Eur J Appl Physiol (2012) 112:32753285 DOI 10.

1007/s00421-011-2291-7

ORIGINAL ARTICLE

Inuence of altitude training modality on performance and total haemoglobin mass in elite swimmers
Clare E. Gough Philo U. Saunders John Fowlie Bernard Savage David B. Pyne Judith M. Anson Nadine Wachsmuth Nicole Prommer Christopher J. Gore

Received: 23 June 2011 / Accepted: 13 December 2011 / Published online: 11 January 2012 Springer-Verlag 2012

Abstract We compared changes in performance and total haemoglobin mass (tHb) of elite swimmers in the weeks following either Classic or Live High:Train Low (LHTL) altitude training. Twenty-six elite swimmers (15 male, 11 female, 21.4 2.7 years; mean SD) were divided into two groups for 3 weeks of either Classic or LHTL altitude training. Swimming performances over 100 or 200 m were assessed before altitude, then 1, 7, 14 and 28 days after returning to sea-level. Total haemoglobin mass was measured twice before altitude, then 1 and 14 days after return to sea-level. Changes in swimming performance in the rst
Communicated by Guido Ferretti. C. E. Gough (&) P. U. Saunders D. B. Pyne C. J. Gore Department of Physiology, Australian Institute of Sport, PO Box 176, Belconnen, Canberra, ACT 2616, Australia e-mail: clare.gough@ausport.gov.au C. E. Gough P. U. Saunders C. J. Gore Faculty of Health, University of Canberra, Canberra, Australia J. Fowlie AIS Swimming, Australian Institute of Sport, Canberra, Australia B. Savage Swimming Australia Ltd, Canberra, Australia J. M. Anson Faculty of Applied Science, University of Canberra, Canberra, Australia N. Wachsmuth N. Prommer Department of Sports Medicine and Sports Physiology, University of Bayreuth, Bayreuth, Germany C. J. Gore Exercise Physiology Laboratory, School of Education, Flinders University, Adelaide, Australia

week after Classic and LHTL were compared against those of Race Control (n = 11), a group of elite swimmers who did not complete altitude training. In addition, a seasonlong comparison of swimming performance between altitude and non-altitude groups was undertaken to compare the progression of performances over the course of a competitive season. Regardless of altitude training modality, swimming performances were substantially slower 1 day (Classic 1.4 1.3% and LHTL 1.6 1.6%; mean 90% condence limits) and 7 days (0.9 1.0% and 1.9 1.1%) after altitude compared to Race Control. In both groups, performances 14 and 28 days after altitude were not different from pre-altitude. The season-long comparison indicated that no clear advantage was obtained by swimmers who completed altitude training. Both Classic and LHTL elicited *4% increases in tHb. Although altitude training induced erythropoeisis, this physiological adaptation did not transfer directly into improved competitive performance in elite swimmers. Keywords Competition Hypoxia Living High:Training High

Introduction Living and/or training in moderate altitude (2,0003,000 m above sea-level) is popular with athletes seeking a legal performance enhancement and an additional training stimulus (Wilber 2007). In recent years, altitude training has been of great interest to researchers and coaches working with swimmers (Robach et al. 2006; Robertson et al. 2010a; Truijens et al. 2008), a reection of its popularity as a training method in the sport. Mid-season altitude exposure periods have become common practice with

123

3276

Eur J Appl Physiol (2012) 112:32753285

the aim of inducing favourable physiological adaptations for enhancing subsequent training and competitive performance (Saunders et al. 2009). Since the concept of altitude training became popular almost 50 years ago, modalities of altitude exposure have diversied greatly (Millet et al. 2010) and it can be difcult to determine which mode of altitude training should be used. Researchers can offer insight by comparing different altitude training methods for their inuence on an athletes physiology and sports performance. Classical altitude training (Classic) involves athletes living and training at moderate altitude (2,0003,000 m), typically for a period of 34 weeks. Live High:Train Low (LHTL) training, where athletes live at moderate altitude but train at a lower altitude, was rst proposed by Levine and Stray-Gundersen (1997) as a way of circumventing the limitations on training intensity encountered by athletes during Classic altitude training. Although both of these approaches are rmly established as popular training methods, Levine and Stray-Gundersen (1997) are the only researchers to have compared them directly for their inuence on physiology and sport performance. They reported that both Classic and LHTL induced signicant red cell mass increases (10.5 and 5.3%, respectively), but only LHTL elicited an improvement (1.3%) in 5 km running performance. In contrast, a recent meta-analysis (Bonetti and Hopkins 2009) concluded that both Classic and LHTL could yield substantial performance improvements, providing the optimal durations and magnitudes of exposure are followed for each method. Similarly, a contemporary review article (Saunders et al. 2009) estimated that changes in sports performance following a three week Classic or LHTL altitude training camp would be of equal magnitude (*2%). Since Classic and LHTL altitude training are both incorporated into the training programs of elite swimmers (Robertson et al. 2010a) and, given the paucity of comparative data, a sport specic investigation is required to conrm whether both methods of altitude training are equally benecial. The optimal dose of altitude exposure has been questioned many times (Rusko et al. 2004; Wilber et al. 2007) because, for a training camp of the same duration, athletes completing LHTL spend approximately half as many hours in hypoxia as those completing Classic altitude training (1214 h day-1 as opposed to 24 h day-1, respectively). One way in which researchers have approached this question in the past is to examine the bodys total haemoglobin mass (tHb) response as a marker of accelerated erythropoeisis. The hormonal response to each form of altitude training is similar (Koistinen et al. 2000) and tHb has been signicantly increased following both Classic (Friedmann et al. 2005; Garvican et al. 2008) and LHTL altitude (Clark et al. 2009; Wehrlin et al. 2006). However, Saunders et al. (2009) estimated that Classic altitude

training was three-times more effective in increasing tHb; 100 h of hypoxia leading to a 3% tHb increase in Classic, but only a 1% increase in LHTL. Although tHb, maximal aerobic power (VO2max) and endurance performance are theoretically linked (Schmidt and Prommer 2010), performance improvements following altitude are not always accompanied by increases in tHb (Brugniaux et al. 2006; Robach et al. 2006) and indeed this issue has been debated vigorously (Gore et al. 2007; Gore and Hopkins 2005; Levine and Stray-Gundersen 2005). As well as investigating whether Classic altitude engenders a greater tHb response than LHTL altitude in elite swimmers, an important contribution of the present study will be to add to the growing collection of studies investigating the relationship between changes in tHb and performance after altitude training. Empirical evidence for the time-course of performance changes following altitude training is sparse, but anecdotal commentaries have reported that athletes perform well for a few days immediately after altitude, followed by a dip in performance, before reaching peak condition 23 weeks after altitude exposure (Dick 1992; Fuchs and Reiss 1990; Millet et al. 2010). Only Levine and Stray-Gundersen (1997) have detailed a timeline of performance following both Classic and LHTL altitude training. They reported no signicant changes in running performance for 3 weeks after the return to sea-level with the exception of the aforementioned 1.3% improvement 1 day after LHTL. Other researchers have reported *2% improvements in swimming performance 10 days after Classic altitude without other time points for comparison (Friedmann et al. 2005), and 1.4% improved running performance immediately after LHTL altitude but a return to pre-altitude levels 2 weeks thereafter (Robertson et al. 2010c). Changes to indirect indicators of performance following altitude training, such as submaximal power output (Schmitt et al. 2006), hypoxic ventilatory response (Katayama et al. 2005) and buffering capacity (Gore et al. 2007), have been examined and suggest the potential for improved sport performance up to 1 month after the return to sea-level. The lack of empirical data charting the time-course of performance changes after altitude training, a key issue frequently debated by coaches, is surprising and warrants systematic investigation. Based on the existent literature (Bonetti and Hopkins 2009; Saunders et al. 2009), we hypothesised that the improvements in swimming performance of elite swimmers in the weeks following Classic and LHTL altitude training would be equal. Both short-term (\4 weeks) and season-long (*20 weeks) changes in performance were examined. A secondary hypothesis was that the increase in tHb would be greater following 21 days of Classic altitude training compared to LHTL altitude training. We also

123

Eur J Appl Physiol (2012) 112:32753285

3277

assessed the strength of the relationship between changes in swimming performance and changes in tHb.

Methods Experimental design The study comprised two inter-related components. The rst was a parallel groups trial comparing changes in swimming performance and tHb for 28 days after Classic and LHTL altitude training. The swimming race results of a group of elite swimmers completing sea-level training during this same period (Race Control) were sourced from the internet to act as a quasi-control group. The second component was a comparison of season-long progression in race performances between swimmers who did or did not complete mid-season altitude training. A key outcome for both components was the swimming performances of internationally competitive swimmers during the 2009 season. The season-long comparison monitored changes in performance from the Australian National Championships in March to the Federation
Table 1 Subject characteristics Group Parallel groups trial LHTL Classic Race Control Altitude Non-altitude 1 14 5 9 9 8 3 6 5 12 9 17 11 14 21 Male (n) Female (n) Total (n)

International de Natation (FINA) World Championships in August. Characteristics of the participants in both components of the study are described in Table 1. This study was approved by the Ethics Committees of the Australian Institute of Sport and University of Canberra (approval numbers: 20090205 and 09-08) and all swimmers in the parallel groups trial provided written informed consent. Subjects and methodology Parallel groups trial The focuses of the parallel groups trial were changes in swimming performance and tHb in the 28 days following a three week period of altitude exposure (Fig. 1). Twenty-six swimmers were assigned by coaching group into one of two altitude-training groups for a 3-week training camp in May 2009. The LHTL group spent 14 h day-1 at a simulated altitude of 3,000 m in normobaric hypoxia and trained in their normal environment (Canberra, Australia, 600 m). Members of the Classic group travelled to one of two different natural moderate altitude destinations for the

Age (years)

Height (m)

Mass (kg)

IPS (a.u.)

20.8 3.3 21.9 2.3 23.0 1.4 22.4 2.8 22.4 3.7

1.75 0.09 1.84 0.09 1.82 0.01 1.81 0.11 1.79 0.10

65.6 9.0 79.6 10.6 75.3 12.5 75.9 13.5 71.6 12.7

827 61 855 39 894 33 887 34 893 34

Season-long comparison

Group mean SD. IPS FINA International Point Score classication of swimming performance recorded at the National Championships in March 2009

KEY:

= Swim performance

= tHb measurement

Season-long comparison

Parallel groups trial

tHb pre tHb 1 tHb 14

National Championships

Rpre //
-9 -4 -3

Altitude camp
-2 -1 0

R1

R7
1

R14
2 3

R28
4 5 6

World Championships

-10

Weeks

Fig. 1 Schematic timeline of study design

123

3278

Eur J Appl Physiol (2012) 112:32753285

duration of the camp: Sierra Nevada, Spain, 2,320 m (5 swimmers), and Flagstaff, USA, 2,135 m (12 swimmers). 17 of the 26 swimmers were part of the Australian national team preparing for the FINA World championships in August, with a further 7 swimmers part of the Australian youth or Australian universities representative teams. The remaining two swimmers were state representatives and former national or youth national team members. To assess the equality of the groups for swimming ability the international points score system (IPS) of the FINA, swimmings international governing body, was used to compare the performances of the groups based on their swimming performance at the National Championships in March 2009. The IPS determines a point score (range 01,100) for each swim performance scaled up or down from 1,000 points, based on the average of the top ten of all time world rankings for that event (Federation Internationale de Natation 2009). Swimming performance Swimming performance was recorded 7 days before altitude (Rpre), then 1, 7, 14 and 28 days (R128) after the end of the altitude exposure (Fig. 1). This research project was arranged around international swimming competitions and the race times of swimmers in the Classic and LHTL groups were collated as a record of swimming performance. At some time points, and depending upon the location of the swimmers, races were not available. In those instances, swimmers completed electronically timed race simulations in an international standard 50 m pool after having rst undertaken their full race-preparation warm-up. The performance of each swimmer was tracked in a designated swimming event over the study period (100 or 200 m, freestyle or form stroke). Swimmers in the Classic and LHTL groups were asked to maintain consistent race preparation (e.g. warm-up and use of nutritional ergogenic aids, such as caffeine and

sodium bicarbonate) and to wear the same swimsuit for all races. In general, the adherence to these requests was excellent with the exception of three swimmers in the Classic group wearing a different swimsuit at R28 because they were trialling a different suit before the World Championships. If swimmers swam the same event multiple times at a race meet (e.g. heats and nals) their best time was used. An internationally accepted correction factor of 0.7 s was added to equate a relay leg (ying start) with an individual swim race (stationary dive start). This adjustment was used for three swimmers in the Classic group at Rpre only. Some swimmers, typically due to illness or injury, were unable to race at selected time points; rates of participation are shown in Table 2. Competition performance data for the Race Control group were collected from ofcial race records, allowing a comparison to be made between the effects of sea-level and altitude training on swimming performance. To select this group, all members of the Australian World Championships swimming team who did not participate in altitude training between May and August 2009 were initially included (n = 24). Then, the selection was narrowed to swimmers who competed in international race meets at Rpre, R1 and R7 (n = 11 remaining). Attempts to narrow the group further by including only swimmers who also competed in races at R14 and R28 caused a major decline in group size; hence, there are no Race Control data for these time points. The swimmers in the Race Control group did not participate in this research study, per se, rather their race results were accessible via the internet. Consequently, their pre-race preparation at Rpre, R1 and R7 was not controlled. Blood testing All swimmers in the Classic and LHTL groups were measured for tHb, using an optimised carbon monoxide (CO) re-breathing technique (Prommer and Schmidt 2007),

Table 2 Percent changes in swimming performance from pre-training to 1, 7, 14 and 28 days after 3 weeks of Classic altitude, LHTL altitude or sea-level (Race Control) training Group Days post-altitude 1 Classic LHTL Race Control Mean change, % 90% CL (n) Qualitative descriptor of change Mean change, % 90% CL (n) Qualitative descriptor of change Mean change, % 90% CL (n) Qualitative descriptor of change 0.4 0.4 (17) Possibly slower 0.7 1.1 (8) Unclear -0.9 1.3 (9) Likely faster 7 -0.2 0.7 (17) Unclear 0.8 0.9 (7) Likely slower -1.1 0.8 (11) Very likely faster 14 -0.3 0.8 (12) Unclear 0.3 1.1 (9) Unclear 28 0.2 0.9 (14) Unclear 0.1 1.0 (9) Unclear

Note that positive and negative values indicate slower and faster swimming performance, respectively Change in swimming performance is shown compared to Rpre and is described qualitatively in terms of the likelihood of change relative to the smallest worthwhile change of 0.4%

123

Eur J Appl Physiol (2012) 112:32753285

3279

at the start and the end of the training camp, and 2 weeks after the end of the camp (Fig. 1). Briey, subjects re-breathed a bolus of CO equivalent to 1.2 mL kg-1 of body mass through a glass spirometer (BloodTec, Germany) for 2 min. Percent carboxyhaemoglobin (%HbCO) in ngertip capillary blood was measured using an OSM3 hemoximeter (Radiometer, Copenhagen, Denmark) before and 7 min after administration of the CO dose. Ten repeat measures of %HbCO were made for improved precision in tHb estimation (Alexander et al. 2011). Duplicate baseline measures of tHb were made within 1 week of the start of the altitude period and averaged for baseline tHb (tHbpre). 1 and 14 days after swimmers returned to sea-level, tHb was also measured (tHb1 and tHb14). Measurements at tHbpre and tHb14 were made in the same laboratory in Canberra, Australia, for all swimmers, but tHb1 measures for the Classic group were made in different locations close to their training camp venue. To assess the effect of having used two different OSM3 hemoximeters for the Classic groups tHb values, quality control blood samples were measured for %HbCO on both hemoximeters (Gough et al. 2011). No differences were found, thus no adjustments of the data at tHb1 were made. A venous blood sample (4 mL) was collected from all swimmers in the Classic and LHTL groups 1 week prior to the start of altitude exposure to check iron status. Serum ferritin concentration was determined by immunoturbidimetric assay on a Hitachii 911 automatic analyser (Boehringer Mannheim, Germany). All subjects in the Classic and LHTL groups, except two whose serum ferritin levels exceeded the normal range, were provided with a daily oral iron supplement (Ferrograd C: containing 105 mg elemental iron, Abbott Australia, Botany, Australia) for the duration of the 3-week altitude camp to ensure adequate iron stores for accelerated erythropoeisis (Stray-Gundersen et al. 1992). Training load The training completed by swimmers in the Classic and LHTL groups throughout the 3-week altitude camp was quantied by sports physiologists accompanying each of the groups. Calculation of training intensity, based on blood lactate response (Mujika et al. 1996), permitted estimation of a training impulse (TRIMP) for each session, which can be interpreted as the integrated training load (Banister et al. 1975). Particular attention was paid to ensuring accurate notation of training intensity in the Classic group using regular blood lactate sampling since the hypoxic environment would have increased the physiological strain at any given swimming speed. The mean session intensity (TRIMP km-1) was calculated by dividing the session TRIMP by the session volume. Details of

training volume (km) were also collected from the coaches for the 28 days following the camp. No training data for the Race Control group were available but we are condent that all swimmers in this group would have been committed to training during this period since they were preparing for the World Championships. Season-long comparison The performances of swimmers at the FINA World Championships in August 2009 were compared to their best time recorded at the Australian National Championships in the preceding March. Differences in performance were compared between swimmers who completed a midseason 3-week altitude sojourn (Altitude, n = 14) and those who remained at or near sea-level (0500 m) (Nonaltitude, n = 21). The Altitude group was a combination of swimmers who completed Classic or LHTL altitude training as part of the parallel groups trial aspect of this study. For swimmers who swam multiple events at both competitions, the event for which they scored the highest IPS points at the World Championships was selected for comparison in this study. Statistical analysis We employed a contemporary analytical approach involving magnitude-based inferences (Hopkins et al. 2009) to detect small effects of practical importance in an elite athlete group. All data were log-transformed to account for non-uniformity of error. The percent changes in the mean swimming performance and the mean tHb response from pre-altitude to each time point after altitude were calculated. The differences between the groups were assessed with an independent t test for unequal variance (Hopkins 2006). Short-term changes in swimming performance were assessed by comparing each of R1, R7, R14 and R28 to Rpre, and season-long changes assessed by comparing performances at the World Championships to the National Championships. The magnitude of changes were assessed in relation to the smallest worthwhile change (SWC) for swimming performance, estimated as one-half of the between-subject standard deviation in swimming race time, 0.4% (Hopkins et al. 2009; Pyne et al. 2004). Preliminary analyses revealed substantial differences between the ability of swimmers in the Classic (855 39 IPS points; mean SD), LHTL (827 61) and Race Control (894 33) groups. Consequently, IPS points were used as a covariate in the analysis of short-term changes in swimming performance. For the season-long comparison, the altitude and non-altitude groups were evenly matched for swimming ability (IPS scores, 892 33 and 887 34), thus, no covariate was used in this analysis. Changes in tHb following altitude

123

3280

Eur J Appl Physiol (2012) 112:32753285

exposure were assessed by comparing tHb1 and tHb14 to tHbpre, using a SWC of 2.0%. We estimated this SWC as 0.2 9 the between-subject standard deviation in tHb in this sample group of swimmers (Hopkins 2004). This estimate compares favourably with similar estimates ranging from 1 to 4% derived from data reported in earlier studies of tHb in elite swimmers (Friedmann et al. 2005; Heinicke et al. 2001; Robertson et al. 2010a). Between-group differences in training volume (km), intensity (TRIMP km-1) and load (TRIMP) were assessed for the rst, second and third weeks of altitude separately and also cumulatively. The SWC for the training variables was calculated as a standardised small effect: one-fth of the between-athlete SD (Cohen 1988). The observed effects were reported as the mean change or difference 90% condence limits (CL). Effects were termed positive, trivial, or negative depending on the magnitude of the change relative to the SWC and were assigned a qualitative descriptor according to the likelihood of the change exceeding the SWC as follows: 5074% possible, 7594% likely, 9599% very likely. Those effects where the 90% condence interval overlapped simultaneously both the substantially positive and negative thresholds were deemed unclear. The strength of the association between changes in tHb and race performance was explored by calculating the Pearsons correlation coefcients between tHb1 and R1, tHb1 and R7, tHb14 and R14, and tHb14 and R28.

decrement for the week following both Classic and LHTL was demonstrated (Table 3). At the World Championships, the altitude groups performance was likely faster than at the National Championships (mean percent change 90% CL; -0.8 0.6%) and the non-altitude groups performance was very likely faster (-1.1 0.6%). The 0.3 0.8% difference between the groups was trivial but unclear. tHb and haematology The mean (SD) tHb in the Classic and LHTL groups before the training camp were 985 (226) g (males 13.0 0.7 g kg-1, females 10.5 0.5 g kg-1) and 705 (132) g (sole male 13.3 g kg-1, females 10.6 0.9 g kg-1), respectively. 1 day after the training camp, the mean (90% CL) tHb was 3.8 1.3% and 4.0 1.1% higher (very likely) than tHbpre in the Classic and LHTL groups, respectively (Fig. 2). Fourteen days after the end of the training camp, tHb was reduced slightly in both Classic and LHTL groups but remained likely higher than tHbpre. The difference between the groups in tHb response to altitude was likely trivial at both tHb1 and tHb14. Changes in tHb were associated weakly with changes in swimming performance, with all correlation coefcients \0.2. All but three swimmers had a serum ferritin concentration within the normal clinical range: group means (SD) of 71 56 lg L-1 in the LHTL group and 175 88 lg L-1 in the Classic group. Two swimmers, one in the Classic and one in the LHTL group, had a serum ferritin concentration above the normal range, 195 lg L-1 for a female swimmer and 355 lg L-1 for a male swimmer, and consequently were not given a daily oral iron supplement. One female swimmer in the LHTL group (16 lg L-1) was below the normal range for serum ferritin concentration, although she was asymptomatic for iron deciency. Training Throughout the 3-week altitude training camp and in the 4 weeks following, the weekly training volume of the

Results Race performance Changes in swimming performance in the 28 days following Classic and LHTL altitude training were mostly unclear compared to Rpre (Table 2). However, slower performances were apparent 1 and 7 days after Classic and LHTL altitude training, respectively, and when compared to the Race Control group, whose swimming performances were faster than Rpre at R1 and R7, a performance

Table 3 Percent difference in the change in swimming performance from pre-training to 1 and 7 days after Classic altitude or LHTL altitude compared to Race Control Comparison Days post-altitude 1 ClassicRace Control LHTLRace Control Mean difference, (% 90% CL) Qualitative descriptor of change Mean change, (% 90% CL) Qualitative descriptor of change 1.4 1.3 Likely slower 1.6 1.6 Likely slower 7 0.9 1.0 Likely slower 1.9 1.1 Very likely slower

Note that positive and negative values indicate slower and faster swimming performance, respectively

123

Eur J Appl Physiol (2012) 112:32753285

10

3281

Classic

80

LHTL Classic

Change in tHb (%)

Volume (km)

60

40

20

-5

Post 1

Post 2

Post 3
LHTL

-10 10

Altitude

LHTL

2.5

Classic

Intensity (TRIMP.km -1)

2.0 1.5 1.0 0.5 0.0

Change in tHb (%)

-5

Alt 1

Alt 2

-10 -28 -21

Altitude

-14

-7

14

160
LHTL

Days from end of altitude exposure

140

Load (TRIMP)

Fig. 2 Percent changes in total haemoglobin mass (tHb) from pretraining to 1 and 14 days after 3 weeks of Classic altitude or LHTL altitude. Solid circles and error bars show group mean response SD slightly off-set for improved clarity, open circles show individual responses

120 100 80 60 40 20 0

Classic group (52 10 km; mean SD) was 27% higher than the LHTL group (41 8 km): a large difference (Fig. 3). During the camp, the intensity of swimming during training was 20% higher in LHTL (1.8 0.1 TRIMP km-1) than in Classic (1.5 0.1 TRIMP km-1): also a large difference. When combined, the difference between the cumulative training load of the Classic (242 37 TRIMP) and LHTL (232 11 TRIMP) was small (4%).

Alt 1

Alt 2

Weeks
Fig. 3 Training volume, intensity and load during three weeks (Alt 1, Alt 2, Alt 3) of Classic and LHTL training. Training volume is also shown for the 4 weeks following altitude (Post 14). Columns are group mean SD

Discussion The current study directly compared Classic and LHTL altitude training in elite athletes for the rst time in the same study, enabling their relative utility to be assessed. The main nding was impaired performance in the rst week after altitude training and, despite substantial

123

Alt 3

Alt 3
Classic

Post 4

Alt 1

Alt 2

Alt 3

3282

Eur J Appl Physiol (2012) 112:32753285

increases in tHb, there were no clear improvements in swimming performance for up to 28 days after either Classic or LHTL altitude training. Although 3 weeks of both Classic and LHTL altitude training induced a substantial erythropoiesis, this physiological adaptation did not appear to transfer directly into improved competitive performance in the elite swimmers. Comparison of Classic and LHTL Although LHTL has often been vaunted (Levine 2002; Rusko et al. 2004) as an improvement upon the Classic method of altitude training because it affords maintenance of sea-level training intensity, the results of the present study did not identify either method as preferable to the other. Based on two meta-analytical analyses (Bonetti and Hopkins 2009; Saunders et al. 2009), we hypothesised that swimming performances after Classic and LHTL altitude training would be improved, and by an equivalent magnitude (*2%). In the present study, Classic and LHTL did engender very similar patterns of performance change, but they were detrimental rather than benecial. Performances were slower in the rst week after altitude, 1 day after Classic and 7 days after LHTL, compared to baseline, followed by small (\0.4%) but unclear variations up to 4 weeks after the return to sea-level. Comparisons to improved swimming performance in the Race Control group during the same period led to even more marked decrements in performance in the rst week following Classic and LHTL. These ndings are in direct contrast to those of the only other study that has compared Classic and LHTL altitude training directly; although, in their case, in sub-elite athletes (Levine and Stray-Gundersen 1997). These authors demonstrated running performance that was 1.3% better immediately after LHTL, but was subsequently not different from baseline in the following 3 weeks. In the same study, performance was not different from baseline in the Classic group at any time point. These ndings also oppose other researchers who have reported performance improvements immediately after Classic (Friedmann et al. 2005) and LHTL (Robertson et al. 2010b) altitude training, although there are examples of others who, like us, found no sport performance improvement *2 weeks after the cessation of LHTL (Robach et al. 2006). The equivalent *4% increases in tHb immediately following Classic and LHTL demonstrate that both forms of altitude training were equally effective in stimulating erythropoeisis. Altitude-induced changes in tHb have been reported to be as large as 9% for Classic (Heinicke et al. 2005) and 7% for LHTL (Robach et al. 2006); although, in agreement with the ndings of the present study increases of 24% have been reported more frequently (Clark et al.

2009; Garvican et al. 2008; Robertson et al. 2010b). The only previous research to directly compare Classic and LHTL altitude training for changes in red cell mass (measured via Evans Blue dye) reported a greater increase in Classic over LHTL (10.4 vs. 5.3%), a difference that was later supported by Saunders et al. (2009) who concluded that Classic altitude training can be three-times more effective than LHTL in increasing tHb. The results of the present study do not support the ndings of these previous researchers, but instead indicate an equivalent erythropoeitic response to Classic and LHTL. Although there were no control data for tHb collected as part of this study, it has consistently been shown that tHb does not change substantially during sea-level training in well-trained athletes (Garvican et al. 2008; Levine and Stray-Gundersen 1997; Robach et al. 2006), unless training volume is increased substantially for 6 weeks or more (Garvican et al. 2010). Therefore, we are condent that the tHb increases observed here, over a 3-week period, are due to the hypoxic stimuli of Classic and LHTL altitude rather than training alone. Time course of performance changes While there are few empirical studies that have reported the time-course of performance changes following altitude training, numerous qualitative reports based on coaches experiences conclude that good performances tend to occur within the rst few days of returning to sea-level, then again 23 weeks later (Dick 1992; Fuchs and Reiss 1990; Jung and Shon 1994). The ndings of the present study do not support this consensus of opinion. Instead, performance was slower for up to 1 week after altitude and no period of peak form was identied. There are, however, two other research studies that have also reported unchanged sports performance in the period of 1421 days after Classic (Levine and Stray-Gundersen 1997) and LHTL (Levine and Stray-Gundersen 1997; Robertson et al. 2010b). However, the similarities end there, since both of the aforementioned studies reported *1.5% improvements in running performance immediately after LHTL. It is evidently a complex task to interpret these results collectively since no clear pattern emerges. However, information about the time course of changes in an athletes physiology following altitude may offer insight into the likely timing of performance peaks. One factor that has received a lot of attention, and on which we may comment directly, is the rate at which altitude-induced gains in tHb decline after the return to sea-level. When measured 14 days after the cessation of Classic and LHTL altitude, tHb had decreased slightly compared to its levels immediately after altitude, although it remained substantially higher than baseline. Previous researchers have demonstrated mixed results about the longevity of increased tHb

123

Eur J Appl Physiol (2012) 112:32753285

3283

following LHTL altitude, with some reporting a decline within 12 weeks of a return to sea-level (Brugniaux et al. 2006) while others found no such decline (Clark et al. 2009). In fact, a similarly equivocal response to Classic altitude training has been demonstrated (Garvican et al. 2008; Heinicke et al. 2005). While the poor correlations between changes in tHb and swimming performance in the present study appear to oppose haematological changes as the primary mediator of performance, small but worthwhile changes in swimming performance should theoretically result from tHb increases of the magnitude (*4%) shown here. A recent comprehensive review conrmed a very large positive relationship between increases in tHb and increases in VO2max to the magnitude of 1 g Hb:3.5 mL of O2 (Schmidt and Prommer 2010). Based on the calculated aerobic contribution to energy production during competitive 100 and 200 m swimming races, the 3.8% increase in tHb we observed could elicit a 0.30.7% improvement in race time (Toussaint and Hollander 1994) [assuming a VO2max of 57 mL kg-1 min-1 (Robach et al. 2006), a mean tHb of 13 g kg-1 and a body mass of 82 kg]. Although improvements of this magnitude are equivalent to the smallest worthwhile change for swimming performance (Hopkins et al. 2009; Pyne et al. 2004), detecting such changes can be difcult due to the variability associated with racing and the modest sample sizes available when targeting an elite athlete population. The role of tHb in mediating performance after altitude exposure has been discussed at length (Gore and Hopkins 2005; Levine and Stray-Gundersen 2005) and whilst tHb, VO2max and endurance performance are theoretically linked (Schmidt and Prommer 2010), previous research has shown that sports performance can remain above baseline levels 15 days after the end of LHTL altitude training when tHb levels have returned to pre-altitude levels (Brugniaux et al. 2006). The time course of changes to non-haematological mechanisms that may also mediate changes in performance following altitude exposure have been documented: hypoxic ventilatory response persisting for over 1 month after continuous (Forster et al. 1971) or intermittent hypoxia (Katayama et al. 2005), and improved submaximal VO2 prole (Schmitt et al. 2006) and reduced submaximal heart rate (Brugniaux et al. 2006) still apparent 15 days after LHTL. Since these, and other (Gore et al. 2007), nonhaematological mechanisms have been demonstrated following altitude it is likely that the time course of performance changes are inuenced by a combination of these rather than one single factor alone. The performances of swimmers in the present study did not demonstrate a benecial effect of altitude training despite the fact that substantial erythropoeisis was observed. In line with previous research, it is likely that the swimmers also experienced non-haematological

adaptations during this period. One contributory factor towards the muted performances could be that training was not tapered in preparation for performance, as it would normally be prior to a major competition. Throughout the Classic and LHTL altitude camps, training loads were high and, as demonstrated by data collected for 28 days following the return to sea-level, remained so throughout this period. Our capacity to evaluate the effects of differences in training on race performance is limited by the lack of any training detail for the Race Control group. However, the effect of the taper leading into competition can be substantial (Mujika et al. 2002) and if altitude training is completed in the weeks before major competition, performance is likely to be a complex interplay between the negative effects of fatigue and the positive physiological adaptations from altitude. Season-long comparison The season-long aspect of the present study, which compared two tapered performances and thereby minimised the inuence of fatigue on the results, revealed that despite the positive haematological adaptation experienced by the altitude group, there was no clear advantage for those swimmers who completed altitude training. These outcomes call into question the utility of altitude training for competition preparation; however, there are many other elements of a swimmers preparation for competition contributing to the overall performance (e.g. technique, polyurethane swimsuits, nutritional ergogenic aids, taper, etc.), which may confound any small benecial effects elicited by altitude training. These multiple inuences make it difcult to determine whether altitude-induced improvements have taken place in this study, and in a wider context, whether it is worthwhile for athletes to complete altitude training if any benets are likely to be confounded by other factors. Limitations The choice to situate this research in a real-world scenario and observe the training practices of elite swimmers preparing for international competition meant that coaching squads were kept together rather than swimmers being allocated to groups randomly. Consequently, training volume and intensities differed between the Classic and LHTL groups, reecting the styles and philosophies of each individual coach (i.e. the predominantly male Classic groups usual training routine consisted of a higher training volume than the LHTL group). In addition, the training of the Race Control group was not quantied and hence, cannot be compared directly to that of the Classic and LHTL groups. Training was prescribed by very

123

3284

Eur J Appl Physiol (2012) 112:32753285 Australian Institute of Sport Swimming program, and the Federal Institute of Sport Science, Germany. Conict of interest The authors declare that they have no conicts of interest.

experienced coaches of Olympic and World championship gold medallists and we consider that their experience in adjusting training loads according to swimmers experience, age and distance speciality enhanced the ecological validity of this investigation of elite athletes training practices. Furthermore, studies of training volume and intensity in swimmers (Costill et al. 1991; Faude et al. 2008) have shown that changes in these variables are perhaps not as critical as often suggested, and there has not yet been any investigations published, to our knowledge, examining whether one method of training is more effective than another during altitude training. Nevertheless, we cannot discount the possibility that differences in training load between the Classic, LHTL, and Race Control groups inuenced swimming performance in the present study. The swimming performances of the Race Control group served as a useful comparison to those of the altitude training groups, but it must be reiterated that the race preparation of athletes in that group was not controlled. Therefore, it is possible that their race results were inuenced by between-race differences in preparation. The allocation of the groups according to coaching squad also resulted in some inequalities in ability of swimmers between Classic, LHTL, and Race Control groups. However, the differences were relatively minor since most athletes were Australian senior or junior national team members and the Race Control group were the top tier of elite swimmers. This difference in swimming ability was accounted for using covariate analyses.

References
Alexander A, Garvican LA, Burge CM, Clark SA, Plowman JS, Gore CJ (2011) Standardising analysis of carbon monoxide rebreathing for application in anti-doping. J Sci Med Sport 14:100105 Banister EW, Calvert TW, Savage MV, Bach TM (1975) A systems model of training for athletic performance. Aust J Sports Med 7:5761 Bonetti DL, Hopkins WG (2009) Sea-level exercise performance following adaptation to hypoxia: a meta-analysis. Sports Med 39:107127 Brugniaux JV, Schmitt L, Robach P, Nicolet G, Fouillot JP, Moutereau S, Lasne F, Pialoux V, Saas P, Chorvot MC, Cornolo J, Olsen NV, Richalet JP (2006) Eighteen days of living high, training low stimulate erythropoiesis and enhance aerobic performance in elite middle-distance runners. J Appl Physiol 100:203211 Clark SA, Quod MJ, Clark MA, Martin DT, Saunders PU, Gore CJ (2009) Time course of haemoglobin mass during 21 days live high:train low simulated altitude. Eur J Appl Physiol 106:399406 Cohen J (1988) Statistical power analyses for the behavioural sciences. Lawrence Erlbaum Associates, Hillsdale Costill DL, Thomas R, Robergs RA, Pascoe D, Lambert C, Barr S, Fink WJ (1991) Adaptations to swimming training: inuence of training volume. Med Sci Sports Exerc 23:371377 Dick FW (1992) Training at altitude in practice. Int J Sports Med 13(Suppl 1):S203S206 Faude O, Meyer T, Scharhag J, Weins F, Urhausen A, Kindermann W (2008) Volume vs. intensity in the training of competitive swimmers. Int J Sports Med 29:906912 FINA Points System (2009) http://www.na.org/pool/index.php?option =com_content&view=article&id=7236&Itemid=297. Accessed on 20th April 2011 Forster HV, Dempsey JA, Birnbaum ML, Reddan WG, Thoden J, Grover RF, Rankin J (1971) Effect of chronic exposure to hypoxia on ventilatory response to CO2 and hypoxia. J Appl Physiol 31:586592 Friedmann B, Frese F, Menold E, Kauper F, Jost J, Bartsch P (2005) Individual variation in the erythropoietic response to altitude training in elite junior swimmers. Br J Sports Med 39:148153 hentraining: das Erfolgskonzept der Fuchs U, Reiss M (1990) Ho Ausdauersportarten. Trainerbibliothek 27:128 Garvican LA, Martin DT, Clark MA, Quod M, Stephens B, Prommer N, SW F, Impellizeri FM, Rampinini E, Sassi A, Gore CJ (2008) The time course of the erythropoetic response to natural altitude training in elite endurance cyclists. Med Sci Sports 40:S52 Garvican LA, Martin DT, McDonald W, Gore CJ (2010) Seasonal variation of haemoglobin mass in internationally competitive female road cyclists. Eur J Appl Physiol 109:221231 Gore CJ, Clark SA, Saunders PU (2007) Nonhematological mechanisms of improved sea-level performance after hypoxic exposure. Med Sci Sports Exerc 39:16001609 Gore CJ, Hopkins WG (2005) Counterpoint: positive effects of intermittent hypoxia (live high:train low) on exercise performance are not mediated primarily by augmented red cell volume. J Appl Physiol 99:20552057

Conclusion Swimming performance was substantially impaired for up to 7 days following 3 weeks of either Classic or LHTL altitude training. Despite *4% increases in tHb resulting from both Classic and LHTL altitude training, there were no clear benecial performance effects in the 28 days following altitude. The failure to transfer positive haematological adaptations to improved performance may relate to the mid-season nature of the altitude training and the lack of a taper for performance tests. A seasonlong comparison of two tapered performances at major championships also did not reveal a benet for athletes who completed mid-season altitude training despite the substantial physiological changes associated with the altitude.
Acknowledgments The authors wish to thank all the swimmers and coaches who took part in this project. We are grateful for the contributions of Prof. Walter Schmidt and Christian Voelzke, from the University of Beyreuth, and Dr Chris Baldi, from the University of Northern Arizona. This project was collectively funded by the Australian Institute of Sport general and collaborative research fund, the

123

Eur J Appl Physiol (2012) 112:32753285 Gough C, Sharpe K, Ashenden M, Anson JM, Saunders PU, Garvican LA, Bonetti DL, Gore CJ, Prommer N (2011) Quality control technique to reduce the variability of longitudinal measurement of hemoglobin mass. Scand J Med Sci Sports 21(6):e365e371 Heinicke K, Heinicke I, Schmidt W, Wolfarth B (2005) A three-week traditional altitude training increases hemoglobin mass and red cell volume in elite biathlon athletes. Int J Sports Med 26:350355 Heinicke K, Wolfarth B, Winchenbach P, Biermann B, Schmid A, Huber G, Friedmann B, Schmidt W (2001) Blood volume and hemoglobin mass in elite athletes of different disciplines. Int J Sports Med 22:504512 Hopkins WG (2004) How to interpret changes in an athletic performance test. Sportscience 8:17 Hopkins WG (2006) Spreadsheets for analysis of controlled trials, with adjustment for a subject characteristic. Sportscience 10:4650 Hopkins WG, Marshall SW, Batterham AM, Hanin J (2009) Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 41:313 Jung K, Shon R (1994) Altitude traininga summary of the main papers presented at the European Athletics Coaches Associations workshop on altitude training at Belmeken, Bulgaria in May 1994. An edited translation from Die Lehre der Leichtathletik 33 Katayama K, Fujita H, Sato K, Ishida K, Iwasaki K, Miyamura M (2005) Effect of a repeated series of intermittent hypoxic exposures on ventilatory response in humans. High Alt Med Biol 6:5059 Koistinen PO, Rusko H, Irjala K, Rajamaki A, Penttinen K, Sarparanta VP, Karpakka J, Leppaluoto J (2000) EPO, red cells, and serum transferrin receptor in continuous and intermittent hypoxia. Med Sci Sports Exerc 32:800804 Levine BD (2002) Intermittent hypoxic training: fact and fancy. High Alt Med Biol 3:177193 Levine BD, Stray-Gundersen J (1997) Living high-training low: effect of moderate-altitude acclimatization with low-altitude training on performance. J Appl Physiol 83:102112 Levine BD, Stray-Gundersen J (2005) Point: positive effects of intermittent hypoxia (live high:train low) on exercise performance are mediated primarily by augmented red cell volume. J Appl Physiol 99:20532055 Millet GP, Roels B, Schmitt L, Woorons X, Richalet JP (2010) Combining hypoxic methods for peak performance. Sports Med 40:125 Mujika I, Busso T, Lacoste L, Barale F, Geyssant A, Chatard JC (1996) Modeled responses to training and taper in competitive swimmers. Med Sci Sports Exerc 28:251258 Mujika I, Padilla S, Pyne D (2002) Swimming performance changes during the nal 3 weeks of training leading to the Sydney 2000 Olympic Games. Int J Sports Med 23:582587

3285 Prommer N, Schmidt W (2007) Loss of CO from the intravascular bed and its impact on the optimised CO-rebreathing method. Eur J Appl Physiol 100:383391 Pyne D, Trewin C, Hopkins W (2004) Progression and variability of competitive performance of Olympic swimmers. J Sports Sci 22:613620 Robach P, Schmitt L, Brugniaux JV, Roels B, Millet G, Hellard P, Nicolet G, Duvallet A, Fouillot JP, Moutereau S, Lasne F, Pialoux V, Olsen NV, Richalet JP (2006) Living high-training low: effect on erythropoiesis and aerobic performance in highlytrained swimmers. Eur J Appl Physiol 96:423433 Robertson EY, Aughey RJ, Anson JM, Hopkins WG, Pyne DB (2010a) Effects of simulated and real altitude exposure in elite swimmers. J Strength Cond Res 24:487493 Robertson EY, Saunders PU, Pyne DB, Aughey RJ, Anson JM, Gore CJ (2010b) Reproducibility of performance changes to simulated live high/train low altitude. Med Sci Sports Exerc 42:394401 Robertson EY, Saunders PU, Pyne DB, Gore CJ, Anson JM (2010c) Effectiveness of intermittent training in hypoxia combined with live high/train low. Eur J Appl Physiol 110:379387 Rusko HK, Tikkanen HO, Peltonen JE (2004) Altitude and endurance training. J Sports Sci 22:928944 discussion 945 Saunders PU, Pyne DB, Gore CJ (2009) Endurance training at altitude. High Alt Med Biol 10:135148 Schmidt W, Prommer N (2010) Impact of alterations in total hemoglobin mass on VO2max. Exerc Sport Sci Rev 38:6875 Schmitt L, Millet G, Robach P, Nicolet G, Brugniaux JV, Fouillot JP, Richalet JP (2006) Inuence of living high-training low on aerobic performance and economy of work in elite athletes. Eur J Appl Physiol 97:627636 Stray-Gundersen J, Alexander C, Hochstein A, deLemos D, Levine BD (1992) Failure of red cell volume to increase with altitude exposure in iron-decient runners. Med Sci Sports Exerc 24:S90 Toussaint HM, Hollander AP (1994) Energetics of competitive swimming. Implications for training programmes. Sports Med 18:384405 Truijens MJ, Rodriguez FA, Townsend NE, Stray-Gundersen J, Gore CJ, Levine BD (2008) The effect of intermittent hypobaric hypoxic exposure and sea level training on submaximal economy in well-trained swimmers and runners. J Appl Physiol 104:328337 Wehrlin JP, Zuest P, Hallen J, Marti B (2006) Live high-train low for 24 days increases hemoglobin mass and red cell volume in elite endurance athletes. J Appl Physiol 100:19381945 Wilber RL (2007) Application of altitude/hypoxic training by elite athletes. Med Sci Sports Exerc 39:16101624 Wilber RL, Stray-Gundersen J, Levine BD (2007) Effect of hypoxic dose on physiological responses and sea-level performance. Med Sci Sports Exerc 39:15901599

123