take into consideration the following factors: (i) Age and sex of the subjects. At prepubertal and young adult ages, sex differences are not so important, but at middle and older age, evidence from the literature suggests that women have to train at higher intensities than men to improve their bone mass, always keeping a security range to avoid injuries. (ii) Choice and order of the exercises. Since bone adaptation is limited to loaded regions, exercise must be chosen to specifically act on the clinically relevant sites, i.e. lumbar and thoracic spine, whole hip, and especially greater trochanter, intertrochanteric and femoral neck regions. The easiest and safest way to load these regions is by using weight-lifting exercises like: leg press, leg extension, leg curl, squats, loaded back extensions, and some shoulder and arm exercises. If not contraindicated, the training programme should include impact exercises like jumping, jogging, stair climbing and sprinting. Impact exercises must be increased progressively to the maximal effort possible according to the subject’s specific capabilities. The kind of impact exercise included in the programme must be appropriate for the age of the participants, trying to keep the risk of fall as low as possible in the elderly. It must be taken into consideration that the osteogenic potential of jumping exercise is reduced in postmenopausal women, but postmenopausal women may respond well to strength training. (iii) Intensity. To enhance bone mass the threshold intensity must be reached. This level has not been unequivocally established and may vary from subject to subject, probably being lower for subjects with already reduced bone mass. Most strength training programmes showing positive effects on bone mass have used intensities of 70–90% 1 RM, always following an appropriate progression from lower to higher intensities. (iv) Frequency. Most studies with positive results have used 2–3 training days per week. However, good responses to jumping exercise sessions with Sports Med 2009; 39 (6)

Guadalupe-Grau et al.

462

frequencies up to 6 days per week have also been reported. Weight-bearing endurance exercise (30–60 minutes) can be carried out three to five times, or even on a daily basis, depending on training experience and tolerance. (v) Volume. In weight-lifting programmes, the major muscle groups of the upper and lower extremities must be trained in a balanced way, without creating imbalance between agonist and antagonist. The number of repetitions per exercise must be close to maximal that can be performed with a given load and 2–3 sets should be completed, with 1–3 minutes’ resting periods in between. With regard to high impact training, there is no consensus in the literature on how many jumps must be performed, but depending on the subject’s tolerance, 50–100 jumps should be carried out each training day. Strength training combined with high impact exercise could have additive effects in some subjects. (vi) Movement velocity. Although a progression from medium to high speed of movement is advocated at the start of the training programme, as soon as subjects are able to carry out the exercise safely, i.e. with proper biomechanical execution, movements must be performed focusing on achieving the maximal execution speed possible. Explosive muscle contractions are expected to elicit a greater osteogenic stimulus.[155] 4. Conclusions The research completed to date indicates that participation in high impact sports, especially prior to puberty, is important for maximizing bone mass accumulation and achieving a greater peak bone mass independent of sex. The effects of loading appear to be limited to the loaded bones. Starting the exercise before puberty has an additional benefit, since exercise elicits geometrical changes in bone, which in turn enhance mechanical competence. Continuing sport practice is associated with fewer bone fragility fractures in old age in both men and women. Several training methods have been used to improve BMD and content in prospective studies. Not all exercise modalities have positive effects on bone mass. For example, unloaded ª 2009 Adis Data Information BV. All rights reserved.

exercise, like swimming and cycling, has no impact on bone mass, while walking or running has limited positive effects. It is not clear which is the best training method for enhancing bone mass, although scientific evidence points to a combination of high impact exercises (i.e. jumping) with weight-lifting exercises. High impact exercise, even a limited amount, appears to be the most efficient to enhance bone mass except in postmenopausal women. Several types of resistance exercise have been tested with positive results when the intensity of the exercise was high and the speed of movement elevated. Resistance training is positively associated with high BMD in both young people and adults, and the effect of resistive exercise is relatively site specific to the working muscles and the bones to which they attach. However, more studies are needed to establish whether there are sex differences in the bone response to training. Although aerobic exercise and weight-bearing physical activity are important in maintaining overall health and healthy bones, resistance exercise has been shown to have a more potent effect on bone density. Studies performed in older adults show a sex discrepancy. Older men respond better to osteogenic training protocols than their female counterparts, although randomized longitudinal studies on the effects of exercise on bone mass in the elderly are still lacking. Old women show only mild increases or just a maintenance or attenuation of BMD losses. It remains to be determined if old women need a different exercise protocol to men of similar age. Impact and resistance exercise should be advocated for the prevention of osteoporosis. For those with osteoporosis, weight-bearing exercise in general, and resistance exercise in particular, as tolerated, along with exercise targeted to improve balance, mobility and posture, should be recommended to reduce the likelihood of falling and its associated morbidity and mortality. There is certainly a need for additional randomized, controlled trials in this research area, which will allow development of criteria for appropriate training loads according to age, sex, actual bone mass and past training history. Sports Med 2009; 39 (6)

Exercise and Bone Mass in Adults

463

Acknowledgements The study was supported financially by Ministerio de Educacio´n y Ciencia (DEP2006-56076-C06-04/ACTI). Consejerı´ a de Educacio´n, Cultura y Deportes del Gobierno de Canarias (2006/179 0001 and FEDER). Borja Guerra is a fellow of the ‘Recursos Humanos y Difusio´n de la Investigacio´n’ Program (ISCIII, MSC, Spain). The authors thank Jose´ Navarro de Tuero for his excellent technical assistance. The specialized advice from Tony Webster in editing the English version of the manuscript is also acknowledged. The authors have no conflicts of interest that are directly relevant to the content of this review.

References 1. Bouxsein ML, Karasik D. Bone geometry and skeletal fragility. Curr Osteoporos Rep 2006; 4 (2): 49-56 2. Viguet-Carrin S, Garnero P, Delmas PD. The role of collagen in bone strength. Osteoporos Int 2006; 17 (3): 319-36 3. Seeman E, Delmas PD. Bone quality: the material and structural basis of bone strength and fragility. N Engl J Med 2006; 354 (21): 2250-61 4. Ottani V, Raspanti M, Ruggeri A. Collagen structure and functional implications. Micron 2001; 32 (3): 251-60 5. Currey JD. Changes in the impact energy absorption of bone with age. J Biomech 1979; 12 (6): 459-69 6. Currey JD. The effect of porosity and mineral content on the Young’s modulus of elasticity of compact bone. J Biomech 1988; 21 (2): 131-9 7. Currey JD, Brear K, Zioupos P. The effects of ageing and changes in mineral content in degrading the toughness of human femora. J Biomech 1996; 29 (2): 257-60 8. Bailey AJ, Wotton SF, Sims TJ, et al. Post-translational modifications in the collagen of human osteoporotic femoral head. Biochem Biophys Res Commun 1992; 185 (3): 801-5 9. Boskey AL, Wright TM, Blank RD. Collagen and bone strength. J Bone Miner Res 1999; 14 (3): 330-5 10. Wang X, Bank RA, TeKoppele JM, et al. The role of collagen in determining bone mechanical properties. J Orthop Res 2001; 19 (6): 1021-6 11. Currey JD. The mechanical consequences of variation in the mineral content of bone. J Biomech 1969; 2 (1): 1-11 12. Niyibizi C, Eyre DR. Structural characteristics of crosslinking sites in type V collagen of bone: chain specificities and heterotypic links to type I collagen. Eur J Biochem 1994; 224 (3): 943-50 13. Eyre DR, Dickson IR, Van Ness K. Collagen cross-linking in human bone and articular cartilage: age-related changes in the content of mature hydroxypyridinium residues. Biochem J 1988; 252 (2): 495-500 14. Fantner GE, Birkedal H, Kindt JH, et al. Influence of the degradation of the organic matrix on the microscopic fracture behavior of trabecular bone. Bone 2004; 35 (5): 1013-22 15. Banse X, Sims TJ, Bailey AJ. Mechanical properties of adult vertebral cancellous bone: correlation with collagen

ª 2009 Adis Data Information BV. All rights reserved.

16.

17. 18.

19.

20. 21.

22. 23. 24.

25. 26.

27.

28.

29. 30.

31. 32.

33. 34.

intermolecular cross-links. J Bone Miner Res 2002; 17 (9): 1621-8 Fantner GE, Hassenkam T, Kindt JH, et al. Sacrificial bonds and hidden length dissipate energy as mineralized fibrils separate during bone fracture. Nat Mater 2005; 4 (8): 612-6 Hert J, Fiala P, Petrtyl M. Osteon orientation of the diaphysis of the long bones in man. Bone 1994; 15 (3): 269-77 Martin RB, Boardman DL. The effects of collagen fiber orientation, porosity, density, and mineralization on bovine cortical bone bending properties. J Biomech 1993; 26 (9): 1047-54 Puustjarvi K, Nieminen J, Rasanen T, et al. Do more highly organized collagen fibrils increase bone mechanical strength in loss of mineral density after one-year running training? J Bone Miner Res 1999; 14 (3): 321-9 Harada S, Rodan GA. Control of osteoblast function and regulation of bone mass. Nature 2003; 423 (6937): 349-55 Pavy-Le Traon A, Heer M, Narici MV, et al. From space to earth: advances in human physiology from 20 years of bed rest studies (1986-2006). Eur J Appl Physiol 2007; 101 (2): 143-94 Chilibeck PD, Sale DG, Webber CE. Exercise and bone mineral density. Sports Med 1995; 19 (2): 103-22 Frost HM. Bone’s mechanostat: a 2003 update. Anat Rec A Discov Mol Cell Evol Biol 2003; 275 (2): 1081-101 Rittweger J. What is new in neuro-musculoskeletal interactions: mechanotransduction, microdamage and repair? J Musculoskel Neuron Interact 2007; 7 (2): 191-3 Rosen CJ. Bone remodeling, energy metabolism, and the molecular clock. Cell Metab 2008; 7 (1): 7-10 Lee KC, Lanyon LE. Mechanical loading influences bone mass through estrogen receptor alpha. Exerc Sport Sci Rev 2004; 32 (2): 64-8 Skerry TM. One mechanostat or many? Modifications of the site-specific response of bone to mechanical loading by nature and nurture. J Musculoskel Neuron Interact 2006; 6 (2): 122-7 Scott A, Khan KM, Duronio V, et al. Mechanotransduction in human bone: in vitro cellular physiology that underpins bone changes with exercise. Sports Med 2008; 38 (2): 139-60 Rubin J, Rubin C, Jacobs CR. Molecular pathways mediating mechanical signaling in bone. Gene 2006; 367: 1-16 Saxon LK, Lanyon LE. Assessment of the in vivo adaptive response to mechanical loading. Method Molec Biol 2008; 455: 307-22 Frost HM. Bone ‘‘mass’’ and the ‘‘mechanostat’’: a proposal. Anat Rec 1987; 219 (1): 1-9 Frost HM. Wolff’s law: an ‘MGS’ derivation of gamma in the Three-Way Rule for mechanically controlled lamellar bone modeling drifts. Bone Miner 1993; 22 (2): 117-27 Bailey CA, Brooke-Wavell K. Exercise for optimising peak bone mass in women. Proc Nutr Soc 2008; 67 (1): 9-18 Shaw J. Exercise for skeletal health and osteoporosis prevention: ACSM’s resource manual for guidelines for exercise testing and prescription. Philadelphia (PA): Lippincott Williams & Wilkins, 1998: 288-93

Sports Med 2009; 39 (6)

464

35. Frost HM. Skeletal structural adaptations to mechanical usage (SATMU): 1, redefining Wolff’s law – the bone modeling problem. Anat Rec 1990; 226 (4): 403-13 36. Burr DB, Forwood MR, Fyhrie DP, et al. Bone microdamage and skeletal fragility in osteoporotic and stress fractures. J Bone Miner Res 1997; 12 (1): 6-15 37. Rubin CT, Lanyon LE. Regulation of bone formation by applied dynamic loads. J Bone Joint Surg Am 1984; 66 (3): 397-402 38. Rubin C, Turner AS, Bain S, et al. Anabolism: low mechanical signals strengthen long bones. Nature 2001; 412 (6847): 603-4 39. Rubin C, Turner AS, Muller R, et al. Quantity and quality of trabecular bone in the femur are enhanced by a strongly anabolic, noninvasive mechanical intervention. J Bone Miner Res 2002; 17 (2): 349-57 40. Borer KT. Physical activity in the prevention and amelioration of osteoporosis in women: interaction of mechanical, hormonal and dietary factors. Sports Med 2005; 35 (9): 779-830 41. Nordstro¨m A, Karlsson C, Nyquist F, et al. Bone loss and fracture risk after reduced physical activity. J Bone Miner Res 2005; 20 (2): 202-7 42. WHO Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. World Health Organ Tech Rep Ser 1994; 843: 1-129 43. Cummings SR, Nevitt MC, Browner WS, et al. Risk factors for hip fracture in white women: Study of Osteoporotic Fractures Research Group. N Engl J Med 1995; 332 (12): 767-73 44. Kanis JA, Melton 3rd LJ, Christiansen C, et al. The diagnosis of osteoporosis. J Bone Miner Res 1994; 9 (8): 1137-41 45. Mein AL, Briffa NK, Dhaliwal SS, et al. Lifestyle influences on 9-year changes in BMD in young women. J Bone Miner Res 2004; 19 (7): 1092-8 46. Vicente-Rodriguez G. How does exercise affect bone development during growth? Sports Med 2006; 36 (7): 561-9 47. Vicente-Rodriguez G, Jimenez-Ramirez J, Ara I, et al. Enhanced bone mass and physical fitness in prepubescent footballers. Bone 2003; 33 (5): 853-9 48. Vicente-Rodriguez G, Dorado C, Perez-Gomez J, et al. Enhanced bone mass and physical fitness in young female handball players. Bone 2004; 35 (5): 1208-15 49. Vicente-Rodriguez G, Dorado C, Ara I, et al. Artistic versus rhythmic gymnastics: effects on bone and muscle mass in young girls. Int J Sports Med 2007; 28 (5): 386-93 50. McVeigh JA, Norris SA, Pettifor JM. Bone mass accretion rates in pre- and early-pubertal South African black and white children in relation to habitual physical activity and dietary calcium intakes. Acta Paediatr 2007; 96 (6): 874-80 51. Vicente-Rodriguez G, Ara I, Perez-Gomez J, et al. High femoral bone mineral density accretion in prepubertal soccer players. Med Sci Sports Exerc 2004; 36 (10): 1789-95 52. Wang Q, Alen M, Nicholson P, et al. Weight-bearing, muscle loading and bone mineral accrual in pubertal girls: a 2-year longitudinal study. Bone 2007; 40 (5): 1196-202

ª 2009 Adis Data Information BV. All rights reserved.

Guadalupe-Grau et al.

53. Bradney M, Pearce G, Naughton G, et al. Moderate exercise during growth in prepubertal boys: changes in bone mass, size, volumetric density, and bone strength: a controlled prospective study. J Bone Miner Res 1998; 13 (12): 1814-21 54. Bass SL. The prepubertal years: a uniquely opportune stage of growth when the skeleton is most responsive to exercise? Sports Med 2000; 30 (2): 73-8 55. Bailey DA, McKay HA, Mirwald RL, et al. A six-year longitudinal study of the relationship of physical activity to bone mineral accrual in growing children: the university of Saskatchewan bone mineral accrual study. J Bone Miner Res 1999; 14 (10): 1672-9 56. Greene DA, Naughton GA. Adaptive skeletal responses to mechanical loading during adolescence. Sports Med 2006; 36 (9): 723-32 57. Bass SL, Saxon L, Daly RM, et al. The effect of mechanical loading on the size and shape of bone in pre-, peri-, and postpubertal girls: a study in tennis players. J Bone Miner Res 2002; 17 (12): 2274-80 58. Seeman E. An exercise in geometry. J Bone Miner Res 2002; 17 (3): 373-80 59. Kannus P, Haapasalo H, Sankelo M, et al. Effect of starting age of physical activity on bone mass in the dominant arm of tennis and squash players. Ann Intern Med 1995; 123 (1): 27-31 60. Heinonen A, Sievanen H, Kannus P, et al. High-impact exercise and bones of growing girls: a 9-month controlled trial. Osteoporos Int 2000; 11 (12): 1010-7 61. Manias K, McCabe D, Bishop N. Fractures and recurrent fractures in children; varying effects of environmental factors as well as bone size and mass. Bone 2006; 39 (3): 652-7 62. Laing EM, Wilson AR, Modlesky CM, et al. Initial years of recreational artistic gymnastics training improves lumbar spine bone mineral accrual in 4- to 8-year-old females. J Bone Miner Res 2005; 20 (3): 509-19 63. Lanyon LE. Using functional loading to influence bone mass and architecture: objectives, mechanisms, and relationship with estrogen of the mechanically adaptive process in bone. Bone 1996; 18 (1 Suppl.): 37S-43S 64. Martyn-St James M, Carroll S. Progressive high-intensity resistance training and bone mineral density changes among premenopausal women: evidence of discordant site-specific skeletal effects. Sports Med 2006; 36 (8): 683-704 65. Forwood MR, Turner CH. The response of rat tibiae to incremental bouts of mechanical loading: a quantum concept for bone formation. Bone 1994; 15 (6): 603-9 66. Pazzaglia UE, Andrini L, Di Nucci A. The effects of mechanical forces on bones and joints: experimental study on the rat tail. J Bone Joint Surg Br 1997; 79 (6): 1024-30 67. Robling AG, Duijvelaar KM, Geevers JV, et al. Modulation of appositional and longitudinal bone growth in the rat ulna by applied static and dynamic force. Bone 2001; 29 (2): 105-13 68. Ehrlich PJ, Lanyon LE. Mechanical strain and bone cell function: a review. Osteoporos Int 2002; 13 (9): 688-700

Sports Med 2009; 39 (6)

Exercise and Bone Mass in Adults

69. Pedersen BK, Hoffman-Goetz L. Exercise and the immune system: regulation, integration, and adaptation. Physiol Rev 2000; 80 (3): 1055-81 70. Van Hall G, Jensen-Urstad M, Rosdahl H, et al. Leg and arm lactate and substrate kinetics during exercise. Am J Physiol Endocrinol Metab 2003; 284 (1): E193-205 71. Mastorakos G, Pavlatou M. Exercise as a stress model and the interplay between the hypothalamus-pituitary-adrenal and the hypothalamus-pituitary-thyroid axes. Horm Metab Res 2005; 37 (9): 577-84 72. Calbet JA, Lundby C, Sander M, et al. Effects of ATPinduced leg vasodilation on VO2 peak and leg O2 extraction during maximal exercise in humans. Am J Physiol Regul Integr Comp Physiol 2006; 291 (2): R447-53 73. Ara I, Perez-Gomez J, Vicente-Rodriguez G, et al. Serum free testosterone, leptin and soluble leptin receptor changes in a 6-week strength-training programme. Br J Nutr 2006; 96 (6): 1053-9 74. Frost HM, Jee WS. On the rat model of human osteopenias and osteoporoses. Bone Miner 1992; 18 (3): 227-36 75. Bourrin S, Palle S, Pupier R, et al. Effect of physical training on bone adaptation in three zones of the rat tibia. J Bone Miner Res 1995; 10 (11): 1745-52 76. Yeh JK, Liu CC, Aloia JF. Effects of exercise and immobilization on bone formation and resorption in young rats. Am J Physiol 1993; 264 (2 Pt 1): E182-9 77. Hagihara Y, Fukuda S, Goto S, et al. How many days per week should rats undergo running exercise to increase BMD? J Bone Miner Metab 2005; 23 (4): 289-94 78. Barengolts EI, Curry DJ, Bapna MS, et al. Effects of endurance exercise on bone mass and mechanical properties in intact and ovariectomized rats. J Bone Miner Res 1993; 8 (8): 937-42 79. Horcajada M, Coxam V, Davicco M, et al. Influence of treadmill running on femoral bone in young orchidectomized rats. J Appl Physiol 1997; 83 (1): 129-33 80. van der Wiel HE, Lips P, Graafmans WC, et al. Additional weight-bearing during exercise is more important than duration of exercise for anabolic stimulus of bone: a study of running exercise in female rats. Bone 1995; 16 (1): 73-80 81. Bourrin S, Genty C, Palle S, et al. Adverse effects of strenuous exercise: a densitometric and histomorphometric study in the rat. J Appl Physiol 1994; 76 (5): 1999-2005 82. Iwamoto J, Takeda T, Sato Y. Effect of treadmill exercise on bone mass in female rats. Exp Anim 2005; 54 (1): 1-6 83. Wronski TJ, Yen CF. The ovariectomized rat as an animalmodel for postmenopausal bone loss. Cell Mater Suppl 1991; 1: 69-74 84. Jarvinen TL, Kannus P, Pajamaki I, et al. Estrogen deposits extra mineral into bones of female rats in puberty, but simultaneously seems to suppress the responsiveness of female skeleton to mechanical loading. Bone 2003; 32 (6): 642-51 85. Sanchis-Moysi J, Dorado C, Vicente-Rodriguez G, et al. Inter-arm asymmetry in bone mineral content and bone area in postmenopausal recreational tennis players. Maturitas 2004; 48 (3): 289-98

ª 2009 Adis Data Information BV. All rights reserved.

465

86. Calbet JA, Dorado C, Diaz-Herrera P, et al. High femoral bone mineral content and density in male football (soccer) players. Med Sci Sports Exerc 2001; 33 (10): 1682-7 87. Morel J, Combe B, Francisco J, et al. Bone mineral density of 704 amateur sportsmen involved in different physical activities. Osteoporos Int 2001; 12 (2): 152-7 88. Calbet JA, Moysi JS, Dorado C, et al. Bone mineral content and density in professional tennis players. Calcif Tissue Int 1998; 62 (6): 491-6 89. Wittich A, Mautalen CA, Oliveri MB, et al. Professional football (soccer) players have a markedly greater skeletal mineral content, density and size than age- and BMImatched controls. Calcif Tissue Int 1998; 63 (2): 112-7 90. Egan E, Reilly T, Giacomoni M, et al. Bone mineral density among female sports participants. Bone 2006; 38 (2): 227-33 91. Nichols JF, Rauh MJ, Barrack MT, et al. Bone mineral density in female high school athletes: interactions of menstrual function and type of mechanical loading. Bone 2007; 41 (3): 371-7 92. Alfredson H, Nordstro¨m P, Lorentzon R. Total and regional bone mass in female soccer players. Calcif Tissue Int 1996; 59 (6): 438-42 93. Ducher G, Prouteau S, Courteix D, et al. Cortical and trabecular bone at the forearm show different adaptation patterns in response to tennis playing. J Clin Densitom 2004; 7 (4): 399-405 94. Karlsson MK. Skeletal effects of exercise in men. Calcif Tissue Int 2001; 69 (4): 196-9 95. Layne JE, Nelson ME. The effects of progressive resistance training on bone density: a review. Med Sci Sports Exerc 1999; 31 (1): 25-30 96. Nichols DL, Sanborn CF, Essery EV. Bone density and young athletic women: an update. Sports Med 2007; 37 (11): 1001-14 97. Karlsson M. Has exercise an antifracture efficacy in women? Scand J Med Sci Sports 2004; 14 (1): 2-15 98. Daly RM, Rich PA, Klein R, et al. Effects of high-impact exercise on ultrasonic and biochemical indices of skeletal status: a prospective study in young male gymnasts. J Bone Miner Res 1999; 14 (7): 1222-30 99. Taaffe DR, Snow-Harter C, Connolly DA, et al. Differential effects of swimming versus weight-bearing activity on bone mineral status of eumenorrheic athletes. J Bone Miner Res 1995; 10 (4): 586-93 100. Hawkey A. The importance of exercising in space. Interdiscip Sci Rev 2003; 28 (2): 130-8 101. Perez-Gomez J, Olmedillas H, Delgado-Guerra S, et al. Effects of weight lifting training combined with plyometric exercises on physical fitness, body composition, and knee extension velocity during kicking in football. Appl Physiol Nutr Metab 2008; 33 (3): 501-10 102. Kellis E, Katis A, Vrabas IS. Effects of an intermittent exercise fatigue protocol on biomechanics of soccer kick performance. Scand J Med Sci Sports 2006; 16 (5): 334-44 103. Hetland ML, Haarbo J, Christiansen C. Low bone mass and high bone turnover in male long distance runners. J Clin Endocrinol Metab 1993; 77 (3): 770-5

Sports Med 2009; 39 (6)

466

104. Kannus P, Haapasalo H, Sievanen H, et al. The site-specific effects of long-term unilateral activity on bone mineral density and content. Bone 1994; 15 (3): 279-84 105. Haapasalo H, Kannus P, Sievanen H, et al. Long-term unilateral loading and bone mineral density and content in female squash players. Calcif Tissue Int 1994; 54 (4): 249-55 106. Ducher G, Tournaire N, Meddahi-Pelle A, et al. Shortterm and long-term site-specific effects of tennis playing on trabecular and cortical bone at the distal radius. J Bone Miner Metab 2006; 24 (6): 484-90 107. Vicente-Rodriguez G. How does exercise affect bone development during growth? Sports Med 2006; 36 (7): 561-9 108. Calbet JA, Dorado C, Diaz-Herrera P, et al. High femoral bone mineral content and density in male football (soccer) players. Med Sci Sports Exerc 2001; 33 (10): 1682-7 109. Egan E, Reilly T, Giacomoni M, et al. Bone mineral density among female sports participants. Bone 2006; 38 (2): 227-33 110. Ng MY, Sham PC, Paterson AD, et al. Effect of environmental factors and gender on the heritability of bone mineral density and bone size. Ann Hum Genet 2006; 70 (Pt 4): 428-38 111. Kemper HC, Post GB, Twisk JW, et al. Lifestyle and obesity in adolescence and young adulthood: results from the Amsterdam Growth And Health Longitudinal Study (AGAHLS). Int J Obes Relat Metab Disord 1999; 23 Suppl. 3: S34-40 112. Armstrong N, Welsman JR. The physical activity patterns of European youth with reference to methods of assessment. Sports Med 2006; 36 (12): 1067-86 113. Tanner JM. Principles of growth standards. Acta Paediatr Scand 1990; 79 (10): 963-7 114. Friedlander AL, Genant HK, Sadowsky S, et al. A twoyear program of aerobics and weight training enhances bone mineral density of young women. J Bone Miner Res 1995; 10 (4): 574-85 115. Schroeder ET, Hawkins SA, Jaque SV. Musculoskeletal adaptations to 16 weeks of eccentric progressive resistance training in young women. J Strength Cond Res 2004; 18 (2): 227-35 116. Hawkins SA, Schroeder ET, Wiswell RA, et al. Eccentric muscle action increases site-specific osteogenic response. Med Sci Sports Exerc 1999; 31 (9): 1287-92 117. Nickols-Richardson SM, Miller LE, Wootten DF, et al. Concentric and eccentric isokinetic resistance training similarly increases muscular strength, fat-free soft tissue mass, and specific bone mineral measurements in young women. Osteoporos Int 2007; 18 (6): 789-96 118. Kato T, Terashima T, Yamashita T, et al. Effect of lowrepetition jump training on bone mineral density in young women. J Appl Physiol 2006; 100 (3): 839-43 119. Bassey EJ, Ramsdale SJ. Increase in femoral bone density in young women following high-impact exercise. Osteoporos Int 1994; 4 (2): 72-5 120. Sinaki M, Wahner HW, Bergstralh EJ, et al. Three-year controlled, randomized trial of the effect of dose-specified loading and strengthening exercises on bone mineral density of spine and femur in nonathletic, physically active women. Bone 1996; 19 (3): 233-44

ª 2009 Adis Data Information BV. All rights reserved.

Guadalupe-Grau et al.

121. Chilibeck PD, Calder A, Sale DG, et al. Twenty weeks of weight training increases lean tissue mass but not bone mineral mass or density in healthy, active young women. Can J Physiol Pharmacol 1996; 74 (10): 1180-5 122. Nindl BC, Harman EA, Marx JO, et al. Regional body composition changes in women after 6 months of periodized physical training. J Appl Physiol 2000; 88 (6): 2251-9 123. Lohman T, Going S, Pamenter R, et al. Effects of resistance training on regional and total bone mineral density in premenopausal women: a randomized prospective study. J Bone Miner Res 1995; 10 (7): 1015-24 124. Vainionpa¨a¨ A, Korpelainen R, Leppaluoto J, et al. Effects of high-impact exercise on bone mineral density: a randomized controlled trial in premenopausal women. Osteoporos Int 2005; 16 (2): 191-7 125. Winters-Stone KM, Snow CM. Site-specific response of bone to exercise in premenopausal women. Bone 2006; 39 (6): 1203-9 126. Heinonen A, Kannus P, Sievanen H, et al. Randomised controlled trial of effect of high-impact exercise on selected risk factors for osteoporotic fractures. Lancet 1996; 348 (9038): 1343-7 127. Kohrt WM, Bloomfield SA, Little KD, et al. American College of Sports Medicine Position Stand: physical activity and bone health. Med Sci Sports Exerc 2004; 36 (11): 1985-96 128. Deschenes MR, Kraemer WJ. Performance and physiologic adaptations to resistance training. Am J Phys Med Rehabil 2002; 81 (11 Suppl.): S3-16 129. Kraemer WJ, Ratamess NA. Fundamentals of resistance training: progression and exercise prescription. Med Sci Sports Exerc 2004; 36 (4): 674-88 130. Kraemer WJ, Mazzetti SA, Nindl BC, et al. Effect of resistance training on women’s strength/power and occupational performances. Med Sci Sports Exerc 2001; 33 (6): 1011-25 131. Kraemer WJ, Nindl BC, Ratamess NA, et al. Changes in muscle hypertrophy in women with periodized resistance training. Med Sci Sports Exerc 2004; 36 (4): 697-708 132. Uusi-Rasi K, Sievanen H, Pasanen M, et al. Association of physical activity and calcium intake with the maintenance of bone mass in premenopausal women. Osteoporos Int 2002; 13 (3): 211-7 133. Snow-Harter C, Bouxsein ML, Lewis BT, et al. Effects of resistance and endurance exercise on bone mineral status of young women: a randomized exercise intervention trial. J Bone Miner Res 1992; 7 (7): 761-9 134. Mayhew TP, Rothstein JM, Finucane SD, et al. Muscular adaptation to concentric and eccentric exercise at equal power levels. Med Sci Sports Exerc 1995; 27 (6): 868-73 135. Gilsanz V, Wren TA, Sanchez M, et al. Low-level, highfrequency mechanical signals enhance musculoskeletal development of young women with low BMD. J Bone Miner Res 2006; 21 (9): 1464-74 136. Fujimura R, Ashizawa N, Watanabe M, et al. Effect of resistance exercise training on bone formation and resorption in young male subjects assessed by biomarkers of bone metabolism. J Bone Miner Res 1997; 12 (4): 656-62 137. Hartman JW, Tang JE, Wilkinson SB, et al. Consumption of fat-free fluid milk after resistance exercise promotes

Sports Med 2009; 39 (6)

Exercise and Bone Mass in Adults

138.

139.

140.

141.

142.

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

greater lean mass accretion than does consumption of soy or carbohydrate in young, novice, male weightlifters. Am J Clin Nutr 2007; 86 (2): 373-81 Ballard TL, Specker BL, Binkley TL, et al. Effect of protein supplementation during a 6-month strength and conditioning program on areal and volumetric bone parameters. Bone 2006; 38 (6): 898-904 Ryan AS, Ivey FM, Hurlbut DE, et al. Regional bone mineral density after resistive training in young and older men and women. Scand J Med Sci Sports 2004; 14 (1): 16-23 Mullins NM, Sinning WE. Effects of resistance training and protein supplementation on bone turnover in young adult women. Nutr Metab (Lond) 2005; 2: 19 Torvinen S, Kannus P, Sievanen H, et al. Effect of 8-month vertical whole body vibration on bone, muscle performance, and body balance: a randomized controlled study. J Bone Miner Res 2003; 18 (5): 876-84 Beck BR, Kent K, Holloway L, et al. Novel, highfrequency, low-strain mechanical loading for premenopausal women with low bone mass: early findings. J Bone Mineral Metab 2006; 24 (6): 505-7 Menkes A, Mazel S, Redmond RA, et al. Strength training increases regional bone mineral density and bone remodeling in middle-aged and older men. J Appl Physiol 1993; 74 (5): 2478-84 Huuskonen J, Vaisanen SB, Kroger H, et al. Regular physical exercise and bone mineral density: a four-year controlled randomized trial in middle-aged men – the DNASCO study. Osteoporos Int 2001; 12 (5): 349-55 Stewart KJ, Bacher AC, Hees PS, et al. Exercise effects on bone mineral density relationships to changes in fitness and fatness. Am J Prev Med 2005; 28 (5): 453-60 Madeo B, Zirilli L, Caffagni G, et al. The osteoporotic male: overlooked and undermanaged? Clin Interv Aging 2007; 2 (3): 305-12 Hui SL, Slemenda CW, Johnston Jr CC, et al. The contribution of bone loss to postmenopausal osteoporosis. Osteoporos Int 1990; 1 (1): 30-4 Jones HH, Priest JD, Hayes WC, et al. Humeral hypertrophy in response to exercise. J Bone Joint Surg Am 1977; 59 (2): 204-8 Province MA, Hadley EC, Hornbrook MC, et al. The effects of exercise on falls in elderly patients: a preplanned meta-analysis of the FICSIT Trials. Frailty and injuries: cooperative studies of intervention techniques. JAMA 1995; 273 (17): 1341-7 Paganini-Hill A, Chao A, Ross RK, et al. Exercise and other factors in the prevention of hip fracture: the Leisure World study. Epidemiology 1991; 2 (1): 16-25 Wyshak G, Frisch RE, Albright TE, et al. Bone fractures among former college athletes compared with nonathletes in the menopausal and postmenopausal years. Obstet Gynecol 1987; 69 (1): 121-6 Nelson ME, Fiatarone MA, Morganti CM, et al. Effects of high-intensity strength training on multiple risk factors for osteoporotic fractures: a randomized controlled trial. JAMA 1994; 272 (24): 1909-14

ª 2009 Adis Data Information BV. All rights reserved.

467

153. Kerr D, Morton A, Dick I, et al. Exercise effects on bone mass in postmenopausal women are site-specific and loaddependent. J Bone Miner Res 1996; 11 (2): 218-25 154. Kohrt WM, Snead DB, Slatopolsky E, et al. Additive effects of weight-bearing exercise and estrogen on bone mineral density in older women. J Bone Miner Res 1995; 10 (9): 1303-11 155. Stengel SV, Kemmler W, Pintag R, et al. Power training is more effective than strength training for maintaining bone mineral density in postmenopausal women. J Appl Physiol 2005; 99 (1): 181-8 156. Chien MY, Wu YT, Hsu AT, et al. Efficacy of a 24-week aerobic exercise program for osteopenic postmenopausal women. Calcif Tissue Int 2000; 67 (6): 443-8 157. Kohrt WM, Ehsani AA, Birge Jr SJ, et al. Effects of exercise involving predominantly either joint-reaction or ground-reaction forces on bone mineral density in older women. J Bone Miner Res 1997; 12 (8): 1253-61 158. Verschueren SM, Roelants M, Delecluse C, et al. Effect of 6-month whole body vibration training on hip density, muscle strength, and postural control in postmenopausal women: a randomized controlled pilot study. J Bone Miner Res 2004; 19 (3): 352-9 159. Pruitt LA, Jackson RD, Bartels RL, et al. Weight-training effects on bone mineral density in early postmenopausal women. J Bone Miner Res 1992; 7 (2): 179-85 160. Maddalozzo GF, Snow CM. High intensity resistance training: effects on bone in older men and women. Calcif Tissue Int 2000; 66 (6): 399-404 161. Ryan AS, Treuth MS, Rubin MA, et al. Effects of strength training on bone mineral density: hormonal and bone turnover relationships. J Appl Physiol 1994; 77 (4): 1678-84 162. Ryan AS, Treuth MS, Hunter GR, et al. Resistive training maintains bone mineral density in postmenopausal women. Calcif Tissue Int 1998; 62 (4): 295-9 163. Bassey EJ, Rothwell MC, Littlewood JJ, et al. Pre- and postmenopausal women have different bone mineral density responses to the same high-impact exercise. J Bone Miner Res 1998; 13 (12): 1805-13 164. Sugiyama T, Yamaguchi A, Kawai S. Effects of skeletal loading on bone mass and compensation mechanism in bone: a new insight into the ‘‘mechanostat’’ theory. J Bone Miner Metab 2002; 20 (4): 196-200 165. Palombaro KM. Effects of walking-only interventions on bone mineral density at various skeletal sites: a metaanalysis. J Geriatr Phys Ther 2005; 28 (3): 102-7 166. Bergmann G, Graichen F, Rohlmann A. Hip joint loading during walking and running, measured in two patients. J Biomech 1993; 26 (8): 969-90 167. Gusi N, Raimundo A, Leal A. Low-frequency vibratory exercise reduces the risk of bone fracture more than walking: a randomized controlled trial. BMC Musculoskel Disord 2006; 7: 92 168. Rubin C, Recker R, Cullen D, et al. Prevention of postmenopausal bone loss by a low-magnitude, highfrequency mechanical stimuli: a clinical trial assessing compliance, efficacy, and safety. J Bone Miner Res 2004; 19 (3): 343-51

Sports Med 2009; 39 (6)

468

169. Zehnacker CH, Bemis-Dougherty A. Effect of weighted exercises on bone mineral density in post menopausal women: a systematic review. J Geriatr Phys Ther 2007; 30 (2): 79-88 170. Englund U, Littbrand H, Sondell A, et al. The beneficial effects of exercise on BMD are lost after cessation: a 5-year follow-up in older post-menopausal women. Scand J Med Sci Sports. Epub 2008 May 22 171. Mackey DC, Lui LY, Cawthon PM, et al. High-trauma fractures and low bone mineral density in older women and men. JAMA 2007; 298 (20): 2381-8

ª 2009 Adis Data Information BV. All rights reserved.

Guadalupe-Grau et al.

172. Frost HM. Some effects of basic multicellular unit-based remodelling on photon absorptiometry of trabecular bone. Bone Miner 1989; 7 (1): 47-65 173. Forwood MR, Burr DB. Physical activity and bone mass: exercises in futility? Bone Miner 1993; 21 (2): 89-112

Correspondence: Prof. Jose A.L. Calbet, Departamento de Educacio´n Fı´sica, Campus Universitario de Tafira, 35017 Las Palmas de Gran Canaria, Canary Islands, Spain. E-mail: [email protected]

Sports Med 2009; 39 (6)

Sports Med 2009; 39 (6): 469-490 0112-1642/09/0006-0469/$49.95/0

REVIEW ARTICLE

ª 2009 Adis Data Information BV. All rights reserved.

Lactate Threshold Concepts How Valid are They? Oliver Faude,1,2 Wilfried Kindermann2 and Tim Meyer1,2 1 Institute of Sports Medicine, University Paderborn, Paderborn, Germany 2 Institute of Sports and Preventive Medicine, University of Saarland, Saarbru¨cken, Germany

Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Historical Remarks on Endurance Performance Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Incremental Exercise Testing and the Interpretation of Blood Lactate Curves . . . . . . . . . . . . . . . . . . . 2.1 The Entire Blood Lactate Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Test Design and Data Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Methodology of Blood Lactate Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 A Framework for Endurance Diagnosis and Training Prescriptions. . . . . . . . . . . . . . . . . . . . . . . . . . 3. Validation of Lactate Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Competition Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 The Maximal Lactate Steady State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Lactate Threshold Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Located Lactate Threshold Concepts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.1 Aerobic Lactate Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.2 Anaerobic Lactate Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Lactate Thresholds and (Simulated) Competition Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Lactate Thresholds and Maximal Lactate Steady State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Conclusions and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

469 470 471 471 472 472 473 474 474 474 475 475 475 476 477 480 484

During the last nearly 50 years, the blood lactate curve and lactate thresholds (LTs) have become important in the diagnosis of endurance performance. An intense and ongoing debate emerged, which was mainly based on terminology and/or the physiological background of LT concepts. The present review aims at evaluating LTs with regard to their validity in assessing endurance capacity. Additionally, LT concepts shall be integrated within the ‘aerobic-anaerobic transition’ – a framework which has often been used for performance diagnosis and intensity prescriptions in endurance sports. Usually, graded incremental exercise tests, eliciting an exponential rise in blood lactate concentrations (bLa), are used to arrive at lactate curves. A shift of such lactate curves indicates changes in endurance capacity. This very global approach, however, is hindered by several factors that may influence overall lactate levels. In addition, the exclusive use of the entire curve leads to some uncertainty as to the magnitude of endurance gains, which cannot be precisely estimated. This deficiency might be eliminated by the use of LTs. The aerobic-anaerobic transition may serve as a basis for individually assessing endurance performance as well as for prescribing intensities in

Faude et al.

470

endurance training. Additionally, several LT approaches may be integrated in this framework. This model consists of two typical breakpoints that are passed during incremental exercise: the intensity at which bLa begin to rise above baseline levels and the highest intensity at which lactate production and elimination are in equilibrium (maximal lactate steady state [MLSS]). Within this review, LTs are considered valid performance indicators when there are strong linear correlations with (simulated) endurance performance. In addition, a close relationship between LT and MLSS indicates validity regarding the prescription of training intensities. A total of 25 different LT concepts were located. All concepts were divided into three categories. Several authors use fixed bLa during incremental exercise to assess endurance performance (category 1). Other LT concepts aim at detecting the first rise in bLa above baseline levels (category 2). The third category consists of threshold concepts that aim at detecting either the MLSS or a rapid/distinct change in the inclination of the blood lactate curve (category 3). Thirty-two studies evaluated the relationship of LTs with performance in (partly simulated) endurance events. The overwhelming majority of those studies reported strong linear correlations, particularly for running events, suggesting a high percentage of common variance between LT and endurance performance. In addition, there is evidence that some LTs can estimate the MLSS. However, from a practical and statistical point of view it would be of interest to know the variability of individual differences between the respective threshold and the MLSS, which is rarely reported. Although there has been frequent and controversial debate on the LT phenomenon during the last three decades, many scientific studies have dealt with LT concepts, their value in assessing endurance performance or in prescribing exercise intensities in endurance training. The presented framework may help to clarify some aspects of the controversy and may give a rationale for performance diagnosis and training prescription in future research as well as in sports practice.

1. Historical Remarks on Endurance Performance Diagnosis As early as 1808, Berzelius observed that lactic acid was produced in the muscles of hunted stags.[1] About a century later, several scientists studied the biochemistry of energy metabolism and muscle contraction in more detail. This led to a much deeper understanding of the formation of lactic acid (lactate and hydrogen ions) during intense exercise.[2-5] At that time, it was common belief that lactic acid is a waste product of glycolysis and will be formed when oxygen delivery to exercising muscles is not sufficient and muscle anaerobiosis occurs.[2,6,7] This view has been challenged considerably during the last two decª 2009 Adis Data Information BV. All rights reserved.

ades. Anaerobic glycolysis and, thus, lactate kinetics rather seem to be an ongoing process – even in the resting individual – which is highly related to the metabolic rate but not necessarily to oxygen availability (for detailed review see Gladden,[1,8] Brooks,[9] Robergs et al.[10]). In the first half of the 20th century the concept of maximum oxygen consumption as the first and probably most common means of evaluating aerobic endurance capacity was developed by the working group of Nobel Laureate AV Hill.[6]
maximal oxygen uptake (VO2max) has been established as a valuable tool to distinguish between fit and unfit subjects. However, several concerns were raised regarding the sensitivity of
VO2max. For instance, it is difficult to discriminate Sports Med 2009; 39 (6)

Validity of Lactate Thresholds

between subjects of homogenous performance
levels by means of VO2max.[11-18] In addition, sufficient effort during whole-body work and, therefore, adequate motivation of the investigated subject is necessary to appropriately de
termine VO2max. Particularly in clinical settings with diseased patients, whole-body exhaustion is difficult to attain or is even avoided because of the risk of adverse events.[19,20] Therefore, attempts have been made to establish sub-maximal parameters to assess cardiorespiratory fitness in patients and athletes. Early research by the working group of Hollmann established the so-called ‘point of optimum ventilatory efficiency’ corresponding to the first increase in the ventilatory equivalent of oxygen and of arterial lactate concentrations during incremental exercise.[19,21] A few years later, Wasserman and McIllroy[22] determined this intensity by plotting ventilation versus oxygen uptake in cardiac patients and named it the ‘anaerobic threshold’ (LTAn). At that time, routine determination of blood lactate concentrations (bLa) was associated with several difficulties and gas exchange measurements were more common – especially in clinical settings. Therefore, it became popular to detect the LTAn by means of gas exchange analysis. In the 1960s, the enzymatic method for measuring lactate concentrations from capillary blood samples was developed. This led to the increasing popularity of using bLa as a parameter to assess endurance capacity as well as for classifying work rate during exercise.[19,23,24] In the following years, numerous lactate threshold (LT) concepts were developed. The number of scientific studies on LTs has increased enormously up to now and the sub-maximal course of bLa during incremental exercise has probably become one of the most important means in the diagnosis of endurance performance in sports practice.[15,16,25,26] However, the variety of different threshold concepts has led to considerable confusion and misinterpretation. An intense and ongoing debate emerged, which was mainly based upon terminology and/or the physiological background of LT concepts.[27] Early assumptions on lactate producª 2009 Adis Data Information BV. All rights reserved.

471

tion and distribution in the organism have been challenged.[1,8-10,28] It has been argued that bLa increase continuously rather than show a clear threshold during incremental exercise. Furthermore, the contribution of aerobic and anaerobic pathways to energy production does not change suddenly but shows a continuous transition and, therefore, the term ‘threshold’ might be misleading.[29] Against this background and to unravel the confusion, it seems valuable to give a summary on published LT concepts. The present review is mainly aimed at evaluating the located LT concepts with regard to their validity in assessing aerobic endurance capacity and prescribing training intensity. A further aim was to try to integrate those concepts into a framework that was originally called the aerobic-anaerobic transition.[30-32] It has to be emphasized that this text focuses on LTs only. Although a close link between lactate and gas exchange markers has often been proposed,[21,31,33-36] there is still controversial debate with regard to the underlying physiological mechanisms.[37] A comprehensive review on gas exchange thresholds has recently been published.[31] Additionally, it is not within the scope of this article to exhaustively review the biochemistry of glycolysis and lactate metabolism. 2. Incremental Exercise Testing and the Interpretation of Blood Lactate Curves 2.1 The Entire Blood Lactate Curve

Usually, graded incremental exercise tests (GXTs) are used to evaluate aerobic endurance performance capacity. Typically, an exponential rise in bLa during incremental exercise testing can be observed (figure 1). The issue of interest is to interpret the resulting lactate curve with regard to endurance capacity. It is generally accepted that a rightward shift of the lactate curve (lower bLa at given workload) can be interpreted in terms of an improved endurance capacity[38-40] and, in contrast, a shift to the left (higher bLa at given workload) is usually considered to represent worsening endurance capacity.[41] Sports Med 2009; 39 (6)

Faude et al.

Blood lactate concentration (mmol/L)

472

10 8

Regenerative/ Moderate-/ Interval low-intensity high-intensity training endurance endurance sessions training training

6 Aerobicanaerobic transition

4 Aerobic threshold

MLSS = anaerobic threshold

2 0 Work intensity

Fig. 1. A typical lactate-workload plot including the aerobicanaerobic transition as a framework to derive endurance training intensities for different intensity zones. MLSS = maximal lactate steady state.

Overall lactate levels are known to be influenced by depleted glycogen stores (due to a low carbohydrate diet or preceding exhaustive exercise).[42-44] For instance, lower bLa at the same work rates have been reported in a glycogendepleted subject compared with a subject in normal condition. This may lead to a downward shift of the lactate curve and it is important that this is not falsely interpreted as an enhancement in endurance capacity.[45] Furthermore, several other factors like muscle fibre composition, glycolytic and lipolytic enzyme activity as well as capillary or mitochondrial density might influence blood lactate curves.[46] Additionally, the entire lactate curve is dependent on several other methodological issues, which should be taken into account when interpreting test results. 2.1.1 Test Design and Data Treatment

It is of note that the specific GXT protocol can vary considerably with regard to starting and subsequent work rates, work rate increments and stage duration. A recent review focused on the influence of varying test protocols on markers usually used in the diagnosis of endurance performance.[47] For instance, varying stage duration or work rate increments may lead to relevant differences in blood lactate curves and LTs.[48-50] A possible reason might be the time allowed for ª 2009 Adis Data Information BV. All rights reserved.

lactate diffusion in the blood until the next work rate increment.[47] In addition, there has been great debate on the best fitting procedure for the obtained bLa data set. For instance, a single-[51] or double-phase model[52] using two or three linear regression segments, a double-log model,[53] a third-order polymonial[54] or an exponential function[55] have been used in previous studies. Up to now, no generally accepted fitting procedure has been established.[47] Thus, it seems appropriate that test design as well as data fitting procedures should be chosen (and reported) as has been originally described for a certain LT. 2.1.2 Methodology of Blood Lactate Determination

From a methodological point of view, the site (earlobe, fingertip) as well as the method (venous, arterial, capillary) of blood sampling[56,57] and the laboratory methods (lactate analyser, analysed blood medium)[58-60] may also affect the test result. Samples taken from the earlobe have uniformly been shown to result in lower bLa than samples taken from the fingertip.[57,61,62] With regard to the analysed blood medium, plasma values were considerably higher than whole venous lactate concentrations, with capillary values lying in between.[48,56,63-65] In addition, several studies reported partly considerable differences between various lactate analysers (portable field vs laboratory analysers, amperometric vs photometric method) and under various climatic conditions.[58,66-69] The analysis of the whole blood lactate curve is a very global approach to evaluating endurance capacity. On the one hand, this approach is affected by the above-mentioned factors on overall lactate levels. On the other hand, the use of the entire curve leads to some uncertainty as to the magnitude of endurance gains that cannot be precisely estimated. However, the use of LTs enables a quantitative evaluation of changes in endurance performance. In addition, the ideal LT concept would not be affected by the abovementioned factors. There is evidence that approaches that analyse relative changes in bLa during GXTs may be favourable compared with the use of absolute lactate values in this regard.[56,67] Sports Med 2009; 39 (6)

Validity of Lactate Thresholds

2.2 A Framework for Endurance Diagnosis and Training Prescriptions

In 1979, Kindermann et al.[30] introduced the concept of the aerobic-anaerobic transition as a framework for performance diagnosis and training prescription in endurance sports (figure 1). Since then, this framework has been adopted, applied and refined by several scientists either using lactate or gas exchange markers.[16,26,31,33,34,46,70-75] This model consists of two typical breakpoints that are passed during incremental exercise. In the low intensity range, there is an intensity at which bLa begin to rise above baseline levels. This intensity was originally determined using gas exchange measurements,[21,22] and Wasserman called it the ‘anaerobic threshold’. This term has since been used for various LTs, particularly those with a different physiological background,[33,75] and, thus, has caused considerable confusion. Kindermann et al.[30] and Skinner and McLellan[34] suggested this intensity be called the ‘aerobic threshold’ (LTAer), because it marks the upper limit of a nearly exclusive aerobic metabolism and allows exercise lasting for hours. This intensity might be suitable for enhancing cardiorespiratory fitness in recreational sports, for cardiac rehabilitation in patients or for lowintensity and regenerative training sessions in high level endurance athletes.[16,25,26,32,70,76-81] Exercise intensities only slightly above the LTAer result in elevated but constant bLa during steady-state exercise and can be maintained for prolonged periods of time (~4 hours at intensities in the range of the first increase in bLa[82-84] and 45–60 minutes at an intensity corresponding to the maximal lactate steady state [MLSS][85,86]). Although anaerobic glycolysis is enhanced, it is speculated that such intensities may induce a considerable increase in the oxidative metabolism of muscle cells.[30,87] Theoretically, a high stimulation of oxidative metabolism for as long a period of time as is possible in this intensity range might be an appropriate load for endurance training. The highest constant workload that still leads to an equilibrium between lactate production and lactate elimination represents the MLSS. ª 2009 Adis Data Information BV. All rights reserved.

473

Some authors suggested that this intensity be called the ‘anaerobic threshold’.[27,30,49,88] It has been shown that the constant bLa at MLSS is not equal in all individuals and can vary considerably (values from 2 up to 10 mmol/L were reported in several studies).[50,72,86,89-93] Beneke and von Duvillard[94] as well as Beneke et al.[95] reported that bLa at MLSS is dependent on the motor pattern of exercise. Therefore, it was suggested that to determine the LTAn, individualized approaches rather than a fixed bLa should be used.[88,96,97] The MLSS represents the upper border of constant load endurance training.[30,49,71,95] Intensities above the MLSS have been used to guide interval training sessions in different endurance sports.[26,31,98-102] The intensity range between LTAer and LTAn is called the aerobic-anaerobic transition. The described thresholds (first increase in bLa and MLSS) have recently also been called ‘lactate threshold and lactate turnpoint’, ‘lactate threshold and anaerobic threshold’, or ‘anaerobic threshold 1 and 2’, respectively.[26,75,103,104] Within the present review, it was decided to stick to the originally introduced nomenclature.[30,31,34] There has been an exhaustive debate whether there exist clear breakpoints in the lactate/work rate relationship or whether lactate increase is rather a continuous function during incremental work.[47] Furthermore, the terms ‘aerobic’ and ‘anaerobic’ threshold may suggest clearly discernible physiological processes. However, these processes are rather of a transitional nature with aerobic and anaerobic energetic pathways always simultaneously contributing to energy production during both low- and high-intensity exercise. However, the proposed model seems appropriate both from a practical and from a didactical point of view. In addition, there is evidence that the described breakpoints may have some exercise physiological relevance. It has been shown that exercise above the MLSS is associated with an over-proportional excretion of stress hormones as well as of immunological markers during constant load exercise.[105,106] Furthermore, Lucia et al.[107] observed changes in electromyographical activity of the vastus lateralis and Sports Med 2009; 39 (6)

Faude et al.

rectus femoris that were coincidental with the aerobic-anaerobic transition in 28 elite male cyclists. The widespread use of this model as well as the absence of an accepted alternative was the rationale for using this framework in the present review to categorize published LT concepts. 3. Validation of Lactate Thresholds 3.1 Competition Performance

It is widely accepted that LTs (and the submaximal course of bLa during incremental exercise) are a criterion measure for aerobic endurance performance.[24,26,30,72,81,108] In particular, it has been shown that LTs are superior to maximal oxygen uptake when assessing endurance performance in homogenous groups of athletes.[11,12,109-111] The obvious gold standard to validate an LT concept is to compare it with the most recent competition performance in an endurance event (concurrent validity) or to assess its value in predicting endurance performance in future events (predictive validity). As an alternative to competition performance, the results of laboratory tests simulating an endurance event can be used. This might have the advantage of a higher standardization and, therefore, these test results may be more reliable. Correlations between the test value (LT) and the validity criterion (competition performance) can be dependent on several confounding factors such as, for example, the chosen competitive event (duration, laboratory or outdoor, athletic track or off-road), the sport that is evaluated as well as sex or age group and its heterogeneity in terms of endurance. 3.2 The Maximal Lactate Steady State

Endurance capacity can – from a metabolic point of view – be regarded as the highest steady state by energy supply from oxidative phosphorylation.[87] Therefore, another approach to assess aerobic endurance performance is the determination of the highest constant exercise intensity that can be maintained for a longer period of time ª 2009 Adis Data Information BV. All rights reserved.

Blood lactate concentration (mmol/L)

474

10 8 6 MLSS

4 2 0 Rest

10 20 Time (min)

30

40

Fig. 2. The blood lactate response to several constant workload exercises with different intensities. The highest workload during which blood lactate concentrations can be still accepted as being steady state is defined as the maximal lactate steady state (MLSS).

without a continuous rise in bLa. This intensity represents the MLSS, which has been shown to be highly related to competition performance in endurance events (r [correlation coefficient] = 0.92 with 8 km running, r = 0.87 with 5 km running and r = 0.84 with 40 km cycling time trial speed, respectively).[112-114] The MLSS has been defined by some authors as the ‘anaerobic threshold’ because it represents an exercise intensity that can be maintained without considerable contribution of anaerobic metabolism.[27,30,50,72,115] Each higher intensity results in a clearly identifiable increase in bLa with time during constant load work.[50,86,88] The gold standard for the determination of the MLSS is performing several constant load trials of at least 30 minutes’ duration on different days at various exercise intensities (in the range of
50–90% VO2max, figure 2).[49,50,86,116,117] An increase in bLa of not more than 1 mmol/L between 10 and 30 minutes during the constant load trials appears to be the most reasonable procedure for MLSS determination.[86,115] MLSS represents a steady state in several but not all physiological parameters. For instance, oxygen uptake, carbon dioxide output, respiratory exchange ratio and bicarbonate concentration were reported to remain nearly constant during constant load exercise at MLSS, but respiratory rate and heart rate significantly increased during this time.[85,118] Sports Med 2009; 39 (6)

Validity of Lactate Thresholds

In several endurance sports it is recommended to aim at a defined metabolic strain when a certain training stimulus is intended.[71,73,119,120] Therefore, it seems likely that training intensities for endurance training can be appropriately described when MLSS is known. For the purposes of this review based on the above-mentioned rationales, LTs are considered valid as performance indicators when there are high linear correlations with (simulated) endurance performance. In addition, a close relationship between LTs and MLSS suggests validity with regard to the prescription of training intensities. Therefore, it is desirable that LTs should fulfil both validity criteria. 4. Lactate Threshold Concepts For the purposes of the present paper, the MEDLINE database PubMed was searched for the search terms ‘lactate threshold’, ‘aerobic threshold’ and ‘anaerobic threshold’ combined with either ‘endurance performance’ or ‘maximal lactate steady state’. Additionally, the references of the selected articles were searched for further relevant papers. The located original publications were searched for papers describing different LT concepts (section 4.1), a correlation between LTs and (simulated) endurance performance (section 4.2) or the relationship between LTs and the MLSS (section 4.3). 4.1 Located Lactate Threshold Concepts

A total of 25 different LT concepts were located. Two studies were excluded from the present analysis because threshold determination was not solely based on bLa but also took gas exchange measurements into account.[121,122] All threshold concepts were divided into three categories. Several authors used so-called fixed blood lactate thresholds (LTfix) during incremental exercise to evaluate aerobic endurance performance. These fixed bLas were set at 2, 2.5, 3 or 4 mmol/L[24,108,123-125] with LT4 (4 mmol/L lactate threshold, originally described by Mader et al.[24] and by others later as the onset of blood ª 2009 Adis Data Information BV. All rights reserved.

475

lactate accumulation [OBLA][108]) being the most frequently used method. 4.1.1 Aerobic Lactate Thresholds

Table I shows an overview of LT concepts that could be categorized as the first rise in bLa above baseline levels (LTAer). Several researchers described the procedure to determine this threshold with terms like ‘‘the first significant/marked/ systematic/non-linear/sharp/abrupt sustained increase in bLa above baseline’’.[30,110,126-133,138] Although the visual determination of the first rise of bLa above baseline levels seems obvious and simple, in practice it is associated with considerable problems because of the only slight changes in bLa on the first steps during GXTs. Yeh et al.[142] demonstrated that the visual detection of the LTAer (in that study called ‘anaerobic threshold’) led to relevant differences between observers. Therefore, it does not seem appropriate to determine this threshold by simple visual inspection. Thus, other methods were developed to make threshold determination more objective. For instance, some authors took the typical error of their lactate analysers into account and

Table I. Lactate threshold concepts that were categorized in the aerobic threshold category. For further explanation see text Method and description Work intensity or oxygen uptake before/at which bLa begins to increase above baseline level[110,126] at which bLa exhibits a marked/systematic/significant/non-linear/ sharp/abrupt sustained increase above baseline value[30,110,127-133] first significant elevation of lactate level (approximately 2 mmol/L)[30,34] before an elevation in bLa above baseline (at least 0.2 mmol/L due to error of lactate analyser)[123,134] rise in delta lactate (onset of plasma lactate accumulation)[109] at minimum lactate equivalent (bLa divided by oxygen uptake or work intensity)[36,135-137] at which plasma lactate concentration begins to increase when log bLa is plotted against log (work intensity)[53] at which bLa increases 0.5 mmol/L above resting concentration[138] at which bLa increases 1 mmol/L above baseline (i.e. lactate at low
intensity corresponding to 40–60% VO2max)[111,139] preceding a bLa increase by 1 mmol/L or more[140,141]
bLa = blood lactate concentrations; VO2max = maximal oxygen uptake.

Sports Med 2009; 39 (6)

Faude et al.

476

Table II. Lactate threshold concepts that were categorized in the anaerobic threshold category. For further explanation see text Threshold concept

Method and description

IAT (Stegmann et al.)[88]

Tangent to bLa curve from recovery curve where bLa is equal to the value at end of GXT

IAT (Keul et al.)[96]

Tangent to bLa curve at 51

IAT (Simon et al.)[97]

Tangent to bLa curve at 45

IAT (Berg et al.)[137]

Intersection point between tangent for the minimum lactate equivalent and the linear function for the final 90 sec of GXT

IAT (Bunc et al.)[143]

Intersection between the exponential regression of the lactate curve and the bisector of the tangents of the upper and lower parts of the lactate curve

IAT (Dickhuth et al.)[36,136]

1.5 mmol/L above minimum lactate equivalent

IAT (Baldari and Guidetti)[144]

The second lactate increase of at least 0.5 mmol/L from the previous value

Dmax (Cheng et al.)[54]

Maximal distance from bLa curve to the line formed by its endpoints

Dmod (Bishop et al.)[140]

Maximal distance from bLa curve to the line formed by the point before the first rise in bLa and the value at cessation of exercise

Lactate turnpoint[103]

The final running velocity before the observation of a sudden and sustained increase in bLa
between LTAer and VO2max

Lactate minimum speed[145]

Minimum in bLa during GXT after high intensity exercise

bLa = blood lactate concentration; GXT = incremental exercise test; IAT = individual anaerobic threshold; LTAer = aerobic threshold;
VO2max = maximal oxygen uptake.

determined this LT as the workload 0.2 mmol/L above the lowest exercise lactate value.[123] Hughson and Green[138] arbitrarily chose an increase of 0.5 mmol/L above resting lactate concentrations. Another work group[111,139] chose a 1 mmol/L increment above lactate levels at low
intensity (~40% to 60% VO2max) because it could be determined objectively and in a standardized manner in all subjects. Furthermore, the lowest value when bLa is divided by work intensity or
VO2 has also been used as a marker for LTAer (minimum lactate equivalent).[36,135-137] Whereas Beaver and colleagues[53] used a log-log transformation to assess the first rise in bLa more objectively as the intersection of two linear regressions, Farrell et al.[109] plotted the difference in bLa between two consecutive stages against work intensity and determined the first rise of this relationship (onset of plasma lactate accumulation). 4.1.2 Anaerobic Lactate Thresholds

All threshold concepts that were assigned either to the MLSS or to a rapid/distinct change in the inclination of the blood lactate curve were categorized as LTAn (table II). Originally, the LT4 was established because it seemed to be the highest bLa that was sustainable for a longer duration and, therefore, was regarded ª 2009 Adis Data Information BV. All rights reserved.

as the upper border for constant load endurance training.[24] It was soon recognized that a fixed bLa does not take into account considerable interindividual differences and that LT4 may frequently underestimate (particularly in anaerobically trained subjects) or overestimate (in aerobically trained athletes) real endurance capacity.[88,96,97,146] Therefore, several so-called ‘individualized’ LT concepts were developed. For instance, Keul et al.[96] and Simon et al.[97] determined the individual anaerobic threshold (IAT) at a certain inclination of the lactate curve (tangent of 51 and 45, respectively). However, it seems questionable whether the use of a fixed inclination may reflect individual lactate kinetics better than a fixed bLa. Stegmann et al.[88] developed a more complicated model that is based on the exercise lactate curve as well as on the lactate course during the early recovery period. This model is based on several assumptions regarding lactate distribution in blood and muscle compartments, lactate diffusion through biological membranes and lactate elimination. However, some of these premises have been challenged.[8,147] Berg et al.[137] defined the LTAn as the intersection point between the tangent at the minimum lactate equivalent and the linear function Sports Med 2009; 39 (6)

Validity of Lactate Thresholds

for the final 90 seconds of GXT. Similarly, Bunc et al.[143] determined the LTAn as the intersection between the exponential regression of the lactate curve and the bisector of the tangents on the upper and lower parts of the regression. A comparable model was established by Cheng et al.[54] and called the Dmax method. Those authors determined the maximal perpendicular distance of the lactate curve from the line connecting the start with the endpoint of the lactate curve. It is obvious that these threshold models are dependent on the start intensity as well as the maximal effort spent by the subjects. To eliminate the influence of the start point of the GXT, Bishop et al.[140] connected the LTAer with the endpoint of the lactate curve and observed that this modified Dmax threshold (Dmod) was also highly correlated with performance during a 1-hour time trial in 24 female cyclists. Tegtbur et al.[145] developed the so-called lactate minimum test. This test starts with a short supramaximal exercise leading to high bLa. A short rest period (about 8 minutes)[145] should allow for an equilibrium between muscle and bLa. Immediately after this rest period, a standard incremental exercise test is conducted. After an initial fall of bLa at low exercise intensities, bLa begins to rise again. The lowest point of the lactate curve, the lactate minimum speed (LMS), is assumed to mark the LTAn. This procedure has recently been criticized because standardization is difficult.[112,148] For instance, the induced acidosis prior to the incremental test is unlikely to be uniform for different subjects. Additionally, initial intensity as well as stage increment and duration seem to affect LMS. Furthermore, supramaximal exercise might be contraindicated in some instances, for example in cardiac patients or in athletes at some time points during their training. Baldari and Guidetti[144] defined the IAT as the workload corresponding to the second lactate increase of at least 0.5 mmol/L with the second increase greater than or equal to the first one. A limitation to this approach is that only discrete stages according to the test protocol can be identified as threshold workload. Additionally, those authors determined the IAT by plotting each lactate value against the preceding workª 2009 Adis Data Information BV. All rights reserved.

477

load. This was claimed to be done because during 3-minute stages no steady-state lactate level could be reached[147] and, therefore, it was hypothesized that a lactate value at a given 3-minute stage would represent the realistic value of the previous stage. From empirical observations, the work group of Dickhuth et al.[36,135,136] estimated the IAT at a blood lactate concentration 1.5 mmol/L above the minimum lactate equivalent (i.e. above LTAer). Finally, the lactate turnpoint (LTP) has been defined as the final running velocity before the observation of a sudden and sustained increase in bLa between LTAer and
VO2max.[103] Reproducibility of the velocity or power output at LTs has been reported to be high (r > 0.9, independent of whether LTfix, LTAer or LTAn
were analysed).[52,149-152] For VO2 at LTs, reliability coefficients seem to be slightly lower (r = 0.8–0.9).[150,152,153] 4.2 Lactate Thresholds and (Simulated) Competition Results

Thirty-eight studies were located that compared LT values with performance in endurance events or simulated competitions. Six studies were excluded from the analysis. Three of these studies compared an LT obtained during cycling exercise with running performance,[110,154,155] two studies only reported LT as a fraction of
VO2max,[11,156] and one study reported correlations with time-to-exhaustion in an open-end interval programme.[157] A total of 32 studies were thus included in this analysis. Eighteen studies evaluated the correlation of
the work intensity (running velocity or VO2) at various LTs with performance in running competitions of different distances (800 m up to marathon; table III).[108,109,112,123,124,129-132,134,135,158-164] Competition distances from 0.8 to 3.2 km, from 5 km to 16.1 km and from 21.1 to 42.2 km were subsumed as correlates of short-, middle- and long-distance endurance events. The main result was that nearly all studies reported high correlation coefficients with (simulated) competition results. These results were confirmed by Weltman Sports Med 2009; 39 (6)

Faude et al.

478

Table III. Correlation coefficients between lactate thresholds and running performance over various distances Threshold concept

LTfix

0.8–3.2 km

5 km–16.1 km

19.3–42.2 km

v


VO2

v


VO2

v


VO2

0.82[135] 0.88[123] 0.86[123] 0.85[123] 0.87[134] 0.85[134] 0.84[134] 0.93[158] 0.78[132] 0.68[131] 0.85[131] 0.88[131]

0.79[123] 0.75[123] 0.75[123] 0.72[134] 0.74[134] 0.75[134] 0.73[158] 0.60[132] 0.51[131] 0.55[131] 0.69[131]

0.88[135] 0.91[135] 0.91[159] 0.93[159] 0.91[159] 0.84[159] 0.91[159] 0.94[159] 0.83[160] 0.81[112] 0.95[163] 0.94[163]

0.90[159] 0.92[159] 0.92[159] 0.83[159] 0.88[159] 0.93[159] 0.86[163] 0.74[163]

0.91[135] 0.81[135] 0.98[124] 0.98[124] 0.98[124] 0.68[129] 0.96[108] 0.91[163] 0.92[163]

0.76[161] 0.83[163] 0.73[163]

Median (min–max)

0.85 (0.68–0.93)

0.73 (0.51–0.79)

0.91 (0.81–0.95)

0.89 (0.74–0.93)

0.92 (0.68–0.98)

0.76 (0.73–0.83)

LTAer

0.74[135] 0.85[123] 0.70[134] 0.93[158] 0.77[132] 0.43[131] 0.65[131] 0.70[131] 0.91[109]

0.77[123] 0.61[134] 0.84[158] 0.69[132] 0.77[131] 0.66[131] 0.64[131] 0.85[109] 0.62[162] 0.66[162] 0.58[162]

0.73[135] 0.79[135] 0.78[160] 0.96[109] 0.97[109] 0.79[130] 0.83[130] 0.79[130] 0.84[130] 0.83[130] 0.81[130] 0.93[112] 0.94[163] 0.92[163] 0.92[163] 0.89[163] 0.87[163] 0.85[163]

0.89[109] 0.91[109] 0.84[162] 0.83[162] 0.79[162] 0.69[162] 0.92[162] 0.79[162] 0.76[130] 0.77[130] 0.84[130] 0.81[130] 0.82[130] 0.88[130] 0.72[163] 0.56[163] 0.66[163] 0.52[163] 0.81[163] 0.69[163]

0.76[135] 0.81[135] 0.78[129] 0.97[109] 0.98[109] 0.90[163] 0.91[163] 0.87[163] 0.86[163] 0.83[163] 0.77[163]

0.91[109] 0.89[109] 0.69[163] 0.52[163] 0.66[163] 0.42[163] 0.80[163] 0.65[163]

Median (min–max)

0.74 (0.43–0.93)

0.66 (0.58–0.85)

0.84 (0.73–0.97)

0.79 (0.45–0.92)

0.86 (0.76–0.98)

0.68 (0.42–0.91)

LTAn

0.88[135]

0.91[135] 0.92[135] 0.86[160] 0.83[112] 0.93[163] 0.91[163] 0.94[163] 0.90[163] 0.76[164] 0.73[164]

0.83[163] 0.70[163] 0.81[163] 0.66[163] 0.45[164] 0.45[164]

0.93[135] 0.93[135] 0.90[163] 0.91[163] 0.90[163] 0.89[163]

0.68[161] 0.83[163] 0.71[163] 0.81[163] 0.67[163]

Median (min–max) 0.88 0.91 (0.83–0.94) 0.76 (0.66–0.83) 0.91 (0.89–0.93)
LTfix = fixed lactate threshold; LTAer = aerobic threshold; LTAn = anaerobic threshold; v = velocity; VO2 = oxygen uptake.

et al.,[123,134] who cross-validated the obtained regression equations and found high correlation coefficients between actual and predicted scores. There is a tendency for higher correlations with longer endurance events (0.43–0.93 in short-term ª 2009 Adis Data Information BV. All rights reserved.

0.71 (0.67–0.83)

events vs 0.68–0.98 over the long-distance competitions). Additionally, correlations tended to be higher for LTfix and LTAn compared with LTAer. This might be due to the average intensity in running events being higher than the intensity Sports Med 2009; 39 (6)

Validity of Lactate Thresholds

479

corresponding to the first increase in bLa. In total, the results of the analysed studies point to a common variance of LTs and competition results in running events between 55% and 85%. In cycling, a total of eight studies evaluated the relationship between LTs and (simulated) cycling time trial performance (table IV).[12,89,140,141,165-168] Only one study analysed the correlation with short-duration time trial performance (4000 m individual pursuit) and found a high correlation coefficient of r = 0.86 in 18 male high-performance track cyclists.[167] Four studies evaluated distances between 13.5 and 20 km or time trial durations between 20 and 30 minutes.[89,165,166,168] The correlation coefficients in these studies were in most cases higher (between 0.8 and 0.9) than for the longer time trials (40 km or 60–90 minutes, r ~ 0.7).[140,141,165] Overall, the results of these studies were more heterogeneous. Correlation coefficients between LTs and (simulated) competition performance varied between r = 0.23[165] and r = 0.93.[89] In total, the results of the analysed studies point to a common variance of LTs and competition results between 35% and 85% in cycling events. However, the low number of studies and the heterogeneous results point to the need for further carefully designed studies to

arrive at more comprehensive conclusions with regard to the relationship of LTs and time trial performance in cycling. Two studies were found that analysed the relationship of LT markers with mountain bike cross-country race performance.[169,170] Such races are usually conducted on a graded terrain with considerable time spent ascending and descending. Impellizzeri et al.[170] observed high correlations between LTAer as well as OBLA and race time during a 31 km mountain bike race. Whereas correlations were about 0.7 when LT was expressed in absolute terms, correlations became considerably higher (~0.9) when power output at LT was expressed relative to body mass. Similarly, Gregory et al.[169] reported higher correlations between LTAer and a crosscountry time trial in 11 male mountain bikers when LTAer was expressed as related to body mass (r ~ 0.5 in absolute terms vs r ~ 0.8 relative to body mass). This finding can be explained with the considerable influence of bodyweight and body composition on performance capacity in cyclists during ascents.[171-173] In addition to the studies in running and cycling, another four studies were detected that evaluated LTs and (simulated) competition

Table IV. Correlation coefficients between lactate thresholds and cycling time trial events over various distances and times Threshold concept

4 km PO

13.5–20 km; 20–30 min
VO2

[165]

LTfix

0.23 0.82[166] 0.90[166]

Median (min–max) LTAer

0.86[167]

Median (min–max)

0.86

LTAn

40 km; 60–90 min
VO2

PO


V O2

PO [165]

0.54 0.60[141] 0.81[140]

0.82 (0.23–0.90)

0.60 (0.54–0.81)

0.67[165] 0.88[166] 0.86[166] 0.91[168] 0.88[168]

0.91[165] 0.59[141] 0.61[140] 0.69[140] 0.65[140]

0.93[12]

0.88 (0.67–0.91)

0.65 (0.59–0.91)

0.93

0.45[165] 0.89[166] 0.91[166] 0.93[89]

0.77[165] 0.58[141] 0.52[141] 0.72[141] 0.84[140] 0.83[140]

Median (min–max) 0.90 (0.45–0.93) 0.75 (0.52–0.84)
LTAer = aerobic threshold; LTAn = anaerobic threshold; LTfix = fixed lactate threshold; PO = power output; VO2 = oxygen uptake.

ª 2009 Adis Data Information BV. All rights reserved.

Sports Med 2009; 39 (6)

Faude et al.

480

performance. Two of these studies analysed competitive race walkers. Yoshida et al.[174] found correlation coefficients for OBLA as well as for LTAer of 0.94 and 0.85, respectively, with walking pace during a 5 km road race in eight female race walkers. Similar results were observed by Hagberg and Coyle[111] in a heterogeneous group of race walkers with correlation coefficients of 0.94 and 0.82 for velocity and oxygen uptake at LTAer in a 20 km race walking performance. Two studies dealt with rowing performance and LTs. Whereas Ingham et al.[175] observed high correlations (r = 0.86–0.92) between work rate at fixed and aerobic LTs and 2000 m ergometer performance in 41 rowers of different categories, Cosgrove et al.[176] found considerably lower correlations (r = 0.39–0.73) in 13 male rowers. To summarize, the overwhelming majority of published studies on the relationship between LTs and endurance performance showed strong correlations, particularly for running events. This supports findings of earlier training studies that found training-induced improvements in competitive performance significantly correlated with improvements in LTs.[130,162] Although it seems likely that other influences such as central nervous system processes may have regulatory and decisive characteristics in endurance events as it was recently claimed,[177] peripheral metabolic adaptations highly related to the LT[46] seem to be a necessary and important prerequisite for aerobic endurance performance.

4.3 Lactate Thresholds and Maximal Lactate Steady State

MLSS determination has become very popular in performance diagnosis in several endurance sports. Thus, numerous studies have dealt with the problem of an adequate estimation of MLSS during one single laboratory visit. For instance, some authors tried to estimate MLSS from performance during all-out time trials (5 km or 40 km)[114,178] from physiological strain (bLa, heart rate, ratings of perceived exertion) during standardized sub-maximal constant-load exercise[179-182] or from gas exchange measurements.[183-189] ª 2009 Adis Data Information BV. All rights reserved.

However, an overview of those studies is beyond the scope of the present review. There are several studies that examined the metabolic responses during steady-state exercise intensities related to LTs but did not analyse exercise intensities slightly above or below. Schnabel et al.[190] observed average steady-state lactate concentrations (~4.5 mmol/L) during 50-minute runs at the IAT according to Stegmann et al.[88] However, no other intensity was analysed in this investigation. Stegmann and Kindermann[146] compared 50-minute cycling exercise in 19 subjects at the IAT as well as at LT4 and found steady-state lactate levels (~4 mmol/L) during IAT trials, whereas exercise at LT4 resulted in continuously rising bLa (up to 9.6 mmol/L) and a premature cessation. This is in line with findings of OyonoEnguelle et al.,[191] who similarly reported no lactate steady state in three out of five subjects during exercise at LT4. In contrast, Loat and Rhodes[189] found continuously increasing bLa (on average from 3.4 mmol/L after 15 minutes to 4.6 mmol/L after 45 minutes) and premature fatigue during 60-minute constant load trials at the IAT. However, those authors did not use the originally described test protocol and Heck[50] has shown that IAT determination is dependent on the protocol used. Baldari and Guidetti[144] compared steadystate running at their IAT determined when lactate values were plotted against the corresponding exercise intensity (IATm) and against the preceding intensity (IATa) and found steady-state lactate levels for IATa (~4 mmol/L-1) but not for IATm. However, due to the determination procedure, the difference between both thresholds was exactly one stage increment and no other intensities in between were evaluated. Ribeiro et al.[192] assessed a 40-minute steady-state cycling exercise at LTAer, between LTAer and LTAn (LTP), at LTAn as well as between LTAn and maximum. Those authors found on average steadystate lactate levels up to LTAn (~5 mmol/L-1), whereas at the highest intensity, bLa increased continuously and exercise had to be terminated prematurely. Bacon and Kern[193] and Tegtbur et al.[145] compared constant load trials at LMS and 5% or 0.2 m/s, respectively, above the LMS. Those Sports Med 2009; 39 (6)

Validity of Lactate Thresholds

authors found that LMS intensity but not the higher intensity on average resulted in a lactate steady state. However, in the study of Bacon and Kern,[193] the average blood lactate increase between minutes 12 and 28 during the constant load trial at the LMS +5% intensity was 1.2 mmol/L, and in four out of ten subjects a lactate steady state according to the recommended criterion[72,115] was present. A total of 11 studies evaluated the relationship between one or more LT concepts and MLSS using the recommended procedure, including several constant load trials of at least 30 minutes’ duration to determine the MLSS (table V). One study determined MLSS with 20-minute constant load trials.[113] Most researchers analysed the relationship of LT4 with MLSS.[49,72,90,92,112,117] For instance, Heck and colleagues[49,50,72] found strong correlations between LT4 and MLSS during running as well as during cycling exercise. However, the fitness level of their subjects was quite heterogeneous and, therefore, the high correlations to some extent might be spurious. Additionally, they observed that the velocity at LT4 was higher than MLSS velocity when stage duration during the GXT was 3 minutes, whereas this was not the case with 5-minute stages. Therefore, these authors concluded that LT4 gives a valuable estimate of the MLSS when stage duration is at least 5 minutes. Also, Jones and Doust[112] found a high correlation between LT4 and the MLSS in a homogenous group of trained runners with LT4 being higher than MLSS (3-minute stages). Lower correlations were found by van Schuylenbergh et al.[92] in elite cyclists as well as by Beneke[117] in a homogenous group of rowers. Also, LT4 and MLSS did not differ significantly with 6-minute stages,[92] whereas LT4 was considerably higher than MLSS with 3-minute stages.[117] Lajoie et al.[90] evaluated whether the intensity corresponding to 4 mmol/L lactate during a GXT with 8-minute stages and 30 W increments is appropriate to estimate the MLSS in nine cyclists. Average power output at MLSS and LT4 was not significantly different. However, because bLa at MLSS differed considerably between subjects, the authors concluded ª 2009 Adis Data Information BV. All rights reserved.

481

that it is unrealistic to rely on a blood lactate value of 4 mmol/L as a universal criterion for MLSS. Unfortunately, a more detailed analysis regarding the correlation or individual differences between LT4 and MLSS was not reported. Heck et al.[49,50] observed high correlations between MLSS and the IAT according to Stegmann et al.[88] In addition, running velocity was not significantly different between IAT and MLSS independent of stage duration (3 or 5 minutes), whereas in cycling IAT was about 8% higher than MLSS. Urhausen et al.[86] found in runners as well as in cyclists that constant load trials at IAT resulted on average in a lactate steady state, whereas a 5% higher intensity led to a continuous rise in bLa. Similarly, McLellan and Jacobs[91] arrived at the conclusion that the IAT is a valid estimate for the MLSS in most subjects, although there exists a considerable difference in a few cases. Unfortunately, these studies reported no measure of correlation between IAT and MLSS or no quantitative data on individual differences between IAT and MLSS. In contrast to the previously mentioned studies, Beneke[117] found the IAT to be considerably higher than MLSS in nine rowers. Additionally, the correlation in this study was lower than was observed by Heck et al.[49] This finding might be due to the more homogenous performance level of the rowers as well as to the slow increment in the chosen test protocol.[50] Heck et al.[49] and Heck[50] found high correlations between the IAT according to Keul et al.[96] and Bunc et al.[143] and the MLSS in running and cycling. However, the high correlations might be partly accounted for by the heterogenous endurance level of the subjects. Furthermore, both thresholds were dependent on the test protocol during the running tests (3-minute vs 5-minute stages). The LMS was evaluated in two studies.[89,112] The results of these studies were contradictory. Jones and Doust[112] found only a low correlation between LMS and MLSS. Additionally, LMS was considerably lower than MLSS. In contrast, LMS was not significantly different from MLSS in the study of MacIntosh et al.[89] These contrasting observations might have been due to Sports Med 2009; 39 (6)

Faude et al.

482

Table V. Comparison of lactate threshold concepts with MLSS determined by several constant load trials of different intensity Threshold concept

Subjects

Main outcome

Reference

LT4, OBLA

16 healthy males (running)

High correlation between LT4 and MLSS (r = 0.98) LT4 on average 0.12 m/s higher than MLSS with 3 min stages but not with 5 min stages during GXT Heterogenous endurance level

Heck et al.[49,72]

22 healthy subjects (cycling)

Significant correlation between LT4 and MLSS (r = 0.92) LT4 on average 19.9 W higher than MLSS Heterogenous endurance level, slow increase in power output (+6 W/min)

Heck[50]

8 trained male runners

High correlation (r = 0.93) between OBLA and MLSS OBLA on average 0.4 km/h higher than MLSS

Jones and Doust[112]

21 elite cyclists

Low correlation (r = 0.71) between LT4 and MLSS No significant difference between LT4 and MLSS (MLSS 15 W higher) Homogenous endurance level

Van Schuylenbergh et al.[92]

9 male rowers

Significant correlation (r = 0.82) between LT4 and MLSS LT4 significantly higher (32 W) than MLSS Homogenous endurance level

Beneke[117]

10 well trained cyclists

Average power output at LT4 and MLSS was not significantly different (282 W vs 277 W) Strong MLSS criterion ( 0.05; 95% CI -0.628, 0.116). 2.6.2 Exercise versus Antidepressant Medication

Three studies[32,86,87] compared exercise versus antidepressants and found an overall effect size of 0.02. This difference was not significant (t = 0.223; p > 0.05; 95% CI -0.152, 0.184). 2.7 Dose Response

Twelve studies provided adequate data to calculate exercise volume. One study yielded an exercise dose that was >3 standard deviations above the mean of all exercise doses and was removed from the analysis. Some of the remaining 11 studies included multiple exercise groups, resulting in 25 exercise groups. The Pearson productmoment correlation for exercise dose and effect sizes was found to be nonsignificant (r = -0.040; p > 0.05). Multiple regression analysis also Sports Med 2009; 39 (6)

Rethorst et al.

500

Table III. Moderating variables for the overall population Category

df

Qb

Level

k

ES

SE

Upper CI

Lower CI

Population

1

19.19883*

Non-clinical

41

-0.58685

0.043453

-0.50049

-0.67082

Clinical Age (y)

3

5.108655

Sex

Intervention duration (wk)

2

4

4.339101

45.9213*

17

-1.03137

0.091975

-0.85110

-1.21164

55

17

-0.65607

0.069295

-0.52026

-0.79189

Male

7

-0.72193

0.128623

-0.46982

-0.97403

Female

7

-0.49163

0.093083

-0.30918

-0.67407

Mixed

44

-0.7031

0.046652

-0.61166

-0.79453

Acute

2

-1.50348

0.220743

-1.07082

-1.93613 -0.79534

4–9

16

-0.64662

0.075876

-0.49791

10–16

26

-0.74666

0.063409

-0.62238

-0.87094

17–26

8

-0.24955

0.121108

-0.01218

-0.48692 -0.61906

>26 Exercise type

2

23.85307*

5

-0.43008

0.096417

-0.24111

48

-0.64056

0.042285

-0.55768

-0.72344

6

-0.54462

0.133631

-0.2827

-0.80653

Combined

4

-1.72853

0.22322

-1.29102

-2.16604

20–29

9

-0.79508

0.092961

-0.61287

-0.97728

30–44

14

-0.65653

0.083749

-0.49238

-0.82068

Aerobic Resistance

Bout duration (min)

Exercise frequency (/wk)

2

2

10.3643*

28.75021*

45–59

11

-0.45277

0.093997

-0.26854

-0.637

‡60

15

-0.46854

0.072793

-0.32586

-0.61121

7

-0.24831

0.089114

-0.07365

-0.42298

44

-0.78172

0.04868

-0.6863

-0.87713

5

4

-0.52009

0.12201

-0.28095

-0.75923

50–60

3

-0.76479

0.154123

-0.16667

-1.06687

61–74

11

-0.33339

0.08506

-0.16667

-0.50011

‡75

10

-0.8478

0.09814

-0.65544

-1.04015

2 3–4

Exercise intensity (%)

2

17.28126*

df = degrees of freedom; ES = effect size; k = number of effect sizes; Qb = measure of homogeneity, see text; SE = standard error; * p < 0.05.

revealed a nonsignificant relationship (F[1,23] = 0.037; p > 0.05). 3. Discussion Aerobic and resistance exercise programmes were hypothesized to significantly alleviate depressive symptoms. The overall effect size indicates an improvement in depression scores of 0.80 standard deviation units following an exercise programme. Analysis of moderating variables indicates aerobic and resistance exercises are equally effective. This finding supports the initial hypothesis and is consistent with the findings of Craft and Landers.[9] In the overall samª 2009 Adis Data Information BV. All rights reserved.

ple, analysis of moderating variables indicated that exercise interventions that combined aerobic and resistance exercise resulted in larger effects than aerobic or resistance exercise alone. It should be noted, however, that this effect size is based on only four trials, and further research must be conducted to confirm this finding. Additionally, within the clinically depressed population, aerobic and resistance exercises were found to be equally effective in alleviating depressive symptoms. The second hypothesis of this study stated that participants with clinical levels of depression would show improvements greater than those of the general population. Improvements within Sports Med 2009; 39 (6)

The Antidepressive Effects of Exercise

501

the clinically depressed population were 1.01 standard deviation units compared with 0.59 in the general population. This difference was found to be statistically significant and in line with the initial hypothesis and the previous findings of Craft and Landers.[9] The final hypothesis of this study was that longer intervention durations would result in greater improvements in depressive symptoms. This hypothesis was based on the findings of North et al.[11] and Craft and Landers[9] in which longer intervention durations resulted in larger decreases in depression scores. Analysis of moderating variables supports this hypothesis within the clinically depressed population, where interventions of 10–16 weeks resulted in larger effects than interventions lasting 4–9 weeks. However, within the overall population, interventions of 4–9 weeks resulted in significantly larger effects than interventions of 17–26 weeks, while interventions of 10–16 weeks resulted in significantly

larger effects than interventions of 16–26 weeks and >26 weeks. One possible explanation for these differences is a floor effect in the general population, where maximum improvements were achieved in the first 16 weeks of exercise training. Analysis also indicated significant differences in moderating variable categories that were not hypothesized. Within the overall population, exercise bouts of 20–29 minutes resulted in larger effects than bouts of ‡45 minutes, while within the clinically depressed population, exercise bouts of 45–49 minutes resulted in larger effects than bouts of 30–44 minutes and of ‡60 minutes. Once again, however, the analysis of moderating variables within the clinically depressed population includes a small number of trials, and more research must be done before conclusions can be drawn on the optimal exercise bout duration. Significant differences were also present across categories of exercise intensity across the overall population, with exercise of 61–74% maximum

Table IV. Planned comparisons for moderating variables (overall population) Category

Comparison

w2

p-Value

Population

Clinical vs non-clinical

19.199**

26 10–16 vs 17–26

Exercise type

Bout duration (min)

Exercise frequency (/wk)

Exercise intensity (%)

0.7339 13.223

0.3916 0.0003**

10–16 vs >26

7.5258

0.0061**

17–16 vs >26

1.3601

0.2435

Aerobic vs resistance

0.4686

0.4934

Aerobic vs combined

22.933**