Correlation of Anxiety and Hostility With Adrenocortical Function
Study of Two Psychologic Variables in Patients With Lung Cancer Theodore Wohl, PhD, Leon J. Sholiton, MD, and Emile E. Werk, Jr., MD, Cincinnati
D URING THE COURSE of their disease, pa- tients with bronchogenic carcinoma manifest adrenocortical hyperfunction as measured by sev- eral parameters. One of the most notable indices of this hyperactivity appears to be the concentration of morning plasma unconjugated 17-hydroxycorti- costeroids (17-OHCS) which incrcase as the dis- ease progresses and the interval to death shortens. Although several metabolic aspects might be sus- pected of influencing the adrenocortical status of these seriously ill patients, a possible factor to be considered is their emotional state, as numerous studies of patients with cancer have indicated a wide range of emotional response. Of the many emotional factors to be considered, anxiety has been most closely associated with changes in adrenocortical function,1, 2 probably through neuro- endocrine pathways affecting corticotrophin release.
The purpose of the present study has been to explore this possible relationship by correlating plasma 17-OHCS levels with anxiety in patients with and without lung cancer. The factor of hos- tility was also included in the study as a means of further validating one of the evaluative instru- ments,3 and has assumed some interest in view of the previous work of Funkenstein and his col- leagues demonstrating the close correlation of hos- tility to the adrenal secretion of epinephrine and norepinephrine.4 In performing this investigation, we have attempted to be explicit and detailed re- garding our selection of cancer patients, the nature of control groups, the nature of measuring instru- ments, terminology, and statistical analysis.3
Fourteen ambulatory alert male patients with in- operable bronchogenic carcinoma represent the group of patients with cancer. All were aware of their diagnosis and were defined ( as well as clinical definition will allow) as to type, location, extent, duration, and survival time from date of the study. An equal number of ambulatory alert males with chronic, moderately stabilized, non-neoplastic dis-
From the Metabolism Section and Psychology Service, Veterans Administration Hospital, and the University of Cincinnati College of Medicine.
Read before the Section on Endocrinology and Bone at the 111th Annual Meeting of the American Medical Association, Chicago, June 28, 1962.
eases were selected to represent the control group. By using male patients from a single public institu- tion, with essentially the same degree of mental alacrity, age, and socioeconomic status, we have hoped to arrive at a fairly homogeneous sampling. All patients were screened carefully for such factors as previous or present psychiatric disorders or drug therapy, which is known to interfere with 17-OHCS determination. Procedure was designed as follows:
Session 1 .- Blood was drawn at 8 am for the initial 17-OHCS (pretest assay). Psychological evaluation ensued from 9 to 10 am, after which a second blood sample was obtained for 17-OHCS (post-test determination). In approximately 2 weeks, a repeat study in an identical fashion was performed.
Session 2 .- Repeat testing afforded an oppor- tunity to evaluate the reliability of these methods and to note changes in responses over a set period. Plasma unconjugated 17-OHCS levels were deter- mined by a modification of the Nelson and Samuels method.6
Five psychometric parameters were utilized in evaluating aspects of anxiety and hostility as fol- lows: (1) A clinical rating scale utilizing such objective findings as motor and autonomic nervous system activity 7; (2) The Institute for Personality and Ability Testing Scale (IPAT),8 utilizing several sources: for evaluating anxiety; (3) The Verbal Anxiety Scale 3; (4) The Verbal Hostility-Inward Scale and the Verbal Hostility-Outward Scale.” The last 3 measures were specifically designed as a brief objective measure of an individual’s emotional level or psychological state at any particular time. These methods are considered particularly appropriate in that they reflect even slight variations in anxiety state, and do not weight too heavily somatic evi- dences of anxiety of one group over the other. The tests were administered by 1 of 3 psychiatrists without previous contact with the subjects and without knowledge of their diagnoses. All psycho- metric measurements for a given patient were made by the same examiner during both testing sessions.
Since multiple variables are obviously included in
the experimental design, these data have been sub- jected to analysis of variance and analysis of co- variance. Among the most pertinent results is the confirmation of previous studies from this labora- tory 9 that a significant difference exists between the plasma 17-OHCS level of the cancer group and the control group both before and after the psycho- metric test period. The decrease in average con- centration of plasma 17-OHCS for pretest and post-test was statistically significant for all persons and sessions, attesting to the presence of diurnal variation. This analysis also demonstrated that, in the control patients, the drop in plasma levels be- tween Session 1 and Session 2 was statistically sig- nificant. A possible hypothesis is that this disparity might be accounted for by a difference in psycho- logical state on the 2 occasions. Accordingly, an analysis of covariance was performed for this group between the average pre- and post-test 17-OHCS level and each of the psychological variables for the 2 sessions. This analysis revealed no significant cor- relation between hydroxycorticosteroid values and anxiety. A possible explanation for this greater de- crease of plasma values in the control group, is that the 2-week interval resulted in improvement in the status of these patients, whereas progression in the cancer patients may have occurred during the same period. It is clear that there is no significant dif- ference in all psychological modalities measured between the cancer and noncancer group, ie, it is concluded that anxiety, and “hostility in” and “hos- tility out” cannot account for the consistent findings of elevated basal plasma unconjugated 17-OHCS found in lung cancer patients in this and previous studies.1º The measure of “hostility outwards” was correlated significantly to the over-all average 17-OHCS level of the subjects, but did not account for differences from session to session. Only “hos- tility inwards” obtained a correlation which was consistent both within and between persons, and thus could account, to a small extent, for the dif- ference in hydroxycorticosteroid levels between the 2 sessions. Analysis of covariance also was per- formed using psychological measures and the aver- age pre- and post-test 17-OHCS levels of both groups in the first session. Only “hostility outwards” and “hostility inwards” yielded a substantial cor-
relation within groups. Neither accounted for the difference between cancer and control groups.
The significant correlation of hostility with the over-all level of 17-OHCS would suggest that this psychological factor, in some as yet unknown man- ner, could affect the physiologic response of the anterior pituitary or the adrenal cortex. Moreover, a highly significant correlation between hostility-in and the verbal anxiety scale indicates that hostility perhaps is elemental in certain measurements of anxiety noted by other workers who have attempted to demonstrate a correlation between anxiety and adrenocortical function.
Chief, Inpatient Psychology, Veterans Administration Hospital, Cincinnati (Dr. Wohl).
We wish to acknowledge the assistance of Dr. Goldine Gleser, Associate Professor of Psychology, University of Cincinnati College of Medicine, in the statistical evaluation. We also wish to thank Drs. Irvin Milowe, Gilbert Morrison, and John Stiefel, Department of Psychiatry, University of Cincinnati, and Members of the Social Service Department, Cincinnati VA Hospital, for their participation in this study.
References
1. Bliss, H. L., et al .: Reaction of Adrenal Cortex to Emotional Stress, Psychosom Med 18:56, 1956.
2. Price, D. B .; Thalter, M .; and Mason, J. W .: Pre- operative Emotional States and Adrenocortical Activity: Studies on Cardiac and Pulmonary Surgery Patients, Arch Neurol 77:646, 1957.
3. Gleser, G. C .; Gottschalk, L. A .; and Springer, K. J .: Anxiety Scale Applicable to Verbal Samples, Arch Gen Psychiat 5:593, 1961.
4. Funkenstein, D. H .; Greenblatt, M .; and Solomon, H. C .: Nor-epinephrine-like and Epinephrine-like Sub- stances and Psychotic and Psychoneurotic Patients, Amer J Psychiat 108:652, 1952.
5. Perrin, G. M., and Pierce, I. R .: Psychosomatic Aspects of Cancer, Psychosom Med 21:397, 1959.
6. Nelson, D. H., and Samuels, L. P .: Method for De- termination of 17-Hydroxycorticosteroids in Blood; 17-Hy- droxycorticosterone in Peripheral Circulation, J Clin Endocr 12:519, 1952.
7. Gottschalk, L. A., et al: Effects of Perphrenezine on Verbal Behavior Patterns, Arch Gen Psychiat 2:632, 1960.
8. Catell, R. B., and Scheir, I. H .: The Meaning and Measurement of Neuroticism and Anxiety, New York City: Ronald Press Co., 1961.
9. Sholiton, L. J .; Werk, E. E., Jr .; and Marnell, R. T .: Diurnal Variation of Adrenocortical Function in Non- endocrine Disease States, Metabolism 10:632, 1961.
10. Werk, E. E., Jr., and Sholiton, L. J .: Adrenocortical Function in Carcinoma of Lung, Cancer 13:469, 1960.
S’ YDENHAM AND GOUT .- He began to suffer from Gout before his thirtieth year; and to this was afterwards added the still more painful disease, Calculus. The attacks were certainly frequent, for in the summer of 1660 (when thirty- seven years old) he was laid up for some months with a severer attack of Gout than he had ever suffered from before, and it was accompanied for the first time by Haematuria. In the early part of 1677 he had a very severe attack of Gout, and symptoms of stone, which made driving in a carriage extremely painful to him. These troubles lasted for three months, and compelled him to retire into the country till the autumn .- Payne, J. F .: Thomas Sydenham, New York: Longmans, Green & Co., 1900.