vTraining and Testing ~..~~~'am?~~~114. s,i; Q~

-"""""""'~"",",""='"',"'-'---""~~

J;B

A. Urhausen, H. H. W. Gabriel, B. Weiler, W. KindermannInstitute of Sports and Preventive Medicine, University of Saarland, Saarbrucken, Germany

r.

1

A. Urhausen, H. H. W. Gabriel, B. Weiler, W. Kindem1ann, Ergo-metric and Psychological Findings During Overtraining: ALong-Term Follow-Up Study in Endurance Athletes, Int. J.Sports Med., Vol. 19, pp.114-120,1998.

tration of the incremental graded exercise as well as the alteredmood profile turned out to be the most sensitive parametersfor the diagnosis of QT.

i1.!

if,;~rre Key words: Anaerobic threshold. heart rate. lactate. mood

profile. staleness.Accepted after revision: August 25,1997

IntroductionIn the present prospective longitudinal study 17 male en-

durance trained athletes (cyclists and triathletes; age 23.4:t6.7years, V02max 61.2 :t7.5 mi. min-l .kg-l; means:tSD) were in-vestigated both during a state of overtraining syndrome (OT:N = 15), mainly induced by an increase of exercise intensity, as

well as several times in a state of regular physical ability (NS:N = 62). Cycle-ergometric and psychological data were com-

pared for a period of approximately 19 months. On 2 separatedays, each subject performed a maximum incremental gradedexercise, two anaerobic tests (10 sand 30 s) as well as a short-endurance "stress test" with the intensity of 110% of the indi-vidual anaerobic threshold until volitional exhaustion. Themood state was recorded by a psychological questionnaire in-cluding 40 basic items. During OT the submaximallactate con-centrations were slightly decreased. The performance of the10 s- and 30 s-tests was unaffected. In contrast, the duration ofthe "stress test" decreased significantly by approximately 27%during OT compared to the individual NS. The submaximal oxy-gen uptake measured during the incremental graded exercisewas slightly higher during OT as compared to NS, whereas thesubmaximal and maximal respiratory exchange ratio, maximalheart rate and maximal lactate concentrations were decreased.At the 10 th minute of the "stress test", ammonia tended to beincreased during OT (P = 0.048). The parameters of mood state

at rest as well as the subjective rating of perceived exertionduring exercise were significantly impaired during OT. In con-clusion, the results indicate a decreased intramuscular utiliza-tion of carbohydrates with diminished maximal anaerobic lact-acid energy supply during OT. Neither the lactate-performancerelationship during incremental graded exercise nor the anaer-obic alactacid performance showed alterations. The duration ofthe short-endurance "stress test", the maximal lactate concen-

A sustained imbalance between actual load and resistance toexercise training and other stressors leads to an overtrainingsyndrome (OT). It is attended by reduced physical perform-ance and early fatigue in comparison to the usual individual si-tuation. At the same time more or less severe vegetative com-plaints are reported (16.18). OT represents a rather frequentand severe. but reversible problem in the monitoring of topathletes, however. valid diagnostic tools are not available.Many parameters are supposed to be altered during OT (11,14.18,24.30.35.37), but base more on anecdotal and single re-ports, respectively. than on experimentally controlled investi-gations (31).

Primarily the monitoring of performance data and parametersof the mood state are supposed to help in the diagnostic of anOT (11,14,18). Israel (16) distinguished between a sympathico-tonic and parasympathicotonic form of an OT. This was due tothe observation of increased vegetative excitability or ratherdepressive mood state, respectively. The latter, parasympathi-cotonic and probably more chronical form of OT is dominatingin endurance athletes and leads to a relatively bad prognosisbecause it often requires prolonged recovery periods, whilethe vegetative symptoms are rather discrete, which makes anearly diagnosis particularly difficult and important. If the de-creased performance as a cardinal symptom during OT has tobe objectified, sport specific and standardized methods are ne-cessary. These methods have to be ba,sed on knowledge aboutenergy delivering pathways limiting performance during OT.Only few studies deal with endurance athletes and use morethan one standardized test protocol. Incremental graded orsubmaximal constant load protocols and anaerobic tests werepreferred investigating less than 10 overtrained athletes (7,10,12,17,18). Usually at the onset of the study the not-overtrainednormal condition has been characterized using one initial test,which, afterwards, was compared to the results obtained dur-ing and after a following intensive training period lasting a few

Int.J. Sports Med.19 (1998) 114-120It) Georg Thieme Verlag Stuttgart. New York

M 1?9t

,fK

Sports Med. 19 ~ fN~it'~Int.

days or weeks (7.12.17.19.21.23). In endurance athletes intra-individual comparisons between different seasonal points intime by using longer periods of observation during trainingand competition periods are an exception (10)..1

During a follow-up lasting approximately 1.5 years the presentprospectively designed study monitored various ergometricand psychological parameters during training and competitionperiods in a group of endurance athletes. Results during over-training periods were intra-individually compared to results ofnot-overtrained state (NS).

23.4:1:6.7

12.5:t 2.1

61.2:1:7.5

19.t3 monthsMethods I

study:(5 investigations)

NS NS OTNS NS

""

Subjects

Initially 23 male endurance athletes (cyclists. triathletes) par-ticipated in this investigation. Seventeen of these individualscompleted the investigation (for anthropometrical data see Ta-ble 1). Drop-outs were due to long-lasting illness or injuries(independent of aT), new professional arrangements and lackof compliance with the time schedule of the study. Each indi-vidual gave informed written consent prior to the start of thestudy, which was approved by the Faculty of Medicine of theUniversity of Saarland.

/r--investigation: day 1

~-incremental graded ex.-all-out lOs-test-all-out 30s-test

3- 7 days ~day2

~-psychologica) questionnaire-"stress test"-(110% IAT 1 c

c

time-to-exhaust!on)

Fig.l Study design (NS = not-overtrained state, OT = overtraining: in-vestigations #1, 2, 3 or 4; IAT = individual anaerobic threshold).

Study design

Within a time of 19:!: 3 months per subject. each individualwas investigated five times. These investigations were ap-proximately 3 -5 months apart. and each investigation includ-ed a sequence of standardized testings on 2 separate days(Fig.1). In agreement with the individual training and compe-tition program of the athlete. a period opportune for the induc-tion of an aT was chosen, while the detailed procedure of in-duction was not strictly defined. If a subject was suspected tobe overtrained. he was enabled to consult our institute withina few days. All investigations were performed at the same timeof the day after an overnight fast (8.00-11.00 a.m.). Prior to la-boratory testings training sessions were protocolled for 2weeks and most of the training sessions were monitored forheart rates. The day before each testing only regenerativetraining sessions were allowed. The last intensive or longerlasting (> 90 min) training was at least 36 h before the testingday.

Two experienced physicians independently diagnosed OT byexclusion of other reasons, i.e. organic diseases, classicalsymptoms such as decrease of performance (decline in resultsat recent competitions, premature interruption of effort or im-possibility to keep the usual training pace), decreased subjec-tive tolerance of effort and early fatigue together with more orless severe vegetative symptoms (14,16,18,24). At the time ofdiagnosis no subject suffered from infectious disease or dimin-ished iron stores, objectified by clinical examination and rou-tine laboratory parameters.

Ergometry

All exercises were performed on electrically braked cycle er-go meters in an upright position (incremental graded exerciseand "stress test": Jager Inc., Wiirzburg, Germany; anaerobictests: K~per-Dynavit Inc., Kaiserslautern, Germany).The first day of each investigation comprised the following

tests: ciinica! and training history, physical examination in-cluding routine blood analysis, anthropometric measurementsincluding skin caliper (41), resting ECG, incremental gradedspiroergometry with ECG and indirect measurement of bloodpressure, two anaerobic all-out exercise tests for 10 s and 30 s.In addition, on the first day of the first investigation heart vol-ume was measured by combined one- and two-dimensionalechocardiography (modified Simpson rule (8); Vingmed CFM700, Sonotron Inc., Norway). On the second day of each inves-tigation. three to seven days later. present history was takenagain, a standardized psychological questionnaire was filledin and a high-intensive short-endurance exercise ("stresstest") was performed to exhaustion.

Incremental graded exercise started at 100W and the work-load (rpm-independent modus) was increased by 50W every3 min to subjective exhaustion. Each athlete was motivated tocomplete the test scheme maximally. Parameter~ of gas ex-change were measured with an open system for every 30 s(Oxycon, Hellige Inc., Freiburg, Germany). Heart rates weremonitored after 10 min of rest in supine position, during thelast 15 s of each step of exercise, 15 s before end of exerciseand at the end of the 3 rd and 5 th min after finishing exerciseby using outprints of the ECG. Blood pressure measurementswere made by the indirect method of Riva-Rocci after 10 minof rest in supine position as well as during the 3rd min of eachstep of exercise up to 250 W. Before, at the end of each step ofexercise 50).11 blood samples from the hyperemized ear lobewere taken for measurements of lactate concentrations. The

u

Sports Med. 19 (1998)- A. Urhausen, H. H. W. Gabriel, 8. Weiler, W. Kinder

,'~~

I t{'JJ~~ ~

kinetic of lactate during exercise and the recovery phase wasused to determine the individual anaerobic threshold (34).

tions were calculated with an ANOVA for repeated measuresand post-hoc analysis with Newman-Keul tests (level of signif-icance 5 %). Relationships between two variables were calculat-ed with linear regression analysis. (55-software (Stat. Soft,Tulsa, Oklahoma, USA) was used for all statistical calculations.

After a recovery of gOmin and a warm-up (100-150W) of5 min followed by 3 min of rest in upright position the subjectexercised a maximal 10 $ test (rpm-dependent modus). Thirtymin after the 10 s test another all-out test lasting 30 s was per-formed. again after warming up following an identical protocolas prior to 10 s tests. Capillary blood samples were taken formeasurements of lactate concentrations before each test aswell as 5 and 6 times during the first 10 and 12 min followingthe 10 and 30 s tests, respectively.

Results

At 15 of 85 examinations OT was diagnosed. An OT could be ex-cluded 62 times'. Within these 15 cases 6 appeared during thecompetition period, in 10 cases at the 4th individual investiga-tion. Each subject had at least one examination in NS during thecompetition period and also at least one examination in NSafter OT. Nine OT's were induced during autumn,l to 3 duringthe other seasons. OT was experimentally induced in 13 casesby a substantial increase of high-intensive training during ap-proximately 2-3 weeks without the usual regenerative days.by increasing the frequency of competitions or by prolongingthe competition period. OT spontaneously appeared twiceac-companied by occupational stress. The athletes had significant-ly increased the amount of training at an intensity above orwithin the range of the individual anaerobic threshold beforeOT (approximately 4.5 h per week) in comparison to NS (ap-proximately 1.5 h per week). Weekly total training volume be-fore OT was not different from NS (approximately 10 and 9 h. re-spectively). From the athletes completing the study two failedto induce an OT despite repeated intensive training sessions.

l".1

ilThe so-called "stress test" consisted in an endurance exerciseabove the maximal lactate steady state. This time-to-exhaus-tion test is performed with an intensity of 110% of the individ-ual anaerobic threshold. During each test the subject wasmaximally motivated by an experienced co-worker to contin-ue until subjective exhaustion. Previous studies have shownthat the individual anaerobic threshold can be regarded as anindicator for the maximum lactate steady state at enduranceexercise lasting 30 min on a cycle ergometer (39).

:~f~.!

The Sport-Tester PW 3000 (Polar Electro Inc., Finland) servedfor measurements of heart rates before, at the end of 10 minintervals during the "stress test", 15 s before the end of exerciseas well as 5 and 10 min post-exercise. Capillary blood sampleswere taken from both hyperemized ear lobes before, at the endof 10 min intervals during exercise immediately and 5 minafter the end of exercise for measurements of lactate, glucoseand ammonia.

Complaints and psychological parameters

The athletes complained about typical OT symptoms: the feel-ing of heavy muscles in the lower limbs at unusually modestexercise intensities was clearly predominant in all subjects.Thirteen athletes complained about intense daily fatigue andlack of mental concentration. 11 reported about sleeping disor-ders (difficulities to get to sleep and/or to sleep through), 4 adiminished appetite and 3 an increased sweating rate over-night and/or during exercise. These complaints had started13 :!: 4 days before the date of the "stress test" and lasted24:!: 10 days in total despite exclusively regenerative trainingafter the day of the "stress test".

.,~~Laboratory methods

Lactate was measured in whole capillary blood enzymatically(Testomar-Laktat Mono Kit, Behring Inc., Marburg, Germany;(15). Glucose was determined enzymatically (UV-test) andammonia reflectometrically (Ammoniak Checker, HEK Inc.,Hamburg, Germany; [13]).

Psychological parameters

Borg scale values (rating of subjective exertion) were signifi-cantly higher after 10 min of the "stress test" during OT, with-out differences to NS at the end of the exercise (Fig. 2). The fol-lowing factors of the self-condition scale according to Nitsch(27' were altered significantly during OT: mean of all 14 binaryfactors as a measure for the global mood profile, fatigue, recov-ery, strain, sleepiness, satisfaction (Fig. 2), as welI as readinessfor effort, motivation and sociability (P < 0.05, respectively;data not shown). Results of ANOVA showed no significant dif-ferences of the psychological parameters between the differ-ent investigations in NS.

The athletes were asked about vegetative complaints (mainlydisorders of sleep, appetite, capacity to concentrate mentally)on each examination day. The psychological profile wasassessed by using a standardized scale of self-condition ac-cording to Nitsch (27). By using a specialized software the rat-ings (on a scale from 1 to 6) of 40 items are transformed into 14binary factors, which describe the present readiness to act(motivation) and capacity to act (strain) on a scale from 1 to9. In 10 min intervals during the "stress test" the athletes esti-mated their grade of subjective rating of exertion by using theBorg-scale (from 6 to 20; [2]).

\~"

Statistics Anthropometric measurementsand parameters of the cardiorespiratory system

Data are shown as means and standard deviations. Medianswere calculated for each individual in case of not-overtrainednormal state (NS). Statistical comparisons between NS and OTwere made using the Wilcoxon test for matched pairs. The lev-el of significance was set at 5% (P < 0.05) and adjusted by usinga Bonferoni-correction if repeated measurements (rest/exer-cise) were considered. Possible differences betweenNS condi-

Body mass increased by 2.6 kg (not significant) during thestudy. In case of aT body weight. percentage of body fat andfat-free body mass were the same as in NS.

Respiratoryratio

Ventila-tionequivalent

Table 2 Cardiorespiratoryparameters at submaximalexercise intensity (250W)of the incremental gradedexercise during overtrainedand not-overtrained state(means.t SO; N = 15;

.P<O.O25;.. P<O.O1).

--Incremental gradedtest (250W)

Heart rate(min-1)

Ventilationrate(min-1)

Minuteventilation(I. min-l)

Oxygenuptake(ml.min-')

j86.0:1:9.9 0.94:1:0.07

86.8:1:9.8 "0.88:1:0.06

26.8:1: 3.0

26.6:1:3.3

Table 3 Measurements during the maximal short-endurance .stresstest" performed with an intensity of 110% of the individual anaerobicthreshold during overtrained (OT) and not-overtrained (NS) state

(meanS:l:SD;N-15;(*) P=0.048,** P<O.Ol).-

.Stress test" Heart rate Lactate

(min-1) (mmol.Glucose(mmol.l.

Ammonia(ILmol.I-')

Rest NSOT

lOthmin NSOT

: NSOT

50:t7 0.97:t0.12 4.77:t0.21 28:t1849:t8 0.92:!:0.28 4.81 to.36 33:t19

173:t10 5.19:!:0.94 4.74tO.39 58t19173:!:11 5.58:!:1.39 4.80:t0.40 {*)68t21182t9 6.98t1.81 5.94t1.68 82t18

**178:!:11 6.80:!:1.71 6.02:t1.82 79:t21Max

lower during OT in comparison to NS. At the submaximal in-tensity of 250 W of the incremental graded exercise oxygenuptake was slightly, but significantly higher during aT com-pared to NS, but no differences were seen at rest and at themaximal exercise intensity (Table 2). At submaximal (Table 2)and maximal (NS: 1.11 :l:0.06,OT: 1.02 :1:0.06; P< 0.01) exerciseintensities of the incremental graded test the respiratory ex-change ratio was significantly lower during OT. Minute venti-lation, respiratory rate and ventilation equivalent did not showsignificant differences between both conditions. There wereno significant changes of the heart rate measurements be-tween the different investigations in NS.

.Nitsch-Scale Fatigue Recovery Strain Sleepiness Salisractjon Borg-ScaJe

Fig. 2 Parameters of mood profile (Nitsch-Scale = mean of all 14 bin-

ary factors) and subjective rating of perceived exertion (Borg-Scale at10th min of the maximal short-endurance "stress test") during over-trained (OT) and not-overtrained (NS) state (means:t:SD; N = 15;

++:P<O.Ol, +++: P<O.OOl).

[min1]-180

170

160

150

140

130

120

110

100

[mmolf'24-

Incremental graded exercise

++10 NslI-or!20

16

12Performance capacity on ergometer

The mean intensity of the "stress test" was 83% of the VO2maxduring both OT and NS. The individual anaerobic threshold aswell as the power-output at 4 mmol.I-1 lactate were slightlybut significantly higher (about + 10W) during OT compared toNS. Maximal power of the incremental graded exercise andmean power at the anaerobic 10 s- and 30 s-tests were not dif-ferent (Fig.4). In contrast, time to exhaustion of the "stresstest" was significantly lower by 27 % during OT in comparisonto NS (Fig. 4). The results of a regression equation comparingthe performance at the individual anaerobic threshold to thetime to exhaustion of the "stress test" during NS shows that areduction of the time to exhaustion by 6% would have been ex-pected, if the somewhat higher mean calculated individualanaerobic threshold during OT is taken into account. Resultsof ANOVA showed a small, but significant improvement of themaximal ergometrical performance of the incremental gradedtest by approximately 10 W (P < 0.05) in NS during the study,the duration of the "stress test" was unchanged.

8

4

0Lactate max Heart rate max

Flg.3 Maximal lactate concentrations and maximal heart rate duringthe incremental graded exercise during overtrained (aT) and not-over-trained (NS) state (means:t: SD; N ~ 15: + + P < 0.01)

II

...

f1Blood pressure (at rest: NS: systolic 129:!: 8 mmHg/diastolic73 :!:8 mmHg; OT: 132:!: 10/73:!: 8 mmHg; exercise data notshown), ECG at rest and during the incremental graded exer-cise as well as heart rates at rest (NS: 50:!: 7 beats. min -1: OT:49:!: 8) were not affected by OT. In contrast, maximal heartrates at the incremental graded exercise and at the end of the"stress test" were significantly reduced during OT by approxi-mately 5 beats. min -1 (Fig. 3, Table 3).

Substrates

The maximal lactate concentration during the incrementalgraded exercise test was significantly decreased in OT (Fig. 3).

At the 5th (P < 0.01) and 10th min (P = 0.033. not significant) of.the post-exercise period of the "stress test" heart rate~ ~re

:"'"8'1,",1'. .;~",~ ~~ Sports Med.19 (1998)-

an individual increase of the frequency of highly intensivetraining sessions (exercise intensities above the individualanaerobic threshqld) or competitions. Similar investigationsusing a comparable follow-up period and a comparable nUm-ber of athletes in OT have not yet been published.

;~j

The predominant symptom during OT was the feeling of heavylegs. These muscular complaints appeared at unusually lowexercise intensities during training, but also were observed indaily routine activities as found in prior studies as well (12,26,31,32). The increased subjective strain during the "stress test"fit to this symptom. Athletes also frequently complained aboutchronic fatigue and sleeping disturbances. Disturbances ofsleep and appetite as well as vegetative lability combined withloss of motivation to work and enhanced irritability were al-ready described in the earliest references about OT (29).

i'q

~

During OT, the self-condition scale according to Nitsch (27)especially revealed altered indices of capacity to act and fa-tigue. At the same time the mood state (degree of satisfaction)already declined. The very small overlap of psychological pa-rameters between OT and non-overtrained status in this studyis remarkable. Impaired parameters of global mood and de-pression were also described in swimmers classified as stale(28,30). A differentiation into sympathicotonic or parasympa-thicotonic type of OT, however, was not possible in the presentstudy. This might be explained by the fact, that endurance ath-letes were investigated, described as more susceptible to de-velop a parasymp~thiocotonic type of OT (16.18), while theOT was not induced by an increase of volume but of intensityof training. However, the clinical relevance of this differentia-tion might be questioned.

;J..,i

!~~,,(

ii:" : ";

,

No differences were found for the lactate increase after theanaerobic 10 s- and 30 s-tests (approximately 4 -5 and 8-9 mmol.I-\ respectively). During the "stress test" the concen-trations of lactate and glucose were similar for aT and NS,while ammonia was tendentially increased at the 10th min inaT (Table 3). Results of ANOVA revealed that from the differentinvestigations in NS the highest maximal lactate concentra~tions after the incremental graded exercise were measured atthe first investigation (by approximately 1.5 mmol.I-\P < 0.05). however, the differences were significantly smallerin comparison to the changes in aT.

:~

A reduction of total and active body mass is described as anearly indicator of aT occasionally (11.31.35), but was not ob-served in this and other prospective studies (12.17). In the lit-erature speculations about altered ECG and blood pressure ex-ist (35). Increased resting heart rates were supposed to contri-bute to the diagnosis of aT (9,16,31), but these findings cannotbe confirmed neither by the present study or the data of otherauthors (5.10,12).

Combination of parameters for diagnosis of overtraining

In 11 subjects in aT (out of 15) the mean of the 14 binary fac-tors of the self-condition scale according to Nitsch was meas-ured lower than the lowest value in NS. Eight athletes in aTshowed their lowest exercise time to exhaustion of the "stresstest", 8 subjects their lowest maxima1lactate concentration atthe incremental graded test, respectively. In 7 cases of aT. themaximal heart rate at the incremental graded test was de-creased in comparison to the lowest value in NS. The best pre-dictor was obtained by using a combination of parameters asfollows: maximal lactate concentration and/or maximal heartrate at the incremental graded exercise (criterion 1). time toexhaustion during the "stress test" (criterion 2) and mean ofthe 14 binary factors of the self-condition scale (criterion 3).In this investigation 13 out of 15 aT conditions could be recog-nized if at least 2 of these criteria were significantly lower thanthe individually lowest value during NS; The two spontaneouscases of aT did not show essential differences to the otherstates of aT and also fulfilled two the above-mentioned crite-ria: each presented its lowest value of self-condition scale aswell as the lowest duration of "stress test" or the lowest meas-urement of maximal heart rate during the graded exercise test.

The increased oxygen uptake at submaximal exercise intensi-ties during OT might be interpreted as an altered economy ofneuromuscular coordination or a reduced degree of efficiency(20,33,35). But it should be kept in mind, that the absolute dif-ferences were only small, and heart rate and lactate concentra-tions at submaximal exerdse intensities were not increased inOT. The significant reduction of the respiratory exchange ratiocan be explained by a shift of the energy delivering processesto an enhanced metabolisation of fat subsequeI)t to a declinedsupply of carbohydrates. This argumentation follows earlierreports (6,33). In addition, a tendency towards increased sub-maximal levels of ammonia was found. This effect might beexplained by an impaired mobilization of carbohydrates lead-ing to a deficit of energy supply and increased activation of thepurine nucleotide cycle, maybe with an enchanced recruit-ment of additional (FT) muscle fibres (3,25,36). There are re-ports about central or peripheral effects of increased ammonialevels leading to a reduction of performance (1), but the smalldifferences measured in this study seem to be of minor -if any-clinical relevance. Furthermore it should be considered thatthe 10th min of the "stress test" where the submaximal ammo-

'":1,,.JI

I.I.{1'1

Discussion

In the present prospective and longitudinal study of 1.5 yearsan early stage of OT or overreaching was induced"mainly by

nia concentrations have been measured, reflects a higher per-centage of total exercise time in OT than in NS. Occassionallyincreased ammonia concentrations at rest were described dur-ing periods of largely increased training volume of weight lif-ters (35.40). Others found reduced ammonia levels in periodsofOT(23).

parison to the not-overtrained state the time to exhaustionwas reduced by 27%. Decreased. performances by 0.7-15%were reported before in overloaded athletes (14). To guaranteehigher demands on carbohydrate metabolism the exercise in-tensity for the "stress test" was defined slightly above themaximal lactate steady state (39). A reduced performance atshort maximal endurance exercise tests was also reported inthe early phase of OT (12,40).The present data show a slight shift of the lactate curve to the

right during OT, which has also been described by others (4,17,35). This can be related to the previously discussed shift of theenergy supplying mechanism combined with a reduced utili-zation of carbohydrates. It should be noticed that during OTan unchanged or even increased anaerobic threshold may pre-tend a false high performance capacity. Although a carbohy-drate depletion caused by intense training of2-3 weeks can-not be completely ruled out (7), it should be emphasized thatneither significantly increased ammonia concentrations andheart rate measurements nor elevated concentrations of glyco-lytic hormones (data not shown; publication in preparation)typical for exercising under conditions of reduced muscle gly-cogen stores were measured in the present study.

The combination of 3 criteria (maximal lactate concentrationand/or heart rate of the incremental graded exercise, time toexhaustion of the "stress test", psychological profile) resultedin a relatively high accordance with the "clinical" diagnosis ofOT. These criteria exclusively require intraindividual compara-tive data. It was not possible to define absolute values of pa-rameters to provide a valid differentiation between OT andnormal state. The usefu))ness of the recently proposed relationbetween subjective rating of exertion and lactate concentra-tions during exercise for the diagnosis of OT (32) may be con-firmed.

The interpretation of the present results requires somemethodical restrictions. A "golden standard" to diagnose OTdoes not exist. The underlying "clinical" definition of an OThad to remain a diagnosis by exclu5:ion. but experience fromsports medical practice and recommendations from the litera-ture were taken into consideration (11,18,22). The "stress test"itself represents an open-end test until subjective exhaustionand thus can be influenced by the knowledge of former test re-sults; it may be speculated if a time-trial protocol would bemore appropriate. In addition, training intensities were evalu-ated by using training protocols and monitoring of heart rates,but this can only lead to orientating data. Evidence about thecarbohydrate mobilization may only be achieved by more sen-sitive methods including magnetic resonance technique andmetabolic tracers. It would have been useful to include de-tailed protocols of nutrition in our study, but their interpreta-tion would be problematical without considering the energyoutput of the training and competitions performed. Althoughit was not the aim of the present study to assess data from therecovery period after OT, this interesting phase should be espe-cially considered i.n future studies. Finally. the present resultsare valid for endurance athletes, but cannot necessarily betransposed to speed- and strength-trained athletes.

In contrast to earlier reports (17,18), maxima! oxygen uptake ofthe incremental graded exercise was unchanged in the presentstudy. At first sight this result seems to be unexpected, but itshould be considered that the athletes were in an early phaseof OT, particularly important to prevent a long-term form ofOT. Other follow-up studies investigating relatively short in-duction periods of OT also showed unchanged (5,7,12) orslightly, but not significantly, reduced maximal oxygen up-takes (23).

In overtrained athletes the impairment of the maximal anaero-bic lactacid capacity combined with a reduced tolerance ofacidosis seems to be of primary importance. During maximalanaerobic tests lasting 1-2 min on a treadmill (18), repeatedhigh-intensive bouts of speed-endurance runs (3,300 m; 5)or maximal swims over 365.8 m performance and/or increaseof lactate concentrations were also reported to decline. In thepresent study maximal lactate concentrations and maximalheart rates of the incremental graded exercise were signifi-cantly reduced in aT. Both were described earlier as well (7)and a reduced sensitivity for catecholamines has been put for-ward to explain the effects (24). It seems, however, that the re-duction of the maximal lactate concentrations seems to bemore sensitive to indicate an aT in comparison to maximalheart rates, because the relative differences in comparison toNS are greater (lactate: approximately 18% lower, heart rateonly 3%). The power output of the 10s- and 30 s-testwas unaf-fected during aT. Although the maximal 30 s-test also requiressignificant anaerobic lactacid energy supply, the duration ofexercise is too short to reach a higher lactacidosis. In a priorstudy we had reported about an unchanged anaerobic alacta-cid capacity of middle-distance runners (18). Other workinggroups found an increased performance in 5 x 50 m sprint testsof overloaded judoka (5), or decreased performance over22.9 m in overloaded swimmers (10). These findings could bedifferent in non-endurance athletes such as sprinters.

In summary, it can be concluded that the diagnosis of OT re-quires carefully chosen standardized exercise tests performedto exhaustion as shown by the presented "stress test", Thecomparison of the test results with pre-existing individualdata is indispensable, Obviously the intramuscular mobiliza-tion of carbohydrates is impaired and attended by a reducedmaximal anaerobic lactacid energy supply. The sub maximalrelation of lactate and performance including the calculationsof anaerobic thresholds are not altered during periods of OT.Performance capacity achieved in short exercise tests withoutmaximallactacidosis also remains unchanged. To diagnose anOT it is necessary to consider the altered psychological profileand typical vegetative complaints presented in this study.

L The duration of the "stress test" -an all-out short-term endur-ance exercise test performed at the intensity of 110% of the in-dividual anaerobic threshold -turned out to be the most sen-

~sitive single objectifiable criterion to diagnose an OT. In CtJm-

Aknowledgements

This study was supported by grants from the Bundesinstitutfur Sportwissenschaft. Cologne.

r\

;W)1~ I t:~, ~~ ,~

~Med.

19 (1998)

t:,-..!

I;

j

~~~~

:"

23. Lehmann M, Dickhuth HH, Gendrisch G, Lazar W, Thurn M, Ka-minski R. Aramendi jF, Peterke E, Wieland W, Keul j. Training -Overtraining. A prospective, experimental study with experi-enced middle- and long-distance runners. Int j Sports Med1991; 12: 444- 52 .

24, Lehmann M, Foster C, Keul j, Overtraining in endurance ath-letes: a brief review. Med Sci Sports Exerc 1993; 25: 854-62

25. Meyer RA, Terjung RL AMP deamination and reamination inworking skeletal muscle. Amj Physiol1980; 239: C32-8

26. Morgan WP, Brown DR. RaglinjS, O'Connor Pj, Ellickson MS. EI-lickson KA. Psychological monitoring of overtraining and stale-ness. Br j Sports Med 1987: 21: 107 -14

27, Nitsch JR. Die Eigenzustandsskala (EZ-Skala) -Ein Verfahrenzur hierarchisch-mehrdimensionalen Befindlichkeitsskalier-ung, In: Nitsch'JR, Udris I (eds,), Beanspruchung im Sport. BadHomburg, Limpert 1976: 81 -102

28. O'Connor Pj, Morgan WP, Raglin jS, Barksdale CM. Kalin NH.Mood state and salivary cortisol levels following overtraining infemale swimmers. Psych Neurol Endocrinol1989; 14: 303 -10

29. Parmenter DC. Some medical aspects of the training of collegeathletes, Boston Med Surgj 1923; 189: 45-50

30. Raglin jS. Exercise and mental health, Beneficial and detrimen-tal effects, Sports Med 1990; 9: 323 -9

31. Ryan AJ, Brown RL, Frederick EC, Falsetti HL. Burke ER. Over-training of athletes. Physician Sports Med 1983; 11: 93 -110

32. Snyder AC, jeukendrup AE, Hesselink MKC, Kuipers H, Foster C.A physiological/psychological indicator of over-reaching duringintensive training. Intj Sports Med 1993; 14: 29 -32

33. Starling RD, Pizza FX, Flynn MG, NaurethM, Davenport R, Salba-ny F, Agostl D. Run training vs cross training: Effects of excessivetraining on running economy. Med Sci Sports Exerc 1992; 24:

(Suppl.) S10134. Stegmann H, Kindermann W, Schnabel A. Lactate kinetics and

individual anaerobic threshold.lntj Sports Med 1981; 2: 160-535. Stone MH, Keith RE, Kearney jT. Fleck Sj, Wilson GO, Triplett NT.

Overtraining: A review of the signs, symptoms and possiblecauses.j Appl Sport Sci Res 1991; 5: 35-50

36. Urhausen A, Kindermann W. Blood ammonia concentrationsduring endurance exercise of differing intensities. Eur j ApplPhysiol1992; 65: 209-14

37. Urhausen A, Kindermann W. Biochemical monitoring of train-ing. Clinj Sports Med 1992: 2: 52-61

38. Urhausen A. Gabriel H, Kindermann W. Blood hormones asmarkers of training stress and overtraining. Sports Med 1995;20: 251 -76

39. Urhausen A, Co en B, Weiler B, Kindermann W.lndividual anaer-obic threshold and maximum lactate steady state. Int J SportsMed 1993; 14: 134-9

40. Warren Bj, Stone MH, Kearney jT, Fleck Sj, Johnson RL, WilsonGO, Kraemer Wj. Performance measures, blood lactate and plas-ma ammonia as indicators of overwork in elite junior weight-lifters. Intj Sports Med 1992; 13: 372-6

41. Wormersley j. Durnin j. A comparison of skinfold method withextent of "overweight" and various weight-height relationshipsin the assessment of obesity. Br j Nutr 1977; 38: 271- 84

r

Corresponding Author

Axel Urhausen. M.D. PhDInstitute of Sports and Preventive MedicineUniversity of Saarland66041 Saarbrticken. GermanyFax: (0)681-3024296E-mail: a.urhausen@rz.uni-sb.dehttp: Ilwww.med-rz.uni-sb.de/med_fak/sport-praev/index.html

References

1. Banister EW. Cameron BJC. Exercise-induced hyperammone-mia: Peripheral and central effects. Int J Sports Med 1990; 11:129-42

2. Borg GAV. Perceived exertion: A note on "history" and methods.Med Sci Sports Exerc 1973; 5: 90-3

3. Broberg S, Sahlin K. Hyperammonemia during prolonged exer-cise: an effect of glycogen depletion? J Appl Physiol1988; 65:2475-7

4. Busse MW. Maassen N, Boning D. The work load lactate curve:measure of endurance capacity or criterion of muscle glycogenstorage? I. Glycogen depletion.lntJ Sports Med 1987; 8: 140

5. Callister R. Callister RJ. Fleck SJ. Dudley GA. Physiological andperformance responses to overtraining in elite judo athletes.Med Sci Sport Exerc 1990; 22: 816-24

6. Costill DL, Bowers R, Branam G, Sparks K. Muscle glycogen utili-sation during prolonged exercise on successive days.J Appl Phy-sio11971; 31: 834-8

7. Costill DL. Flynn MG, Kirwan JP, Houmard JA, Mitchell JB. Tho-mas R. Park SH. Effects of repeated days of intensified trainingon muscle glycogen and swimming performance. Med SciSports Exerc 1988; 20: 249-54

8. Dickhuth HH, Urhausen A. Huonker M. Heitkamp H. Kinder-mann W, Simon G, Keul J. Die echokardiographische Herz-groBenbestimmung in der Sportmedizin. Dtsch Z Sportmed1990; 41: 4-12

9. Dressendorfer RH, Wade CE, Scaff JJ. Increased morning heartrate in runners: a valid sign of overtraining? Physician SportsMed 1985; 13: 77-86

10. Flynn MG, Pizza FX. Boone JB, Andres FF, Michaud TA, Rodri-guez-Zayas JR. Indices of training stress during competitive run-ning and swimming seasons.lntJ Sports Med 1994; 15: 21-6

11. Fry RW; Morton AR, Keast D. Overtraining in athletes. SportsMed 1991; 12: 32-65

12. Fry RW. Morton AR. Garcia-Webb P, Crawford GPM, Keast D.Biological respones to overload training in endurance sports.Eur J Appl Physiol1992; 64: 335-44

13. Hageloch W. Weicker H. Methodik der Blutammoniakbestim-mung unter sportmedizinischen Aspekten. Dtsch Z Sportmed1988; 39: 180 -8

14. Hooper SL. Mackinnon LT. Monitoring overtraining in athletes.Recommendations. Sports Med 1995; 20: 321 -7

15. Hohorst HJ. L-( +) Laktat. Bestimmung mit Laktatdehydrogen-ase und DPN. In: Bergmeyer HU (eds.). Methoden der enzyma-tischen Analyse. Weinheim. Chemie 1962: 266- 77

16. Israel S. Die Erscheinungsformen des Obertrainings. Sportmed1958; 9: 207-9

17. Jeukendrup AE. Hesselink MKC. SnyderAC. Kuipers H, Keizer HA.Physiological changes in male competitive cyclists after twoweeks of intensified training.lntJ Sports Med 1992; 13: 534-41

18. Kindermann W. Das Obertraining -Ausdruck einer vegetativenFehlsteuerung. Dtsch Z Sportmed 1986; 37: 238 -45

19. Kirwan JP. Costill DL, Flynn MG. Mitchell JB. Fink WJ. Neufer PD,Houmard JA. Physiological responses to successive days of in-tense training in competitive swimmers. Med Sci Sports Exerc1988; 20: 255 -9

20. Kirwan JP. Costill DL, Houmard JA. Mitchell JB, Flynn MG. FinkWJ. Changes in selected blood measures during repeated daysof intense training and carbohydrate control. Int J Sports Med

1990; 11: 362-621. Kraemer WJ. Fleck SJ. Callister R. Shealy M. Dudley GA. Maresh

CM, Marchitelli L, Cruthirds C. Murray T. Falkel JE. Training re-sponses of plasma beta-endorphin. adrenocorticotropin, andcortisol. Med Sci Sports Exerc 1989: 21: 146 -53

22. Kuipers H. Keizer HA. Overtraining in elite athletes. Sports Med

1988; 6: 79-92