phisiological response anxiety cardiodd

7/30/2019 Phisiological Response Anxiety Cardiodd

1/13

Endocrine and Cardiovascular Responses DuringPhobic AnxietyRANDOLPH M. NESSE, MD, GEORGE C. CURTIS, MD, BRUCE A. THYER,P H D , DAISY S. M C C ANN, P H D ,MARLA J. HUBER-SMITH, BA, ANDRALPH F. KNOPF, MD

In vivo exposure therapy for phobias is uniquely suited for controlled studies of endocrine andphysiologic responses during psychologic stress. In this study, exposure therapy induced sig-nificant increases in subjective anxiety, pulse, blood pressure, plasma norepinephrine, epi-nephrine, insulin, cortisol, and growth hormone, but did not change plasma glucagon or pan-creatic polypeptide. Although the subjective and behavioral manifestations of anxiety wereconsistent and intense, the magnitude, consistency, timing, and concordance of endocrine andcardiovascular responses showed considerable variation.

INTRODUCTIONStress has been implicated in the etiol-ogy of many diseases, including myocar-dial infarction, cancer, and psychiatricconditions (1-3]. With the advent of ra-dioimmunoassay (RIA) and related tech-

niques, reliable measurements of hor-mones in small samples have becomepractical, and endocrine research on themechanisms of stress response has, in turn,been stimulated (4-9). Although manyphysiologic variables are now known torespond to psychologic stimuli, the prin-ciples that organize these responses haveremained elusive. One view is that sub-

From the Departments of Psychiatry and InternalMedicine, TheUniversity of Michigan Medical School,Ann Arbor, MI.Address reprint requests to: Randolph M. Nesse,M.D., Department of Psychiatry, The University ofMichigan Medical School, B2917 C.F.O.B, Box 056,1045 East Ann Street, Ann Arbor, MI 48109.Presented in part at the American PsychosomaticSociety Annual Meeting, March 11, 1984, on HiltonHead Island, SC.Received for publication February 9, 1984; revi-sion received July 30, 1984.

jective, behavioral, and physiologic arousalare tightly linked. In some human studies,however, physiologic responses to psy-chologic stress have proved weak or un-reliable. Research has been hampered bythe difficulty of reliably inducing severeand sustained stress in human subjects ina laboratory setting. For this purpose, wehave proposed exposure therapy for pho-bias as a research tool (10). In this proce-dure, the patient is confronted with theactual feared object (snake, spider, etc.) andis encouraged to approach and touch theobject as rapidly as possible (11). The anx-iety induced is very intense, but patientsare motivated by the knowledge that rel-atively few sessions are usually necessaryfor the treatment to be effective.

In our initial work with this approach,plasma growth hormone was not changedduring adaptation to the laboratory, buttwo thirds of subjects showed some in-crease during treatment (12).Plasma cor-tisol increased during laboratory adapta-tion butshowed no statistically significantresponse to treatment (13, 14).When datafrom individual subjects were analyzedseparately, a few subjects treated during

320 Psychosomatic Medicine Vol . 47, No. 1 (July/August 1985)Copyright 1985 by the American Psychosomatic Society, IncPublished by Elsevier Science Publishing Co , Inc52 Vanderbilt Ave , New York, NY 1001 7

0033-3174/8S/S3.30


2/13

PHYSIOLOGICAL RESPONSES DURING ANXIETY

the early morning showed mild cortisolelevations, whereas subjects treated dur-ing the evening showed no increase at all.Plasma prolactin and thyroid stimulatinghormone levels were not affected by ex-posure therapy (15,16). These results sug-gested three possible explanations: 1) thatthe carefully controlled conditions elimi-nated confounding effects such as posture,exercise, diet, time of day, and novelty,which may have influenced previous stud-ies; 2) that the anxiety induced by expo-sure therapy is fundamentally differentfrom other kinds of psychologic stress; or3) that the various aspects of the stressresponse are not, in fact, closely coupled.The present study was designed to ex-tend this inquiry by incorporating newvariables and more stringent controls, byrandomly assigning subjects to time oftreatment, and by analyzing the data totake full advantage of the simultaneousfrequent measurement of multiple vari-ables. Pulse and blood pressure were mea-sured because they are known to increaseduring anxiety (17, 18) and because theymay mediate disease. Subjective anxietywas measured with two widely used scales.Plasma epinephrine (19-22), norepineph-rine (19-22), growth hormone (12, 23), andcortisol (13, 24) were measured becausethey also are known to increase duringstress. This protocol allows study of theintensity, timing, and coordination of theirresponses to acute psychologic stress un-der strictly controlled experimental con-ditions. The influence of stress on insulinhas been uncertain (25) but of obvious im-portance for understanding glucose regu-lation. Glucagon and pancreatic polypep-tide are the other two islet cell hormoneswhose plasma levels mainly reflect pancrea-tic secretion (26, 27).Both increase duringhypoglycemia, and glucagon releasemight be adaptive during stress because

it induces glycogenolysis and gluconeo-genesis (26, 27). Repeated simultaneousmeasurements of these 12 variables duringperiods of rest and anxiety make possiblea detailed and coordinated view of the re-sponses to psychologic stress.

METHODSTen women requesting treatment from the Uni-versity of Michigan Phobia Clinic were selected forstudy according to the following criteria. All hadsimple animal phobias rated 4 (severe) or 5 [verysevere) on the 5-point Gelder and Marks Phobia Se-verity Scale (28) and had no other psychiatric dis-orders. They were 2543 years old, took no medi-cation, and reported good physical health and normalmenstrual cycles. History, physical examination, andblood analyses (screening panel, CBC, T3, T4) con-firmed their good health. All gave informed consent.The research protocol included four 3-hr sessionsfor each subject. Sessions 1 and 4 were control pe-riods in which subjects sat and read quietly. Sessions2 and 3 were treatment sessions: during the middlehour of these sessions subjects received rapid in vivoexposure therapy for their phobias; during the firstand third hours subjects again sat and read. Subjectsunderstood this schedule before the protocol began.Individual sessions were scheduled at weekly inter-vals, always at the same time of day. Stage in themenstrual cycle was not a factor in session sched-uling. Between-session variance may have been in-creased for this reason, but systematic bias is un-likely. To control for circadian effects, half of thesubjects were randomly assigned to morning ses-sions, the other half to evening sessions. Researchsessions started 3 or 15 hr after the individual sub-

ject's mean time of mid sleep. Starting times wereapproximately 6:00 A.M. OT 6:00 PM Subjects wereinstructed not to eat, smoke, exercise, or drink caf-feinated or alcoholic beverages for 7 hr before eachsession and they sat quietly for at least 20 min beforethe start of each session.The 12 dependent variables chosen for study wereSubjective Units of Distress (SUDS), state anxiety,pulse, systolic and diastolic blood pressure, plasmaepinephrine, norepineph rine, growth hormone, cor-tisol, insulin, glucagon, and pancreatic polypeptide.The SUDS scale is a self-rating of subjective anxietyon a scale of 0 ("no anxiety") to 100 ("the most anx-ious it is possible to feel") (29). State anxiety was

Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985) 32 1


3/13

R. M. NESSE et al.

measured using the Spielberger State Anxiety Inven-tory (30). Pulse was measured by palpating the radialartery for 1 min. Blood pressure was measured usinga standard sphymomanometer. Data for all variableswas obtained at the start of each session (time 1) andevery 20 min thereafter, except for state anxiety, whichwas rated hourly.Blood samples were taken via a needle insertedinto an antecubital vein at the beginning of each ses-sion (time 1) and kept patent by a slow normal salineinfusion. Blood samples were placed immediatelyinto chilled tubes containing glutathione and eth-ylene glycol tetraacetic acid (for epinephrine andnorepinephrine assay) or heparin (for other assays).Plasma was rapidly separated using a refrigeratedcentrifuge, and aliquots were frozen at -70C. Nor-epinephrine and epinephrine concentrations wereassayed by a modification of an enzymatic single-isotope derivative procedure that is accurate at con-centrations greater than 20 pg/ml (31). For norepi-nephrine, the within-assay coefficient of variation(WACV) is 3.9% and th e between-assay coefficientof variation (BACV) is 10.7%, at 300 pg/ml. For epi-nephrine, the WACV is 8.6% and the BACV is 17.9%at 85 pg/ml. Growth hormone was assayed by an RIA

with a sensitiv ity of 17 pg/tube, a WACV of 8.1%, aBACV of 7.8%, and 50% inhibition at 393 pg. Insulinwas assayed with an RIA sensitive to 0.1 u,U/tube,with a WACV of 6.4%, a BACV of 8.4%, and 50%inhibition of 2.2 (JLU. Thegluc agon RIA was sensitiveto 2 pg/tub e, had a WACV of 7.5%, a BACV of 10.2%,and 50% inhibition at 59 pg. The pancreatic poly-peptide RIA was sensitive to 2.9 pg/tube, had an WACVof 3.5%, a BACV of 9.1 %, and 50% in hibitio n at 66pg. The cortisol assay by competitive protein bindingwas sensitive to 1.0 mg/dl. with a WACV of 7.0%, aBACV of 10.0%, and 70%-80% binding of cortisol.The data were transformed to standardized scoresbased on each individual subject's standard devia-tion from her median for all 40 measurements ofeach variable (standardized score = raw value -med ian -= SD). The med ian was chosen as the in-dicator of centra] tendency because some variableshad skewed distributions, and because the medianbest reflected baseline levels for labile variables. Thismethod of analysis minimizes the variance resultingfrom baseline differences between subjects, equalizesthe contributions of subjects to the analysis for thosevariables where subjects show widely differing base-line levels or amounts of variation, allows morningand evening results to be conveniently analyzed bothtogether and separately, and allows uniform graphicand statistical analysis of diverse variables and in-teractions between variables. For each variable, a one-

way analysis of variance (ANOVA) was used to com-pare the mean of all standardized values obtainedduring treatment periods to the mean of all stan-dardized values obtained during nontreatment pe-riods. These analyses compared, for each variable,the mean of all values for all subjects during anxietyperiods (six per subject; three values obtained in eachof the two treatment hours) to the mean of all othervalues obtained (34 per subject).

RESULTSDuring exposure therapy, all subjectsmanifested intense anxiety by their verbalreports, by our observation of their agita-tion, tremors, piloerection, and crying, andby their scores on the SUDS and StateAnxiety scales. Ten of the 12 variablesshowed a treatment effect significant atp < 0.0001. Only glucagon and pancreatic

polypeptide did not respond. Figure 1 il-lustrates the response for each variable.Table 1 summarizes the statistical analysisof these results.All variables were compared on thestrength of response to the anxiety-pro-voking stimulus . Strength of response canbe calculated in three different ways; eachhas different implications (Table 2). First,response strength can be estimated by thecorrelation ratio (T)2, eta-squared), whichestimates the percent of the total varianceexplained by the treatment effect. Thisranking method incorporates four factors:the amount, the consistency, and the tem-poral specificity of a variable's response totreatment, and the relative stability ofbaseline levels. A second system ranksvariables according to the percent of datapoints that are above that subject's overallmedian for that variable; here the ordermainly reflects the reliability of response.A third perspective is provided by rankingthe variables according to the mean per-cent-increase during anxiety, above mean

32 2 Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985)


4/13


S.D.

EPINEPHRINE

/ / \

NOREPINEPHRINE GROWTH HORMONE

5 6 7 8 9 10 2 3 4 5 6 2 3 0 5 6 7 8 9 10

2 3 4 5 6 7 8 9 10 6 7 8 9 10STATE ANXIETY PANCREATIC POLYPEPTIOE

2 3 4 5 6 7 8 9TIME

56TIME

2 3 4 5 6 7 B 9 1 0T I M E

C on t ro l Session 2 1Session 3 J Exper imenta l

Fig. 1. Mean standardized scores forall subjects, at each time in each session. Y axis in allcases is scaledin standard deviation units. The X axis is divided into nine intervals of 20 min each. Anxiety-inducing treatment occurred during sessions 2 and 3 only, where it began at time 4 and ended attime 7, as indicated by the shading.

baseline levels. This widely used method is best appreciated if multiple simultane-reflects the absolute magnitude of the re- ous samples formany variables are avail-sponse but is relatively insensitive to the able foranalysis.reliability of response and the stability of Repeated simultaneous measurementsbaseline levels. The relative ranking ofeach ofmultiple variables also make it possiblecomponent of a stress response depends to compare the relative timing of variouson how response strength is denned. This responses (see Fig. 1). Pulse and SUDS in-Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985) 32 3


5/13

R. M. NESSEetal.

TABLE 1. Summ ary of Means and ANOVAs"

SUDSPulseSystolic BPNorepinephrineState an xietyEpinephrineInsulinDiastolic BPCortisolGrowth hormonePancreaticpolypeptideClucagon

ExperimentalTimes

N = 6 per subjectMEAN SD1 944 1.196(63.8)1 434 1.068(85.4 bpm)1.174 1.030(115.6 mm Hg)1.190 1.190(466.9 pg/ml)1.080 1.260(46.9)0.898 1.360(127.0 pg/ml)b0.828 0.985{12.2 mil/ml)0.704 1.117(73.2 mm Hg)0.664 1.041(117 mg/dl)0.922 1.323(5.5 ng/ml)0.128 1.010(93 3 pg/ml)0.201 1.131(43.5 pg/ml)

Control TimesN = 34 per subjectMEAN SD

0.122 0.698(20.3)-0.005 0.831(74.4 bpm)- 0 . 1 1 6 0.858(105.5 mm Hg)-0.046 1.216(336.2 pg/ml)- 0 . 0 1 0 0.0823(39.4)0.056 0.864(107.9 pg/ml)b0.030 0.953(11.5 mU/ml)- 0 . 0 5 8 0.942(67.8 mm Hg)-0.005 0.971(9.4 mg/dl)0.310 0.897(3.2 ng/ml)0.349 0.998(97.3 pg/ml)0.028 0.979(39.7 pg/ml)

F189.09131.4191.1271 7123.01129.8628.2526.4521.3019.19

2.121.21

P


6/13


crease in an ticipation of the treatment pe-riod, peak in mid-treatment, and begin de-creasing before treatment ends. Stateanxiety increases markedly before andduring treatment. The other variables, withthe possible exception of norepinephrinein session 3, demonstrate a notable lack ofresponse immediately before treatment.Blood pressure, norepinephrine, cortisol,and insulin increase promptly when pho-bic anxiety begins, and stay elevated dur-ing treatment. The rapid insulin increase,and the low or negative correlations of in-sulin values with values of other hor-mones during treatment suggest that in-sulin response is not secondary to otherhormone response. More detailed work isneeded, however, to better consider the

relationships between insulin and otherresponses. Growth hormone remains atbasal levels for at least 20 min after treat-ment begins and peaks only at the end ofthe hour of treatment, a finding that is con-sistent with the pattern of growth hormoneresponse to other stimuli (33). At time 8,20 min after treatment has ended, all vari-ables except growth hormone and insulinhave returned substantially toward base-line. Forty minutes after treatment, allvariables are back to baseline levels.The circadian effect on responses dur-ing treatment was assessed by comparing,for each variable, all standardized scoresobtained during treatment times for morn-ing subjects to those for evening subjects(see Fig. 2) . Responses for the two groups

I l Morning subjectsEvening subjectsISUDS PULSE SVS BP NOREPI. EPI. INSU LIN DIA. BP COR T. G.H.

Fig. 2. Mean scores during treatment for morn ing subjects vs. evening subjects.Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985) 32 5


7/13

R. M . NESSE et al .

were comparable except that, when com-pared to evening subjects using a one-wayANOVA, morning subjects had, duringtreatment, higher blood pressure scores(systolic: F = 3.32, p < 0.07; diasto lic:F = 5.95, p < 0.02) and much lower cor-tisol scores (F = 15.39, p < 0.0003}. Theresponse of cortisol to stress during thecircadian period of maximal secretion isminimal (F = 3.24, p < 0.07) compared tothe response in the evening (F = 27.50,p < 0.0001). The mean increase duringstress for morning subjects was from 12.4to 13.8 (i-g/dl, while evening subjects in-creased from 5.6 to 9.2 u.g/dl.

A similar technique was used to com-pare anxiety responses in session 2 to thosein session 3 (Fig. 3). Mean scores duringtreatment averaged 31% higher in session

2, except for systolic and diastolic bloodpressure, which responded more in ses-sion 3. The difference in response betweensessions 2 and 3 was not statistically sig-nificant for any variable (all p > 0.10). Thegeneral decline in response from session2 to session 3 may reflect the rapid efficacyof flooding treatment, and suggests thatpulse and hormone responses decreaseconcordantly with anxiety. The failure ofblood pressure to follow this pattern is dif-ficult to explain. Controlling for the men-strual cycle stage might have clarified theseresults.

Despite the clear effect of exposure ther-apy on 10 of the 12 dependent variables,it was not the case that all subjects showeda reliable and substantial increase in allvariables during the entire treatmen t time.

SUDS PULSE SYS. BP NOR EPI. EPI. INSULIN DIA. BP CORT. G HFig. 3. Mean scores during treatment in session 2 vs. session 3.

326 Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985)


8/13


Instead, examination of graphs of individ-ual variables for individual subjects (forone example, see Figure 4) reveals rela-tively few instances of sustained eleva-tions of multiple hormones throughouttreatment. More often, a hormone peaksrelatively sharply during some treatmenttimes and is near baseline at other treat-ment times. The increased amplitude andfrequency of these peaks, during treat-ment, averaged for many subjects, com-prise the seemingly consistent elevationsin Figure 1.An individual subject's peaks for dif-ferent variables sometimes concur, butveryoften do not. Confirming this relative lackof concordance are the generally low cor-relations between variables for all simul-taneous values obtained during treatmentperiods (see Table 3). For example, epi-nephrine and growth hormone values dur-ing treatment are poorly correlated

(r = 0.043), despite the fact that eachstrongly responds to phobic anxiety. Theonly variables whose values are signifi-cantly correlated during treatment(p < 0.05) are those of norepinephrine withgrowth hormone, epinephrine, and corti-sol. If any hormone can be regarded as"central" to the stress response in this study,it is norepinephrine. Anxiety explains moreof its variance than that of any other hor-mone, and it is the only hormone whosevalues during treatment are significantlycorrelated with more than one other hor-mone.We examined the data to determine ifthose individuals who showed a particu-larly strong response in one hormone alsotended to respond strongly in another hor-mone (Table 4). A response strength foreach variable for each subject in both treat-ment sessions was estimated by the cor-relation ratio from a separate ANOVA that

Con trol Session

2

D

_

7~.i

s. /

1

t;\! \i i

I

T reatment Sessior

/

1 1

A/ \

A/ ^uX

1

\

\ \ ,-V. H"S\\ I''

1 1 11 2 3 4 5 6 7 8 9 10

Times ^ Growth Horn . _ Epinephrme^ ^ ^ ^ - Norepinephri

Cortisol

Fig. 4. Endocrine responses in a control and a treatment session for a single subject. Anxiety was inducedduring times 4-7 of sessions 2 and 3. Y axis is scaled in standard deviation units.Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985) 32 7


9/13

R. M. NESSE et al .

TABLE 3. Correlations Between Variables for Values During Anxiety Periods in Individual SubjectsPulseSystolic BPNorepinephrineEpinephrineInsulinDiastolic BPCortisolGrowth hormone

0.086- 0 . 1 1 50.0540.110- 0 203- 0 . 1 6 2- 0 2050.093SUDS

1.0000.311 a0.027- 0 . 0 0 20.0220.0730.316a- 0 . 1 8 7Pulse

1.000- 0 . 0 1 40.000- 0 . 1 4 20.311"0.0080.160

SystolicB.P.

1.0000.396"0.101-0.405-"0.319a0.398b

Norepinephrine

1.000- 0 . 3 1 0- 0 . 1 9 60.2290.043

Epinephrine

1.000- 0 . 0 1 50.1120.188

Insulin

1.000- 0 . 2 2 80.134DiastolicB.P.

1.000- 0 . 1 3 8Cortisol

"p < 0.05; bp < 0.0 1, uncorrected for the number of correlations performed.

TABLE 4. Corre lations Between Response Strengths in Pairs of Variables in Individu al SubjectsPulseSystolic BPNorepinephrineEpinephrineInsulinDiastolic BPCortisolGrowth hormone

0.3950.0310.6400.279- 0 . 2 8 4- 0 . 3 8 80.443

- 0 . 2 8 6SUDS

1.0000.4340.421- 0 . 3 9 50.5370.2600.003

- 0 . 3 9 0Pulse

1.000- 0 . 2 0 30.0310.4250.6260.116

- 0 . 3 8 3SystolicB P.

1.0000.779*- 0 . 5 8 5- 0 . 7 2 00.0600.203Norepinephrine

1.000- 0 . 2 8 4- 0 . 3 8 80.443

- 0 . 2 8 6Epinephrine

1.000- 0 . 5 4 1- 0 . 1 2 7- 0 . 4 3 7Insulin

1.000- 0 . 3 0 7- 0 . 1 8 6DiastolicB.P.

1.000- 0 . 3 4 6Cortisol

"p =s0.023, uncorrected for the number of correlations performed.

compared the 6 treatment values to the34control values. These response strengthswere then correlated. Only epinephrine andnorepinephrine responses significantlypredicted one another (r = 0.779,p 0.023).Correlations between the strengths ofresponse for each individual in each vari-able in session 2 with that in session 3 (seeTable 5) were performed to see if a sub-ject's relative response strength in a vari-able in one session could predict the re-sponse in another session. The resultssuggest that subjects may have individ-ually consistent patterns of responsedur-ing separate episodes of stress a week apart,especially for cortisol, pulse, and insulin,but a stronger conclusion will require astudy designed to answer this question.

TABLE 5. Corre lations Between the Strength of aSubject's Response in Session 2 with that Subject's

Response in Session 3

CortisolPulseInsulinSUDSDiastolic BPEpinephrineGrowth hormoneSystolic BPNorepinephrine

N998787987

r0.440.420.350.290.260.14- 0 . 0 2- 0 . 1 8- 0 . 1 9

DISCUSSIONSome of these results directly confirmprior work whereas some are quite sur-prising. As expected, pulse, blood pres-

328 Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985)


10/13


sure, epinephrine, norepinephrine, corti-sol, and growth hormone all increasedduring anxiety, thus confirming m uch pre-vious work (4-8, 17-24) in a setting thatallows the rigorous control of experimen-tal conditions during acute and intensepsychologic anxiety. The definite increaseof insulin deserves emphasis because pre-vious studies have been inconclusive (25).The failure of glucagon and pancreaticpolypeptide to respond is of interest be-cause both increase during hypoglycemicstress (26, 27) and because their clear ab-sence of response helps to validate thepositive resu lts for other variables. The def-inite responses of these ten variables tophobic anxiety increase confidence in ourprevious reports, based on a similar method,that plasma TSH and prolactin are notchanged by phobic anxiety (15, 16).These results help to explain why ourprevious work on cortisol and growth hor-mone showed weaker effects than someother studies. The reaction of both hor-mones during stress is particularly clearin the present study because of the con-trolled conditions, the large number of datapoints, and because the data transforma-tion efficiently eliminated variance fromsubject's differences in baseline levels, sothat even small individual responses arereflected in the analysis. When comparedto other hormones, however, cortisol andgrowth hormone responses are relativelyinconsistent, even though the magnitudeof increase is substantial when peaks dooccur.The subjects were all women with se-vere, specific, simple phobias. Though theywere in all other ways healthy and typicalof the general population, the results can-not necessarily be generalized to men orto people without phobias. The stress em-ployed was particularly acute and intenseanxiety relatively uncontaminated by other

emotions or physical factors. The body maywell respond differently to stress that ischronic or induced by other emotions orsituations.Could anxiety-induced changes in glu-cagon and pancreatic polypeptide havebeen missed because their half-lives inplasma are only about 5 min (26, 27)? Al-though single brief peaks might escape de-tection, repeated peaks would have beenreflected in the analysis, as they were forthe catecholamines which have evenshorter half-lives (33). Cortisol (34) andgrowth hormone (35) are cleared slowlyenough from plasma that every substantialsecretory burst should be observed withthe 20-min sampling interval.Not surprisingly, the SUD measure re-sponded most strongly to exposure ther-apy, no matter how response strength wasestimated. The ranking of other variablesdepended substantially on the measure ofresponse strength used. The catechol-amines consistently ranked high, whereascortisol and growth hormone, traditionalstress hormones, were the two lowestranked variables in the analyses that em-phasized reliability of response. Thoughgrowth hormone showed the third highestmean percent-increase during stress, datafrom any of the other responding variableswould better predict treatment and non-treatment times.

The timing of various aspects of the re-sponse is relatively uncoordinated. Onlypulse and anxiety levels were elevated inanticipation of treatment. Growth hor-mone showed a characteristic delayed re-sponse and, along with insulin, had ele-vations sustained past the end of treatment.It is clear, even from these group data, thatthe stress response is not a simple on-offphenomenon.The rapid growth of knowledge aboutcircadian effects (36) makes considerationPsychosomatic Medicine Vol. 47, No. 4 (July/August 1985) 329


11/13

R. M. NESSEeta l .

of their influence essential to stress re-search. The comparable responses of mostvariables at the circadian extremes was ex-pected, but the minimal cortisol responsein the morning conflicts w ith previous work(13, 14) and cannot be regarded as a firmconclusion, although blunted cortisol re-sponses might be expected at times whenlevels are already high. Systolic and dia-stolic blood pressure were the only otherdivergent variablesboth were more re-sponsive in morning subjects. If con-firmed, this finding has implications forstudies involving blood pressure.When responses in session 2 were com-pared to those in session 3 (Fig. 3), bloodpressures again opposed the trendtheywere the only variables that respondedmore in session 3 than in session 2. Thoughno firm conc lusions can be d rawn from thelimited data available, the possibility thatblood p ressure may mediate stress-relateddisease draws attention to these divergentpatterns.Several interesting results emerge fromthe patterns of interactions between vari-ables (Table 4). The streng th of growth hor-mone response was negatively correlatedwith that of every other variable exceptnorepinephrine. The strength of a sub-ject's norepinephrine response predicts thestrength of epinephrine responsethis isnot the case for any of the other 35 pairsof variables. This correlation suggests thatlinked mechanisms may control the re-lease of norepinephrine and epinephrineduring anxiety.Subjects who had a cortisol, insu lin, orpulse response during session 2 tended toshow the same pattern in session 3. It wouldbe possible to determine if individuals havecharacteristic patterns of response by ex-panding the same method to involve fouror more episodes of treatment. This mightbegin to explain why different individua ls

develop different symptoms in responseto stress.Finally, subjects did not show reliable,substantial, and sustained increases inmultiple variables during treatment. Theendocrine responses to psychologic stressmay be definite, but they are not reliable,sustained, and coordinated in the way thata simple theory of stress might predict.Instead, they consist of relatively incon-sistent, brief and seemingly uncoordi-nated changes. The elegant analyses byWard et al. (20) of individual subjects'variable catecholamine responses to a va-riety of stimuli point to the same conclu-sion. The concept of "s tress" as a consist-ent pattern of response to a variety ofstressors needs to be considered in the lightof these findings.

SUMMARYEndocrine, cardiovascular, and subjec-tive responses during psychologic stresswere studied by using exposure therapyfor phobias to induce intense acute anxi-ety. This technique allowed rigorous con-trol of confounding variables, and madepossible the simultaneous measurement of11 variables at 10 times during each of four3-hr sessions. Ten healthy female subjectswith severe simple phobias received rapid

exposure treatment during the middle hoursof sessions 2 and 3. All experienced severeanxiety during this treatment.Results were transformed to standardscores. Scores during treatment w ere thencompared to scores during control periodsusing one-way ANOVAs. Ten variablessignificantly incr eas ed (F > 19.0,p < 0.0001) during the anxiety periods. Thedesign and analysis make it possible tocompare and list these variables in orderof response strength as measured by the33 0 Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985)


12/13


proportion of variance explained by theanxiety periods: subjective anxiety, pulse,systolic blood pressure, norepinephrine,state anxiety, epinephrine, insulin, dia-stolic blood pressure, cortisol, and growthhormone. The marked insulin response isof note because previous studies have beeninconclusive on this point. Glucagon andpancreatic polypeptide did not respond.Compared to evening subjects, morningsubjects showed much less cortisol re-sponse and somewhat more blood pres-sure response. Time of day did not affectother responses. Responses were de-creased in session 3, compared to session2, except for blood pressure. Graphs of meanvalues show the relative timing as well asthe magnitude of responses in each vari-able. Graphs of individual's responses re-veal that sustained substantial responsesin multiple variables were the exceptionrather than the rule. For different vari-

ables, peaks very often did not concur. Re-sponse patterns were more complex andvariable than expected. These results con-firm and extend our knowledge of the en-docrine and cardiovascular responses dur-ing psychologic stress, and they illustratethe advantages of a model that employsphobic anxiety as a stressor.

We thank the University of MichiganClinicaJ Research Center and its staff fortheir assistance, Dr. Daniel Rourke and theUniversity of Michigan Statistical Re-search Laboratory for statistical consul-tation, Dr. Bernard /. Carroll and Dr. OliverG. Cameron /or support and advice, thestaff of the Word Processing Center in theUniversity of Michigan Department ofPsychiatry, for manuscript preparation,and the proprietors of Ann Arbor Pet Sup-ply for their indispensable assistance.

REFERENCES

1. Riley V: Psychoneuroendocrine influences on immunocompetence and neoplasia. Science 212:1100-1109,19812. Steptoe A: Psychological factors in cardiovascular research. New York, Academic, 19813. Anisman H, Zacharko RM: Depression: The predisposing influence of stress. Behav Brain Sci 5: 84-137,19824. In Levi L (ed), EmotionsTheir Parameters and Measurement. New York, Raven, 1975, pp 143-1825. Selye H (ed]: Selye's Guide to S tress Research. New York, Van Nostrand R einhold, 19806. Rose RM, Jenkins CD, Hurst M: Air Traffic Controller Health Change Study. Galveston, Texa s, Universityof Texas Medical Branch, 19787. Ursin H, Baade E, Levine S (eds): Psychobiology of Stress. New York, Academic, 19788. Mason JW, Maher JT, Harley LH, et al: Selectivity of corticosteroid and cate cholamin e respon ses tovarious natural stimuli. In Serban G [ed), Psychopathology of Human Adaptation. New York, Plenum,1976, pp 147-1729. Mason JW: A historical view of the stress field. J Hum Stress 1: 6-12, 2 2-3 6, 197510. Curtis GC, Nesse RM, Buxton M, et al: Flooding in-vivo as a research tool and treatment for phobias.Compr Psychiatry 17: 153-16 0, 197611. Marks I: Behavioral treatments of phobic and obsessive-compulsive disorders: a critical appraisal. InHersen M, Eisler RM, Miller PM (eds), Progress in Behavior Modification. New York, Academic, 1975,Vol 1, pp 66-15812. Curtis GC, Nesse RM, Buxton M, et al: Plasma growth hormone: Effect of anxiety during flooding-in-vivo. Am J Psychiatry 136: 410-11 4, 1979

Psychosomatic Medicine Vol. 47, No. 4 (July/August 1985) 331


13/13

R. M. NESSE et al.

13. Curtis GC, Nesse RM, Buxton M, et al: Anxiety and plasma cortisol at the crest of the circadian cycle:reappraisal of a classical hypothesis. Psychosom Med 40: 368-378, 197814. Curtis GC, Buxton M, Lippman D, et al: "Flooding in vivo" during the circadian phase of minimalcortisol section: Anxiety and therapeutic success through adrenal cortical activation. Biol Psychiatry11: 101-107, 197615. Nesse RM, Curtis GC, Brown GM: Phobic anxiety does not affect plasma levels of thyroid stimulatinghormone in man. Psychoneuroendocrinology 7: 69-74, 198216. Nesse RM, Curtis GC, Brown GM, et al: Anxiety indu ced by flooding therapy for phobias does not e licitprolactin secretory response. Psychosom Med 42: 25-31, 198017. Moss A, Wynar B: Tachycardia in house officers presenting cases at grand roun ds. Ann Intern Med 72:255-256, 197018. Bonelli J: Stress, catecholamines and beta-blockade. Acta Med Scand [Suppl] 660: 214-218, 198219. Dimsdale JE, Moss J: Plasma c atecholamines in stress and exercise. JAMA 243: 340-342, 198020. Ward MM, Mefford IN, Parker SD, et al: Epinephrine and norepinephrine responses in continuouslycollected plasma to a series of stressors. Psychosom Med 45: 471486, 198321 . Frankenhauser M: Experimental approach to the study of catecholamines and emotion. In Levi L fed],EmotionsTheir Parameters and Measurement. New York, Raven, 1975, pp 209-23422. Asian S, Nelson L, Carruthers M , et al: Stress and age effects on catecholam ines in normal subjects. JPsychosom Res 25: 33-41, 198123. Miyabo S, Asato T, Mizoshima N: Psychological correlates of stress-induced cortisol and growth hor-mone releases in neurotic patients. Psychosom Med 41: 515-523, 197924. Vernikos-Danellis J, Heybach JP: Psychophysiologic mechanisms regulating the hypotha-lamic-pituitaryadrenal response to stress. In Selye H (ed), Selye's Guide to Stress Research. NewYork, Van Nostrand Reinhold, 1980, pp 206-25125. Lustman P, Carney R, Amado H: Acute stress and metabolism in diabetes. Diabetes Care 4: 658-659,198126. Porte D, Halter J: The endocrine pancreas and diabetes mellitus. In Williams RH (ed), Textbook ofEndocrinology, 6th ed. Philadelphia, Saunders, 1981, pp 716-84327. Lonovics J, Devitt P, Watson LC: Pancreatic polype ptide. Arch Surg 116: 1256-1264, 198128. Gelder MG, Marks IM: Severe agoraphobia: a controlled prospective trial of behavior therapy. Br JPsychiatry 112: 309-319, 196629. Wolpe J: The Practice of Behavior Therapy, 2nd ed. New York, Pergamon, 197330. Spielberger CD: The measurement of state and trait anxiety; conceptual and methodological issues. InLevi L (ed), EmotionsTheir Param eters and Measurement. New York, Raven, 1975, pp 713-72531. McCann DS, Huber-Smith MJ: Plasma catecholamines. J Clin Immunoassay 6: 308-312, 198432. Hays WL: Statistics, 3rd ed. New York, Holt, Rinehart, and Winston, 1981, p 34933. Cryer PE: Physiology and pathophysiology of the human sympathoadrenal neuroendocrine system. NEngl J Med 30 3: 436-444, 198034. Hellman L, Nakada F, Curtis J, et ai: Cortisol is secreted episodically by normal m an. J Clin Endocrinol30: 411-422, 197035. Daughaday WH: The adrenohypophysis. In Williams RH (ed), Textbook of Endocrinology, 6th d.Philadelphia, Saunders, 1981, pp 73-11636. Moore-Ede MC: The Clocks That Time Us: Physiology of the Circadian Timing System. Cambridge,MA, Harvard University Press, 1982

332 Psychosomatic Me dicin e Vo l. 47 , No . 4 (July/August 1985)

phisiological response anxiety cardiodd

Documents