vTraining and Testing.w""-'ii(

iUi~~~,:7~;,~f'~

45

A. L. T. Uusitalol, A. J. Uusitalo2, H. K. Ruskol1 KIHU-Research Institute for Olympic Sports, jyvaskyla, Finland

2 Department of Clinical Physiology, Tampere University Hospital, Tampere. Finland

Uusitalo ALT. Uusitalo AJ. Rusko HK. Heart Rate and Blood Pres-sure Variability During Heavy Training and Overtraining in theFemale Athlete. Int J Sports Med 1999; 20: 45 -53

Introduction

Overtraining state in athletes results from long-term stress orexhaustion due to prolonged imbalance between training andother external and internal stressors and recovery. Stress re-searchers have identified two hormonal stress response types[7,29): the adrenal medullar response with elevated function-ing of the sympathetic nervous system [14] and the pituitaryadrenal cortical response which is connected to the activity ofthe parasympathetic nervous system [32]. Two types of over-training state (the sympathetic and the parasympathetic) havealso been reported to exist [17]. It has also been suggested thatintrinsic sympathetic activity, receptor sensitivity to catechol-amines and adrenal sensitivity to ACTH decrease in the over-training state, resulting in performance incompetence andhigh fatigue ratings [21). On the other hand, endurance train-ing with poSitive training effects (enhanced endurance per-formance) induces an increase in vagal tone in relation to sym-pathetic tone [10) which can be detected by increased heartrate variability [28,34).

Accepted after revision: July 25.1999

~ We investigated heavy training- and overtraining-inducedchanges in heart rate and blood pressure variability during su-pine rest and in response to head-up tilt in female enduranceathletes. Nine young female experimental athletes (ErG) in-creased their training volume at the intensity of 70-90% ofmaximal oxygen uptake (VO2max) by 125% and training volumeat the intensity of < 70% ofVO2max by 100% during 6-9 weeks.The corresponding increases in 6 female control athletes were5% and 10%. The VO2max of the ETG and the control athletesdid not change, but it decreased from 53.0:!: 2.2 ml x kg-1 x min-1to 50.2:!: 2.3 ml x kg-1 x min-1 (mean:!:SEM, p < 0.01) in five

overtrained experimental athletes. In the ErG, low-frequencypower of R-R interval (RRI) variability during supine rest in-creased from 6:!:1 ms2xl02 to 9:!:2ms2xl02 (p<0.05). The30/15 index (= RRlmax 30/RRImin 15' where RRlmax 30 denotes thelongest RRI close to the 30th RRI and RRlmln 15 denotes the short-est RRI close to the 15th RRI after assuming upright position inthe head-up tilt test), decreased as a result of training (analysisof variance, p = 0.05). In the ETG, changes in VO2max were relat-

ed to the changes in total power of RRI variability during stand-ing (r = 0.74, P < 0.05). Heart rate response to prolonged stand-

ing after head-up tilt was either accentuated or attenuated inthe overtrained athletes as compared to the normal trainingstate. We conclude that heavy training could increase cardiac

sympathetic modulation during supine rest and attenuated bi-f.~lasic baroreflex-mediated response appearing just after shift-ing to an upright position. Heavy-training-/overtraining-induceddecrease in maximal aerobic power was related to decre~sedheart rate variability during standing. Physiological responsesto overtraining were individual.

Heart rate and blood pressure variability (HRV and BPV) havebeen presented as indicators of cardiac and vascular auto-nomic modulation [13.25J. Previous studies on the influenceof endurance exercise training on heart rate and blood pres-sure variability in healthy young adults and in young athletesare mostly cross-sectional (e.g. [8.12.18.22J). There are nolongitudinal studies on heart rate and blood pressure variabil-ity either in overtraining state or in female athletes. Morillo etal. [24J proposed that healthy humans with neurally mediatedsyncope symptoms have increased vagal tone and decreasedsympathetic tone during standing after head-up tilt. but therewas no difference in supine heart rate variability when theywere compared to the control group. These position-depen-dent findings could also be possible in endurance athleteswho are susceptible to syncope symptoms during standing.

To study possible changes in autonomic function in responseto heavy training and overtraining we measured heart rateand blood pressure variability in supine and standing positionsand in response to head-up tilt during a 6-9 week heavy en-durance training period in young female athletes. On the basisof the two stress response types reported by stress researchers,we hypothesized that different physiological types to respondto overtraining could exist. These types could also representdifferent stages of stress response.

n Key words: Overtraining, autonomic nervous system, peak

oxygen consumption, head-up tilt test,

I

4~: ~orts Med 2000: 21 A. L. T. Uusitalo et--The study was reviewed and approved by the Ethical Commitee of the University of jyvaskyla.

Methods

SubjectsTrainingFifteen healthy female endurance athletes with no smoking

history gave their informed written consent to participate inthe study (Table 1). They were familiarized with the experi-mental procedure and purpose of the study in one preliminarysession. All athletes had trained regularly for at least one yearbefore the experiment. The athletes did not take any medica-tion, including oral contraceptives. They were in differentmenstrual phases at baseline (eight in the follicular phase, sixin the luteal phase and one in-between [day 15]) and had nomenstrual irregularities.

During the experimental training period, the ETG athletetrained 7 days a week. Their training was supervised by the research staff. The intensity of training was determined based 0:the lactate threshold (T lacJ. The training consisted of intensiv.training (in.tensity ~ T1acJ which included interval running (5-12 km with a 2-minute rest in between) and continuous fasrunning (5 -12 km), and of low-intensity training (intensit~< T1acJ which was mainly long-term running (50 minutes to 3hours), but also cycling, cross-country skiing and swimming.The volume of intensive training was increased by one exercisesession each week, starting with one session and a treadmilltest during the first week. The volume of low-intensity train-ing was planned to increase by 7 -10% each week. The CGathletes were allowed to train according to their own trainingprogramme. In both groups, training was light (low-intensitytraining for less than 1 hour) for two days before each meas-urement. All training sessions were controlled by heart ratemonitors (Polar Electro Sport testerTM, Kempele. Finland) andsubjective feelings were recorded every day.

The athletes were divided by the researchers into two homoge-nous groups: an experimental training group (ETG, four long-distance runners, one cross-country skier, two triathletes, twoorienteers) and a control group (CG, one long-distance runner,three cross-country skiers, one triathlete, one orienteer). Thepurpose of the experimental training period was to overtrainthe ETG athletes. The criteria of overtraining were as follows:decreased maximal oxygen uptake by at least 2 ml x kg-I xmin-\ decreased maximal treadmill performance: unwilling-ness to train and the feeling of inability to go on training incombination with some of the following overtraining signsand symptoms: mood disturbances (decreased positive feel-ings: energetic, helpful, calm, vigorous, relaxed, confident andincreased negative feelings: irritable, depressed, moody, fa-tigued, anxious, confused, excited, desperate, unable to con-centrate), sleeping problems, menstrual irregularities, poorappetite, shaky hands, sweating or other psychosomatic symp-toms, and no illness, injury or other explaining factor for theperformance decrement [11,15]. Five ETG athletes were diag-nosed as being overtrained and they formed a subgroup ofovertrained athletes (OA subgroup).

Heart rate and blood pressure variabilityand the head-up tilt test

The athletes entered a quiet laboratory room with constanttemperature (20-23°C) and dimmed lights at 8.00. 9.00 or10.00 a.m. one hour after a light (300 kcal) bre~kfast. Theywere requested not to drink coffee. tea, chocolate, or coladrinks that morning or the previous evening, or to drink alco-hol during the previous 48 hours. Athletes rested in the supineposition on the tilt table with foot support for 20 minutes be-fore the 5-minute supine recording and the head-up tilt testfollowing that. The head-up tilt test consisted of 25 minutesof rest in the supine position and of 5 minutes of standing at70". Orthostatic intolerance was not studied.

Identical series of measurements were repeated at baseline,after 4 weeks of training (4 wk). when the athletes met theovertraining criteria or were physically and/or mentally ex-hausted after 6-9 weeks of training (END) and after 4-6weeks of recovery training (RE). The measurements consistedof 1) heart rate and blood pressure and heart rate and bloodpressure variability during a 5-minute supine rest and in re-sponse to a head-up tilt test and 2) heart rate and blood lactateconcentration during a submaximal treadmill running test,and a maximal oxygen uptake (VO2max) test on a treadmill.

Three ECG electrodes were attached to the chest and connect-ed to the ECG transducer (M9407. Medikro Oy, Kuopio. Fin-land). Continuous arterial blood pressure was registered fromeither the left or the right middle finger with 2300 NIBP Moni-tor (Ohmeda. Inc., Englewood. CO, USA). A finger cuff waswrapped around the second phalanx of the middle finger. Dur-ing the study the finger with the cuff was kept at heart level. I

Table 1 Physical characteristics of I

n Age

(years)

Ie athlete!

Weight (kg)1 Height (cm) .Fat %1 ..Training (years)

ETG 9 62.7:t6.2 17.1.O4:t4.9 19.9:!:3.2

( (50.6-70.5) (164.0-177.0) (16.4-24.3) I

C 6, 67.5:t5.2 171.9:t4.9 20.4:1:4.6

( (58.9-73.9) (163.7-178.0) (16.0-26.8) I~~. 5 66.7:!:3.7 173.6:t3.0 20.5:1:2.6

( (60.5-70.5) (169.0-177.0) (17.9-24.3) .-

Results are expressed as means:t SD (range). ETG, experimental training group; CG. control group; OA subgroup, overtrained athletes subgroup. "Calculated fromskin folds according to Durnin & Womersley [9]. ""Training years before entering the experiment. 1. there were no significant changes in these parameters duringthe experimental period.

7.9:1:5.3

(1-16)9.2t3.7

[3 -13)7.6 t 5.2

(1-12)

23.9:1:3.0

'.19.0-27.3)23.1:1:

2.7

:20.7-27.5)25.1

:!: 2.3

'22.4- 27.3)

vA

An EXP~~~!~I Overtraining Study lilt J Sports Med 2000; 21 ,47'.,,""

The finapres device was in SERVO OFF mode (continuousmode without set-point adjustments) during the measure-ments, and a controlled computer-generated signal was usedto pace the breathing rate of 0.20 Hz (2.5 seconds for both in-spiration and expiration). The analogue outputs .of the ECGtransducer and the Finapres device were connected to the ana-logue interface (M9401 Transducer Interface, Medikro Oy,Kuopio, Finland) of an IBM PC/AT-compatible microcomputer.The microcomputer was equipped with a software package(CAFfS, Medikro Oy, Kuopio, Finland) for evaluating heart rateand blood pressure variability in time and frequency domains(31]. Both the ECG and the blood pressure signal were ana-logue-to-digital converted (200 Hz, 12 bits) and saved on thehard disk for subsequent off-line analysis. The R-R intervalswere detected with a temporal resolution of better than 2 msand followed by evaluating systolic (SAP) and diastolic (OAP)arterial pressure in each RRI. The following variables were cal-culated in time domain from the 5-minute supine rest and the4-minute standing (first standing minute excluded because ofbiphasic heart rate and blood pressure response just after as-suming the upright position): mean! 1 SO of RRI, HR, SAP andOAr and the square root of the mean of the sum of the squaresof differences between adjacent R-R intervals (RRI RMSSO).The regions of interest were selected by excluding any ectopicbeats and by visual judgment of stationarity.

1.8 km x h-l until 13.5 km x h-l. Thereafter, the inclination wasincreased by 1.4 c to induce a 6 km x kg-l x min-1 increase in

the theoretical oxygen demand (V02demand [4]). In the ski-walking the starting velocity was 6.0 km x h-l and the inclina-tion 2.3 o. Exercise intensity was increased evety 3 minutes toinduce a 6 ml x kg-l x min-1 increase in the V02demand [4) byincreasing both inclination (mean: 1.0', range.: O. 9-1.1 O) andvelocity (mean: 0.3 mts, range: 0- 0.5 mts) during the first 4stages and then only inclination (mean: 1.6", range: 1.5-1.7 ')until physical exhaus~ion. Blood samples were taken from thefingertip immediately after each exercise intensity and fivetimes (immediately and at 1,4,7, and 10 min) after exhaustionto analyze the peak blood lactate concentration. Lung ventila-tion (VE), V02 and CO2 production were measured during theexercise test at 20-seconds intervals using Sensormedics2900z gas analyzer (Sensormedics, Yorba Linda, CA, USA) con-nected on line to a computer and calibrated before and aftereach test. V02max was calculated as the highest mean of threeconsecutive 20-second determinations. The criterion forV02max was that oxygen uptake increased less than half ofthe calculated increase in oxygen demand. If the levelling offcri~erion was not met the additional criteria were maximalblood lactate concentration (did not change significantly inany group) and respiratory exchange ratio (was over 1.0 in ev-ery case). Maximal treadmill performance was calculated asthe oxygen demand of exercise (V02max demand) based onthe velocity and the inclination of the treadmill during the lastminute before exhaustion [4).

The same time series of RRI, SAP and OAP were subjected topower spectral density analysis in frequency domain. Modifiedcovariance autoregressive modeling with a fixed model orderof 18 was used. Total powers, i.e. variances, of RRI, SAP andOAP variability (RRI TP, SAP TP and OAP TP) were generatedafter linear detrending of the signals. The powers of RRI, SAPand OAP variability in the three frequency bands (power invery low-frequency band [VLFP: 0.00 Hz ...0.07 Hz), power inlow-frequency band [LFP: 0.07Hz ...0.15 Hz) and power inhigh-frequency band [HFP: 0.15 Hz ...0.40 Hz) were calculatedby integration and the ratio of the RRI powers in the low-fre-quency band by that in the high-frequency band was calculat-ed and expressed as a percentage (RRI LFP/HFP). In addition,the following variables from the head-up tilt test were calcu-lated to describe the changes between supine and standing po-sition: the ratio between maximal and minimal R-R intervalduring standing, the 30/15 index (= RRlmax 30/RRImin IS whereRmax30 denotes the longest RRI close to the 30th RRI and RRlmin15 denotes the shortest RRI close to the 15th RRI after assumingthe upright position), instantaneous increase in heart rate andchanges in HR, SAP and OAP at 30 sec, 1, 2, 3, 4 and 5 min afterassuming the upright position.

Lactate threshold was defined as the starting point of bloodlactate increase over the initial steady level (1- 2 mmol x 1-1 )observed at the lowest exercise intensities and was confirmedby the first non-linearity in the VE/VO2 ratio corresponding tothe "anaerobic threshold" [3.35).

Statistics

Distribution of each variable was calculated, and to normalizethe observed distribution of some frequency domain meas-ures, 10-based logarithms of the values were taken and usedin subsequent analysis. Parametric tests (two-way analysis ofvariance for repeated measures, Student's t-test for pairedsamples, Bonferroni test and Pearson correlation coefficients)were used to analyze the results (SPSS release 6.1, SPSS. Inc.,Chicago, lL, USA). Results in the text are expressed as means :!:SEM (95 % confidence intervals [0]). A P value of 0.05 was usedas a critical level of significance.

ResultsSubmaximal and maximal performance on a treadmill

-During

submaximal treadmill running the velocity was9.9 km x h-1 for 5 minutes and 11.7 km x h-1 for another 5 ffiin-

roo-Lutes at the inclination of 1 °. Heart rate during the run wasmeasured by ECG electrodes placed on CM2, CM6, and theneck. Fingertip blood samples were taken after the 10-minuterrun in order to analyze the blood lactate concentration (Ep-pendorfEBIO 6666, Eppendorf, Hamburg, Germany).

Training

The athletes trained for 6-9 weeks. The reasons for finishingthe experimental training were overtraining according toaforementioned overtraining criteria in five athletes. desire tostop the experimental training in two athletes. leg injury inone athlete and fever 'in another. Two CG athletes sufferedfrom health problems during the experimental period. Onehad a common cold and another had abdominal problems.One CG athlete was distressed not only because of trainingbut also because of other life stressors (Table 4). During the ex-perimental period. six athletes (5 ETG and 1 CG athletes) hadlate periods. The total training volume of the ETG increased

Incremental treadmill running/ski-walking until physical ex-U haustion was performed 5 minutes after the submaximal test.

Running was started with an inclination of 10 and a velocity of6.3 km x h-I, The velocity was increased every 3 minutes by

48

A.

L. T. Uusitalo et a--by 80 % (p < 0.01). the volume of low-intensity training by 98 %(p < 0.01) and the volume of intensive training by 130%(p < 0.01) while strength training volume decreased by 54%(p < 0.05) from the first training week to the END (Fig. 1). Thecorresponding increases in the CG were 6%. 5%,10%. and 21 %(Fig.1). Strength training volume was higher (p < 0.05) on thesecond. fifth. sixth and the last training week in the CG as com-pared to the ETG. The training of the OA did not differ from thetraining of the other ETG athletes.

Supine and standing heart rate and blood pressure variabilit}and the head-up tilt test

RRI LFP during supine rest increased during the training perio((p < 0.05) in the ETG (Table 2a). Other significant changes irheart rate variability were not found in either of the groups 01in the OA subgroup (Tables 2a and 2 b). Standing SAP and OAFVLFP had a tendency to increase in the ETG from 12.3! 3.4(4.4-20.3) mmHg2 to 22.4!6.1 (8.0-36.8) mmHg2 and from6.9!1.7 (4.9-10.9) mmHg2 to 12.2!2.8 (5.6-18.7) mmHg2and to decrease in the CG from 33.1 !12.3 (1.5-64.7) mmHg2to 12.3 !2.5 (5.8-18.9) mmHg2 and from 17.6! 6.1 (1.9-33.3)mmHg2 to 10.0! 2.8 (2.8 -17.2) mmHg2, respectively, duringthe experimental training (interactions, p < 0.05). Other signif-icant changes in blood pressure variability were not found(baseline values in Table 3). Table 4 shows the individualchanges in the selected parameters as percentages. The limitsof individual retest variation according to our unpublished re-peatability study are based on the coefficient of variation of re-peated measurements in five physically active humans duringone week. The limits were 7% (0 95%, 5-8%) for standingheart rate,18% (12%-24%) for standing RRI SO, 33% (22%-45%) for standing RRI TP, 72% (47%-98%) for standing RRILFPI/HFP and 34% (30%-39%) for HR3diff.

~.r1

I

--

~-t--

C'" ~UJ

Training weeksThe 30/15 index decreased during the training period accord-ing to the analysis of variance (p = 0.05). In the ETG, it de-creased by 5:1:3 ([-11 )-[+1]) % from the baseline level of1.12:1: 0.05 (1.01-1.24) to the END, and in the CG, it was un-changed 0:1:4 ([-8]-[+8]) % from the baseline level of1.08:1: 0.08 (0.89 -1.28). Although we did not find any changesin the heart rate difference between supine and standing posi-tion in ,the head-up tilt, the individual findings after 3 minutesof standing are presented in Table 4. The Table shows thatthere were both accentuated and attenuated heart rate re-sponses to head-up tilt in both overtrained and not-over-trained athletes. In the END, one CG athlete (CG6) felt dizzi-ness, her blood pressure fell and she was tachycardic after 2minutes of standing, and the test was interrupted.

c Submaximal and maximal treadmill running/ski-walking

There was a group by training interaction between the ETG andthe CG athletes in heart rate at submaxiinal work rate (HRsubmaxp < 0.01). The HRsubmax decreased in the ETG athletes duringthe training period (p < 0.001). being 176:t: 3 (167 -184) bpmat baseline. 170:t:4 (160-179) bpm after 4 weeks of training(p<0.01). and 167:t:4 (158-176) in the END. It returned tothe baseline level after the recovery period. being 173:t: 5(163 -184) bpm (p = 0.05). In the OA subgroup. HRsubmax had atendency to decrease during the training period. being 171 :t: 1(167 -175) bpm at baseline and 166:t: 1 (162 -169) bpm in theEND. In the CG athletes. the HRsubmax remained at the same lev-el throughout the experimental period (167t3 [158-176] atbaseline and 165:t:4 [155-175] in the END).

u

-L-

t'-J/

0.: t0.5

--I I I. -,0N~~""'~"""OW, , Z ~

Training weeks .W

Fig.1 a Total training volume, b Low-intensity training volumeand c Intense training volume as (hoursfweek [mean :tSD]). ETG,experimental training group (solid lines), (G, control group (dottedlines), END. the end (last training week) of the experimental train-ing period: RE. 4th recovery week.-week before training period.'f"f'" significantly different between the groups. 'p<O.OS,, p<O.Ol. p<O.OOl

---[

VO2max, VO2max demand and maximal heart rate (HRmax) didnot change in the ETG and theCG during the training period. Inthe ETG, the VO2max, VO2max demand and HRmax were51.9:t2.2 (46.7-57.1) mlxkg-lxmin-l, 52:t2 (48-56)ml x kg-l x min-1 and 190:t 4 (186 -200) bpm at baseline and52.1 :t 2.3 (46.7 -58.3) ml x kg-l x min-\ 51:t 1 (48 -54)mlxkg-lxmin-l and 189:!:4 (180-199) bpm in the END,

i.J

I

...

v

'1

r

'~~!::','sb'11

Int J Sports Med 2000; 21;00""" --

Table 3 Blood pressure and blood pressure variability variables at baseline

Variable SupineErG

StandingETG CGCG

SAP (mmHg) 113:i:4

(105-122)52:i:2

(47 -57)26.41 :i:6.47

(11.10-41.72)20.32:i:5.15

(8.16-32.49)3.71:i:1.42

(0.36- 7.06)2.31 :i:0.42

(1.33 -3.30)173:i:62

(27 -320)8.10:i:1.58

(4.37 -11.83)5.74:i:1.32

(2.61-8.87)1.31 :i:0.45

(0.25-2.37)1.01 :i:0.28

(0.34-1.68)178:i:61

(33-323)

108:1:7

(91-126)52:1:6

(37-67)31.55:1:7.21

(13.02-50.08)25.68:1:5.78

(4.34-30.82)2.88:1:0.95

(0.43-5.33)2.95:1:0.72

(1.10-4.80)90:1:17

(42-137)7.08:1:1.16

(4.11-10~06)5.33:f:0.98

(3.05-8.47)

1.Q2:f:0.21

(0.47-1.57)0.70:f:0.17

(0.26-1.14) .

188:f:73

(1-375)

125 t 6

(110-140)75t4(65-85)45.21t14.10

(11.88- 78.55)12.35t3.38

(4.36- 20.34)15.01 t3.71

(6.24-23.79)14.56t4.08

(4.92-24.20)217t73

(44-390)1 5.60t 2.94

(8.65 -22.55)6.87t 1.68

(2.89-10.86)8.47 t 2.39

(2.81-14.14)2.99tO.96

(0.71-5.26)796t323(33-1560)

DAP (mmHg)

118:t4

(107 -129)74:t5

(60-89)53.08:t15.15

(14.13-92.04)33.1 O:t 12.28

(1.53-64.67)12.23:t2.81

(5.02-19.45)7.40:t2.15

(1.88-12.92)203:t67

(32-374)26.68:t6.78

(9.25-44.12)17.60:t6.11

(1.89-33.31)7 .38:t 1.89

(2.51-12.25)1.57:t0.48

(0.32-2.81)742:t 391

(- 265-1748)

SAP TP (mmHg2)

SAP VlFP (mmHg2)

i

ii

J~;!1~~,%{~SAP LFP (mmHg2)

SAP HFP (%)

SAP lFP/HFP (%)

DAP TP (mmHg2)

DAP VLFP (mmHg2)

DAP LFP (mmHg2)

DAP HFP (mmHg2)

DAP LFP/HFP (mmHg2)

[

Results are expressed as means:!: SEM (O 95%). SAP, systolic arterial pressure; DAP, diastolic arterial press\!re. For other abbreviations see Table 2a

Table 4 Individual changes as percentages in selected standing heart rate and heait rate variability variablesrHR3diff61

HR~1

RRI SOIII

RRI TP6.1

Subjecth.2 62 ~2~2 t.2

~'~,-25

-43'

-68'+400'

+107'

-62'

-46'

+28

+40'

+145.

+20

+12

0

0

+ 337"

-93"

-12

-13

-41

+186"

-14

+11

+101"

+457"

-22

-23

-66

+380"

+101.

+365'-42

-72

-34

-54'

-60'

+39'

+13

-57'

-77'

0

+65"+125"

-9

+40.

0

-69.

-6

-14.+2

+15.+15.-17.-21.+17.

+5+13.-10.-10.-11.+3

+17.

-2

-34--12

-6

-18

+91-+62-

+14

+6

-15

-28-+51-

+4

-6

-63-

+17

-30'

-7

-10

-40"

-23

-12

-52"

+ 268"

+186"

+4

+ 31

-56.-12

OAlc

OA2c

OA3c

OA4c3

OASc

ETG6

ETG7

ETGS#

ETG9#

CG11

CG21

CG3

CG4

CGS

CG62

-5

-19'

-10'

+18.

-18.-21"

+14"

+5

+10"

+4

+4

-16"

+6+ 22"

-13+122.+34.

+11+10-34.-18

+27.+26.-11

-37.

-55'+400'

+87'

+5

+19

-56'

-43'

+100'+50'

-20

-53'

+14

+15

-55

+11

+227.+ 84.

-37

+52

-80.-45

+ 227.

-1

-19

-43"

+163"

0+ 88"-86 "

ErG, experimental training group; (G, control group; HR, heart rate; RRI SD of R-R interval variability; RRI TP, total power of R-R interval variability; RRI LFPjHFP,:

ratio of low-frequency and high-frequency powers of R-R interval variability; HR3dill' heart rate difference between supine rest and after 3 min standing; 61, change ~during the entire training period of six to nine weeks; 62, change during the last two to five training weeks, +, increase, -, decrease, " change Is outside the;

retest variation (see text) and at least 4 bpm, c, overtrained athlete; #. experimental training finished because of illness or injury; 1, athlete had health problerr.s; 2 t

athlete felt distressed and demonstrated orthostatic intolerance after 2 min standing in the END; 3, 4 wk data not available. j!

[

No significant changes in maximal or submaximal blood lac-tare in any of the groups were found. Maximal blood lactatehad a tendency to decrease in the OA subgroup from 10.2 :t 1.4(6.2-14.2) mmol/l to 8.2 :t1.6 (3.9-12.5) mmoi/i.

increased performance capacity, although not in this case,and also at the same time indicate decreased sympathetic acti-vation during the same absolute submaximal exercise load.

Heavy endurance training seemed to induce increase in RRILFP during supine rest. RRI LFP has been thought to be mostlysympathetically mediated [25], both parasympathetically andsympathetically mediated [1] and only parasympatheticallymediated [13]. In our studies, RRI LFP has been shown to be

mostly parasympathetically mediated, however, with a slightsympathetic component [34]. In this case, increased cardiacsympathetic modulation could be supposed to be behind theincreased RRI LFP, because there was no change in RRI HFPwhich is clearly the best frequency component to reflect cardi-ac parasympathetic modulation [13,34].

Correlation analyses

Correlation between the change in (6) VO2max and RRI TP dur-ing standing in the ETG is presented in Fig. 2. In addition, 6 su-pin!.' RRI TP from baseline to 4 wk (r = -0.71. P < 0.05), 6 supinelog RRI HFP from baseline to 4 wk (r= -0.75. P < 0.05). 6 supinelog SAP HFP from baseline to 4 wk and to END and from 4 wk to

END (r=-0.79, p<0.05, r=-0.67. p=0.05 and r=-0.70.p = 0.05. respectively), and 6 standing DAP VLFP from 4 wk toEND (r = 0.82, P < 0.05) correlated with the corresponding 6

VO2max in the ETG.

Experimental training attenuated immediate biphasic heartrate response to head-up tilt. This may be a sign of decreasedbaroreflex function and increased plasma volume which re-sults in increased ventricular compliance. and again attenu-ated baroreflex responsiveness [26]. In the present study, theattenuated heart rate response was not specific to overtrain-ing, but was also typical during normal training of the controlgroup.

r= .74, p<O.O5

Hormonal findings of previous studies indicate that autonomicnervous function changes in the overtraining state [19.20.21].Changes in cardiac autonomic function have not been pre-viously studied in the overtraining state. The positive correla-tion between the changes in VO2max and standing RRI TP andthe individual changes in the overtrained athletes (see Table 4)in the present study showed that heart rate variability in thestanding position had a tendency to decrease with overtrain-ing. This is partly consistent with the study on patients withneurally-mediated syncope [24], and indicates pronounced va-gal withdrawal and possibly decreased sympathetic excitabil-ity in the overtrained athletes during standing. However. therewas one overtrained athlete (OA1) whose standing RRI TP didnot change due to increased RRI VLFP. At the same time, herRRI HFP and RRI LFP decreased as in the other four over-trainedathletes. An identical increase in RRI VLFP was also observed inthree other athletes (OA 3. ETG 8 and CG 5). The reason for theincreased RRI VLFP during standing remains obscure. Metho-dological bias can also confuse findings concerning VLFP com-ponent in short recordings.

I ..I i-3 -1 1 3 5 7

6VO2max (ml/kg/min)

Fig.2 Significant correlations in the experimental training groupbetween d maximal oxygen uptake and d total power of standingR-R interval variability (1O-based logarithm) during the experimentaltraining period of 6-9 weeks. VO2max. maximal oxygen uptake; RRITP. total power of R-R interval variability.

Discussion

Increase in VO2max is used as a primary sign of training-in-duced cardiorespiratory fitness [2]. The present heavy trainingperiod resulted in the overtraining state in five out of nine ETGathletes whose VO2max decreased by 4-9%. Of the four not-overtrained ETG athletes one was injured, another one becomeill and two felt distressed with no decrease in VO2max whichwere the reasons why their experimental training was fin-ished. The different reasons indicate that the responses tohE'avy training are individual. Local overstrain, illness or men-tal exhaustion can "preserve" an athlete from the physiologicalovertraining syndrome. Decreases in VO2max were also foundin one not-overtrained ETG athlete (by 2 %) and in two C.G ath-letes (by 2 % and 9 %). However, these athletes did not meet theother overtraining criteria.

In many subjects the changes in supine and standing heart ratevariability seemed to be rather contrary. The exact reason forthe conflicting position-dependent findings remains obscure.In the standing position. the regulation of haemodynamics ismore complicated. The body has to adapt to the new positionby increasing sympathetic activity [23]. like during exercise. Inaddition, there is a number of other factors which strongly in-fluence the cardiovascular response during standing-up [36J,e.g. the renin-angiotensin system and the Clistribution of bloodvolume (changes in preload).

b

~Decreased maximal heart rate in overtrained athletes could bea sign of their decreased cardiac sympathetic modulation dur-ing maximal physical effort. That could also be related to de-creased ~-receptor sensitivity and/or density in the heart [211.or be a question of mechanical and metabolical changes of thecardiac muscle or changes in preload. Decreased heart rateduring submaximal exercise in the ETG might be a marker of

Since there were not any other uniform findings in the over-trained athletes in the head-up tilt test, we also looked at indi-vidual results. To find out whether there were differences in

sympathetic excitability after the tilt. we looked at the heartrate responses during prolonged standing. Both accentuatedand attenuated heart rate responses to continued upright posi-

r

lJ

n-:-.

0.6 r ... 05 0.4

0.30.2

0.1 I0 .

-0.1 1-0.2 I

-0.3 i-0.4 .

-5

52 Int J Sports Med 2000; 21 A. L. T. Uusitalo et i

tion were found in the overtrained athletes as compared totheir values in the normal training state. This is a sign of eitherincreased or decreased ability to increase sympathetic dis-charge during standing [6,23] and corresponds to the twoovertraining types [17]: sympathetic and parasympathetic(exhaustion) (see Table 4). However, the changes were notspecific to overtraining because there also were similar chang-es in the not-overtrained athletes.

Acknowledgements

This study was supported by grants from the Finnish Ministry .1of Education. and the Medical Research Fund ofTampere Uni- :tversity Hospital. We thank Leena Paavolainen, M.Sc., and llkka ~Vaananen, M.Sc., for submaximal and maximal treadmill test imeasurements and Tiina Hoffman for assistance in heart rate iand blood pressure variability analyses. !t

We have previously reported that the nature of blood pressurevariability is obscure because so many factors determine bloodpressure [34]. In the present study, training-induced changesin heart rate and blood pressure variability had a tendency tobe identical. Changes in the high-frequency power of RRI andSAP variability during supine rest correlated to the cor-responding VO2max changes. The reported changes in SAPand OAP VLFP may result from reduced baroreflex function[30], decreased vascular sympathetic effect [13,30], or otherunidentified factors. On the other hand, during short time-periods the values of very low-frequency variability may notbe reliable.

References

This is the first experimental study in which training was con-tinued until the overtraining state or other signs and symp-toms of long-term overload (local overstrain, immunologicaldeficit or mental distress compelling the athlete to interruptexperimental training). In previous studies training has beencontinued for a predetermined period [20) or athletes haveovertrained accidentally during the normal training season(5,16,19). The main weakness of the present study is the s~allnumber of overtrained subjects. Another weakness is great in-ter- and intra individual variability in the measured heart rateand blood pressure variability variables (33] emphasizing theimportance of evaluating individual changes and decreasingthe value of cross-sectional studies. When heart rate and bloodpressure variability and VO2max are used to observe training-induced physiological changes, other possible reasons thatcould influence the results such as illness, mental distress, lackof sleep, food intake and menstruation should always be takeninto consideration. Sympathetic nervous activity has been sug-gested to be higher during the luteal than the follicular phaseboth at rest and during a complex tracing task [27].ln the pres-ent study heart rate and blood pressure variability was not sys-tematically higher in either of the phases.

We conclude that heavy endurance training induced an in-crease in RRI LFP during supine rest as.a possible marker of in:'creased cardiac sympathetic modulation and that attenuationin the biphasic heart rate response immediately after head-uptilt might indicate training-induced decrease in baroreflexfunction. Overtraining state seemed to be related to the tend-ency of heart rate variability to decrease in the standing posi-tion as a sign of pronounced vagal withdrawal and in somecases decreased sympathetic excitability. Changes in respons-es to prolonged standing after head-up tilt in heavily trained/overtrained athletes seemed to be individual reflecting differ-ences in sympathetic excitability during standing and possiblydifferent types to respond to heavy training/overtraining.

1 Akselrod S, Gordon D, Madved JB, Snidman NC, Shannon DC, Co- i

hen RJ. Hemodynamic regulation: investigation by spectal analy- Hi"sis.AmJ Physiol1985; 249: H867-H875 .i,

2 Astrand PO, Rohdahl K. Textbook of Work Physiology: Physio- ,,\~

logical Basis of Exercise. New York: McGraw-Hili Book Company, }1977; 2nd ed.: 324 ..i)i;

3 Aunola S, Rusko H. Aerobic and anaerobic thresholds determined ;1

from venous lactate or from ventilation and gas exchange in re!a- f;tion to muscle fiber composition. IntJ Sports Med 1986; 7: 161 -W166 ~~

4 Balke B, Ware RW. An experimental study of physical fitness of air j,

force personnel. U. S. Armed Forces Med 1959; 10: 675 -688 ~:. ~5 Barron JL, Noakes TD, Levy W, Smith C, Millar RP. Hypothalamic.. ;

dysfunction in overtrained athletes. J Clin Endocrinol Metab ~;1985; 60: 803-806 !Ii

6 Borst C, Wieling W, Van Brederode JFM, Hond A, de Rijk LG, Dun- ~

ning AJ. Mechanism of initial heart rate response to postural 'P'change. AmJ Physiol1982; 243: H676-H681 1

7 Cannon WB. Stress and strains of homeostasis. Am J Med Sci l

1935; 89: 1-14 ~8 Dixon EM, Kamtah MV, McCartney N, Fallen EL. Neural regulation I

of heart rate variability in endurance athletes and sedentary con- J:trois. Cardiovasc Res 1992; 26: 713 -719 j

9 Durnin J, Womersley J. Body fat assessed from total body densiry r~and its estimation from skin fold thickness: measurements on481 men and women aged from 16 to 72 years. Br J Nutr 1974:32: 77-97 :.

10 Ekblom B, Kilbom A, Soltysiak J. Physical training, bradycardia ~'

and autonomic nervous system. Scand J Clin Lab Inv 1973; 32: .r251-256 J

11 Fry RW, Morton AR, Keast D. Overtraining in athletes: An update. r

SportsMed1991;12:32-65 ,12 Gregoire J, Tuck S, Yamamoto Y, Hughson RL Heart rate at rest:

and exercise influence of age, gender and physical training. Ca.'! J iAppl Physiol1996; 21: 455 -470 \

13 Hedman AE, Hartkainen JEK, Tahvanainen KUO, Hakumaki MOK. ~

Power spectral analysis of heart rate and blood pressure variabil- tity in anaesthetized dogs. Acta Physiol Scand 1992; 146: 155-164

14 Henry JP. Present concept of stress theory. In: Us din E, Kver-

nansky R, Kopin Ij (eds). Catecholamines and Stress: Recent Ad-vances. North-Holland, NY: Elsevier,1980: 557 -571

15 Hooper SL, McKinnon LT. Monitoring overtraining in athletes.

Sports Med 1995; 20: 321-32716 Hooper SL, McKinnon LT, Gordon RD, Bachmann AW. Hormonal

responses of elite swimmers to ov,ertraining. Med Sci Sports Ex-erc 1993; 25: 741- 747

17 Israel S. 2ur Problematik des Obertrainings aus internistischer

und leistungsphysiologischer Sicht. Med u Sport 1976; XVI: 1 -12 ,18 Janssen MJA, de Bie J, Swenne CA, Oudhof J. Supine and standing:

sympathovagal balance in athletes and controls. Eur J Appl Phys- (io11993; 67: 164-167 '

19 Kindermann W. Overtraining-expression of a disturbed autO-

nomic regulation. Dtsch 2ft Sportmed 1986; 8: 238 -244 :'

":n (,'(per~r~~~_n~ ~:_~r~ng S!udy

Corresponding Author:

Arja Uusitalo, M.D.KIHU -Research Institute for Olympic SportsRautpohjankatu 6

SF-40700jyvaskylaFinland

Phone:Fax:E-mail:

+ 358 (14) 603170+ 358 (14) 603171

Uusitalo@kinu,jyu.fi

U&: :~ l;

'~

r"t.I

~,f

U'

lPlehmann M, Gastmann U, Petersen KG. Bachl N. Seidel A, Khalaf"N. Fischer S. KeulJ. Training-overtraining:'performance and hor-;11011e levels after a defined increase in training volume ve~sus in-tensity in experienced middle- and long-distance runners. Br JSports 1\.1ed 1992; 26: 233-:-242 .

~I Lehmann M, Lormes W, OpItz-Gress A. SteInacker J, Netzer N, Fos-ter C. Gastmann U. Training and overtraining: an overview andexperiml'ntal results in endurance sports.J Sports Med Phys Fit-ness 1997; 37: 7-17

/':1 Macor F. Fagard R, Amery A Power spectral analysis ofRR interval

dnd blood pressure short-term variability at rest and duringdynamic exercise: comparison between cyclists and controls. IntJ Sports Med 1996; 17: 175 -181

!1 Marin Neto JA. Gallo L, Manco JC. Rassi A, Amorim DS. Mecha-

nisms of tachycardia on standing: studies in normal individualsJnd in chronic Chagas' heart patients. Cardiovasc Res 1980; 14:541-550

;14 Morillo CA. Klein GJ, Jones DL, Yee R. Time and frequency domain

.lnalyses of heart rate variability during orthostatic stress in pa-tients with neurally mediated syncope. Am J Cardiol 1994; 74:1258 -1262

;.~. P.ll'dl1i M, Lombardi F, Guzetti S, Rimoldi 0, Furlan R. Pizzinelli P,

Sa;~drone G. Malfatto G, Dell'Orto S. Piccaluga E. Turiel M. BaselliG. Cerutti S. Malliani A. Power spectral analysis of heart rate andarterial pressllre variabilities as a marker of sympatho-vagal in-teraction in man and conscious dogs. Circ Res 1986; 59: 178 -193

~6 Raven PB, Pawelczyk JA. Chronic endurance' exercise training: a

condition of inadequate blood pressure regulation and reducedtolerance to LBNB. Med Sci Sports Exerc 1993; 25: 713- 721

27 Sato N, Miyake S, AkatsuJ, Kumanshiro M. Power spectral anal-

ysis of heart rate variability in healthy young women during thenormal menstrual cycle. Psychosom Med 1995; 57: 331-335

28 Seals DR, ChasePB. Influence of physical training on heart rate

variability and baroreflex circulatory control. J Appl Physiol1989; 66: 1886 -1895

29 Selye HA. Syndrome produced by diverse noxious agents. Nature

1936; 138: 3230 Sleight P, Rovere MT, Mortara A, Pinna G. Maestri R. Leuzzi S,

Bianchini B. Travazzi L, Bernardi L Physiological and pathophys-iology of heart rate and blood pressure variability in humans: ispower spectral analysis largely an index of baroreflex gain? ClinSci 1995; 88: 103 -109

31 Tahvanainen K, L'insimies E, Tikkanen P, Hartikainen J, Karki T.

lyyra T, Mantysaari M. Microcomputer-based monitoring of car-diovascular functions in simulated microgravity. Adv Space Res1992; 12: 227 -236

32 Toivanen H, L'insimies E. Jokela V. Helin P. Penttila I. Hanninen O.

Plasma levels of adrenal hormones in working women during aneconomic recession and the threat of unemployment: Impact ofregular relaxation training. J Psychophysiol1996; 10: 36 -48

33 Toyry J. Mantsysaari M. Hartikainen J, L'insimies E. Day to day

variability of cardiac autonomic regulation parameters in normalsubjects. Clin Physiol1995; 15: 39-46

"11!'.;;',;10 ALT. Tahvanainen KUO, Uusitalo AJ. Rusko HK. Non-in-v.lsive evaluation of sympathovagal balance in athletes by timeand frequency domain analyses of heart rate and blood pressurevariability. Clin Physiol1996: 16: 575-588 .

35 Wasserman K. Whipp BJ. Koyal SN, Beaver Wl. Anaerobic thresh-

old and respiratory gas exchange during exercise. J Appl Physiol

1973; 35: 236-24336Wieling W. Standing. orthostatic and autonomic function. In:

Bannister R (ed). Autonomic Failure: A Textbook of Clinical Dis-orders of the Autonomic Nervous System. Oxford: Oxford Uni-

versity Press. 1989: 308-320

~'

r