INTERNATIONAL JOURNAL OF BEHAVIORAL MEDICINE, 3(3), 266-279 Copyright �9 1996, Lawrence Erlbaum Associates, Inc.

Secretory Immunoglobulin A and Cardiovascular Responses to Acute

Psychological Challenge

Douglas Carroll, Christopher Ring, Jo Shrimpton, Phil Evans, Gonneke Willemsen, and Frank Hucklebridge

Cardiovascular activity and secretory immunoglobulin A (slgA) in saliva were recorded at rest, during a 30-rain computer game task, and during subsequent recovery. Blood pressure (BP) rose and remained elevated during the task and returned to resting levels during recovery. This pressor response was produced by increased total peripheral resistance rather than increased cardiac output. SIgA secretion rate also increased dur/ng the task, although the effect proved significant only toward the end of the task. As such, the data provide preliminary indication that slgA is sensitive to acute psychological challenge in the laboratory. Although correlational analyses revealed that slgA reactions were stable, they were not signifi- cantly correlated with pressor reactions. The influence of task uncertainty was explored by comparing individuals who had previously played the computer game with those who had not. Task-induced increases in BP and slgA were a feature of individuals new to the computer game. In contrast to these novice players, experi- enced players showed minimal increases in BP and no increases in sIgA.

Key words: acute psychological challenge, blood pressure, secretory immunoglobu- lin A (slgA), task uncertainty

Enhanced cardiovascular reactions to acute psychological challenge have been implicated in the etiology and clinical expression of cardiovascular disease (for

Douglas Carroll. Christopher Ring, Jo Shrimpton, and Gonnekc H. M. Willemsen, School of Sport and Exerci~ Sciences, University of Birmingham, Birmingham, England; Phll Evans, Division of Psychology. University of Westminster. London, England; Frank Hueklebridge, Division of Physiology and Health Sciences, University of Westminster, London, England.

We thank Shirley Lambert for performing the enzyme-linked immunoabsorbent (EL|SA) assays. Correspondence concerning this article should be addressed to Douglas Carroll. School of Sport and

Exercise Sciences, University of Birmingham, Edgbaston, Birmingham B [5 2TT, England. E-maih r @sportex,bham.ac.uk.

IMMUNE AND CA ~ DIO VASCUI :AR RESPONSES

review, see Manuck, I994). However, the precise nature of the risk conferred by enhanced reactivity remains to be determined. Nevertheless, it is possible that cardiovascular reactivity acts in a synergistic manner, aggravating the effects of renal (Lovallo & Wilson, 1992), hemostatic (Allen & Patterson, 1995), or immu- nological processes (Mills et al., 1995).

With regard to the last of these possibilities, recent evidence suggests that some measures of immune function are responsive to acute psychological challenges (Cacioppo, 1994; Herbert et at., 1994). Moreover, immunological reactions appear to be more pronounced in individuals exhibiting greater cardiovascular reactivity (Herbert et al., 1994; Manuck, Cohen, Rabin, Muldoon, & Bachen, 1991; Zak- owski, Cohen, Hall, Wollman, & Baum, 1994; Zakowski, McAllister, Deal, & Baum, 1992). Such findings are certainly consistent with the view that the cardio- vascular and immune systems act in concert, and are regulated by interrelated mechanisms. An obvious candidate here is the sympathetic nervous system. For instance, Benschop et al. (1994) found that the changes in cellular immunity associated with acute psychological stress were abolished by sympathetic blockade, as were the changes in cardiac activity.

To date, however, evidence derives from a restricted sample of enumerative and functional cellular immune indices. As yet, secretory immunoglobulins have not been examined in this context, although there are several reasons for expanding the scope of laboratory-based psychoimmunological research to include secretory immunoglobulin A (slgA). First of all, slgA is revealed, in a recta-analysis (Herbert & Cohen, 1993), to be responsive to naturalistic stressors, although more persua- sively for concentration than for secretion rate. However, Zeier, Brauchli, and Joller-Jemelka (1996) recently measured immunoglobulin A secretion rates in air-traffic controllers before and immediately after 100-min work shifts; secretion rates were significantly higher at the end of the shift.

Secondly, McClelland, Ross, and Patel (1985) found that, compared to sub- sequent resting baseline levels, sIgA and noradrenaline were elevated immediately following an academic examination, which suggests that the sympathetic nervous system may be implicated in slgA reactions to acute stress. Studies that simultane- ously record slgA and cardiovascular reactions to laboratory challenges could help clarify sympathetic nervous system involvement.

Thirdly, preliminary empirical evidence exists linking slgA to measures of iltness. Iemmot and McClelland (1989) reported the outcome of a meta-analysis showing that slgA levels were related to the incidence of upper respiratory tract infection, whether measured subjectively or objectively by assessing virus-specific antibody. A more recent review (Evans, Hucklebridge, Clow, & Doyle, I995) cites further evidence, which implicates low levels of slgA in the incidence of ear infections.

Finally, it has been reported that those exceptional individuals who lack slgA tend to suffer more illness, notably respiratory, than normal, although often the slgA deficit is somewhat compensated for by an increase in other immunoglobulins

268 CARROLL, RING. SHRIMPTON. EVANS. WILLEMSEN, HUCKLEBRIDGE

(Kuby, 1994). This study was undertaken to determine whether sIgA is sensitive to acute laboratory challenge, and to examine the relation between the patterns of sIgA and cardiovascular activity.

METHOD

Participants

Twenty-eight male students, with a mean age of 2! years (SD = 1.23), mean weight of 77.18 kg (SD = 9.86), and mean height of 180.25 cm (SD = 6.54), participated in the study. They were informed that the most improved task performance would earn a s book token. In reality, the recipient of the book token was chosen at random.

Cardiovascular Measures

Impedance cardiography (ICG) and electrocardiography (ECG) were recorded using the VU-AMD system (Vrije Universiteit Amsterdam; see Willemsen, de Geus, Klaver, van Doomen, & Carroll, 1996). The ICG and ECG signals were sampled at 250 Hz and 1000 Hz, respectively. The ECG was recorded using disposable pregelled Ag/AgC1 spot electrodes (Marquette Electronics, Inc., Jupiter, FL): The recording electrodes were placed 4 cm above the jugular notch of the sternum and at the apex of the heart over the ninth rib, and the ground electrode was placed above the right iliac crest. The ICG recordings were obtained using a spot electrode configuration. One of the inner recording electrodes was shared with the ECG electrode placed over the jugular notch. The other recording electrode was placed directly over the tip of the xiphoid process of the sternum. The two outer current electrodes were placed on the back, at the base of the neck and over TS--T9 vertebrae, at least 4 cm higher and lower, respectively, than the recording electrodes on the front.

Ensemble averages of the ICG were obtained over 60-sec intervals that were scored and edited using an interactive software program. The first derivative (dZ/dt) of the change in transthoraoic impedance was used to identify the opening of the aortic valve (B-point), the maximum rate of change of blood flow (dZ/dt,,,,), and the closing of the aortic valve (X-point). The onset of left ventricular contraction (R-wave) was obtained from the ECG.

The 60-sec ensemble averages were used to calculate the following measures: heart rate (FIR; bpm) was obtained from the R-R interval using the formula HR -- 60,000 / R-R interval; left ventricular ejection time (LVET; ms) was calculated as the time between the B-point and the X-point; stroke volume (SV; ml), was calculated using the Kubicek et al. (1974) formula SV = rho * (ED/Z0) ~ * LVET *

IMMUNE AND CARDIOVASCULAR RESPONSES 269

dZldtm~, where blood resistivity (rho) = 135 (Ohms.era), ED = electrode distance (cm), and Z0 = basal thoracic impedance (Ohms); cardiac output (CO; l/min), was calculated as the product of HR and SV.

Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were recorded using an Accutraker II auscultatory monitor (Model t04, Suntech Medical Instru- ments, Inc., Raleigh, NC), The monitor also recorded ECG using a spot electrode configuration. The first recording electrode was placed above the right clavicle bone. The second recording electrode was placed adjacent to the ECG electrode of the ICG system at the apex of the heart. A ground electrode was placed over the ninth rib on the fight-hand side, directly opposite the second ECG electrode. A microphone was centered over the brachial artery on the nondominant arm, and the cuff was attached over the microphone. The monitor operated in a dynamic pressure inflation mode: the cuff was inflated to 30 mmHg more than the prev{ous SBP, and then deflated in steps of three mmHg. Mean arterial pressure (MAP; mmHg) was calculated using the formula MAP = DBP + ((SBP-DBP) 1 3). Total peripheral resistance (TPR; dynes/slcm ~) was derived using the formula TPR = (MAP t CO) x 80.

Immunological Measures

SIgA was collected from unstimulated saliva samples using Salivettes (Sartstedt Ltd., Numbrecht, Germany). Participants were instructed to abstain from eating or drinking for the 30 min prior to the session. For each sample, the participant first swallowed to dry his mouth, and then a cotton wooE swab was placed under his tongue, where it remained for 2 min. The swab was then placed in a Salivette tube, which was sealed and immediately frozen at -20 ~ The samples remained frozen until the time of assay.

On thawing, saliva was extracted from the cotton by centrifugation at 3.5 x 10 ~ rpm for l0 rain. The concentration of sIgA in saliva was determined by a double- antibody ELISA. Saliva atiquots (100 ol) were assayed at a dilution of 1 in 1,000. The coating antibody was antihuman IgA (Sigma I9506, Sigma Diagnostics, St. Louis, MO) at 4 pg per well. Assays were performed in triplicate against a range of standards (Human igA Sigma 10633), 0-350 ng/ml. A reference sample was incorporated into each plate. The enzyme conjugate was anfihuman IgA-alkaline phosphatase (Sigma A3400) and the substrate was p. nitrophenyl phosphate (Sigma N2765). A sample was accepted if it met the following two criteria: a coefficient of variation of less than 10% between triplicate repeats, and a standard correlation coefficient greater than .96.

Each 2-rain sample provided a measure of the saliva sIgA concentration (lag/ml) and saliva volume (m]). As recommended by Brandtzaeg, Fjellanger, and Gjeruld- sen (1970) and Herbert and Cohen (1993), secretion rate, rather than concentration, was preferred as the index of sIgA activity. Herbert and Cohen's (I 993) meta-analy- sis identified only two studies that had employed secretion rate, with mixed results;

CNNROEL RING~ $ItN|M~TONi EVANS~ WILEEMSEN~ ~:IUCKLEN|~:ff?GE

they concluded that more were ne.eded. The slgA secretion rate 0ag/min) was calculated from the formula secretion rate = (concentration * volume) t 2.

Procedure

The study consisted of an initial 6-min rest period, a 30-rain computer game task, and a 20-rain recovery period. The task was a commercially available computer game (Doom, I.D. Software, Greeley, CO). Previous research has shown that computer games produce substantial cardiodynamic changes (Miller & Ditto, 1989; Turner & Carroll, 1985). This game required the participant to navigate a series of rooms, and confront hostile creatures that had to be shot and destroyed.

The participant sat in a comfortable chair in a light-attenuated room, and faced a computer monitor. The participant was asked if he had experience with the game, and was then informed how to play the computer game. The participant then practiced for 5 min. This practice period allowed the initial level of difficulty to be tailored for each participant. The electrodes and recording monitors were then attached, signals checked, and the saliva collection protocol explained.

ICG and ECG were recorded continuously during the session. The first BP measurement was initiated manually 1 rain into the rest period, the second 4 rain into the rest period, and the third 1 min into the task. Subsequent BP measurements were taken automatically every 5 min. Saliva samples were taken four times during the study: 4 rain into the rest period, 6 min into the task, 24 min into the task, and 18 rain into the recovery period.

During the recovery period participants were asked to rate, on a 7-point scale, how tense they felt when they started the task, how tense they felt during the task, how difficult they found the task, and how well they performed the task.

Data Reduction and Analysis

The l-rain means for CO were used to calculate an average over the 6-rain rest period, six 5-rain averages over the 30-rain task period, and four 5-rain averages over the 20-min recovery period. The two BP measurements taken during the rest period were used to calculate average resting SBP, DBP, MAP, and TPR. In the task and recovery periods, MAP and CO associated with each 5-rain interval were used to calculate TPR for that interval. Hence, l I values for each cardiovascular variable were available. The sIgA concentration and the saliva volume obtained in each of the four 2-rain samples were used to calculate the sIgA secretion ratethat occurred in intervals 1, 3, 6, and 11 of the study.

A series of one-way repeated measures analyses of variance (ANOVAs) were used to examine each cardiovascular variable (11 intervals), and the sIgA secretion rate (4 intervals) during the rest, task, and recovery periods of the study. All analyses were adjusted using the Greenhouse-Geisser correction: the original degrees of

:[MM U NE AND CAP.DIOV ASCULAR RESPONSES 27~

freedom (F) and the associated epsi}on value (c) are reported. The degrees of freedom vary among analyses as a result of occasional unscoreable data. Significant effects were followed by post hoc comparisons using the Newman-Keuls method. A 5% significance level was adopted in all tests.

RESULTS

SBP and DBP

Figure 1 displays the mean SBP and DBP during the rest, task, and recovery periods. Both SBP and DBP rose at the start of the task, remained elevated throughout the 30-min task period, and returned to resting levels during the recovery period. ANOVA applied to SBP yielded a signifi cant effect for intervals, F(10, 230) = 9.30, p < .05, ~--- 0.3747. Post hoe eornpari sons revealed that SB P was sign ificantly higher during the task relative to both the rest and recovery periods, SBP values did not differ significantly during the task, nor between the rest and recovery periods. Similarly, ANOVA indicated that DBP varied significantly across intervals, F(I 0, 240) = 6.79, p < .05, ~ = 0.5933. Post hoc comparisons revealed that DBP was significantly elevated during the task when compared to rest and recovery.

MAP, CO, and TPR

As would be expected, ANOVA and post hoe comparisons indicated that MAP was higher during the task than duringrest or recovery, F(10, 230) = 11.38, p < .05, e = 0.4676 (see Figure 2, top). In contrast; CO did not change significantly during the sessi on, F( 10, 240) = 3.00, p > .05, e = 0.1951, (see Figure 2, m iddl e). Clearly, the pressor response to the task was driven by peripheral rather than central hemodynamic mechanisms. ANOVA applied to the TPR data yielded a significant effect for intervals, F(10, 220) = 7.33, p < ,05, I~ = 0.2591. TPRs during the rest, task, and recovery periods are displayed in Figure 2, (bottom). Post hoc compari- sons revealed that TPR increased from rest to task. Further, TPR did not return fully to the restiag level during the recovery period.

SlgA

Average sIgA secretion rates are presented in Figure 3. Because of the large interindividual variability in these data, a square root transformation was performed to increase homogeneity of variance (Myers, 1972). ANOVA on the transformed data yielded a significant effect for Intervals, F(3, 66) = 3.24, p < .05, e = 0.6802.

" ~ " i 4 0 I

E 1 3 5

m

m ~25 0

~20. 0

t

i

T �84

i

i i 5 ............ ......

1 t 2 3 ~ 6 I 2 3 4

REST TASK RECOVERY

i E E

" J 8 0 i w

7 5 w

o 70 . 0

R 6 5

iT 0

v L

a !'�84184 .................................. . . . . . . . . . . . . . . . . . . . . .

2 3 4 5 2 3 4

REST TASK RECOVERY

FIGURE 1 Mean (SE) systo]ic blood p~.&~urr (top panel) and diastolic blood pressure (bouom panel) during the rest period (I x 6-rain interval), compuler game ta.~k (6 x 5.rnin intervals), and subsequent recovery period (4 x S-rain intervals).

N

oL 2 3 ~ 5 5 ~ 3

B , 5

�84184184184

t I ! I I I 1 f i ! l 1 1 2 ,.3 4. .5 B 1 2 3 4.

REar TASK R E ~

~176

0

i 2 3 5 5 2 3;

FIGURE 2 Me, an (5"K) arlerla~ pressure (top panel), cardiac output (middle panel), and toted peripheral resistance (bottom panel) during the rest, ta~qk, and recovery periods.

2 7 3

274 CARROLL, RING, SHRIMPTON, EVANS, W[LLEMSEN, HUCKLEBRIDGE

Subsequent post hoc comparisons revealed that the sIgA secretion rate at the end of the task was significantly greater than the rate during rest.

Relation Between Pressor and SIgA Reactions

SBP, DBP, and sIgA reactions to the task were calculated as the difference between resting and task values. It is important to note that individual differences in slgA reaction were stable from early to late in the task, r(21) = .82, p < .05. Pearson correlation coefficients were also computed between the slgA reactions and both the SBP and DBP reactions. These analyses yielded, in all, 24 coefficients, none of which approached the criterion for statistical significance. In other words, individ- ual differences in the pressor reactions to psychological challenge were unrelated to changes in sIgA secretion rate.

Pressor and SIgA Reactions in Novice and Experienced Players

Participants were classified either as experienced players, if they had previously played the computer game (n = 9), or as novice players (n = 17) if they had not. Novice and experienced players did not differ in age, body mass index, and resting physiological activity. Pressor and slgA reactions during the task were then

w

0 E 1 W

1

REST

FIGURE 3

I e

r i

1 2 3 4 5 6 1 2

TASK RECOVERY

Mean(SE) slgA during the rest, task, and recovery periods.

4.

E, perienc~

12- A . . . . . A

' " [ I I T , , , + 1

Q & 6 o_+~+~ [ [ + 1 "' 6 ! ~ ~ i 1 1 ~ 0

+ 2 ; . . . . . . . . ..... = = ~ . . . . + ~ ............................................................. ..................................................................... .......

2 3 4: 5 6

T~K iNTERVAL

!80

+0

!,~___~ 20 O c+

Z r,,.30

~ �84184 <_+,~

Experienced A . . . . A Novice

i

-40 1

-60 4 t 2 3 # 5 6

TASK IN~RVAL

275

276 CARROLL. RING, SHRIMFrl'ON, EVANS, WILLf~MSEN, HUCKLEBR|DGE

analyzed as a function of participants' experience status. ANOVA (2 x 6; Groups • Intervals) on SBP reactivity data yielded a significant Groups x Intervals interaction effect, F(5, 115) = 2.98, p < .05, g = 0.8438. The data arc summarized in Figure 4 (top), which shows that the novices were more reactive than the experienced players during the early and middle stages of the task. Similar analysis performed on the DBP reactivity data indicated that neither main nor interaction effects were statistically significant. ANOVA (2 x 2; Groups x Intervals) on the slgA reactions during the task (see Figure 4, bottom) revealed only a significant main effect for Group, F(1, 21) = 7.38, p < .05. The novices showed substantial increases in slgA secretion rates (M = ~0.95, SD = I 1.34), whereas the experienced players showed decreased secretion rates (M =-20.89, SD = 21.47).

Participants" ratings of task difficulty, performance, and tension did not correlate with either the pressor or the slgA reactions to the task. However, a 2 Group (Novice, Experienced) x 2 Occasions (Start of Task, During Task) ANOVA on the tension ratings indicated that novice players (M = 3.53, SD = 0.80) reported greater tension than experienced players (M= 2.83, SD =0.71) throughout, F(I, 24) = 4.81, p < .05. In addition, all participants reported that they were more tense during the task (M = 4.09, SD = 1.46) than at the start of the task (M = 2.27, SD = 1.06), F( 1, 24) = 20.10, p < .05. Novice players did not differ from experienced players in their ratings of either task difficulty or performance. Correlational analyses between slgA reactions and pressor reactions, applied to the novice group alone, failed to detect any significant relations.

DISCUSSION

The computer game challenge used in this study elicited sustained pressor reactions that appeared to be mediated primarily by peripheral vasoconstrictive mechanisms, Cardiac output was unchanged during the task, whereas TPR increased markedly. In contrast, Turner and Carroll (1985) found substantial increases in HR dxaring a brief computer game. Using a more protracted computer game challenge, Miller and Ditto (1988, 1989) also reported increases in HR that appeared to underlie the BP increases, particularly during the initial stages of the task. However, with continued exposure to the task the pressor response increasingly came under vascular control. In a subsequent pharmacological blockade study, Miller and Ditto (1991) concluded that the increasing contribution of vascular resistance to BP reactivity was a function of increasingly dominant alpha-adrenergic activity.

In the case of slgA in saliva, secretion rates increased significandy by the end of the task relative to the resting rate. This fi nding agrees with the results ofa nu tuber of other studies that report rises in slgA in response to acute naturalistic challettges, such as oral examinations (Evans, Bristow, Hucldebridge, Clow, & Pang, 1994), written examinations (McClelland et al., I985), film presentations (McCtelland & Kirshnit, 1988), and air-traffic control (Zeier et at., 1996). In contrast, studies exploring the link

IMMUNE ~N:D eNRaiovAse~JLaR R~PONSgS 277

between self-reported exposure to more chronic stress and slgA have almost uniformly found an inverse relation, that is, greater life stress associated with reduced levels of slgA (Evans, Bristow, Hucklebridge, Clow, & Waiters, 1993; Graham, Chiron, Battholomeusz, Taboonpong, & La Brooy, 1988; McClelland, Alexander, & Marks, 1982). Taken together, these studies suggest that different mechanisms may mediate acute and chronic regulation of slgA activity (Evans et al., I994).

Recent studies of immune responses to acute laboratory challenge have focused on cellular measures (see Herbert & Cohen, 1993). They paint a reasonably consistent picture. In terms of enumerative measures, most studies report increases in natural killer cells and CD8 ~ cells during the challenge (e.g., Cacioppo et al., 1995; Herbert et al., 1994; Mills et al., 1995). In terms of functional measures, studies generally find a reduced blastogenic response to at least some mitogen provocations during psychological challenge (Herbert et al., 1994; Manuck et al., 1991; Zakowski et al., 1994). Evidence from beta-adrenergie blockade suggests that these changes are mediated by sympathetic nervous system activation (Ben- schop et al. 1994). Additional evidence of sympathetic involvement comes from the finding in some studies that immune responses are greater in those individuals who show greater cardiovascular reactions to the challenge (Cacioppo, 1994; Herbert et al., 1994; Manuck et al., 1991; Zakowski et at., 1992, I994).

However, in this study, correlational analyses, although confirming that immune reactions were reliable, failed to provide any evidence that they were related to the magnitude of the pressor reactions to the computer game challenge. Further, the current pattern of cardiovascular adjustments, although suggestive of sympathetic nervous system involvement, may owe more to alpha-adrenergic than to beta-ad- renergic activation. In this context, it is worth noting that the administration of labetol, a combined alpha- and beta-adrenoceptor antagonist, attenuated the usual cellular enumerative and functional changes associated with acute psychological challenge (Bachen et al., t995), although this could reflect primarily the effects of beta-adrenergie blockade (Benschop et al., 1994). Nevertheless, the immunoglobu- lin enhancement in response to cold stress in mice was found to be blocked by phentolamine, an alpha-adrenoceptor antagonist, but not by propanolol, a beta-ad- renoeeptor antagonist (Carr, Woolley, & Blalock, 1992). Clearly, pharmacological blockade studies are required to determine the physiological mechanisms underly- ing slgA responses to psychological stress in humans. It is also important to determine whether the observed increase in slgA reflects increased release of already preformed immunogEobulin A or local synthesis of immunoglobulin A (Butler, Waidmann, Rossen, Douglas, & Couch, 1970).

Task uncertainty has been shown to be an important factor in determining cardiovascular responseS to psychological challenge (Elliott, 1966; Light & Obrist, 1980). In this study, increases in BP and slgA were a feature of individuals new to the computer game. In contrast to these novice players, experienced players Showed

2 7 8 CARROLL. RING, SHRIMPTON, EVANS, WILLEMSEN, HUCKLEBRIDG15

minimal increases in BP and an increases in sIgA. Task uncertainty, then, which undoubtedly characterizes oraI and written examinations (Evans et al., 1994; McClelland et al., 1985) and the demands of air-traffic control (Zeier et al., 1996), appears to affect immune reactions as well as cardiovascular reactions.

In conclusion, this study showed that an acute psychological challenge that elicits a sustained pressor response also stimulates slgA in saliva. As such, it offers partial support for the view that cardiovascular and immune reactions to psycho- logical challenge are mediated by interrelated mechanisms~ Critically, this study indicates that it is entirely feasible to investigate the effects of stress on sIgA in a controlled laboratory setting using short-term psychological challenges. It also suggests that task uncertainty would be a worthwhile focus for further study.

REFERENCES

Allen, M. T., & Patterson, S. M. (1995). Hemoconcentration and ,~'ess: A review of the physiological mechanisms and relevance for cardiovascular disease risk, Biological P.vychology, 41, 1-28.

Bachen, E. A., Manuck, S. B., Cohen, S., Maldoon, M. F., Raible, R., Herbert, T. B., & Rabin, B. S. (1995). Adrenergic blockade ameliorates cellular immune responses to mental stress in humans. Psychosomatic Medicine, 5Z 366-372.

Benschop, R. J., Nieuwenhuis, E. E. S., Ttoml~, E. A. M., Godea~, G., Ballieux, R., & van Doomen, L. J. P. C 1994). The effects of beta-adrenergic blockade on immunologic and cardiovascular changes induced by mental stress. Circulation, ~9, 762-769.

Brandtzaeg, P., Fjellanger, 1., & Gjeruldsen, S. T. 0970). Human secretory immunoglobulins. I. Salivary secretions from individuals with normal or low levels of serum immunoglobulins. Scandi- navian Journal of Haematolngy, 12(Suppl.), I-g3:

Butler, W. T,, Waldmann, T. A., Rossen, R. D., Douglas, R. G., & Couch, R. B. (1970). Changes in IgA and IgG concentrations in nasal secretions prior to the appearance of antibody during viral respiratccy infection in man. Journal oflramunology, lOS, 584--591.

Cacioppo, J. T. (1994). Social neuroscience: Autonomic, nenro--endocrine, and immune responds to stress, Psychaphysiology, 31, l 13-128.

Cacioppo, J. T., Malarkey, W. B., KiecoFt-Glaser, .1. K., Uchino, B. N., Sgoutas-Emch, S. A., Sheridan, J. F., Bernston, G. G., & Giaser, R. (1995). Heterogeneity in neuroendocrine and immune responses Io brief psychological stressors as a function of autonomic cardiac activation. Psyclmsomatic Medicine, 57, 154-164.

Can', D. J., Woolley, T. W., & Blalock, J. E. (1992). Phentolamine but not propranolol blocks the immnnopotentiating effect of cold stress on antigen-specific lgM production in mice orally immu- nized with sheep red blood cells. Brain, Behavior, and Immuni~., 6, 50--63.

Elliott. R. (I 966). Effects of uncertainty about the nature and advent of a noxious stimulus (shock)upon heart rate. Journal of Personality and Social Psychalogy. 3. 353-356.

Evans, P., Bristow, M., Hucklebridge, F.. Clow, A., & Pang, F,-Y. ([994). Stress, arousal,.eortisol and secretory immunoglobulin A in students undergoing asses~nmnt. British Journal of Clinical Psy- chology, 33, 575--576.

Evans, P., Btistow, M., Hucklebridge, F., Clow, A., & Waiters, N. (t993). The relationship between secretory immunity, mood and life events. British Journal of Clinical Psychology, 32, 227-236.

Evans, P., Hueklebridge, F., Clow, A., & Doyle, A. (1995). Secretory i mmunoglobulin A as a convenient biomarker in health survey work. In L Rodriguez-Marin ted.), Health psychology andquality of life research (Vol, I1. pp. 541-549). Alicante, Spain: University of Alieante Press.

IMMUNE AND CARDIOVASCULAR RESPONSES 27Q

Graham, N. M. H.. Charon, R., Banholoncusz. A., Taboo~pong, N., & La Bmoy, J. T, (1988). Does anxiety reduce the secretion role of secretory ]gA in saliva? The Medical J~Jurnal of Australia, 148, 131 ~[ 33.

Herbert, T. B., & Cohen, S, 0993), Stress and immunity in humans: A recta-analytic review. Pyv. chosomatic Medicine, 55. 364-379,

Hedx:rt, T. B, Cohen. S,, Marsland, A. L,, Beg:hen. E: A , Robin, E. S., Muldo~n, M, F., & Manuel S. B. (1994). Cardiovascular reactivity and the course of immune response to an acute psychological stressor. Pyychosomatic Research. 56, 337-344.

Jemmot, J. M., & MeC[elland, D. C. (1989). Secretory lgA as a measure of resistance to infectious di.,~a~: Comments on Stone, Cox, Valdismatsdottih and Neale. l~ehavioral Medicine, 12, 63-71.

Kubicek, W. G., Kottke, F. J., Ram0s, M. V., Patterson, R. P, Labree, J. W., Remole. W. Layman, T. E., Sehocning, H., & Ga.ramela, J. T. (1974). The Minnesota impedance eardiograph---Theo~ and imp|[cations. Biomedical Engineering, 9, 410-417.

Kuby, 1. ( 1994 ). Immunology (2od ed,; pp. 516--517 ). New Yon/c: Freeman. Light, K. C., & Obrist, P. A. (1980). CardioVascular respon~ to stress: Effects of opportunity to avoid

shock ex~denee and performance feedbacL p~Chot~hysiotogy, 17. 243-252.

Tu~*eL A g a ~ , & L~g~t (Eds~s l~di~Ms~al cr everdm "~g~ede*~ re:;,ovmss to Stress (pp. 26-~280), New York: Plenum.

Manock, S. B. (199g), Cardiovascular reactivity in cardiovascular disease: "Once more unto the breach." International Journal of Behm,ioral Medicine, l, 4-3 l

Manaek, S. B., Cohen, S., Rabin, B, S., Muldoon, M. F., & Baehen. E. A. (1991). Individual differences in cellular immune responses to stress. P,eychologicalScience. 2. I 11-115.

MeClellaad, D. C.. Alexander, C., & Marks, E. (1982). The need for power, stress, immune function and illness among male prisoners. Journal of Abnormal P.~chology, 91, 61-70.

McClelland, D. C.. & Kirshnit, C. (t988). The effect of rtmtiva~ionat arousal through films o~ salivary immunogIobutin A. P.vychology and Health. 2, 31-52.

McClelland, D. C., Ross. G., & Patel, V. (1985). The effect of an academic examination on salivary norepinephrine and immunogiobUlin levels. Journal o/Human Stress. t I, 52-59.

Miller, S. B., & Ditto, B. (1988). Cardiovasctffar responses to an extet~led avcesive video game task. Psychophysiology, 25. 200-208.

Miller, S. B., & DRto, B. (1989). Individual differences in heart rate and peripheral vascular responses to an extended stress task. P~chophysiofogy, 26, 506-513.

Miller, S. B., & Dilto. B. (1991). Exaggerated sympathetic nervou~ system response to extended psychological stress in offspring of hypertensives. Psychophysioh~gy, 28, t03~ I 13.

Mills, P. L, Berry, C, C,, Dimsdale, J. E., Ziegler, M. G,, Nelesen, R. A., & Kennedy, B. P. (1995). Lymphocyte subset redistribution in response to acuteexperimental stregs; Effects of gender, ethaicity, hypertension, and the sympathetic nervous system. Brain, Behavior, and Immunity, 9, 61--69.

Myers, J+ L, (1972). Fundamentals of experimental design. Boston: Allyn & Bacon. Tamer, J. R., & Carroll. D (1985). Heart rate and oxygen consumption during mental arithmetic, a

video game and graded exercise: Further evidence of metabolically-exaggerated cardiac adjust- meats? Psychop~,siology, 22., 261-267.

Wil|em~n, G. H. M., de Geus, E. J. C., Klaver, C. H. A. M., van Doomen, L. J. P., & Carroll, D. (1996). Ambulatory monitoring of the impedance cardiogram. Psychophysiofogy, 33, 184-193.

Zakowski, S. G., Cohen, L., Hall, M. H,, Wollman, K., & Baum, A. (1994), Differential effects of active and passive laboratory stre~qors on ima'nune function in tw, aRhy m~n. In~rnational Journal r Belutvioral Medicine; 1, 163-184.

Zakowski, S. G.. McAllister, C. G., Deal, M., & Bourn, A. (I992). Stress, reactivity and immune function. Health P.W. chology, 11, 223-232.

Zeier, H., 8rauch[i, P., & loller-J'eme[ka, H. |. ([996). Effe~cts of wock demands on immunoglobulia A and r in air traffic controllers. Biological P.~,chotogy, 42, 413--423.