Eur J Appl Physiol (1993) 67:540-544 European ,oum.,o, A p p l i e d P h y s i o l o g y

and Occupational Physiology © Springer-Vedag 1993

The combined effects of orthostatic and mental stress on heart rate, T-wave amplitude, and pulse transit time

Attila Szabo

D~partement d'6ducation physique, Universit6 de Montr6al, C.P. 6128, succ. A, Montreal, Quebec, Canada, H3C 3J7

Accepted August 14, 1993

Abstract. The purpose of the present study was to de- termine whether cardiovascular reactivity to mental stress may be comparable in sitting and standing pos- tures. Fif teen healthy males per formed two 1-min men- tal arithmetic tasks, either while sitting or while stand- ing, in counterbalanced order. Hear t rate, pulse transit time, and T-wave amplitude were recorded in the last 10 s of the minute before, during, and after the mental arithmetic. Reactivity scores for each of the dependent measures were computed by calculating the percentage change from baseline values. Data were analyzed with multivariate and univariate repeated measures analysis of variance. Hear t rate reactivity to the combination of orthostatic and mental stress was greater than to ei ther stressor alone. Cardiac-sympathetic reactivity was greater in response to orthostatic than to mental stress as revealed by greater changes in T-wave amplitude and pulse transit t ime in response to the former. No additional decreases in T-wave amplitude, in response to mental stress, were observed during standing, but pulse transit t ime decreased in the same situation. However , no changes in pulse transit t ime in response to mental stress were observed in the sitting position. These results demonstrate that cardiovascular reactivi- ty to mental stress depends on the body position in which the stressor is encountered.

Key words: Orthostatic stress - Mental stress - Hear t rate - T-wave amplitude - Pulse transit t ime

Introduction

The occurrence of an autonomic responses to both or- thostatic Bevegard et al. 1960; Stafford et al. 1970; Hatch et al. 1986) and mental stress (Levi 1985; Obrist et al. 1978; LeBlanc et al. 1979) is well established. However , it remains unclear whether a change in body position could affect cardiovascular reactivity to men-

Correspondence to: A. Szabo (c/o F. P6ronnet)

tal stress. To date, to the best of my knowledge, this question has not been addressed directly by research- ers in the field. 1

Physiological data, derived from experiments exam- ining the impact of mental stress on physiological var- iables, are largely based on recordings obtained in an armchair-sitting position (Sinyor et al. 1983; Seragan- ian et al. 1987; Brown et al. 1988). Encounters with mental stress, however, are not confined to this pos- ture. Consequently, generalization from findings ob- tained in one physical position may be questionable.

The scope of the present study was, therefore, to ex- amine whether mental stress encountered during standing (which itself may be considered orthostatic stress) elicits comparable reactivity, in heart rate, T- wave amplitude, and pulse transit time, to reactivity observed in the customarily adopted sitting position, or whether a postural change affects reactivity.

Methods

Subjects. Fifteen male graduate students gave their informed written consent to participate in the present study. Their age, mass, and height were [mean (SD)] 30.1 (6.7) years, 74.5 (7.7) kg, and 176.6 (6.06) cm, respectively. On the Exercise Behavior Questionnaire (Godin et al. 1986), subjects reported a healthy, drug- and tobacco smoke-flee lifestyle and participation in an av- erage of two bouts of vigourous exercise activities per week. None of them suffered from orthostatic hypotension.

Recording apparatus. A three-output channel computer software driven Marquette electrocardiogram (Max-l) was pre-program- med for cardiovascular recordings. Signals from a photoplethys- mograph (optical-sensor pulse meter, model CT-4600, IBS) were fed into one of the input channels of the electrocardiogram (ECG) and were simultaneously recorded with the ECG. Special- ly prepared Marquette Ag/AgC1 electrodes (with ECG electro- lyte paste) were used.

1 The reverse question, i.e., how mental stress could modify car- diovascular response during orthostatic stress, has been raised by Goldstein and Shapiro (1988). However, changes in cardiovascu- lar reactivity, i.e., the magnitude of the response, to mental stress as a function of body position were not examined in this study

541

Mental stressor. The arithmetic task, used as mental stressor in the present study, was described in more detail elsewhere (Szabo and Gauvin 1992). In brief, subjects were expected to solve four arithmetic problems of approximately equal difficulty level in 1 min. Each problem consisted of three basic operations involv- ing single, double, and triple digit numbers (e.g., 189 x 2 + 10-7 ) . The problems were presented visually, required verbal answers, and could be solved in subjects' order of preference. The answers were recorded on audio tape. Three seven-point rating scales, ranging from 0 (not difficult at all) to 6 (extremely difficult), were used to gauge the subjectively perceived difficulty of each prob- lem set, the effort exerted by the subjects to do well on each problem set, and the importance to the self, or self-challenge, to perform well on each problem set.

Procedure. All tests were performed in the morning. Subjects were asked to refrain from alcoholic or caffeinated beverages for 12 h before the testing. Following electrode placement, initial trial ECG recordings, and explanation of the arithmetic task, sub- jects rested in a reclined armchair for 5 min. Subsequently, a trial arithmetic problem was presented to habituate the subjects to the stressor, which was followed by a 3-min rest period in the arm- chair. Next, the subjects were asked to either stand or remain seated (counterbalanced order) for 1 rain which was followed by the presentation of the arithmetic task for 1 min and a 1-min re- covery period. The ECG was recorded during the last 10 s of each minute during this interval. After 3 min of rest, these steps were repeated, with the subject either in the standing or sitting position, opposite to the first part of the testing.

ECG recordings. Skin preparation consisted of the light abrasion, washing with 70% alcohol, and drying of the skin. The electrodes were placed on the subject in a 12-lead recording arrangement. The bipolar lead II records yielded the highest (3-9 mm) and clearest T-waves, and hence lead II was selected for output chan- nel 1. The signals from the pulse meter were recorded from out- put channel 2, and the aVf unipolar lead record was read from channel 3. The chart paper speed was set to 25 r am-s - l , and the recorder sensitivity was set to 10 m m ' m V -a.

Data reduction. Heart rate data were quantified by measuring the successive R-R intervals on the ECG chart. The T-wave ampli- tude was determined by measuring the distance between the isoelectric P-Q interval and the peak of the T-wave (Heslegrave and Furedy 1979; Scher et al. 1984) to the nearest 0.25 mm accu- racy. Pulse transit time was determined by computing the interval between the peak of the ECG R-wave and the midpoint of the next photoplethysmograph deflection (Roth et al. 1990). All quantifications were performed manually with the aid of a high- resolution magnifying glass and the obtained values were aver- aged across the 10-s recording interval.

Absolute values or raw data of heart rate, T-wave amplitude, and pulse transit time were converted to relative values, or per- centage change from baseline scores, using the formula below (Fillingim et al. 1992) and by adopting the last 10-s records, pre- ceding the mental task, as baseline (Shulhan et al. 1986).

Percentage (%) change = response-basel ine x 100 from baseline baseline

The change scores (referred to beneath as reactivity scores) were correlated with their corresponding baselines to account for the possible influence of the latter, or the law of the initial values (LIV), on reactivity scores (Jamieson and Howk 1992). Statistical analyses were performed with the Systat computer software (Wil- kinson 1989).

Results

Prior to the examination of reactivity differences in the sitting and standing position, it was essential to estab- lish that the presented mental stressor was potent enough to elicit a measurable response in the depend- ent measures. This preliminary analysis was carried out by examination of the raw data (Table 1). Separate one-way analyses of variance (ANOVA) were per- formed, but they were adjusted using the Bonferroni corrections (Wilkinson 1989) for multiple tests. Ac- cordingly, the acceptable probability of significance level was set to 0.008 (0.05/6=0.008). A significant heart rate response to mental arithmetic was revealed in both the sitting and standing positions [F(1,14) = 13.9; P < 0.003; and F(1,14) = 10.7, P < 0.006, respectively]. Significant T-wave amplitude decrease in response to mental stress was found in the sitting [F(1,14)=13.4, P<0.003] but not in the stand- ing position. On the other hand, a significant decrease in pulse transit time was observed in the standing posi- tion [F(1,14)= 11.1, P<0.005], but not during sitting. These analyses established that, depending on the pos- ture, the mental stressor elicited a response in all de- pendent measures.

The LIV did not influence reactivity as revealed by non-significant correlations between the baseline val- ues of heart rate, T-wave amplitude, and pulse transit time and the corresponding reactivity scores. There- fore, the subsequent examination of the reactivity scores, as described below, was deemed feasible.

The raw data were used to establish that a response to mental stress had taken place in both the sitting and standing positions. To test whether cardiovascular reactivity (i.e., the magnitude of the response) to men- tal stress in the sitting and standing positions was com- parable, reactivity scores (percentage change from sitting baseline vs percentage change from standing ba- seline) observed in the two body postures were exam- ined with a multivariate repeated measures ANOVA (MANOVA) by using heart rate, T-wave amplitude, and pulse transit time reactivity scores as the multivar- iate dependent measures. This analysis yielded a signif- icant main effect [Wilks' lambda=0.31, F(3,12)=8.9, P<0.002]. Univariate tests revealed that reactivity in

Table 1. Mean (SE) heart rate, T-wave amplitude, and pulse transit time before, during, and after mental arithmetic perform- ed in the sitting and standing positions (n = 15)

Test period Heart rate T-wave Pulse (beats- rain - 1) amplitude transit

(mV) time (ms)

Pre-sitting 69.0 (2.38) 0.72 (0.07) 263.7 (4.29) Sitting 79.9 (3.93)* 0.63 (0.08)* 263.5 (5.04) Post-sitting 65.0 (2.44) 0.71 (0.07) 265.3 (5.26) Pre-standing 82.9 (2.32) 0.58 (0.07) 283.2 (4.61) Standing 88.2 (3.21)* 0.56 (0.08) 268.7 (6.35)* Post-standing 80.0 (2.94) 0.60 (0.07) 287.0 (5.93)

* Significantly different from pre- (and post-) values (P < 0.008)

2 2 I I I 18

1 4

U,I 1 0 Z i _1 u.i 6 u) <c m 2

0 - 2 n- I.i.

- 6 u.I

Z - 1 0 <

-!. - 1 4

- 1 8

- 2 2

l

=~lU I

542

I I I

H R T W A P T T

Fig. 1. Percentage changes [mean (SE)] from baseline (dashed line at 0) in heart rate (HR), T-wave amplitude (TWA), and pulse transit time (PTT) in response to a 1-min mental arithmetic task in sitting (n) and standing (N) position

3 5

2 8 A

2 1 LU Z 1 4 ..I I,IJ r,n 7 ,,< Q2 :Z 0 0 r,- iJ. - 7

- 1 4

"I- -21 O

- 2 8

-35

I I

I I

H R T W A

I i .... ........

I P T T

Fig. 2. Percentage changes [mean (SE)] from baseline assessed in a sitting position (dashed line at 0) in heart rate (HR), T-wave amplitude (TWA), and pulse transit time (PTT) in response to standing (R), mental arithmetic (N), and to mental arithmetic during standing (N)

all three cardiovascular indices was different (Fig. 1) in the sitting and standing positions [F(1,14)=5.5, P<0.03, for heart rate; F(1,14)=6,9, P<0.02, for T- wave amplitude; and F(1,14)=14.6, P<0.002, for pulse transit time].

In the subsequent analysis, data were re-examined from a double stressor perspective, in which the magni- tude and direction of cardiovascular reactivity (using the pre-mental stress sitting values as baseline) to or- thostatic stress (standing), to mental stress, and to the two of them combined were determined. A MA- NOVA, with the above three conditions as the re- peated measures and heart rate, T-wave amplitude, and pulse transit time as the multivariate dependent measures, was performed. This test yielded a signifi- cant main effect [Fig. 2; Wilks' lambda=0.18, F(6, 9) = 7.0, P < 0.005]. Univariate repeated measures ANOVAs [using the Greenhouse-Geisser corrected probability levels (G-G P)] showed that heart rate, T- wave amplitude, and pulse transit time all contributed to the multivariate main effect [F(2,28)=6.0, G-G P<0.02, for heart rate; F(2,28)=7.8, G-G P<0.004, for T-wave amplitude; and F(2,28)=11.8, G-G P<0.001, for pulse transit time].

Post-hoc simple contrasts (Wilkinson 1989) for heart rate revealed that reactivity in the combined stressor situation was greater than in response to standing [F(1,14)=9.5, P<0.008], or to mental stress [F(1,14) = 12.1, P < 0.004] alone. Reactivities to the lat- ter two were not significantly different,

Post hoc simple contrasts for T-wave amplitude re- vealed that significantly greater T-wave amplitude de- pression occurred in response to standing than to men- tal stress [F(1,14) = 7.8, P < 0.02]. The combined stress

situation also elicited greater T-wave amplitude changes than mental stress [F(1,14) = 10.8, P < 0.005]. The T-wave amplitude reactivity to the combined stress situation was not different from T-wave ampli- tude reactivity to standing.

Simple contrasts revealed that pulse transit time was greater during standing than during mental stress and during the combined stress situation [F(1,14)=25.0, P < 0.001; and F(1,14) = 11.1, P < 0.005, respectively]. The latter two were not significantly different from one another.

Subjective and objective data (i.e., the number of answers, the number of correct answers, task difficulty appraisal, the reported performance effort, and impor- tance to the self to perform well, or self-challenge), as- sociated with the arithmetic problems, were not differ- ent in the standing and sitting body postures (Ta- ble 2).

D i s c u s s i o n

The results of this study clearly demonstrate that car- diovascular reactivity to mental stress is different in sitting and in standing postures. Accordingly, the mag- nitude of reactivity in heart rate and T-wave amplitude was lower, while pulse transit time was higher during standing than during sitting. These results may be attri- buted to the fact that orthostatic stress itself elicits a cardiovascular response and when combined with men- tal stress an interaction may take place between the stressors which may modify the reactivity patterns. Therefore, in an upright posture, a combined stress sit- uation may have emerged in which the reactivity to

543

Table 2. Objective and subjective measures associated with men- tal arithmetic in the sitting and standing positions [mean (SE); n = 14]

Measure Sitting Standing

Number of answers a 2.0 (0.23) 2.2 (0.28) Correct answers a 1.2 (0.30) 1.4 (0.33) Task difficulty b 3.2 (0.21) 2.9 (0.28) Performance effort b 4.4 (0.25) 3.8 (0.33) Self-challenge b 4.0 (0.39) 3.6 (0.46)

a Objective measures; out of four possible b Subjective measures; out of six possible

mental stress was modified by the reactivity to the or- thostatic stress.

This hypothesis was further examined and substan- tiated when the data were re-analyzed from a com- bined stress perspective. While standing evoked similar magnitudes of heart rate reactivity as mental stress, when the two stressors were combined an additional increase in heart rate was observed that was signifi- cantly greater than reactivity to either type of stressor alone. Therefore, two types of qualitatively distinct stressors (physical and mental) may elicit synergistic responses in heart rate. While no reports are available concerning the combined effects of standing and men- tal stress, these findings may be reconciled with pre- vious studies. For example, increased heart rate reac- tivity to mental stress was reported when the latter was superimposed on a physical stressor that consisted of either the cold pressor task or light physical exercise (LeBlanc et al. 1979; Myrtek and Spital 1986; Roth et al. 1990). Consequently, it is conceivable that physical and mental stress have synergistic influences on heart rate reactivity which may be independent of the type of physical stressor. However, this synergistic action is not the simple summation of the separate responses to the two stressors. Examination of the T-wave re- sponses may shed more light on the mechanism in- volved.

Apparently, T-wave amplitude, which has been sug- gested to be highly sensitive to mental stress (Hesle- grave and Furedy 1979, 1983; Hijzen et al. 1984; Scher et al. 1984; Furedy 1987; Rau 1991), may be even more sensitive to orthostatic changes. Nevertheless, T-wave amplitude was sensitive to mental stress as well, as in- dicated by the significant decrease in the sitting posi- tion. Accordingly, T-wave amplitude was responsive to both orthostatic and mental stress; however, the stronger stimulus (standing) may have cancelled out the T-wave amplitude reactivity to the weaker stimulus (mental arithmetic). One possible mechanism by which this interaction could have taken place may be related to sympathetic parasympathetic involvement ratios. For example, heart rate is a composite index of both autonomic activities, but T-wave amplitude may be a more specific measure of sympathetic activity (Furedy and Heslegrave 1983; Rau 1991). It has also been sug- gested that the lack of changes in T-wave amplitude, accompanied by concurrent changes in heart rate, may

be viewed as changes in parasympathetic activity (Scher et al. 1984). In the present study, heart rate rose in the combined stress situation (as opposed to stand- ing alone), but T-wave amplitude did not decrease sig- nificantly in the same situation. It is therefore possible that an increase in heart rate during mental stress en- countered in upright posture reflects an addtional de- crease in parasympathetic activity without further in- crease in sympatetic activity.

Pulse transit time, which is a measure of contractili- ty (Weiss et al. 1980), increased in response to stand- ing, showed no change in response to mental stress, and showed only a slight increase to the combination of the two stressors. Therefore, changes in pulse transit time in resonse to mental stress may be easier to ob- serve when the subject is standing, because no changes in pulse transit time may be observed in the sitting po- sition, which is consistent with at least one previous re- port (Roth et al. 1990). An increase in peripheral re- sistance (Ziegler 1980) may have contributed to an in- crease in pulse transit time during standing. However, when mental stress was superimposed on standing this increase was significantly less than during standing alone. Increased peripheral resistance during standing takes place in order to maintain blood pressure (Zieg- ler 1980). A likely elevation in blood pressure in re- sponse to mental arithmetic (Goldstein and Shapiro 1988) may trigger a somewhat lesser increase in pe- ripheral resistance, which may be reflected through a lesser increase in pulse transit time. However, blood pressure was not measured in the present study and thus this hypothetical mechanism remains to be tested.

Performance on the mental arithmetic task was not different in sitting and standing postures. This is con- sistent with previous research (Myrtek and Spital 1986; Szabo and Gauvin 1992), which has revealed no differ- ences in mathematical performance in sitting and dur- ing cycling (which may be considered a more attention- distracting event than standing). Furthermore, no dif- ferences in the subjective appraisal of task difficulty, effort exerted to perform well, and self-challenge indi- cated that the observed cardiovascular changes in reac- tivity to mental stress resulted from the postural change and were not attributable to changes in cogni- tive psychological measures.

In conclusion, a postural change affects cardiovascu- lar reactivity to mental stress. Standing alone may elicit greater cardiovascular reactivity than mental arithmet- ic, which is a frequently utilized laboratory mental stressor (Pdronnet and Szabo 1993). Therefore, the presentation of mental stressors in upright body pos- ture creates a combined stressor situation in which car- diovascular reactivity to either of the stressors is mod- ified by the reactivity to the other stressor. The present study draws attention to the fact that the results of studies that have examined the effects of mental stress on cardiovascular variables may be restricted to the or- thostatic position in which those results were ob- tained.

544

Acknowledgements. The author is thankful to Arthur Long, Dan- iel Desrochers, and Andrea Szalai for their technical assistance, and to Dr. Franqois P6ronnet and Dr. Lise Gauvin for their com- ments on various aspects of the manuscript.

References

Bevegard S, Holmgren A, Jonsson B (1960) The effect of body position on the circulation at rest and during exercise, with special reference to the influence on the stroke volume. Acta Physiol Scand 49: 279-298

Brown TG, Szabo A, Seraganian P (1988) Physical versus psy- chological determinants of heart rate reactivity to mental ar- ithmetic, Psychophysiology 25: 532-537

Fillingim RB, Roth DL, Cook EW (1992) The effects of aerobic exercise on cardiovascular, facial EMG, and self-report re- sponse to emotional imagery. Psychosom Med 54:109-120

Furedy JJ (1987) Beyond heart rate in the cardiac psychophysio- logical assessment of mental effort: the T-wave amplitude component of the electrocardiogram. Hum Factors 29:183- 194

Furedy JJ, Heslegrave RJ (1983) A consideration of recent criti- cisms of the T-wave amplitude index of myocardial sympa- thetic activity. Psychophysiology 20:204-211

Godin G, Jobin J, Bouillon J (1986) Assessment of leisure time exercise behavior by self-report: a concurrent validity study. Can J Publ Health 77:359-362

Goldstein IB, Shapiro D (1988) Cardiovascular responses to mental arithmetic and handgrip during different conditions of postural change. Psychophysiology 25:127-136

Hatch PJ, Klatt K, Porges SW, Schroeder-Jasheway L, Supik JD (1986) The relation between rhythmic cardiovascular variabil- ity and reactivity to orthostatic, cognitive and cold pressor stress. Psychophysiology 23 : 48-56

Heslegrave RJ, Furedy JJ (1979) Sensitivities of heart rate and T-wave amplitude for detecting cognitive and anticipatory stress. Physiol Behav 22:17-2.3

Heslegrave RJ, Furedy JJ (1983) On the utility of T-wave ampli- tude: a reply to Schwartz and Weiss. Psychophysiology 20: 702-708

Hijzen TH, Van Der Gugten J, Bouter L (1984) Active and pas- sive coping under different degrees of stress: effects on urina- ry and plasma catecholamines and ECG T-wave. Biol Psychol 18: 23-32

Jamieson J, Howk S (1992) The law of initial values: a four factor theory. Int J Psychophysiol 12:53-61

LeBlanc J, C6t~ J, Jobin M, Labrie A (1979) Plasma catechol-

amines and cardiovascular responses to cold and mental activ- ity. J Appl Physiol 47:1207-1211

Levi L (1965) The urinary output of adrenalin and noradrenalin during pleasant and unpleasant emotional states. Psychosom Med 27: 80-86

Myrtek M, Spital S (1986) Psychophysiological response patterns to single, double, and triple stressors. Psychophysiology 23: 663-671

Obrist PA, Gaebelein CJ, Teller ES, Langer AW, Grignolo A, Light KC, McCubbin JA (1978) The relationship among heart rate, carotid dP/dt, and blood pressure in humans as a func- tion of the type of stress. Psychophysiology 15:102-115

P6ronnet F, Szabo A (1993) Sympathetic resonses to acute psy- chosocial stressor in humans: linkage to physical exercise and training. In: Seraganian P (ed) Exercise psychology: the in- fluence of physical exercise on psychological processes. Wiley, New York, pp 172-217

Rau H (1991) Response of the T-wave amplitude as a function of active and passive tasks and beta-adrenergic blockade. Psy- chophysiology 28:231-239

Roth DL, Bachtler SD, Fillingim RB (1990) Acute emotional and cardiovascular effects of stressful mental work during aerobic exercise. Psychophysiology 27: 694-701

Scher H, Furedy JJ, Heslegrave RJ (1984) Physic T-wave ampli- tude and heart rate changes as indices of mental effort and task incentive. Psychophysiology 21:326-333

Seraganian P, Roskies E, Hanley JA, Oseasohn R, Collu R (1987) Failure to alter psychophysiological reactivity in type A men with physical exercise or stress management programs. Psychol Health 1 : 195-213

Shulhan D, Scher H, Furedy JJ (1986) Phasic cardiac reactivity to psychological stress as a function of aerobic fitness level. Psy- chophysiology 23: 562-566

Sinyor D, Schwartz SG, P6ronnet F, Brisson G, Seraganian P (1983) Aerobic fitness level and reactivity to stress: physiolog- ical, biochemical and subjective measures. Psychosom Med 45: 205-217

Stafford RW, Harris WS, Weissler AM (1970) Left ventricular systolic time intervals as indices of postural circulatory stress in man. Circulation 41 : 485-492

Szabo A, Gauvin L (1992) Mathematical performance before, during, and following cycling at workloads of low and moder- ate intensity. Percept Mot Skills 75:915-918

Weiss T, Del Bo A, Reichek N, Engelman K (1980) Pulse transit time in the analysis of autonomic nervous system effects on the cardiovascular system. Psychophysiology 17:202-207

Wilkinson L (1989) SYSTAT: the system for statistics. Systat Inc., Evanston

Ziegler MG (1980) Postural hypotension. Ann Res Med 31 : 239- 245