Prolactin and Growth Hormone Responses to Psychological Stress in Normal and Neurotic Subjects SUSUMU MIYABO, TADASHI ASATO, AND NORIAKI MIZUSHIMA Department of Medicine, Kanazawa University, Kanazawa, Japan ABSTRACT. In order to study the response of plasma prolactin (PRL) to acute psychological stress and to compare it with that of growth hormone (GH), the mirror drawing test (MDT) was performed in 20 normal controls (11 male, 9 female) and 22 neurotic patients (12 male, 10 female). Plasma PRL and GH were measured serially before, during and after the test. In controls, the test caused no significant change in plasma levels of either hormone. In neurotic males, the response of PRL to the test was not consistent, whereas, in neurotic females, plasma PRL
level rose significantly following the test. Increase of GH, on the other hand, was apparent in the neurotics of both sexes. The correlation between the responses of the two hormones in the neurotics was low and non-significant. The results indicate that although the psychoendocrine coping mechanism in the neurotics works less effectively for both PRL and GH, the two hormones may have different psychological correlates. (/ Clin Endocrinol Metab 44: 947, 1977)
S
All subjects were medically healthy, non-obese and without medication for at least one month prior to the examination. All women, except one normal (age 53) and one neurotic (age 51), were regularly menstruating and the tests were performed within one week following menstrual flow. Informed consent was obtained from all subjects. The test procedures were as described in a previous paper (3). Briefly, physical activity was kept minimal before the test and a standard meal was served at noon. At 1400 h, an 18 gauge needle was inserted into a forearm vein. After 40 min rest in bed, the subjects were given 2 min instructions and then started tracing a test figure (five pointed star) on the apparatus as fast and as Materials and Methods accurately as possible with an electric stylus Twenty normal subjects and 22 neurotic sub- for 8 min, while looking at the mirror reflection jects were studied. Normal subjects included 11 of the figure. After the test, the subjects returned men (34.2 ± 9.6 yrs, mean ± SD) and 9 women to bed rest for an additional 30 min. Serial blood (34.5 ± 8.9 yrs). Neurotic subjects consisted of samples were obtained at —40, - 2 0 , 0 (start of 12 men (37.3 ± 10.8 yrs) and 10 women (37.0 instructions), 10,15,20,25,30 and 40 min. Plasma ± 9.8 yrs). There were no significant age dif- PRL was measured according to the method of ferences among the four groups. The neurotic Sinha et al. (8), using human PRL radioimmunopatients were diagnosed by two psychiatrists, assay kit (VLS #3) kindly supplied by NIAMD. who were not concerned with the experiment, The limit of sensitivity of the assay was 1.0 ng/ml according to the DSM II criteria (4). The psycho- of plasma. The intra- and inter-assay coefficients logical tests: The Cornell Medical Index (CMI) of variation, using plasma containing 8.4 ng/ml of (5), Minnesota Multiphasic Personality Inventory PRL, were 8.9% and 12.3%, respectively. All (MMPI) (6) and Taylor's Manifest Anxiety Scale samples from one subject were run simul(MAS) (7) were used to corroborate the clinical taneously. The data for each time point were impressions. Patients for whom another diagnosis analyzed by the Wilcoxon test and the Mannwas suspected were excluded. The characteristics Whitney U test (9). Analysis of variance and covariance (10) was performed for the main of neurotic patients are given in Table 1. source of variation in the overall responses. The association between the responses of the two Received September 8, 1976.
ERUM prolactin (PRL) is well known to rise after a variety of somatic stimuli (1,2). However, there has been no definite evidence relating the effect of purely psychological stress to plasma PRL level in man. In a previous study, we demonstrated that the mirror drawing test (MDT) caused a definite rise in plasma growth hormone (GH) in neurotic subjects, but not in control subjects (3). The same technique was used to explore the response of PRL and to compare it with that of GH.
947
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 22:00 For personal use only. No other uses without permission. . All rights reserved.
948
JCE & M • 1977 Vol 44 • No 5
MIYABO, ,ASATO AND MIZUSHIMA TABLE
Age (yr)
Diagnoses
31M 37M 40M 44M
Anxiety Anxiety Anxiety Anxiety Anxiety Depressive Depressive Depressive Depressive Hypochondriacal Hypochondriacal Other
57M 23M 41M 42M 46M 21M 42M
23M 34F 35F
1. Patients' characteristics CMI* Emotional disturbance + + + + + + + +
+ + +
+ + +
Anxiety Anxiety Anxiety Anxiety Depressive Depressive Depressive Hypochondriacal Neurasthenic Other
44F
51F 20F 36F
48F 31F 25F 34F
+ +
+ + + + + +
MMPI-Profilef 137-9 027-X 127-X 139-2 027-X 079-X 123-X 178-X 689-0 135-4 137-X 016-4 13-4 135-0 145-0 027-4 135-0 348-0 359-X 134-9 13-0 348-0
139-8
MASf 28 33 18 17
32 38 21 38 32 135-7 016-8 13-9 135-4 027-6 136-6 348-9 13-7 348-5
14 44 29 19
29 21 31
15 5 32 10 12 30
* For criteria of emotional disturbance see reference 5. t For interpretation of profile code see reference 6. | See reference 7.
hormones was tested by the Spearman's coefficient of rank correlation (9).
15,
Results
10
V-1-'.... P< 0.0 B
.MOT
-40
-20
0 10 20 30 40-40
-20
E0 30 40 6 10 20
FlG. 1. The effect of the mirror drawing test on plasma PRL (mean ± SE). Dashed line; normal control, solid line; neurotic. *F < 0.05, **? < 0.01; statistical significance versus 0-min values. C vs. N; the comparison between the control and neurotic groups.
No apparent change in PRL level was noted in the control subjects of both sexes throughout the experiment. In the neurotic males, the MDT caused no consistent change in plasma PRL. Although the mean values were higher in the neurotic males compared to their controls, the difference was not significant at any sampling time. In the neurotic women, on the other hand, plasma PRL was higher upon entering the room and fell significantly (P < 0.05) after 40 min rest. The MDT produced a significant rise from the pre-test level (P < 0.05) and the difference from the control female values also became significant after the test (P < 0.05) (Fig. 1). GH levels were quite stable in the control subjects of both sexes. But, in contrast to PRL, both male and female neurotics
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 22:00 For personal use only. No other uses without permission. . All rights reserved.
PSYCHOLOGICAL STRESS AND PRL AND GH
showed definite increases in plasma GH after the MDT (P < 0.05 or less), confirming the previous report (3) (Fig. 2). The data on basal values and the responses of both hormones to the test—maximal level (Max), maximal increment from basal level (Max A) and sum of increments from basal level (XA)—are listed in Table 2. Since the values of postmenopausal women were comparable with those of menstruating women, their data were included for analysis. Analysis of variance showed that the basal hormone levels tend to be higher in the neurotic subjects (P < 0.1 for PRL, P < 0.05 for GH). To remove the bias introduced by the differences in basal level (11), analysis of covariance was performed for evaluation of hormonal responses. Significant differences (P < 0.05) existed between the control and neurotic groups for all parameters of both hormone responses, except XA PRL (P < 0.1). Sex was not a major source of variation. But the mean PRL response was larger in women than in men (P < 0.1 for Max and Max A, P < 0.05 for XA), whereas the sex difference in GH response was far below the significant level. The correlation coef-
949
OH ng/ml
151
10
-40^2
0 10 20 30 40 -40
-20
0 10 20 30 40
FIG. 2. The effect of the mirror drawing test on plasma GH (mean± SE). Dashed line; normal controls, solid line; neurotic. *P < 0.05, **P < 0.01; statistical significance versus 0-min values. C vs. N; the comparison between the control and neurotic group.
ficients between the responses of the two hormones in the neurotics was uniformly low (r = 0.01 for Max, r = 0.17 for Max A and r = 0.06 for XA) and non-significant.
TABLE 2. Response of PRL and GH to the MDT
Basal level
Maximal level (Max)
Male (M)
Female
4.4 ± 0.6
5.2 ± 1.3
(N)
7.0 ± 1.7
Cos. N
F = 3.74 F = 0.60 F = 0.07
NS
F = 2.88 F = 0.82
Control (C) Neurotics
M vs. F
Interaction
Maximal increment (Max A) Male (M)
Male (M)
Female
(F)
PRL, ng/ml (Mean ± SE) 4.8 ± 0.4 6.5 ± 1.2 0.3 ± 0.4
1.3 ± 0.7
-3.4 ± 2.8
-2.3 ± 5.1
8.5 ± 1.9
9.2 ± 2.2
13.6 ± 3.1
2.2 ± 1.0
5.1 ± 1.4
-5.7 ± 6.3
13.8 ± 5.0
P
level rose significantly following the test. Increase of GH, on the other hand, was apparent in the neurotics of both sexes. The correlation between the responses of the two hormones in the neurotics was low and non-significant. The results indicate that although the psychoendocrine coping mechanism in the neurotics works less effectively for both PRL and GH, the two hormones may have different psychological correlates. (/ Clin Endocrinol Metab 44: 947, 1977)
S
All subjects were medically healthy, non-obese and without medication for at least one month prior to the examination. All women, except one normal (age 53) and one neurotic (age 51), were regularly menstruating and the tests were performed within one week following menstrual flow. Informed consent was obtained from all subjects. The test procedures were as described in a previous paper (3). Briefly, physical activity was kept minimal before the test and a standard meal was served at noon. At 1400 h, an 18 gauge needle was inserted into a forearm vein. After 40 min rest in bed, the subjects were given 2 min instructions and then started tracing a test figure (five pointed star) on the apparatus as fast and as Materials and Methods accurately as possible with an electric stylus Twenty normal subjects and 22 neurotic sub- for 8 min, while looking at the mirror reflection jects were studied. Normal subjects included 11 of the figure. After the test, the subjects returned men (34.2 ± 9.6 yrs, mean ± SD) and 9 women to bed rest for an additional 30 min. Serial blood (34.5 ± 8.9 yrs). Neurotic subjects consisted of samples were obtained at —40, - 2 0 , 0 (start of 12 men (37.3 ± 10.8 yrs) and 10 women (37.0 instructions), 10,15,20,25,30 and 40 min. Plasma ± 9.8 yrs). There were no significant age dif- PRL was measured according to the method of ferences among the four groups. The neurotic Sinha et al. (8), using human PRL radioimmunopatients were diagnosed by two psychiatrists, assay kit (VLS #3) kindly supplied by NIAMD. who were not concerned with the experiment, The limit of sensitivity of the assay was 1.0 ng/ml according to the DSM II criteria (4). The psycho- of plasma. The intra- and inter-assay coefficients logical tests: The Cornell Medical Index (CMI) of variation, using plasma containing 8.4 ng/ml of (5), Minnesota Multiphasic Personality Inventory PRL, were 8.9% and 12.3%, respectively. All (MMPI) (6) and Taylor's Manifest Anxiety Scale samples from one subject were run simul(MAS) (7) were used to corroborate the clinical taneously. The data for each time point were impressions. Patients for whom another diagnosis analyzed by the Wilcoxon test and the Mannwas suspected were excluded. The characteristics Whitney U test (9). Analysis of variance and covariance (10) was performed for the main of neurotic patients are given in Table 1. source of variation in the overall responses. The association between the responses of the two Received September 8, 1976.
ERUM prolactin (PRL) is well known to rise after a variety of somatic stimuli (1,2). However, there has been no definite evidence relating the effect of purely psychological stress to plasma PRL level in man. In a previous study, we demonstrated that the mirror drawing test (MDT) caused a definite rise in plasma growth hormone (GH) in neurotic subjects, but not in control subjects (3). The same technique was used to explore the response of PRL and to compare it with that of GH.
947
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 22:00 For personal use only. No other uses without permission. . All rights reserved.
948
JCE & M • 1977 Vol 44 • No 5
MIYABO, ,ASATO AND MIZUSHIMA TABLE
Age (yr)
Diagnoses
31M 37M 40M 44M
Anxiety Anxiety Anxiety Anxiety Anxiety Depressive Depressive Depressive Depressive Hypochondriacal Hypochondriacal Other
57M 23M 41M 42M 46M 21M 42M
23M 34F 35F
1. Patients' characteristics CMI* Emotional disturbance + + + + + + + +
+ + +
+ + +
Anxiety Anxiety Anxiety Anxiety Depressive Depressive Depressive Hypochondriacal Neurasthenic Other
44F
51F 20F 36F
48F 31F 25F 34F
+ +
+ + + + + +
MMPI-Profilef 137-9 027-X 127-X 139-2 027-X 079-X 123-X 178-X 689-0 135-4 137-X 016-4 13-4 135-0 145-0 027-4 135-0 348-0 359-X 134-9 13-0 348-0
139-8
MASf 28 33 18 17
32 38 21 38 32 135-7 016-8 13-9 135-4 027-6 136-6 348-9 13-7 348-5
14 44 29 19
29 21 31
15 5 32 10 12 30
* For criteria of emotional disturbance see reference 5. t For interpretation of profile code see reference 6. | See reference 7.
hormones was tested by the Spearman's coefficient of rank correlation (9).
15,
Results
10
V-1-'.... P< 0.0 B
.MOT
-40
-20
0 10 20 30 40-40
-20
E0 30 40 6 10 20
FlG. 1. The effect of the mirror drawing test on plasma PRL (mean ± SE). Dashed line; normal control, solid line; neurotic. *F < 0.05, **? < 0.01; statistical significance versus 0-min values. C vs. N; the comparison between the control and neurotic groups.
No apparent change in PRL level was noted in the control subjects of both sexes throughout the experiment. In the neurotic males, the MDT caused no consistent change in plasma PRL. Although the mean values were higher in the neurotic males compared to their controls, the difference was not significant at any sampling time. In the neurotic women, on the other hand, plasma PRL was higher upon entering the room and fell significantly (P < 0.05) after 40 min rest. The MDT produced a significant rise from the pre-test level (P < 0.05) and the difference from the control female values also became significant after the test (P < 0.05) (Fig. 1). GH levels were quite stable in the control subjects of both sexes. But, in contrast to PRL, both male and female neurotics
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 22:00 For personal use only. No other uses without permission. . All rights reserved.
PSYCHOLOGICAL STRESS AND PRL AND GH
showed definite increases in plasma GH after the MDT (P < 0.05 or less), confirming the previous report (3) (Fig. 2). The data on basal values and the responses of both hormones to the test—maximal level (Max), maximal increment from basal level (Max A) and sum of increments from basal level (XA)—are listed in Table 2. Since the values of postmenopausal women were comparable with those of menstruating women, their data were included for analysis. Analysis of variance showed that the basal hormone levels tend to be higher in the neurotic subjects (P < 0.1 for PRL, P < 0.05 for GH). To remove the bias introduced by the differences in basal level (11), analysis of covariance was performed for evaluation of hormonal responses. Significant differences (P < 0.05) existed between the control and neurotic groups for all parameters of both hormone responses, except XA PRL (P < 0.1). Sex was not a major source of variation. But the mean PRL response was larger in women than in men (P < 0.1 for Max and Max A, P < 0.05 for XA), whereas the sex difference in GH response was far below the significant level. The correlation coef-
949
OH ng/ml
151
10
-40^2
0 10 20 30 40 -40
-20
0 10 20 30 40
FIG. 2. The effect of the mirror drawing test on plasma GH (mean± SE). Dashed line; normal controls, solid line; neurotic. *P < 0.05, **P < 0.01; statistical significance versus 0-min values. C vs. N; the comparison between the control and neurotic group.
ficients between the responses of the two hormones in the neurotics was uniformly low (r = 0.01 for Max, r = 0.17 for Max A and r = 0.06 for XA) and non-significant.
TABLE 2. Response of PRL and GH to the MDT
Basal level
Maximal level (Max)
Male (M)
Female
4.4 ± 0.6
5.2 ± 1.3
(N)
7.0 ± 1.7
Cos. N
F = 3.74 F = 0.60 F = 0.07
NS
F = 2.88 F = 0.82
Control (C) Neurotics
M vs. F
Interaction
Maximal increment (Max A) Male (M)
Male (M)
Female
(F)
PRL, ng/ml (Mean ± SE) 4.8 ± 0.4 6.5 ± 1.2 0.3 ± 0.4
1.3 ± 0.7
-3.4 ± 2.8
-2.3 ± 5.1
8.5 ± 1.9
9.2 ± 2.2
13.6 ± 3.1
2.2 ± 1.0
5.1 ± 1.4
-5.7 ± 6.3
13.8 ± 5.0
P
