- Email: [email protected]
Atenolol treatment of late luteal phase dysphoric disorder
Journal of Affective Disorders, 15 (1988) 141-147
141
Elsevier J A D 00564
Atenolol treatment of late luteal phase dysphoric disorder J e f f r e y L. R a u s c h 1 D a v i d S. J a n o w s k y 2, S h a h r o k h G o l s h a n 1, K a r e n K u h n 1 a n d S.C. R i s c h ~ I Department of Psychiatry M-O03, San Diego VAMC, School of Medicine, University of California, San Diego, La Jolla, CA 92093, U.S.A. and 2 Department of Psychiatry, University of North Carolina, Chapel Hill, NC, U.S.A. (Received 1 October 1987) (Accepted 11 March 1988)
Summary Activation of the renin-angiotensin-aldosterone system has been hypothesized as a potential pathophysiological factor in premenstrual tension syndrome (PMS). Atenolol is a predominate ill-blocker which can decrease plasma renin activity and inhibit the urinary excretion of aldosterone. Sixteen women meeting provisional diagnoses of late luteal phase dysphoric disorder were treated for symptoms of PMS with atenolol (50 mg once daily) in a randomized placebo-crossover double-blind design. The data indicated significant improvements on the irritability, vigor, elation, and friendliness scores in response to atenolol compared to placebo. Significant changes were not found for several other ratings scales, indicating that atenolol improved only selected symptoms in the group as a whole. However, the women who had premenstrual tension symptoms for more than 5 years (n = 8) were improved on most of the rating scales. Atenolol decreased premenstrual plasma aldosterone to a limited extent. There was also a trend in the data toward higher luteal progesterone levels during the month subjects took atenolol. Plasma renin activity and aldosterone correlated with estrogen and progesterone levels during the placebo month but not during the active month.
Key words." Beta-blockers; Atenolol; Premenstrual tension; Late luteal phase dysphoric disorder; A1dosterone; Progesterone
Introduction Several hypotheses have been constructed to explain a biological etiology of premenstrual af-
Address for correspondence: Jeffrey L. Rausch, M.D., Department of Psychiatry M-003, UCSD, La Jolla, CA 92093, U.S.A.
fective symptoms (Rausch and Janowsky, 1982). One such theory holds that activation of the renin-angiotensin-aldosterone system may be an etiologic factor (Janowsky et al., 1973). Support for the hypothesis derives from indications that plasma renin activity and aldosterone may increase during the luteal phase of the menstrual cycle (Gray et al., 1968; Reich, 1972; Katz and Romth, 1972) as well as from findings that
0165-0327/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
142 menstrual cycle changes in negative affect may covary with urinary s o d i u m / p o t a s s i u m ratios (Janowsky et al., 1973). Although studies have differed on whether premenstrual tension syndrome (PMS) patients have elevated premenstrual serum aldosterone levels compared to controls (for review, see Rausch and Janowsky, 1982), it is possible that some women may have greater behavioral sensitivities to ordinary luteal increases in mineralocorticoid activity. If premenstrual tension were related to a sensitivity to excess mineralocorticoid activity, a selective/~-blocking drug such as atenolol, which could inhibit the excess mineralocorticoid activity (Aberg, 1974; Sassard et al., 1976) could theoretically attenuate the increase in the premenstrual behavioral symptoms related to mineralocorticoidinduced changes in salt and water balance. No previous such studies are available. Here we report the results of a pilot, doubleblind, placebo-crossover trial of low-dose atenolol. Atenolol was chosen as a predominate/~-blocker which could decrease plasma renin activity, and inhibit the urinary excretion of aldosterone (Hespel et al., 1986). The study's purpose was to assess which, if any, PMS symptoms would respond to the low dose of atenolol. To specify the dependent variables prior to the experiment, three hypotheses were formulated about the effects of atenolol in premenstrual tension sufferers. Firstly, it was postulated that irritability would improve. Irritability has been found to be the most severe premenstrual symptom (Magos et al., 1987) and it has been postulated that irritability may reflect subtle changes in sodium and water balance in the central nervous system (e.g., Smith, 1976). Furthermore, propranolol has been effectively used for the control of irritable behavior following acute brain damage (Elliott, 1977). Secondly, it was theorized that atenolol may work for premenstrual tension by virtue of its /3-adrenergic properties in the sympathetic nervous system. Since fl-blockers have been used with some degree of efficacy in the treatment of anxiety disorders, it was postulated that decreases in anxiety ratings might be found. Atenolol has less lipid solubility than propranolol, and crosses the blood-brain barrier less readily, but it does reach the CNS to at least some limited extent, and
several signs and symptoms of anxiety may be manifest through involvement of the peripheral sympathetic nervous system. Thirdly, the possibility that depression would increase in these subjects was considered. It is known that the side effects of fl-blockers include depression (Griffin and Friedman, 1986), and it was hypothesized that this effect might potentially be seen as a side effect in the women with PMS symptoms. Method
Atenolol (50 mg once daily) was evaluated as a treatment for various symptoms of premenstrual tension in 16 women in a 2-month randomized, placebo-crossover, double-blind design. Subjects were recruited for the study from campus advertising and consisted largely of students and employees at the University of California (San Diego) and the San Diego Veterans' Administration Medical Center. Subjects ranged in age from 23 to 46 (mean = 33.8 + 6.5 SD). Before entry into the study, subjects were given a physical exam, and their medical and psychiatric history was taken. To meet entrance criteria for the study, the women had to have regular menstrual cycles ( + 3 days) for 3 months prior to entry into the study, had to be physically healthy, without previous psychiatric histories, and had to meet Carroll criteria for a diagnosis of premenstrual tension (Steiner et al., 1980). The intensities of premenstrual symptoms for the 2 previous months were recorded by Carroll criteria, and subjects were observed prospectively for premenstrual tension symptoms for 1 month prior to entry into the study. Carroll criteria symptoms were used to generate a mean severity assignment by an a priori decision to assign those patients with a score of 18-27, out of a highest possible score of 36, a severity rating of moderate; those patients with a score higher than 27 were rated as severe. Only those subjects who had an intensity rating of at least moderate for both of the previous 2 months were entered into the study. All patients met provisional D S M - I I I R diagnostic criteria for late luteal phase dysphoric disorder. Behavioral measures and blood samples were collected twice weekly. Mood rating instruments
143
included the Beck Depression Inventory (BDI), the Menstrual Distress Questionnaire (MDQ) (Moos, 1984), the Self-Rating Scale (SRS) for PMS (Steiner et al., 1980), the Profile of Mood States (POMS), Hamilton Rating Scale for Depression (HRSD), and Hamilton Anxiety Scale (HAS). Blood samples were assayed for progesterone, estrogen, aldosterone, and plasma renin activity. All subjects were given atenolol 50 mg or placebo once each morning for 10 days before the estimated start of menses. Since it could only be approximated when menstruation would begin, it occurred that some women took study drug capsules 11 days before menstruation. The study drug was administered in all cases through the fourth day after the onset of menses. The subjects were randomly assigned to either placebo or active drug the first month, in double-blind fashion, and changed to placebo or active drug during the second month. The atenolol and placebo capsules were identical in appearance. Baseline behavioral scores were derived from an average of the two mid to late follicular phase
PMTS CHANGE SCORE
visits occurring 14-22 days before menstruation. These were then compared to the mean of the three premenstrual visits occurring between 1 and 11 days before menstruation. The latter three visits were in the premenstrual phase of the cycle when subjects were taking capsules. Results
All mean premenstrual change scores except for POMS depression, M D Q autonomic regulation and control were globally lower with atenolol than with placebo; however, most of the items did not reach statistical significance. Multivariate analysis of variance (MANOVA) indicated significant differences between active and placebo months only in the premenstrual changes on the irritability score of the PMTS, and three of the POMS scores. These findings indicated that atenolol significantly ( P < 0.02) prevented premenstrual reductions in vigor, elation, and friendliness (df= 1, 14, F = 7.26, 6.90, 7.51) and significantly ( P <0.02) diminished premenstrual irritability (df= 1, 14, F = 10.01) in comparison to placebo (Fig. 1). No other
M D Q MEAN CHANGE SCORES
POMS CHANGE SCORES O
I~active
I--Iplacebo
Y, o
1 ~
o
~
oo
~
2
~ I~
o
O I
~ <
um
o)
a 0
m
~
o
D
u_
Z
13_
Fig. 1. The changes in symptoms between the follicular phase and premenstrual phase (see text) are illustrated for the active (dark bars) and placebo months (open bars). Increases in negative affects and symptoms are graphed as positive values, and decreases in favorable affects are graphed as negative values.
144 TABLE 1 SIDE EFFECTS OF ATENOLOL
vs. P L A C E B O
All side effect w e r e r a t e d as m i l d except for d e p r e s s i o n , o n e case o f active d r u g dizziness, o n e case of active d r u g h e a d a c h e a n d o n e case o f p l a c e b o h e a d a c h e , all o f w h i c h w e r e r a t e d as m o d e r a t e . N o n e were r a t e d as severe. Side effect
Headache Depression Dizziness o r lightheadedness Insomnia Fatigue F r e q u e n c y of urination Forgetfulness Total
Atenolol
Placebo
n
n
Percent
Percent
3 1
18.75 6.25
2 0
12.50
1 2 ]
6.25 12.50 6.25
1 1 0
6.25 6.25
1 0
6.25
0 1
16
6,25
16
significant changes were f o u n d in the m o n t h b y d a y i n t e r a c t i o n s for any of the other scores. Because of the possibility of d e p r e s s i o n as a side effect of fl-blockers, it was interesting that the d e p r e s s i o n item went in the o p p o s i t e direction of the general global p a t t e r n of lower m e a n s y m p t o m scores on active drug; however, the d e p r e s s i o n scores were not significantly higher in the active d r u g c o n d i t i o n . O n l y one subject r a t e d d e p r e s s i o n as a side effect a n d there was not a significantly greater incidence of a n y side effects of a t e n o l o l c o m p a r e d to p l a c e b o ( T a b l e 1). H a l f of the 16 subjects in our s a m p l e h a d p r e m e n s t r u a l tension for m o r e than 5 years. W h e n the d u r a t i o n of p r e m e n s t r u a l s y m p t o m s was used as a g r o u p factor in the M A N O V A , it was f o u n d that subjects who had p r e m e n s t r u a l tension for m o r e than 5 years i m p r o v e d on most scales over those with a shorter d u r a t i o n of P M S s y m p t o m s . T h e r e were statistically significant reductions over p l a c e b o in the m o n t h b y d a y by d u r a t i o n g r o u p i n t e r a c t i o n in P O M S fatigue scores (df= 1, 14, F=5.34, P < 0 . 0 4 ) , total B D I score ( F = 9 . 0 2 , P < 0.01), total H R S D score ( F = 11.49, P < 0.01), total H A S score ( F = 7.03, P < 0.02), M D Q water r e t e n t i o n scores ( F = 9.13, P < 0.01), M D Q beh a v i o r a l change scores ( F = 8.59, P < 0.02), M D Q c o n t r o l scores ( F = 6.85, P < 0.03) a n d P M T S p h y s i c a l s y m p t o m s scores ( F = 5.15, P < 0.04).
There were trends ( P < 0,1) t o w a r d several other items being m o r e i m p r o v e d in the > 5-year d u r a tion group, i n c l u d i n g P O M S tension, M D Q concentration, P M T S irritability, tension, efficiency, d y s p h o r i a , a n d m e n t a l - c o g n i t i v e functioning. T h e differences in the s y m p t o m d u r a t i o n g r o u p s c o u l d not be e x p l a i n e d by s y m p t o m severity; no significant i n t e r a c t i o n s were f o u n d when severity ratings were used as a group factor. A two-way r e p e a t e d m e a s u r e s analysis of variance i n d i c a t e d that serum a l d o s t e r o n e levels increased ( F = 9 . 2 8 , P=0.01) d u r i n g the luteal p h a s e of the m e n s t r u a l cycle as a time effect, with a similar t r e n d ( F = 4.52, P = 0.06) also f o u n d for p l a s m a renin activity (Fig. 2). However, there were no significant drug x time i n t e r a c t i o n s for either serum a l d o s t e r o n e ( F = 0.25, P > 0.6) or p l a s m a renin activity ( F = 1.05, P > 0.3), i n d i c a t i n g that the 50-mg dose of a t e n o l o l d i d not b l u n t the luteal increases in a l d o s t e r o n e or p l a s m a renin activity, overall. In o r d e r to evaluate w h e t h e r f l u c t u a t i o n s in the h o r m o n e levels might o b s c u r e a d r u g effect on the r e p e a t e d measures A N O V A , u n i v a r i a t e A N O V A s were c o n d u c t e d to d e t e r m i n e w h e t h e r single time p o i n t s were different between active a n d p l a c e b o conditions. O n l y on d a y - 2 was a l d o s t e r o n e significantly lower d u r i n g the active drug m o n t h in c o m p a r i s o n to p l a c e b o (Fig. 2). Significant decreases for active d r u g were not found for p l a s m a renin activity. However, there was a significant difference in s u b g r o u p s of w o m en having m o d e r a t e vs. severe intensity of s y m p toms with respect to the way that levels of p l a s m a renin activity were affected b y a t e n o l o l ( F = 10.33, P < 0.015). The g r o u p of w o m e n with m o d e r a t e intensity s y m p t o m s showed larger decreases of p l a s m a renin activity in the active m o n t h a n d had s m a l l e r p r e m e n s t r u a l i n c r e a s e s in H a m i l t o n anxiety scores, c o m p a r e d to those with severe intensity s y m p t o m s ( F = 7.03, P < 0.02). The m a x i m u m luteal p r o g e s t e r o n e levels were significantly (df= 1, 14, F = 4.79, P < 0.05) higher d u r i n g the atenolol m o n t h in c o m p a r i s o n to the p l a c e b o m o n t h on r e p e a t e d m e a s u r e s A N O V A (Fig. 2). T h e r e was also a trend for large decreases in p r o g e s t e r o n e to be c o r r e l a t e d with high M D Q p a i n ratings in the p l a c e b o m o n t h ( r = - 0 . 5 3 , P = 0.06) a n d not in the active m o n t h ( r = - 0 . 1 7 , P = 0.58). The group of w o m e n who h a d P M S
145
HORMONAL FLUCTUATIONS Atenolol vs Placebo Plasma Renin Activity
Aldosterone 35.0 25.0
15.00
• Active o Placebo
] I 10.00
15.0 5.00 5.0 > (D .J
0.00 -28
E (y) E"
w"
0.0 -20
-14
-8
-2 T
-28
20
Progesterone
-8
-2 1"
-8
-2 "1"
Estrogen
20.0
550.00
15.0
-14
l I~\
r-'~
10.0
275.00
5.0 0.0
0.00 -28
-20
-14
-8
-2 T
-2g
-20
-14
Menstrual Cycle Day Fig. 2. Menstrual cycle fluctuations of aldosterone, plasma renin activity, progesterone, and estrogen are depicted across the menstrual cycle for both active (closed squares) and placebo (open circles) months. The time points are plotted as averages ( + 2 days) of the number of days prior to menstruation. symptoms for more than 5 years had significantly ( F = 10.0, P < 0.001) greater increases in progesterone during the active month (compared to placebo) than did the women who had symptoms for less than 5 years. For the whole group, POMS vigor, elation, and friendliness significantly ( P < 0.05) correlated negatively with maximum premenstrual estrogen levels during the placebo but not during the active month (Rausch et al., in preparation). Plasma renin activity and aldosterone correlated with estrogen and progesterone levels during the placebo month but not during the active month. Discussion
Complaints in PMS sufferers of premenstrual bloating, weight gain, and irritability have been postulated to reflect subtle changes in sodium and water balance in the central nervous system (Smith,
1976). Early versions of this theory suggested that increases in sodium and water retention could be linked to a syndrome analogous to a mild subclinical form of cerebral edema with emotional instability presenting as a soft neurological sign (Thorn and Harrop, 1937). Diuretics have since been studied as treatments for different premenstrual emotional symptoms. Early studies which were not placebo-controlled reported success with various diuretic treatments including ammonium chloride, chlorothiazide, chlorthalodone and quinethazone (Stielglitz and Kimble, 1949; Rees, 1953; Kramer, 1962; Baden and Lizcano, 1963). Since then, four placebo-controlled studies have indicated an ammonium chloride/caffeine combination (Hoffman, 1979), metalozone (Werch and Kane, 1976), spironolactone (O'Brien et al., 1979), and a triamterene/benzethiazide combination (Coppen et al., 1969) to be more effective than placebo in the
146
treatment of PMS. Two other controlled studies did not find chlorthalodone (Mattson and Von Schoultz, 1974) or potassium chloride (Reeves et al., 1971) to be better than placebo. In the present pilot study, atenolol had limited therapeutic effects in the 16 women evaluated in the group as a whole. Atenolol significantly prevented premenstrual reductions in vigor, elation and friendliness, and significantly diminished premenstrual irritability for the group as a whole. Although nearly all the other mean premenstrual change scores indicated lower mean symptom scores during the active month, the changes were not statistically significant. There was no improvement in anxiety ratings for the group as a whole. There were also no significant changes in the direction of worsening symptoms, including depression. There was a clear-cut alleviation of most symptoms in the subgroup of eight women who had a duration of premenstrual tension symptoms of more than 5 years. It is not clear why this was the case. No significant differences were found on the ratings, severity of symptoms before entry, or age between the two duration groups. It is possible that the > 5-year duration group presented a more distilled representation of the syndrome, since the possibility cannot be excluded that, despite our attempts to rule out other psychiatric disorders, a portion of the < 5-year duration group might present with alternative diagnoses after extended longitudinal observation (e.g., see DeJong et al., 1985). We hypothesized that atenolol would be effective for premenstrual tension symptoms because of its/31-blocking effects, decreasing plasma renin activity and decreasing aldosterone secretion. The data indicated, however, that 50 mg of atenolol did not significantly decrease plasma renin activity or, for the most part, plasma aldosterone. Previous work by Hespel and colleagues (1986) indicated that 50 mg of atenolol once daily for 3 days decreased plasma renin activity but not plasma aldosterone, although exercise-induced increases in plasma renin activity were not blocked by the 50-mg dose of atenolol in that study. The same may be true for luteal phase-induced increases in plasma renin activity: 50 mg of atenolol may decrease basal levels of plasma renin
activity more readily than increases induced by stimulation of the mineralocorticoid axis. In other words, it is possible that the 50-rag dose of ateno1ol was not large enough to decrease plasma renin activity during a luteal increase in the premenstrual women we studied. It is not known whether a higher dose of atenolol would have had greater effects on premenstrual tension symptoms. A surprising finding was the trend for atenolol to increase premenstrual progesterone levels. The group of women who had the greater increases in progesterone during the active drug month were those women with a history of PMS symptoms for more than 5 years. It was these women who had a number of significant improvements in their symptoms. It is not known whether these improvements could be attributed to the increases found in progesterone in this group. Progesterone itself has natriuretic properties (Losert et al., 1985: Wambach et al., 1979), and the higher progesterone levels in this group could conceivably have contributed to the overall antimineralocorticoid activity in the group. In summary, the data from this study indicated that atenolol blunted symptoms of premenstrual tension to a limited extent in the subject group as a whole. This alleviation was much more dramatic in subjects who had premenstrual tension for more than 5 years. Since atenolol is a B-blocking agent, it is possible that the alleviation may be due in part to the limited antagonism of the renin-angiotensin-aldosterone system found with the 50-rag dose, although it is also possible that part of the alleviation may be due to the higher progesterone levels found during the active month, or due to p-blocking effects in the nervous system per se. However, the sample size in this pilot study is small, and a larger study is needed for replication of the findings. The results indicate that it may be best to test women with premenstrual tension symptoms present for more than 5 years. In addition, higher doses of atenolol may more effectively suppress luteal increases in plasma renin activity and aldosterone.
References Aberg, H. (1974) Letter: Beta receptors and renin release. N. Engl. J. Med. 290, 1026.
147 Baden, W.F. and Lizcano, H.R. (1963) Evaluation of a new drug (quinethazone) in the premenstrual syndrome. J. New Drugs 3, 167. Coppen, A.J., Milne, H.B., Outrum, D.W. and Weber, J.C.T, (1969) Diazide, norethisterone, and placebo in the premenstrual tension syndrome. Clin. Trials J. 6, 33. DeJong, R., Rubinow, D.R., Roy-Byrne, P., Hoban, M.C.., Grover, G.N. and Post, R.M. (1985) Premenstrual mood disorder and psychiatric illness. Am. J. Psychiatry 142, 1359-1361. Elliott, F.A. (1977) Propranolol for the control of belligerent behavior following acute brain damage. Ann. Neurol. 1, 489-491. Gray, M.J., Strausfeld, K.S., Watanabe, M., Sims, E.A.H. and Solomon, S. (1968) Aldosterone secretory rates in the normal menstrual cycle. Clin. Endocrinol. Metab. 28, 1269. Griffin, S.J. and Friedman, M.J. (1986) Depressive symptoms in propranolol users. J. Clin. Psychiatry 47, 453 457. Hespel, P., Lijnen, P., Vanhees, L., Fagard, R. and Amery, A. (1986) Beta-adrenoceptors and the regulation of blood pressure and plasma renin during exercise. J. Appl. Physiol. 60, 108 113. Hoffman, J.J. (1979) A double blind crossover trial of an OTC diuretic in the treatment of premenstrual symptoms and weight gain. Curr. Ther. Res. 26, 575-580. Janowsky, D.S., Berens, S.C. and Davis, J.M. (1973) Correlations between mood, weight, and electrolytes during the menstrual cycle: a renin-angiotensin-aldosterone hypothesis of premenstrual tension. Psychosom. Med. 35, 143-154. Katz, F.H. and Romth, P. (1972) Plasma aldosterone and renin activity during the menstrual cycle. J. Clin. Endocrinol. 34, 819-821. Kramer, H. (1962) A long acting diuretic suitable for outpatient treatment of fluid retention in oedema of pregnancy and premenstrual tension syndrome. African Med. J. 36, 4. Losert, W., Casals-Stenzel, J. and Buse, M. (1985) Progestogens with antimineralocorticoid activity. Arzneim.-forsch. 35, 459 471. Magos, A.L., Jones, D.E.J., Heydeman, T.E. and Studd, J.W.W. (1987) Development of a premenstrual distress questionnaire (PDQ). Proceedings of the 2nd International Symposium of Premenstrual, Postpartum, and Menopausal Mood Disorders.
Mattson, B. and Von Schoultz, B. (1974) A comparison between lithium, placebo, and a diuretic in premenstrual tension. Acta Psychiatr. Scand. (Suppl.) 255, 75-84. Moos, R.H. (1984) Menstrual Distress Questionnaire. Stanford University Medical Center, Palo Alto, CA. O'Brien, P.M., Craven, D., Selby, C. and Symonds, E.M. (1979) Treatment of premenstrual syndrome by spironolactone. Br. J. Obstet. Gyneacol. 86, 142-147. Rausch, J.L. and Janowsky, D.S. (1982) Premenstrual tension: etiology. In: R.C. Friedman (Ed.), Behavior and the Menstrual Cycle. Marcel Dekker, New York, NY, pp. 397 427. Rees, L. (1953) The premenstrual syndrome and its treatment. Br. Med. J. 1, 1014-1016. Reeves, B.D., Garvin, J.E. and McElin, T.W. (1971) Premenstrual tension: Symptoms and weight changes related to potassium therapy. Am. J. Obstet. Gynecol. 109, 1036-1041. Reich, M. (1972) The variations in urinary aldosterone levels of normal females during the menstrual cycle. Aust. Ann. Med. 11, 41-49. Sassard, J., Pozet, N., Vincent, M. and Zech, P.Y. (1976) Letter: Atenolol and renin release. N. Engl. J. Med. 294, 787. Smith, S.L. (1976) The menstrual cycle and mood disturbances. Clin. Obstet. Gynecol. 19, 391-397. Steiner, M., Haskett, R.F. and Carroll, B.J. (1980) Premenstrual tension syndrome: the development of research diagnostic criteria and new rating scales. Acta Psychiatr. Scand. 62, 177 190. Stielglitz, E.J. and Kimble, S.T. (1949) Premenstrual intoxication. Am. J. Med. Sci. 218, 616. Thorn, G.W. and Harrop, G.A. (1937) The sodium-retaining effect of the sex hormones. Science 86, 40-41. Wambach, G., Higgins, J.R., Kern, D.C. and Kaufmann, W. (1979) Interaction of synthetic progestogens with renal mineralocorticoid receptors. Acta Endocrinol. 92, 560-567. Werch, A. and Kane, R.E. (1976) Treatment of premenstrual tension with metalozone: a double blind evaluation of a new diuretic. Curr. Ther. Res. 19, 565-572.
141
Elsevier J A D 00564
Atenolol treatment of late luteal phase dysphoric disorder J e f f r e y L. R a u s c h 1 D a v i d S. J a n o w s k y 2, S h a h r o k h G o l s h a n 1, K a r e n K u h n 1 a n d S.C. R i s c h ~ I Department of Psychiatry M-O03, San Diego VAMC, School of Medicine, University of California, San Diego, La Jolla, CA 92093, U.S.A. and 2 Department of Psychiatry, University of North Carolina, Chapel Hill, NC, U.S.A. (Received 1 October 1987) (Accepted 11 March 1988)
Summary Activation of the renin-angiotensin-aldosterone system has been hypothesized as a potential pathophysiological factor in premenstrual tension syndrome (PMS). Atenolol is a predominate ill-blocker which can decrease plasma renin activity and inhibit the urinary excretion of aldosterone. Sixteen women meeting provisional diagnoses of late luteal phase dysphoric disorder were treated for symptoms of PMS with atenolol (50 mg once daily) in a randomized placebo-crossover double-blind design. The data indicated significant improvements on the irritability, vigor, elation, and friendliness scores in response to atenolol compared to placebo. Significant changes were not found for several other ratings scales, indicating that atenolol improved only selected symptoms in the group as a whole. However, the women who had premenstrual tension symptoms for more than 5 years (n = 8) were improved on most of the rating scales. Atenolol decreased premenstrual plasma aldosterone to a limited extent. There was also a trend in the data toward higher luteal progesterone levels during the month subjects took atenolol. Plasma renin activity and aldosterone correlated with estrogen and progesterone levels during the placebo month but not during the active month.
Key words." Beta-blockers; Atenolol; Premenstrual tension; Late luteal phase dysphoric disorder; A1dosterone; Progesterone
Introduction Several hypotheses have been constructed to explain a biological etiology of premenstrual af-
Address for correspondence: Jeffrey L. Rausch, M.D., Department of Psychiatry M-003, UCSD, La Jolla, CA 92093, U.S.A.
fective symptoms (Rausch and Janowsky, 1982). One such theory holds that activation of the renin-angiotensin-aldosterone system may be an etiologic factor (Janowsky et al., 1973). Support for the hypothesis derives from indications that plasma renin activity and aldosterone may increase during the luteal phase of the menstrual cycle (Gray et al., 1968; Reich, 1972; Katz and Romth, 1972) as well as from findings that
0165-0327/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
142 menstrual cycle changes in negative affect may covary with urinary s o d i u m / p o t a s s i u m ratios (Janowsky et al., 1973). Although studies have differed on whether premenstrual tension syndrome (PMS) patients have elevated premenstrual serum aldosterone levels compared to controls (for review, see Rausch and Janowsky, 1982), it is possible that some women may have greater behavioral sensitivities to ordinary luteal increases in mineralocorticoid activity. If premenstrual tension were related to a sensitivity to excess mineralocorticoid activity, a selective/~-blocking drug such as atenolol, which could inhibit the excess mineralocorticoid activity (Aberg, 1974; Sassard et al., 1976) could theoretically attenuate the increase in the premenstrual behavioral symptoms related to mineralocorticoidinduced changes in salt and water balance. No previous such studies are available. Here we report the results of a pilot, doubleblind, placebo-crossover trial of low-dose atenolol. Atenolol was chosen as a predominate/~-blocker which could decrease plasma renin activity, and inhibit the urinary excretion of aldosterone (Hespel et al., 1986). The study's purpose was to assess which, if any, PMS symptoms would respond to the low dose of atenolol. To specify the dependent variables prior to the experiment, three hypotheses were formulated about the effects of atenolol in premenstrual tension sufferers. Firstly, it was postulated that irritability would improve. Irritability has been found to be the most severe premenstrual symptom (Magos et al., 1987) and it has been postulated that irritability may reflect subtle changes in sodium and water balance in the central nervous system (e.g., Smith, 1976). Furthermore, propranolol has been effectively used for the control of irritable behavior following acute brain damage (Elliott, 1977). Secondly, it was theorized that atenolol may work for premenstrual tension by virtue of its /3-adrenergic properties in the sympathetic nervous system. Since fl-blockers have been used with some degree of efficacy in the treatment of anxiety disorders, it was postulated that decreases in anxiety ratings might be found. Atenolol has less lipid solubility than propranolol, and crosses the blood-brain barrier less readily, but it does reach the CNS to at least some limited extent, and
several signs and symptoms of anxiety may be manifest through involvement of the peripheral sympathetic nervous system. Thirdly, the possibility that depression would increase in these subjects was considered. It is known that the side effects of fl-blockers include depression (Griffin and Friedman, 1986), and it was hypothesized that this effect might potentially be seen as a side effect in the women with PMS symptoms. Method
Atenolol (50 mg once daily) was evaluated as a treatment for various symptoms of premenstrual tension in 16 women in a 2-month randomized, placebo-crossover, double-blind design. Subjects were recruited for the study from campus advertising and consisted largely of students and employees at the University of California (San Diego) and the San Diego Veterans' Administration Medical Center. Subjects ranged in age from 23 to 46 (mean = 33.8 + 6.5 SD). Before entry into the study, subjects were given a physical exam, and their medical and psychiatric history was taken. To meet entrance criteria for the study, the women had to have regular menstrual cycles ( + 3 days) for 3 months prior to entry into the study, had to be physically healthy, without previous psychiatric histories, and had to meet Carroll criteria for a diagnosis of premenstrual tension (Steiner et al., 1980). The intensities of premenstrual symptoms for the 2 previous months were recorded by Carroll criteria, and subjects were observed prospectively for premenstrual tension symptoms for 1 month prior to entry into the study. Carroll criteria symptoms were used to generate a mean severity assignment by an a priori decision to assign those patients with a score of 18-27, out of a highest possible score of 36, a severity rating of moderate; those patients with a score higher than 27 were rated as severe. Only those subjects who had an intensity rating of at least moderate for both of the previous 2 months were entered into the study. All patients met provisional D S M - I I I R diagnostic criteria for late luteal phase dysphoric disorder. Behavioral measures and blood samples were collected twice weekly. Mood rating instruments
143
included the Beck Depression Inventory (BDI), the Menstrual Distress Questionnaire (MDQ) (Moos, 1984), the Self-Rating Scale (SRS) for PMS (Steiner et al., 1980), the Profile of Mood States (POMS), Hamilton Rating Scale for Depression (HRSD), and Hamilton Anxiety Scale (HAS). Blood samples were assayed for progesterone, estrogen, aldosterone, and plasma renin activity. All subjects were given atenolol 50 mg or placebo once each morning for 10 days before the estimated start of menses. Since it could only be approximated when menstruation would begin, it occurred that some women took study drug capsules 11 days before menstruation. The study drug was administered in all cases through the fourth day after the onset of menses. The subjects were randomly assigned to either placebo or active drug the first month, in double-blind fashion, and changed to placebo or active drug during the second month. The atenolol and placebo capsules were identical in appearance. Baseline behavioral scores were derived from an average of the two mid to late follicular phase
PMTS CHANGE SCORE
visits occurring 14-22 days before menstruation. These were then compared to the mean of the three premenstrual visits occurring between 1 and 11 days before menstruation. The latter three visits were in the premenstrual phase of the cycle when subjects were taking capsules. Results
All mean premenstrual change scores except for POMS depression, M D Q autonomic regulation and control were globally lower with atenolol than with placebo; however, most of the items did not reach statistical significance. Multivariate analysis of variance (MANOVA) indicated significant differences between active and placebo months only in the premenstrual changes on the irritability score of the PMTS, and three of the POMS scores. These findings indicated that atenolol significantly ( P < 0.02) prevented premenstrual reductions in vigor, elation, and friendliness (df= 1, 14, F = 7.26, 6.90, 7.51) and significantly ( P <0.02) diminished premenstrual irritability (df= 1, 14, F = 10.01) in comparison to placebo (Fig. 1). No other
M D Q MEAN CHANGE SCORES
POMS CHANGE SCORES O
I~active
I--Iplacebo
Y, o
1 ~
o
~
oo
~
2
~ I~
o
O I
~ <
um
o)
a 0
m
~
o
D
u_
Z
13_
Fig. 1. The changes in symptoms between the follicular phase and premenstrual phase (see text) are illustrated for the active (dark bars) and placebo months (open bars). Increases in negative affects and symptoms are graphed as positive values, and decreases in favorable affects are graphed as negative values.
144 TABLE 1 SIDE EFFECTS OF ATENOLOL
vs. P L A C E B O
All side effect w e r e r a t e d as m i l d except for d e p r e s s i o n , o n e case o f active d r u g dizziness, o n e case of active d r u g h e a d a c h e a n d o n e case o f p l a c e b o h e a d a c h e , all o f w h i c h w e r e r a t e d as m o d e r a t e . N o n e were r a t e d as severe. Side effect
Headache Depression Dizziness o r lightheadedness Insomnia Fatigue F r e q u e n c y of urination Forgetfulness Total
Atenolol
Placebo
n
n
Percent
Percent
3 1
18.75 6.25
2 0
12.50
1 2 ]
6.25 12.50 6.25
1 1 0
6.25 6.25
1 0
6.25
0 1
16
6,25
16
significant changes were f o u n d in the m o n t h b y d a y i n t e r a c t i o n s for any of the other scores. Because of the possibility of d e p r e s s i o n as a side effect of fl-blockers, it was interesting that the d e p r e s s i o n item went in the o p p o s i t e direction of the general global p a t t e r n of lower m e a n s y m p t o m scores on active drug; however, the d e p r e s s i o n scores were not significantly higher in the active d r u g c o n d i t i o n . O n l y one subject r a t e d d e p r e s s i o n as a side effect a n d there was not a significantly greater incidence of a n y side effects of a t e n o l o l c o m p a r e d to p l a c e b o ( T a b l e 1). H a l f of the 16 subjects in our s a m p l e h a d p r e m e n s t r u a l tension for m o r e than 5 years. W h e n the d u r a t i o n of p r e m e n s t r u a l s y m p t o m s was used as a g r o u p factor in the M A N O V A , it was f o u n d that subjects who had p r e m e n s t r u a l tension for m o r e than 5 years i m p r o v e d on most scales over those with a shorter d u r a t i o n of P M S s y m p t o m s . T h e r e were statistically significant reductions over p l a c e b o in the m o n t h b y d a y by d u r a t i o n g r o u p i n t e r a c t i o n in P O M S fatigue scores (df= 1, 14, F=5.34, P < 0 . 0 4 ) , total B D I score ( F = 9 . 0 2 , P < 0.01), total H R S D score ( F = 11.49, P < 0.01), total H A S score ( F = 7.03, P < 0.02), M D Q water r e t e n t i o n scores ( F = 9.13, P < 0.01), M D Q beh a v i o r a l change scores ( F = 8.59, P < 0.02), M D Q c o n t r o l scores ( F = 6.85, P < 0.03) a n d P M T S p h y s i c a l s y m p t o m s scores ( F = 5.15, P < 0.04).
There were trends ( P < 0,1) t o w a r d several other items being m o r e i m p r o v e d in the > 5-year d u r a tion group, i n c l u d i n g P O M S tension, M D Q concentration, P M T S irritability, tension, efficiency, d y s p h o r i a , a n d m e n t a l - c o g n i t i v e functioning. T h e differences in the s y m p t o m d u r a t i o n g r o u p s c o u l d not be e x p l a i n e d by s y m p t o m severity; no significant i n t e r a c t i o n s were f o u n d when severity ratings were used as a group factor. A two-way r e p e a t e d m e a s u r e s analysis of variance i n d i c a t e d that serum a l d o s t e r o n e levels increased ( F = 9 . 2 8 , P=0.01) d u r i n g the luteal p h a s e of the m e n s t r u a l cycle as a time effect, with a similar t r e n d ( F = 4.52, P = 0.06) also f o u n d for p l a s m a renin activity (Fig. 2). However, there were no significant drug x time i n t e r a c t i o n s for either serum a l d o s t e r o n e ( F = 0.25, P > 0.6) or p l a s m a renin activity ( F = 1.05, P > 0.3), i n d i c a t i n g that the 50-mg dose of a t e n o l o l d i d not b l u n t the luteal increases in a l d o s t e r o n e or p l a s m a renin activity, overall. In o r d e r to evaluate w h e t h e r f l u c t u a t i o n s in the h o r m o n e levels might o b s c u r e a d r u g effect on the r e p e a t e d measures A N O V A , u n i v a r i a t e A N O V A s were c o n d u c t e d to d e t e r m i n e w h e t h e r single time p o i n t s were different between active a n d p l a c e b o conditions. O n l y on d a y - 2 was a l d o s t e r o n e significantly lower d u r i n g the active drug m o n t h in c o m p a r i s o n to p l a c e b o (Fig. 2). Significant decreases for active d r u g were not found for p l a s m a renin activity. However, there was a significant difference in s u b g r o u p s of w o m en having m o d e r a t e vs. severe intensity of s y m p toms with respect to the way that levels of p l a s m a renin activity were affected b y a t e n o l o l ( F = 10.33, P < 0.015). The g r o u p of w o m e n with m o d e r a t e intensity s y m p t o m s showed larger decreases of p l a s m a renin activity in the active m o n t h a n d had s m a l l e r p r e m e n s t r u a l i n c r e a s e s in H a m i l t o n anxiety scores, c o m p a r e d to those with severe intensity s y m p t o m s ( F = 7.03, P < 0.02). The m a x i m u m luteal p r o g e s t e r o n e levels were significantly (df= 1, 14, F = 4.79, P < 0.05) higher d u r i n g the atenolol m o n t h in c o m p a r i s o n to the p l a c e b o m o n t h on r e p e a t e d m e a s u r e s A N O V A (Fig. 2). T h e r e was also a trend for large decreases in p r o g e s t e r o n e to be c o r r e l a t e d with high M D Q p a i n ratings in the p l a c e b o m o n t h ( r = - 0 . 5 3 , P = 0.06) a n d not in the active m o n t h ( r = - 0 . 1 7 , P = 0.58). The group of w o m e n who h a d P M S
145
HORMONAL FLUCTUATIONS Atenolol vs Placebo Plasma Renin Activity
Aldosterone 35.0 25.0
15.00
• Active o Placebo
] I 10.00
15.0 5.00 5.0 > (D .J
0.00 -28
E (y) E"
w"
0.0 -20
-14
-8
-2 T
-28
20
Progesterone
-8
-2 1"
-8
-2 "1"
Estrogen
20.0
550.00
15.0
-14
l I~\
r-'~
10.0
275.00
5.0 0.0
0.00 -28
-20
-14
-8
-2 T
-2g
-20
-14
Menstrual Cycle Day Fig. 2. Menstrual cycle fluctuations of aldosterone, plasma renin activity, progesterone, and estrogen are depicted across the menstrual cycle for both active (closed squares) and placebo (open circles) months. The time points are plotted as averages ( + 2 days) of the number of days prior to menstruation. symptoms for more than 5 years had significantly ( F = 10.0, P < 0.001) greater increases in progesterone during the active month (compared to placebo) than did the women who had symptoms for less than 5 years. For the whole group, POMS vigor, elation, and friendliness significantly ( P < 0.05) correlated negatively with maximum premenstrual estrogen levels during the placebo but not during the active month (Rausch et al., in preparation). Plasma renin activity and aldosterone correlated with estrogen and progesterone levels during the placebo month but not during the active month. Discussion
Complaints in PMS sufferers of premenstrual bloating, weight gain, and irritability have been postulated to reflect subtle changes in sodium and water balance in the central nervous system (Smith,
1976). Early versions of this theory suggested that increases in sodium and water retention could be linked to a syndrome analogous to a mild subclinical form of cerebral edema with emotional instability presenting as a soft neurological sign (Thorn and Harrop, 1937). Diuretics have since been studied as treatments for different premenstrual emotional symptoms. Early studies which were not placebo-controlled reported success with various diuretic treatments including ammonium chloride, chlorothiazide, chlorthalodone and quinethazone (Stielglitz and Kimble, 1949; Rees, 1953; Kramer, 1962; Baden and Lizcano, 1963). Since then, four placebo-controlled studies have indicated an ammonium chloride/caffeine combination (Hoffman, 1979), metalozone (Werch and Kane, 1976), spironolactone (O'Brien et al., 1979), and a triamterene/benzethiazide combination (Coppen et al., 1969) to be more effective than placebo in the
146
treatment of PMS. Two other controlled studies did not find chlorthalodone (Mattson and Von Schoultz, 1974) or potassium chloride (Reeves et al., 1971) to be better than placebo. In the present pilot study, atenolol had limited therapeutic effects in the 16 women evaluated in the group as a whole. Atenolol significantly prevented premenstrual reductions in vigor, elation and friendliness, and significantly diminished premenstrual irritability for the group as a whole. Although nearly all the other mean premenstrual change scores indicated lower mean symptom scores during the active month, the changes were not statistically significant. There was no improvement in anxiety ratings for the group as a whole. There were also no significant changes in the direction of worsening symptoms, including depression. There was a clear-cut alleviation of most symptoms in the subgroup of eight women who had a duration of premenstrual tension symptoms of more than 5 years. It is not clear why this was the case. No significant differences were found on the ratings, severity of symptoms before entry, or age between the two duration groups. It is possible that the > 5-year duration group presented a more distilled representation of the syndrome, since the possibility cannot be excluded that, despite our attempts to rule out other psychiatric disorders, a portion of the < 5-year duration group might present with alternative diagnoses after extended longitudinal observation (e.g., see DeJong et al., 1985). We hypothesized that atenolol would be effective for premenstrual tension symptoms because of its/31-blocking effects, decreasing plasma renin activity and decreasing aldosterone secretion. The data indicated, however, that 50 mg of atenolol did not significantly decrease plasma renin activity or, for the most part, plasma aldosterone. Previous work by Hespel and colleagues (1986) indicated that 50 mg of atenolol once daily for 3 days decreased plasma renin activity but not plasma aldosterone, although exercise-induced increases in plasma renin activity were not blocked by the 50-mg dose of atenolol in that study. The same may be true for luteal phase-induced increases in plasma renin activity: 50 mg of atenolol may decrease basal levels of plasma renin
activity more readily than increases induced by stimulation of the mineralocorticoid axis. In other words, it is possible that the 50-rag dose of ateno1ol was not large enough to decrease plasma renin activity during a luteal increase in the premenstrual women we studied. It is not known whether a higher dose of atenolol would have had greater effects on premenstrual tension symptoms. A surprising finding was the trend for atenolol to increase premenstrual progesterone levels. The group of women who had the greater increases in progesterone during the active drug month were those women with a history of PMS symptoms for more than 5 years. It was these women who had a number of significant improvements in their symptoms. It is not known whether these improvements could be attributed to the increases found in progesterone in this group. Progesterone itself has natriuretic properties (Losert et al., 1985: Wambach et al., 1979), and the higher progesterone levels in this group could conceivably have contributed to the overall antimineralocorticoid activity in the group. In summary, the data from this study indicated that atenolol blunted symptoms of premenstrual tension to a limited extent in the subject group as a whole. This alleviation was much more dramatic in subjects who had premenstrual tension for more than 5 years. Since atenolol is a B-blocking agent, it is possible that the alleviation may be due in part to the limited antagonism of the renin-angiotensin-aldosterone system found with the 50-rag dose, although it is also possible that part of the alleviation may be due to the higher progesterone levels found during the active month, or due to p-blocking effects in the nervous system per se. However, the sample size in this pilot study is small, and a larger study is needed for replication of the findings. The results indicate that it may be best to test women with premenstrual tension symptoms present for more than 5 years. In addition, higher doses of atenolol may more effectively suppress luteal increases in plasma renin activity and aldosterone.
References Aberg, H. (1974) Letter: Beta receptors and renin release. N. Engl. J. Med. 290, 1026.
147 Baden, W.F. and Lizcano, H.R. (1963) Evaluation of a new drug (quinethazone) in the premenstrual syndrome. J. New Drugs 3, 167. Coppen, A.J., Milne, H.B., Outrum, D.W. and Weber, J.C.T, (1969) Diazide, norethisterone, and placebo in the premenstrual tension syndrome. Clin. Trials J. 6, 33. DeJong, R., Rubinow, D.R., Roy-Byrne, P., Hoban, M.C.., Grover, G.N. and Post, R.M. (1985) Premenstrual mood disorder and psychiatric illness. Am. J. Psychiatry 142, 1359-1361. Elliott, F.A. (1977) Propranolol for the control of belligerent behavior following acute brain damage. Ann. Neurol. 1, 489-491. Gray, M.J., Strausfeld, K.S., Watanabe, M., Sims, E.A.H. and Solomon, S. (1968) Aldosterone secretory rates in the normal menstrual cycle. Clin. Endocrinol. Metab. 28, 1269. Griffin, S.J. and Friedman, M.J. (1986) Depressive symptoms in propranolol users. J. Clin. Psychiatry 47, 453 457. Hespel, P., Lijnen, P., Vanhees, L., Fagard, R. and Amery, A. (1986) Beta-adrenoceptors and the regulation of blood pressure and plasma renin during exercise. J. Appl. Physiol. 60, 108 113. Hoffman, J.J. (1979) A double blind crossover trial of an OTC diuretic in the treatment of premenstrual symptoms and weight gain. Curr. Ther. Res. 26, 575-580. Janowsky, D.S., Berens, S.C. and Davis, J.M. (1973) Correlations between mood, weight, and electrolytes during the menstrual cycle: a renin-angiotensin-aldosterone hypothesis of premenstrual tension. Psychosom. Med. 35, 143-154. Katz, F.H. and Romth, P. (1972) Plasma aldosterone and renin activity during the menstrual cycle. J. Clin. Endocrinol. 34, 819-821. Kramer, H. (1962) A long acting diuretic suitable for outpatient treatment of fluid retention in oedema of pregnancy and premenstrual tension syndrome. African Med. J. 36, 4. Losert, W., Casals-Stenzel, J. and Buse, M. (1985) Progestogens with antimineralocorticoid activity. Arzneim.-forsch. 35, 459 471. Magos, A.L., Jones, D.E.J., Heydeman, T.E. and Studd, J.W.W. (1987) Development of a premenstrual distress questionnaire (PDQ). Proceedings of the 2nd International Symposium of Premenstrual, Postpartum, and Menopausal Mood Disorders.
Mattson, B. and Von Schoultz, B. (1974) A comparison between lithium, placebo, and a diuretic in premenstrual tension. Acta Psychiatr. Scand. (Suppl.) 255, 75-84. Moos, R.H. (1984) Menstrual Distress Questionnaire. Stanford University Medical Center, Palo Alto, CA. O'Brien, P.M., Craven, D., Selby, C. and Symonds, E.M. (1979) Treatment of premenstrual syndrome by spironolactone. Br. J. Obstet. Gyneacol. 86, 142-147. Rausch, J.L. and Janowsky, D.S. (1982) Premenstrual tension: etiology. In: R.C. Friedman (Ed.), Behavior and the Menstrual Cycle. Marcel Dekker, New York, NY, pp. 397 427. Rees, L. (1953) The premenstrual syndrome and its treatment. Br. Med. J. 1, 1014-1016. Reeves, B.D., Garvin, J.E. and McElin, T.W. (1971) Premenstrual tension: Symptoms and weight changes related to potassium therapy. Am. J. Obstet. Gynecol. 109, 1036-1041. Reich, M. (1972) The variations in urinary aldosterone levels of normal females during the menstrual cycle. Aust. Ann. Med. 11, 41-49. Sassard, J., Pozet, N., Vincent, M. and Zech, P.Y. (1976) Letter: Atenolol and renin release. N. Engl. J. Med. 294, 787. Smith, S.L. (1976) The menstrual cycle and mood disturbances. Clin. Obstet. Gynecol. 19, 391-397. Steiner, M., Haskett, R.F. and Carroll, B.J. (1980) Premenstrual tension syndrome: the development of research diagnostic criteria and new rating scales. Acta Psychiatr. Scand. 62, 177 190. Stielglitz, E.J. and Kimble, S.T. (1949) Premenstrual intoxication. Am. J. Med. Sci. 218, 616. Thorn, G.W. and Harrop, G.A. (1937) The sodium-retaining effect of the sex hormones. Science 86, 40-41. Wambach, G., Higgins, J.R., Kern, D.C. and Kaufmann, W. (1979) Interaction of synthetic progestogens with renal mineralocorticoid receptors. Acta Endocrinol. 92, 560-567. Werch, A. and Kane, R.E. (1976) Treatment of premenstrual tension with metalozone: a double blind evaluation of a new diuretic. Curr. Ther. Res. 19, 565-572.