- Email: [email protected]
Invited review hormones, behaviour and the menstrual cycle
JournoloJPsychosomoric Punted in Great Britain.
Research,
Vol. 29, No. 3, pp. 225-233.
W22-3999/85 $3.M)+ .@I Pergamon Press Ltd.
1985
INVITED REVIEW HORMONES,
BEHAVIOUR
AND THE MENSTRUAL
CYCLE
ANTHONY W. CLARE* (Received
14 November
1984)
INTRODUCTION THE POSSIBLE influence
of the menstrual cycle on the physical, psychological and social functioning of women has become a subject of considerable interest over the past two decades [l-3]. Research aimed at clarifying the precise relationship between hormonal changes, physical symptoms, psychological alterations and behaviour has intensified as increasingly sophisticated methods of hormone assay have been developed [4]. However, attempts to relate specific hormonal changes within the menstrual cycle to specific behaviours or symptoms have in the main foundered on the many methodological rocks which litter this terrain [5]. Whereas the cyclical patterns of the major hormones, including LH, FSH, oestrogen, progesterone, testosterone and prolactin, have been accurately described, the problems surrounding the precise description and measurement of mood, behaviour, subjective physical symptomatology, sexual interest and activity, continue to bedevil efforts at clarification. Most attention, over the years, has been paid to the premenstrual phase of the menstrual cycle, a phase variously defined but invariably including the three to four days immediately preceding the onset of the menses. A remarkable array of physical and psychological symptoms has been reported to occur or be exacerbated during this phase [6, 71. These include feelings of reduced well-being and performance, symptoms of anxiety, depression and irritability, complaints of breast tenderness, swelling, weight gain, headache and a reduction in sexual drive. This brief review considers the present status of the main hormones with regard to premenstrual symptomatology and behaviour change.
HORMONES From the very outset, Frank, in his initial description of the premenstrual syndrome [8], assumed that the cyclical nature of the symptoms was due to alterations in hormonal levels. Since that time each of the main hormones has come under suspicion (Table I). The majority of recent studies have focused on an alleged progesterone deficiency and a number of reports, from studies utilising plasma radio-immunassay of hormones and urinary hormonal excretion analyses, have provided some support [9-l 11. Emotional symptoms have been related to decreasing levels of progesterone in that negative affective symptoms often appear to begin early in the luteal phase and increase in intensity as progesterone drops from its luteal *Professor and Head of Department of Psychological Medicine, St. Bartholomew’s Hospital Medical College, West Smithfield, London. 225
226
ANTHONY W. CLARE TABLE I.-THEORIES OF HORMONALCAUSATIONIN THE PREMENSTRUAL SYNDROME Lowered progesterone Increased oestrogen Altered oestrogen/progesterone Oestrogen withdrawal Increased aldosterone Excess androgenic steroids Increased prolactin
ratio
Dalton [ 191 Cullberg [22] Frank [8] Somerville [ 151 Janowsky [36] Wakoh et al. [41] Carroll and Steiner
1301
peak to its low premenstrual level. In addition, depressive symptoms have been provoked by the withdrawal of exogenous progesterone treatment [12]. However, a number of recent, methodologically careful studies have failed to find a deficiency of progesterone [13, 141 and it is now being suggested that premenstrual affective symptoms might be better understood not in terms of an absolute progesterone deficiency but of withdrawal of the hormone. Such a theory requires an additional causal or aggravating factor given that while every ovulating woman experiences some degree of progesterone withdrawal only a proportion of women complain of symptoms. An alternative theory implicates excess oestrogen. During the luteal phase, oestrogen levels are indeed high but the levels are falling when premenstrual symptoms are generally at their most severe. In turn, oestrogen withdrawal has been proposed and it has certainly been implicated in cases of menstrual migraine [15, 161. The use of subcutaneous implants of oestradiol has recently been reported [ 171 but the method employed to diagnose PMS was not made explicit, the outcome measures were not specified and the treatment was not blind. In the light of failure to find a consistent excess or deficiency of either oestrogen or progesterone, attention has switched to the possibility that there is a disturbance in the oestrogen-progesterone ratio. Greene and Dalton suggested such a cause over thirty years ago [18] and Dalton has since argued in a series of influential papers and a book devoted to the subject [19] that progesterone, by suppository or intramuscular injections, is the treatment of choice in this condition. However, this claim is hotly contested [20, 211. Theories linking oestrogen and progesterone with the premenstrual syndrome derive some support from studies of oral contraceptives. Cullberg [22] found that women with a history of premenstrual irritability and who took progesterone-dominant pills had significantly lower incidences of adverse psychological effects compared with women who took oestrogen-dominant pills. In contrast, women without a pre-treatment history of irritability, noted more adverse reactions with the progesterone-dominant pills, suggesting that these women might have been hormonally balanced before treatment or were even concealing an Cullberg speculated that premenstrual endogenous progesterone dominance. irritability might at least be partly due to an endogenous oestrogen dominance or susceptibility to endogenous oestrogen around the premenstrual phase. Steiner and Carroll [23] have, in turn, suggested that premenstrual irritability, anxiety and hostility may be a separate cluster of symptoms from premenstrual depression and there is a suggestion that the former cluster might be related to a high oestrogenprogesterone ratio whereas the latter cluster might be related to a low oestrogenprogesterone ratio. To date this remains a speculation for despite efforts to
227
Hormones, behaviour and the menstrual cycle
distinguish between the different affective symptoms reported premenstrually and the development of more detailed symptom classifications and measures [24, 251 no study reporting specific hormonal abnormalities consistently linked with particular premenstrual symptoms or symptom clusters has been published. Nor is it clear quite what the effect of oral contraceptives on premenstrual mood and behaviour is. While there is some evidence that many women with premenstrual symptoms complain of adverse side-effects with oral contraceptive use [26, 271 there have been a number of reports recently reviewed by Click and Bennett [28] to the effect that such substances improve global functioning and mood during the premenstrual syndrome. Prolactin is another hormone which has been implicated in the premenstrual syndrome [29, 301. Its levels fluctuate throughout the cycle with a peak at ovulation and later in the mid and late-luteal phases [ 3 1 ] . Raised prolactin levels were reported in one series of 28 patients [32] while Brush [33] reported finding a number of premenstrual complainers with ‘high normal’ prolactin levels which appeared to be associated with low progesterone levels. Benedek-Jaszmann and Hearn-Sturtevant [34], believing prolactin to be ‘the key to the aetiology of the syndrome’, used the prolactin-suppressant drug, bromocriptine, in a double-blind, controlled study and found it superior to placebo in relieving breast symptoms, swelling, weight gain and mood disturbances. However, the reported levels of pre-treatment prolactin levels were within normal limits and the sample was composed of women attending for treatment of prolonged infertility. A subsequent double-blind, controlled trial of the drug did not find it superior to placebo [35]. It has been suggested that premenstrual mood disturbance and symptoms such as weight gain and fluid retention might be related to activation of the reninangiotensin-aldosterone system [ 361. In normal women, a slight mid-cycle elevation of urinary aldosterone excretion followed by a luteal rise which peaked during the premenstrual phase and fell just before or soon after the onset of the period has been reported [37]. Quite why aldosterone rises in the luteal phase is not clear but it has been suggested that the sodium-wasting effect exercised by progesterone might be the stimulus [38]. Janowsky and his colleagues [36] did report alterations in sodium and potassium excretion during the premenstruum and suggested that premenstrual affective symptoms might be more closely related in time to changes in fluid and electrolytes than to changes in oestrogen and progesterone. This group did report increased aldosterone excretion in the premenstrual phase of symptomatic women but other studies have failed to replicate this finding [39,40]. It has also been suggested that the manifestations of the premenstrual syndrome may be related to the fluctuations of androgens [41]. Interest persists not least because of the suggested association between androgens and both sexual drive and aggression although the basis for such a link is difficult to establish [42]. No differences have been reported in androstenedione levels between women with high and low degrees of cyclical mood change [14] although there is some evidence that women with marked cyclical mood change report an associated cyclical pattern of sexual feelings [42]. The common exacerbation of acne which occurs in some women who manifest other signs and symptoms of the premenstrual syndrome [43] also suggests to some observers a possible causal role for androgenic steroids. Other
hormones
implicated
in the syndrome
include
insulin
and melatonin.
An
228
ANTHONY
W.
CLARE
inverse relationship has been reported between insulin binding to monocytes and levels of 17-beta-oestradiol, progesterone and 17 alpha-hydroxyprogesterone, indicating that sex steroids may play a role in the control of insulin receptors. Rausch and Janowsky [44], reviewing this data, suggest that it may in part explain the reduction of glucose tolerance in the second half of the menstrual cycle which has been reported. Given that anxiety and irritability are symptoms often seen in transient hypoglycaemia, it is suggested that such symptoms occurring in the premenstruum may, in part, be provoked by changes in insulin-binding and glucose tolerance during this phase. The pineal hormone, melatonin, which exercises effects on gonadal function and sleep and which increases the synthesis of progesterone in the corpus luteum, varies with the phases of the menstrual cycle and is at its highest during the premenstrual and menstrual phases. There is at least one report of increased sleep duration in women in the premenstrual phase [ 451. It has also been reported that melatonin can exacerbate dysphoria if given to depressed patients, a report of particular interest in the light of the fact that women suffering from affective disorder tend to experience exacerbations of such disorder in the premenstrual phase [46, 471. To date, melatonin has not been studied in premenstrually complaining women. The growth of interest in neuroendocrinological changes occurring in psychiatric, and particularly affective illness, coupled with the growing body of data documenting the complex interaction of the sex hormones and neurotransmitter systems in the brain [48, 491 have focused attention on the possible role in premenstrual affective symptoms of serotonin, noradrenaline and dopamine [44]. Interest in a possible role for pyridoxine, on account of its part in the synthesis of dopamine and serotonin, has been a long-standing feature of premenstrual research [ 501 but to date there is no evidence indicating differences in pyridoxine absorption or metabolism in women with and without premenstrual symptoms [51]. Summarising the current status of hormones in the genesis of cyclical changes in mood, physical functioning and behaviour, there is no detectable and consistent hormonal abnormality differentiating those women who suffer from various premenstrual symptoms from those who do not nor are there detectable alterations in hormonal ratios in such women. While the consistency with which women report cyclical mood changes in the very best conducted, prospective studies is strongly suggestive of some hormonal role, quite which hormones are involved and in what way remain unclear. In the years since Frank first provided a systematic description of the syndrome, biological research has only served to reveal how complicated the underlying neuroendocrinological framework of the menstrual cycle is. In consequence, theories based on a single hormonal abnormality look increasingly frail.
BEHAVIOUR
A variety of behaviours have been reported to vary in relation to the phases of the menstrual cycle, including aggression, illness behaviour, accidents, examination and other test performances, sporting performance and sexual interest and behaviour [27]. However, before briefly referring to the more substantial studies in this area, it is worth reflecting on the methodological hazards involved in such work for it serves to explain why so many of the claims are contradictory in fact and anecdotal in
Hormones,
behaviour
and the menstrual
cycle
229
substance. First, the behaviour in question is rarely defined clearly. Aggressive behaviour, for example, may encompass everything from mild irritability to actual physical violence, sporting prowess may include quite different forms of physical skill and activity while studies of suicidal behaviour may include quite trivial acts of self-harm with successful acts of self-destruction. Secondly, many studies rely on retrospective data and self-reporting, both sources of considerable error, lack any objective assessment of the behaviour in question and do not independently confirm the precise phase of the woman’s cycle at the time the behaviour under examination occurred. Thirdly, there is the problem of attribution. Given that many of the behaviours are fairly frequent in occurrence, e.g. examinations, sporting activities, it is inevitable that many women will remember those occasions when poor performance coincided with the imminent arrival of their periods and will forget those occasions when it coincided with ovulation, the actual onset of the menses or their cessation. Fourthly, merely establishing a temporal relationship between a behaviour and a menstrual cycle phase says little about (a) whether they are causally related and (b) what the direction of the relationship is. The fact, for example, that women may be more prone to accidents in their premenstruum might well owe something to the fact that in a substantial number of women the experience of a stress such as an accident could serve to bring forward their period, thereby accentuating the statistical relationship between the two phenomena but leaving open the issue of cause. Recently, there has been a renewal of interest in the long-standing hypothesis that there is a statistically significant correlation between the premenstrual or menstrual phase of the menstrual cycle and the commission of violent crimes. While there are numerous studies reported, only two repay serious consideration. The first involved a study of the relationship between aggressive behaviour and menstruation in female inmates in a correction centre in North Carolina [52]. The second involved 50 women charged with crimes of violence in London [53]. Both studies reported a significant association between premenstrual and menstrual phase (the so-called ‘paramenstruum’) and aggressive behaviour and crime but neither found a relationship between premenstrual complaint and aggressive behaviour or violence. More recently, three cases have been described in which women successfully pleaded diminished responsibility or mitigation in British courts due to premenstrual syndrome in crimes of manslaughter, arson and assault [54]. The report of these cases provoked a substantial legal and public as well as medical response [55-581. Dalton has been moved to comment on the need ‘to differentiate those women who are jumping on the bandwagon and falsely claiming premenstrual syndrome from the genuine sufferer’ [59]. To date, evidence is lacking as to whether the premenstrual phase can provoke significant aggressive or violent behaviour in women who are not predisposed to such behaviours from the outset. A considerable literature has developed in relation to the claim that women are more prone to accidents during their premenstrual phase although again caution is required in interpreting such a claim. A largely anecdotal account of three fatal aircraft crashes [60] and Dalton’s account of 84 women hospitalised as a result of accidents [61] provide much of the basis for the view of women as accident-prone during the premenstruum and for the need for women to take special care when driving during this phase [62, 63). However, Dalton’s study was retrospective, no attempt was made to check the accuracy of each subject’s phase at the time of the
230
ANTHONYW.CLARE
accident, and no distinction was drawn between being an active or a passive participant in the accident itself. The sample was not a consecutive series of accident admissions, no control group of women who had an accident but who had not been hospitalised was employed, and crucial details concerning the nature of the accidents (e.g. whether the woman was a passenger or a driver) were omitted. The belief that a woman’s ability to function intellectually during the premenstrual phase is impaired is likewise a popular one [64]. A widely cited study is that of girls taking school examinations [65]. Lower examination scores were reported in the premenstruum but the lack of statistical data makes the study difficult to interpret satisfactorily. However, as Parlee has recently pointed out [5], even if a temporal relationship exists there is an alternative explanation ‘which opens up and is related to an extremely interesting set of empirical and theoretical issues.’ Parlee points out that in this study the girls were more likely to be in the premenstrual and menstrual phases than in the remainder of the cycle. From the material presented in the paper, it would appear that the menstrual days are not randomly distributed and the possibility that the stress of the impending examinations affected the timing of the menstrual onset is raised. Parlee suggests that ‘the correlation between cycle phase and the act of taking the examination could be the result of the relationship of each to the common factor of stress occurring before and during the examination’. More recently, Walsh and his colleagues reported on the results obtained by 244 female medical and paramedical students in all examinations taken in one academic year and found no statistically significant relationship between examination performance, personality profiles, dysmenorrhoea and menstrual cycle phase [66]. This negative finding is in line with the results of studies which have attempted to demonstrate impairment on various psychological tests in women in the premenstrual phase [67-701. The overall consensus appears to be that gross changes in performance do not occur in most women although it is still possible that some women are adversely affected. In the opinion of the authors of one extensive review of women in sport, ‘the majority of women do tend to perform less well during the premenstruum and most female athletes are able by observation over a significant number of cycles to pinpoint the time phases of each cycle during which they are at their best and worst’ [71]. However, the remainder of the literature does not provide any consistent pattern regarding the effects of menstruation on athletic performance. Indeed, the bulk of the literature is concerned with the impact of the menstrual period itself rather than the premenstruum. In so far as the premenstruum itself is concerned, the results appear equivocal. One study in this area, notable for its attempt to control some of the more contaminating variables, found different effects on different sporting behaviours of the various phases of the menstrual cycle in basketball and volleyball players at the University of Ontario [72]. The menstrual phase appeared to exercise the most negative impact whereas the impact of the premenstrual phase was very varied; some players performed better on certain tests of coordination and athletic ability during this phase than during any other while others performed more poorly. This failure to find a consistent impact of the premenstruum is in line with the anecdotal report that many a female athlete’s best performance has been obtained just before or during menstruation itself, even to the extent that records have been broken and Olympic titles won [73].
Hormones,
behaviour
and the menstrual
cycle
231
SUMMARY
To date therefore, no consistent hormonal abnormality has been linked with premenstrual affective, physical or behavioural changes and, in turn, no consistent adverse behaviour has been positively linked with any particular menstrual cycle phase. At the same time, the proportion of actively menstruating women reported to identify premenstrual changes of a severity sufficient to warrant the illness label continues to fall so that a figure somewhere between 2 and 8% seems to be established [74] where once figures in excess of 70% were claimed [75, 431. The implication would seem to be that more thorough assessment of the relatively small number of women who are clearly identifiable as sufferers from cyclically-mediated symptoms might well cast light on possible hormonal-behavioural links where larger, less careful studies of heterogeneous samples of women have failed to do so. The great majority of women appear to negotiate the premenstrual phase of their menstrual cycles with little or no impairment or difficulty. A small number appear to be vulnerable and the possibility that it is, at least in part, a biological vulnerability is hinted at in those reports which tentatively suggest that some women who are especially prone to post-partum ‘blues’ and who are particularly sensitive to oral contraceptives may be especially prone to premenstrual tension [76, 141. Finally, the ubiquity of premenstrually perceived ‘changes’ as distinct from ‘symptoms’ [77, 271 raises the possibility that the extent to which a woman experiences various premenstrual changes as complaint or illness may depend ‘more on her basic personality and the circumstances in which she experiences the cyclical changes than on the underlying cyclical mechanism’ [4]. Such a view has important implications for the direction of future research in this area.
REFERENCES 1. FRIEDMAN RC. (Ed.) Behnvior and the Menstrual Cycle. New York: Marcel Dekker, 1982. John Wiley, 1983. 2. Asso D. The Real Menstrual Cycle. Chichester: 3. GOLUB S. Lifting the Curse ofMenstruation. New York: Haworth Press, 1983. 4. SANDERS D, WARNER P, BACKSTROMT, BANCROFT J. Mood, sexuality, hormones and the menstrual cycle. 1. Changes in mood and physical state: description of subjects and methods. Psychosom Med 1983; 45: 487-501. 5. PARLEE MB. The psychology of the menstrual cycle; biological and psychological perspective, Chap. 5. In Behovior and the Menstrual Cycle (Edited by FRIEDMAN RC), pp. 77-99. New York: Marcel Dekker, 1982. 6. SOUTHAM AL, GONZACA FP. Systematic changes during the menstrual cycle. Am J Obstet Gynecol 1965;91: 142-165. 7. Moos RH. Typology of menstrual cycle symptoms. Am J Obstet Gynecoll969; 103: 390-402. 8. FRANK RT. The hormonal basis of premenstrual tension. Arch NeurolPsychiat 1931; 26: 1053SlO57. 9. BACKSTROM T, CARSTENSEN H. Estrogen and progesterone in plasma in relation to premenstrual tension. JSteroid Biochem 1974; 5: 257-260. between progesterone, oestradiol and 10. MUNDAY MR, BRUSH MC, TAYLOR RW. Correlations aldosterone levels in the premenstrual syndrome. Clin Endocrin 1981; 14: l-9. 11. DENNERSTEINL, SPENCER-GARDNER C, BROWN JB, SMITH MA, BURROWSCD. Premenstrual tensionhormonal profiles. JPsychosom Obstet Gynaecoll984; 3: 37-51. 12. HAMBURG D. Effects of progesterone on behavior. Res Pub1 Assoc Res Nerv Ment Dis 1966; 43: 251-265. profile in PMT: 13. ANDERSCH B, ABRAHAMSSON L, WENDESTAM C, OHMAN R, KAH~I L. Hormone effects of bromocriptine and diuretics. Clin Endocrin 1979; 11: 657-664. 14. BACKSTROM T, SANDERS D, LEASK R, DAVIDSON D, WARNER P, BANCROFT J. Mood, sexuality, hormones and the menstrual cycle. II. Hormone levels and their relationship to the premenstrual syndrome. Psychosom Med 1983; 45: 503-507.
232
ANTHONY
W.CLARE
15. SOMERVILLE B. The influence of progesterone and oestradiol upon migraine. Headache 1972; 12: 93-102. 16. MAILS AL, ZILKHA KJ, STUDD JW. Treatment of menstrual migraine by oestradiol implants. JNeurolNeurosurg Psychiat 1983; 46: 10441046. 17. MAG~S AL, COLLINS WP, STUDD JWW. Management of the premenstrual syndrome by subcutaneous implants of oestradiol. JPsychosom Obstet Gynaecoll984; 3: 93-99. 18. GREENE R, DALTONK. The premenstrual syndrome. BrMed J 1953; 1: 1007-1013. 19. DALTON K. The Premenstrual Syndrome and Progesterone Therapy. London: Heinemann, 1977. 20. SMITH SL. Mood and the menstrual cycle. In Topics in Psychoneuroendocrinology (Edited by SACHAR EJ). New York: Grune & Stratton, 1975. 21. SAMPSONGA. Premenstrual syndrome: a double-blind controlled trial of progesterone and placebo.
Br JPsychiat 1979; 135: 209-215. 22.
CULLBERG J. Mood changes
and menstrual
symptoms
with different
gestagen-oestrogen
combinations.
Acta Psychiat Stand 1972; 236: l-86. STEINER BJ, CARROLL BJ. The psychobiology of premenstrual dysphoria: review of theories and treatments. Psychoneuroendocrinology 1977; 2: 321-325. 24. HALBREICH U, ENDICOTT J, NEE J. Premenstrual depressive changes: value of differentiation. Arch
23.
Gen Psychiat 1983; 40: 535-542. 25. RIJBINOW DR, ROY-BYRNE P, HOBAN MC, GOLD PW, POST RM. Prospective assessment of menstrually related mood disorders. Am JPsychiat 1984; 141: 684-686. 26. KU~NER SJ, BROWN WL. History of depression as a risk factor for depression with oral contraceptives and discontinuance. J Nerv Mental Dis 1972; 155: 163- 169. 27. CLARE AW. Psychiatric and social aspects of prementrual complaint, Psychological Medicine Monograph, Suppl. 4. 28. CLICK I, BENNETT SE. Oral Contraceptive and the menstrual cycle. In Behavior nnd the Menstrual Cycle(Edited by FRIEDMAN RC), Chap. 17, pp. 345-365. New York: Marcel Dekker, 1982. 29. HORROBIN D. Prolactin: Physiology and Clinical Significance. Lamcaster: MTP Press, 1973. 30. CARROLL BJ, STEINER M. The psychobiology of premenstrual dysphoria: the role of prolactin.
Psychoneuroendocrinology 1978; 3: 17l-l 80. 3 1. MCNEILLY AS, EVANS GE, CHARD T. Observations In Human Prolactin. Amsterdam: Excerpta Medica, 32.
33. 34. 35. 36.
37.
38. 39. 40. 41.
42.
43. 44. 45.
on prolactin levels during the menstrual cycle. 1973. I&LBREICH U, ASSAELM, BEN-DAVID M, BORNSTEINR. Serum prolactin in women with premenstrual tension. Lancer 1976; 2: 654-655. BRUSH MG. The possible mechanics causing the premenstrual tension syndrome. Current Med Res Opinion 1977; 4: Suppl. 4,9-15. BENEDEK-JASZMANN LF, HEARN-STURTEVANT MD. Premenstrual tension and functional infertility: aetiology and treatment. Lancer 1976; 1: 1095-1098. GHOSEK, COPPEN A. Bromocriptine and premenstrual syndrome: a controlled study. Br Med J 1977; 1: 147-148. JANOWSKY DS, BERENS SC, DAVIS JM. Correlations between mood, weight and electrolytes during the menstrual cycle: a renin-angiotensin-aldosterone hypothesis of premenstrual tension. Psychosom Med 1973; 35: 143-154. GRAY MJ, STRAUSFELDKS, WANTANABE M, SIMS EAW, SOLOMONS. Aldosterone secretory rates in the normal menstrual cycle. JCIin Endocrin Metabolism 1968; 28: 1269-1275. SUNDSFORDJA, AAKVAAC A. Plasma renin activity, plasma renin substrate and urinary aldosterone excretion in the menstrual cycle in relation to the concentration of progesterone and oestrogens in the plasma. Acta Endocrin 1972; 71: 519-529. MUNDAY M, BRUSH MG, TAYLOR RW. Progesterone and aldosterone levels in the premenstrual tension syndrome. JEndocrin 1977; 75: 21-25. O’BRIEN PMS, CRAVEN D, SELBY C, SYMONDS EM. Treatment of premenstrual syndrome by spironolactone. Br JObstet Gynaecoll979; 86: 142-147. WAKOH T, TAKEHOSH~ A, YOSHIMOTO S. Pathophysiological study of the periodic psychosis (atypical endogenous psychosis) with special reference to the comparison with chronic schizophrenia. MieMedJl960; 10:317-336. BANCROFT J, SANDERS D, DAVIDSON D, W~L~NER P. Mood, sexuality, hormones and the menstrual cycle. III. Sexuality and the role of androgens. Psychosom Med 1983; 45: 509-516. SUTHERLAND H, STEWART I. A critical analysis of the premenstrual syndrome. Lancer 1965; 1: 1180~1183. RAUSCH J, JANOWSKY DS. Premenstrual tension: etiology. In Behavior and the Menstrual Cycle (Edited by FRIEDMAN RC), Chap. 19, pp. 397-427. New York: Marcel Dekker, 1982. PA~KAI P, JOHANNSON G, POX B. Mood, alertness and sympathetic-adrenal medullary activity during the menstrual cycle. Psychosom Med 1974; 36: 503-512.
Hormones,
behaviour
and the menstrual
cycle
233
46. KASHIWAGI T, MCCLURE JN, JR, WETZEL KD. Premenstrual affective syndrome and psychiatric disorder. Dis Nerv Sys 1976; 37: 116-l 19. 47. DIAMOND SB, RUBINSTEIN AA, DUNNER DL, FIEVE RR. Menstrual problems in women with primary affective illness. Camp Psychiut 1976; 17: 541-548. 48. CHECKLEY S. Endocrine changes in psychiatric illness. In Ctinical Neuroendocrinology (Edited by BESSER GM, MARTINI L), Chap. 6, pp. 266-282. London: Academic Press, 1982. 49. KLAIBER EL, BROVEPMAN DM, V&EL W, KENNEDY JA, NADEAU CJL. Estrogens and central nervous system function: electroencephalography, cognition and depression. In Behuvior ond the Menstrual Cycfe(Edited by FRIEDMAN RC), Chap. 13, pp. 267-289. New York: Marcel Dekker, 1982. 50. WINSTON F. Oral contraceptives, pyridoxine and depression. Am J Psychiut 1973; 130: 1217-1221. 51. GREEN J. Recent trends in the treatment of premenstrual syndrome: a critical review. In Behavior and the Menstrual Cycle (Edited by FRIEDMAN RC), Chap. 18, pp. 367-395. New York: Marcel Dekker, 1982. 52. ELLIS DP, AUSTIN P. Menstruation and aggressive behavior in a correction center for women.
JCriminul Law Criminal PoliceSci 1971; 62: 388-395. 53. D’ORBAN PT, DALTON J. Violent crime and the menstrual cycle. PsycholMed 1980; 10: 353-359. 54. DALTON K. Cyclical criminal acts in premenstrual syndrome. Luncet 1980; 11: 1070-1071. 55. BERLINS M, SMITH T. Should PMT be a woman’s all-purpose excuse? The Times, 12 November 1981, p. 12. 56. BRAHAMSD. Premenstrual syndrome: a disease of the mind? Lancet 1981; 11: 1238-1240. 57. Lancet Editorial: Premenstrual syndrome, 1981; 11: 1393-1394. 58. RICHARDS JR. PMT-an obstacle in the fight for female equality? The Listener, 29 April 1982, pp. 9-10. 59. DALTON K. Violence and the premenstrual syndrome. Police Surgeon 1982; 21: S-15. 60. WHITEHEAD RE. Women pilots. JAviution Med 1934; 5: 47-49. 61. DALTON K. Menstruation and accidents. BrMed J 1960; 11: 1425-1426. 62. WILSON EW, RENNIE PIC. TheMenstrual Cycle. London: Lloyd-Luke, 1976. 63. Royal Society for the Prevention of Accidents (1977) PMT-Premenstrual Tension. Pamphlet. 64. TIGER L. Male dominance? Yes, alas. A sexist plot? No. Times Mug 25 October 1970, p. 132. 65. DALTON K. Menstruation and examinations. Luncet 1986; 11: 1386-1388. 66. WALSH RN, BUDTZ-OLSEN B, LEADER C, CUMMINS RA. The menstrual cycle, personality and academic performance. Arch Gen Psychiut 1981; 38: 219-221. 67. WICKHAM M. The effects of the menstrual cycle on test performance. Br J Psycho/ 1958; 49: 39-41. 68. SUMMERB. Menstrual cycle changes and intellectual performance. Psychosom Med 1972; 35: 515-534. 69. GOLUB S. The effect of premenstrual anxiety and depression on cognitive function. JPersonulify Sot Psycho1 1976; 34: 99-104. 70. SLADE P, JENNER FA. Results from performance tests in different phases of the menstrual cycle.
JPsychosom Res 1980; 24: 6-8. 71. DALTON K, WILLIAMSJGP. Women in sport. In Sports Medicine (Edited by WILLIAMSJGP, SPERRYN PN), pp. 210-215. London: Edward Arnold, 1976. 72. WEARING MP, YOHOSZ MD, CAMPBELL R, LOVE EJ. The effect of the menstrual cycle on tests of physical fitness. J Sports Med Physicul Fitness 1972; 12: 38-41. 73. BAUSENSWEINI. Sportarztl. Praxis 1960; 3: 12. 74. ANDERSCH B. Epidemiological, hormonal and water balance studies on premenstrual tension. Thesis, University of Goteborg, 1980. 75. LAMB WM, ULETT GA, MASTERS WH, ROBINSON DW. Premenstrual tension, EEG, hormonal and psychiatric evaluation. Am JPsychiut 1953; 109: 840-848. 76. STEIN G. The maternity blues. In Motherhood and Mental Illness (Edited by BROCK~NGTON I, KUMAR R), pp. 119-154. London: Academic Press, 1982. 77. VAN KEEP PA, LEHERT P. The premenstrual syndrome-an epidemiological and statistical exercise. In The Premenstrual Syndrome (Edited by VAN KEEP PA, UTIAN WH), pp. 31-42. Lancaster: MTP Press, 1981.
Research,
Vol. 29, No. 3, pp. 225-233.
W22-3999/85 $3.M)+ .@I Pergamon Press Ltd.
1985
INVITED REVIEW HORMONES,
BEHAVIOUR
AND THE MENSTRUAL
CYCLE
ANTHONY W. CLARE* (Received
14 November
1984)
INTRODUCTION THE POSSIBLE influence
of the menstrual cycle on the physical, psychological and social functioning of women has become a subject of considerable interest over the past two decades [l-3]. Research aimed at clarifying the precise relationship between hormonal changes, physical symptoms, psychological alterations and behaviour has intensified as increasingly sophisticated methods of hormone assay have been developed [4]. However, attempts to relate specific hormonal changes within the menstrual cycle to specific behaviours or symptoms have in the main foundered on the many methodological rocks which litter this terrain [5]. Whereas the cyclical patterns of the major hormones, including LH, FSH, oestrogen, progesterone, testosterone and prolactin, have been accurately described, the problems surrounding the precise description and measurement of mood, behaviour, subjective physical symptomatology, sexual interest and activity, continue to bedevil efforts at clarification. Most attention, over the years, has been paid to the premenstrual phase of the menstrual cycle, a phase variously defined but invariably including the three to four days immediately preceding the onset of the menses. A remarkable array of physical and psychological symptoms has been reported to occur or be exacerbated during this phase [6, 71. These include feelings of reduced well-being and performance, symptoms of anxiety, depression and irritability, complaints of breast tenderness, swelling, weight gain, headache and a reduction in sexual drive. This brief review considers the present status of the main hormones with regard to premenstrual symptomatology and behaviour change.
HORMONES From the very outset, Frank, in his initial description of the premenstrual syndrome [8], assumed that the cyclical nature of the symptoms was due to alterations in hormonal levels. Since that time each of the main hormones has come under suspicion (Table I). The majority of recent studies have focused on an alleged progesterone deficiency and a number of reports, from studies utilising plasma radio-immunassay of hormones and urinary hormonal excretion analyses, have provided some support [9-l 11. Emotional symptoms have been related to decreasing levels of progesterone in that negative affective symptoms often appear to begin early in the luteal phase and increase in intensity as progesterone drops from its luteal *Professor and Head of Department of Psychological Medicine, St. Bartholomew’s Hospital Medical College, West Smithfield, London. 225
226
ANTHONY W. CLARE TABLE I.-THEORIES OF HORMONALCAUSATIONIN THE PREMENSTRUAL SYNDROME Lowered progesterone Increased oestrogen Altered oestrogen/progesterone Oestrogen withdrawal Increased aldosterone Excess androgenic steroids Increased prolactin
ratio
Dalton [ 191 Cullberg [22] Frank [8] Somerville [ 151 Janowsky [36] Wakoh et al. [41] Carroll and Steiner
1301
peak to its low premenstrual level. In addition, depressive symptoms have been provoked by the withdrawal of exogenous progesterone treatment [12]. However, a number of recent, methodologically careful studies have failed to find a deficiency of progesterone [13, 141 and it is now being suggested that premenstrual affective symptoms might be better understood not in terms of an absolute progesterone deficiency but of withdrawal of the hormone. Such a theory requires an additional causal or aggravating factor given that while every ovulating woman experiences some degree of progesterone withdrawal only a proportion of women complain of symptoms. An alternative theory implicates excess oestrogen. During the luteal phase, oestrogen levels are indeed high but the levels are falling when premenstrual symptoms are generally at their most severe. In turn, oestrogen withdrawal has been proposed and it has certainly been implicated in cases of menstrual migraine [15, 161. The use of subcutaneous implants of oestradiol has recently been reported [ 171 but the method employed to diagnose PMS was not made explicit, the outcome measures were not specified and the treatment was not blind. In the light of failure to find a consistent excess or deficiency of either oestrogen or progesterone, attention has switched to the possibility that there is a disturbance in the oestrogen-progesterone ratio. Greene and Dalton suggested such a cause over thirty years ago [18] and Dalton has since argued in a series of influential papers and a book devoted to the subject [19] that progesterone, by suppository or intramuscular injections, is the treatment of choice in this condition. However, this claim is hotly contested [20, 211. Theories linking oestrogen and progesterone with the premenstrual syndrome derive some support from studies of oral contraceptives. Cullberg [22] found that women with a history of premenstrual irritability and who took progesterone-dominant pills had significantly lower incidences of adverse psychological effects compared with women who took oestrogen-dominant pills. In contrast, women without a pre-treatment history of irritability, noted more adverse reactions with the progesterone-dominant pills, suggesting that these women might have been hormonally balanced before treatment or were even concealing an Cullberg speculated that premenstrual endogenous progesterone dominance. irritability might at least be partly due to an endogenous oestrogen dominance or susceptibility to endogenous oestrogen around the premenstrual phase. Steiner and Carroll [23] have, in turn, suggested that premenstrual irritability, anxiety and hostility may be a separate cluster of symptoms from premenstrual depression and there is a suggestion that the former cluster might be related to a high oestrogenprogesterone ratio whereas the latter cluster might be related to a low oestrogenprogesterone ratio. To date this remains a speculation for despite efforts to
227
Hormones, behaviour and the menstrual cycle
distinguish between the different affective symptoms reported premenstrually and the development of more detailed symptom classifications and measures [24, 251 no study reporting specific hormonal abnormalities consistently linked with particular premenstrual symptoms or symptom clusters has been published. Nor is it clear quite what the effect of oral contraceptives on premenstrual mood and behaviour is. While there is some evidence that many women with premenstrual symptoms complain of adverse side-effects with oral contraceptive use [26, 271 there have been a number of reports recently reviewed by Click and Bennett [28] to the effect that such substances improve global functioning and mood during the premenstrual syndrome. Prolactin is another hormone which has been implicated in the premenstrual syndrome [29, 301. Its levels fluctuate throughout the cycle with a peak at ovulation and later in the mid and late-luteal phases [ 3 1 ] . Raised prolactin levels were reported in one series of 28 patients [32] while Brush [33] reported finding a number of premenstrual complainers with ‘high normal’ prolactin levels which appeared to be associated with low progesterone levels. Benedek-Jaszmann and Hearn-Sturtevant [34], believing prolactin to be ‘the key to the aetiology of the syndrome’, used the prolactin-suppressant drug, bromocriptine, in a double-blind, controlled study and found it superior to placebo in relieving breast symptoms, swelling, weight gain and mood disturbances. However, the reported levels of pre-treatment prolactin levels were within normal limits and the sample was composed of women attending for treatment of prolonged infertility. A subsequent double-blind, controlled trial of the drug did not find it superior to placebo [35]. It has been suggested that premenstrual mood disturbance and symptoms such as weight gain and fluid retention might be related to activation of the reninangiotensin-aldosterone system [ 361. In normal women, a slight mid-cycle elevation of urinary aldosterone excretion followed by a luteal rise which peaked during the premenstrual phase and fell just before or soon after the onset of the period has been reported [37]. Quite why aldosterone rises in the luteal phase is not clear but it has been suggested that the sodium-wasting effect exercised by progesterone might be the stimulus [38]. Janowsky and his colleagues [36] did report alterations in sodium and potassium excretion during the premenstruum and suggested that premenstrual affective symptoms might be more closely related in time to changes in fluid and electrolytes than to changes in oestrogen and progesterone. This group did report increased aldosterone excretion in the premenstrual phase of symptomatic women but other studies have failed to replicate this finding [39,40]. It has also been suggested that the manifestations of the premenstrual syndrome may be related to the fluctuations of androgens [41]. Interest persists not least because of the suggested association between androgens and both sexual drive and aggression although the basis for such a link is difficult to establish [42]. No differences have been reported in androstenedione levels between women with high and low degrees of cyclical mood change [14] although there is some evidence that women with marked cyclical mood change report an associated cyclical pattern of sexual feelings [42]. The common exacerbation of acne which occurs in some women who manifest other signs and symptoms of the premenstrual syndrome [43] also suggests to some observers a possible causal role for androgenic steroids. Other
hormones
implicated
in the syndrome
include
insulin
and melatonin.
An
228
ANTHONY
W.
CLARE
inverse relationship has been reported between insulin binding to monocytes and levels of 17-beta-oestradiol, progesterone and 17 alpha-hydroxyprogesterone, indicating that sex steroids may play a role in the control of insulin receptors. Rausch and Janowsky [44], reviewing this data, suggest that it may in part explain the reduction of glucose tolerance in the second half of the menstrual cycle which has been reported. Given that anxiety and irritability are symptoms often seen in transient hypoglycaemia, it is suggested that such symptoms occurring in the premenstruum may, in part, be provoked by changes in insulin-binding and glucose tolerance during this phase. The pineal hormone, melatonin, which exercises effects on gonadal function and sleep and which increases the synthesis of progesterone in the corpus luteum, varies with the phases of the menstrual cycle and is at its highest during the premenstrual and menstrual phases. There is at least one report of increased sleep duration in women in the premenstrual phase [ 451. It has also been reported that melatonin can exacerbate dysphoria if given to depressed patients, a report of particular interest in the light of the fact that women suffering from affective disorder tend to experience exacerbations of such disorder in the premenstrual phase [46, 471. To date, melatonin has not been studied in premenstrually complaining women. The growth of interest in neuroendocrinological changes occurring in psychiatric, and particularly affective illness, coupled with the growing body of data documenting the complex interaction of the sex hormones and neurotransmitter systems in the brain [48, 491 have focused attention on the possible role in premenstrual affective symptoms of serotonin, noradrenaline and dopamine [44]. Interest in a possible role for pyridoxine, on account of its part in the synthesis of dopamine and serotonin, has been a long-standing feature of premenstrual research [ 501 but to date there is no evidence indicating differences in pyridoxine absorption or metabolism in women with and without premenstrual symptoms [51]. Summarising the current status of hormones in the genesis of cyclical changes in mood, physical functioning and behaviour, there is no detectable and consistent hormonal abnormality differentiating those women who suffer from various premenstrual symptoms from those who do not nor are there detectable alterations in hormonal ratios in such women. While the consistency with which women report cyclical mood changes in the very best conducted, prospective studies is strongly suggestive of some hormonal role, quite which hormones are involved and in what way remain unclear. In the years since Frank first provided a systematic description of the syndrome, biological research has only served to reveal how complicated the underlying neuroendocrinological framework of the menstrual cycle is. In consequence, theories based on a single hormonal abnormality look increasingly frail.
BEHAVIOUR
A variety of behaviours have been reported to vary in relation to the phases of the menstrual cycle, including aggression, illness behaviour, accidents, examination and other test performances, sporting performance and sexual interest and behaviour [27]. However, before briefly referring to the more substantial studies in this area, it is worth reflecting on the methodological hazards involved in such work for it serves to explain why so many of the claims are contradictory in fact and anecdotal in
Hormones,
behaviour
and the menstrual
cycle
229
substance. First, the behaviour in question is rarely defined clearly. Aggressive behaviour, for example, may encompass everything from mild irritability to actual physical violence, sporting prowess may include quite different forms of physical skill and activity while studies of suicidal behaviour may include quite trivial acts of self-harm with successful acts of self-destruction. Secondly, many studies rely on retrospective data and self-reporting, both sources of considerable error, lack any objective assessment of the behaviour in question and do not independently confirm the precise phase of the woman’s cycle at the time the behaviour under examination occurred. Thirdly, there is the problem of attribution. Given that many of the behaviours are fairly frequent in occurrence, e.g. examinations, sporting activities, it is inevitable that many women will remember those occasions when poor performance coincided with the imminent arrival of their periods and will forget those occasions when it coincided with ovulation, the actual onset of the menses or their cessation. Fourthly, merely establishing a temporal relationship between a behaviour and a menstrual cycle phase says little about (a) whether they are causally related and (b) what the direction of the relationship is. The fact, for example, that women may be more prone to accidents in their premenstruum might well owe something to the fact that in a substantial number of women the experience of a stress such as an accident could serve to bring forward their period, thereby accentuating the statistical relationship between the two phenomena but leaving open the issue of cause. Recently, there has been a renewal of interest in the long-standing hypothesis that there is a statistically significant correlation between the premenstrual or menstrual phase of the menstrual cycle and the commission of violent crimes. While there are numerous studies reported, only two repay serious consideration. The first involved a study of the relationship between aggressive behaviour and menstruation in female inmates in a correction centre in North Carolina [52]. The second involved 50 women charged with crimes of violence in London [53]. Both studies reported a significant association between premenstrual and menstrual phase (the so-called ‘paramenstruum’) and aggressive behaviour and crime but neither found a relationship between premenstrual complaint and aggressive behaviour or violence. More recently, three cases have been described in which women successfully pleaded diminished responsibility or mitigation in British courts due to premenstrual syndrome in crimes of manslaughter, arson and assault [54]. The report of these cases provoked a substantial legal and public as well as medical response [55-581. Dalton has been moved to comment on the need ‘to differentiate those women who are jumping on the bandwagon and falsely claiming premenstrual syndrome from the genuine sufferer’ [59]. To date, evidence is lacking as to whether the premenstrual phase can provoke significant aggressive or violent behaviour in women who are not predisposed to such behaviours from the outset. A considerable literature has developed in relation to the claim that women are more prone to accidents during their premenstrual phase although again caution is required in interpreting such a claim. A largely anecdotal account of three fatal aircraft crashes [60] and Dalton’s account of 84 women hospitalised as a result of accidents [61] provide much of the basis for the view of women as accident-prone during the premenstruum and for the need for women to take special care when driving during this phase [62, 63). However, Dalton’s study was retrospective, no attempt was made to check the accuracy of each subject’s phase at the time of the
230
ANTHONYW.CLARE
accident, and no distinction was drawn between being an active or a passive participant in the accident itself. The sample was not a consecutive series of accident admissions, no control group of women who had an accident but who had not been hospitalised was employed, and crucial details concerning the nature of the accidents (e.g. whether the woman was a passenger or a driver) were omitted. The belief that a woman’s ability to function intellectually during the premenstrual phase is impaired is likewise a popular one [64]. A widely cited study is that of girls taking school examinations [65]. Lower examination scores were reported in the premenstruum but the lack of statistical data makes the study difficult to interpret satisfactorily. However, as Parlee has recently pointed out [5], even if a temporal relationship exists there is an alternative explanation ‘which opens up and is related to an extremely interesting set of empirical and theoretical issues.’ Parlee points out that in this study the girls were more likely to be in the premenstrual and menstrual phases than in the remainder of the cycle. From the material presented in the paper, it would appear that the menstrual days are not randomly distributed and the possibility that the stress of the impending examinations affected the timing of the menstrual onset is raised. Parlee suggests that ‘the correlation between cycle phase and the act of taking the examination could be the result of the relationship of each to the common factor of stress occurring before and during the examination’. More recently, Walsh and his colleagues reported on the results obtained by 244 female medical and paramedical students in all examinations taken in one academic year and found no statistically significant relationship between examination performance, personality profiles, dysmenorrhoea and menstrual cycle phase [66]. This negative finding is in line with the results of studies which have attempted to demonstrate impairment on various psychological tests in women in the premenstrual phase [67-701. The overall consensus appears to be that gross changes in performance do not occur in most women although it is still possible that some women are adversely affected. In the opinion of the authors of one extensive review of women in sport, ‘the majority of women do tend to perform less well during the premenstruum and most female athletes are able by observation over a significant number of cycles to pinpoint the time phases of each cycle during which they are at their best and worst’ [71]. However, the remainder of the literature does not provide any consistent pattern regarding the effects of menstruation on athletic performance. Indeed, the bulk of the literature is concerned with the impact of the menstrual period itself rather than the premenstruum. In so far as the premenstruum itself is concerned, the results appear equivocal. One study in this area, notable for its attempt to control some of the more contaminating variables, found different effects on different sporting behaviours of the various phases of the menstrual cycle in basketball and volleyball players at the University of Ontario [72]. The menstrual phase appeared to exercise the most negative impact whereas the impact of the premenstrual phase was very varied; some players performed better on certain tests of coordination and athletic ability during this phase than during any other while others performed more poorly. This failure to find a consistent impact of the premenstruum is in line with the anecdotal report that many a female athlete’s best performance has been obtained just before or during menstruation itself, even to the extent that records have been broken and Olympic titles won [73].
Hormones,
behaviour
and the menstrual
cycle
231
SUMMARY
To date therefore, no consistent hormonal abnormality has been linked with premenstrual affective, physical or behavioural changes and, in turn, no consistent adverse behaviour has been positively linked with any particular menstrual cycle phase. At the same time, the proportion of actively menstruating women reported to identify premenstrual changes of a severity sufficient to warrant the illness label continues to fall so that a figure somewhere between 2 and 8% seems to be established [74] where once figures in excess of 70% were claimed [75, 431. The implication would seem to be that more thorough assessment of the relatively small number of women who are clearly identifiable as sufferers from cyclically-mediated symptoms might well cast light on possible hormonal-behavioural links where larger, less careful studies of heterogeneous samples of women have failed to do so. The great majority of women appear to negotiate the premenstrual phase of their menstrual cycles with little or no impairment or difficulty. A small number appear to be vulnerable and the possibility that it is, at least in part, a biological vulnerability is hinted at in those reports which tentatively suggest that some women who are especially prone to post-partum ‘blues’ and who are particularly sensitive to oral contraceptives may be especially prone to premenstrual tension [76, 141. Finally, the ubiquity of premenstrually perceived ‘changes’ as distinct from ‘symptoms’ [77, 271 raises the possibility that the extent to which a woman experiences various premenstrual changes as complaint or illness may depend ‘more on her basic personality and the circumstances in which she experiences the cyclical changes than on the underlying cyclical mechanism’ [4]. Such a view has important implications for the direction of future research in this area.
REFERENCES 1. FRIEDMAN RC. (Ed.) Behnvior and the Menstrual Cycle. New York: Marcel Dekker, 1982. John Wiley, 1983. 2. Asso D. The Real Menstrual Cycle. Chichester: 3. GOLUB S. Lifting the Curse ofMenstruation. New York: Haworth Press, 1983. 4. SANDERS D, WARNER P, BACKSTROMT, BANCROFT J. Mood, sexuality, hormones and the menstrual cycle. 1. Changes in mood and physical state: description of subjects and methods. Psychosom Med 1983; 45: 487-501. 5. PARLEE MB. The psychology of the menstrual cycle; biological and psychological perspective, Chap. 5. In Behovior and the Menstrual Cycle (Edited by FRIEDMAN RC), pp. 77-99. New York: Marcel Dekker, 1982. 6. SOUTHAM AL, GONZACA FP. Systematic changes during the menstrual cycle. Am J Obstet Gynecol 1965;91: 142-165. 7. Moos RH. Typology of menstrual cycle symptoms. Am J Obstet Gynecoll969; 103: 390-402. 8. FRANK RT. The hormonal basis of premenstrual tension. Arch NeurolPsychiat 1931; 26: 1053SlO57. 9. BACKSTROM T, CARSTENSEN H. Estrogen and progesterone in plasma in relation to premenstrual tension. JSteroid Biochem 1974; 5: 257-260. between progesterone, oestradiol and 10. MUNDAY MR, BRUSH MC, TAYLOR RW. Correlations aldosterone levels in the premenstrual syndrome. Clin Endocrin 1981; 14: l-9. 11. DENNERSTEINL, SPENCER-GARDNER C, BROWN JB, SMITH MA, BURROWSCD. Premenstrual tensionhormonal profiles. JPsychosom Obstet Gynaecoll984; 3: 37-51. 12. HAMBURG D. Effects of progesterone on behavior. Res Pub1 Assoc Res Nerv Ment Dis 1966; 43: 251-265. profile in PMT: 13. ANDERSCH B, ABRAHAMSSON L, WENDESTAM C, OHMAN R, KAH~I L. Hormone effects of bromocriptine and diuretics. Clin Endocrin 1979; 11: 657-664. 14. BACKSTROM T, SANDERS D, LEASK R, DAVIDSON D, WARNER P, BANCROFT J. Mood, sexuality, hormones and the menstrual cycle. II. Hormone levels and their relationship to the premenstrual syndrome. Psychosom Med 1983; 45: 503-507.
232
ANTHONY
W.CLARE
15. SOMERVILLE B. The influence of progesterone and oestradiol upon migraine. Headache 1972; 12: 93-102. 16. MAILS AL, ZILKHA KJ, STUDD JW. Treatment of menstrual migraine by oestradiol implants. JNeurolNeurosurg Psychiat 1983; 46: 10441046. 17. MAG~S AL, COLLINS WP, STUDD JWW. Management of the premenstrual syndrome by subcutaneous implants of oestradiol. JPsychosom Obstet Gynaecoll984; 3: 93-99. 18. GREENE R, DALTONK. The premenstrual syndrome. BrMed J 1953; 1: 1007-1013. 19. DALTON K. The Premenstrual Syndrome and Progesterone Therapy. London: Heinemann, 1977. 20. SMITH SL. Mood and the menstrual cycle. In Topics in Psychoneuroendocrinology (Edited by SACHAR EJ). New York: Grune & Stratton, 1975. 21. SAMPSONGA. Premenstrual syndrome: a double-blind controlled trial of progesterone and placebo.
Br JPsychiat 1979; 135: 209-215. 22.
CULLBERG J. Mood changes
and menstrual
symptoms
with different
gestagen-oestrogen
combinations.
Acta Psychiat Stand 1972; 236: l-86. STEINER BJ, CARROLL BJ. The psychobiology of premenstrual dysphoria: review of theories and treatments. Psychoneuroendocrinology 1977; 2: 321-325. 24. HALBREICH U, ENDICOTT J, NEE J. Premenstrual depressive changes: value of differentiation. Arch
23.
Gen Psychiat 1983; 40: 535-542. 25. RIJBINOW DR, ROY-BYRNE P, HOBAN MC, GOLD PW, POST RM. Prospective assessment of menstrually related mood disorders. Am JPsychiat 1984; 141: 684-686. 26. KU~NER SJ, BROWN WL. History of depression as a risk factor for depression with oral contraceptives and discontinuance. J Nerv Mental Dis 1972; 155: 163- 169. 27. CLARE AW. Psychiatric and social aspects of prementrual complaint, Psychological Medicine Monograph, Suppl. 4. 28. CLICK I, BENNETT SE. Oral Contraceptive and the menstrual cycle. In Behavior nnd the Menstrual Cycle(Edited by FRIEDMAN RC), Chap. 17, pp. 345-365. New York: Marcel Dekker, 1982. 29. HORROBIN D. Prolactin: Physiology and Clinical Significance. Lamcaster: MTP Press, 1973. 30. CARROLL BJ, STEINER M. The psychobiology of premenstrual dysphoria: the role of prolactin.
Psychoneuroendocrinology 1978; 3: 17l-l 80. 3 1. MCNEILLY AS, EVANS GE, CHARD T. Observations In Human Prolactin. Amsterdam: Excerpta Medica, 32.
33. 34. 35. 36.
37.
38. 39. 40. 41.
42.
43. 44. 45.
on prolactin levels during the menstrual cycle. 1973. I&LBREICH U, ASSAELM, BEN-DAVID M, BORNSTEINR. Serum prolactin in women with premenstrual tension. Lancer 1976; 2: 654-655. BRUSH MG. The possible mechanics causing the premenstrual tension syndrome. Current Med Res Opinion 1977; 4: Suppl. 4,9-15. BENEDEK-JASZMANN LF, HEARN-STURTEVANT MD. Premenstrual tension and functional infertility: aetiology and treatment. Lancer 1976; 1: 1095-1098. GHOSEK, COPPEN A. Bromocriptine and premenstrual syndrome: a controlled study. Br Med J 1977; 1: 147-148. JANOWSKY DS, BERENS SC, DAVIS JM. Correlations between mood, weight and electrolytes during the menstrual cycle: a renin-angiotensin-aldosterone hypothesis of premenstrual tension. Psychosom Med 1973; 35: 143-154. GRAY MJ, STRAUSFELDKS, WANTANABE M, SIMS EAW, SOLOMONS. Aldosterone secretory rates in the normal menstrual cycle. JCIin Endocrin Metabolism 1968; 28: 1269-1275. SUNDSFORDJA, AAKVAAC A. Plasma renin activity, plasma renin substrate and urinary aldosterone excretion in the menstrual cycle in relation to the concentration of progesterone and oestrogens in the plasma. Acta Endocrin 1972; 71: 519-529. MUNDAY M, BRUSH MG, TAYLOR RW. Progesterone and aldosterone levels in the premenstrual tension syndrome. JEndocrin 1977; 75: 21-25. O’BRIEN PMS, CRAVEN D, SELBY C, SYMONDS EM. Treatment of premenstrual syndrome by spironolactone. Br JObstet Gynaecoll979; 86: 142-147. WAKOH T, TAKEHOSH~ A, YOSHIMOTO S. Pathophysiological study of the periodic psychosis (atypical endogenous psychosis) with special reference to the comparison with chronic schizophrenia. MieMedJl960; 10:317-336. BANCROFT J, SANDERS D, DAVIDSON D, W~L~NER P. Mood, sexuality, hormones and the menstrual cycle. III. Sexuality and the role of androgens. Psychosom Med 1983; 45: 509-516. SUTHERLAND H, STEWART I. A critical analysis of the premenstrual syndrome. Lancer 1965; 1: 1180~1183. RAUSCH J, JANOWSKY DS. Premenstrual tension: etiology. In Behavior and the Menstrual Cycle (Edited by FRIEDMAN RC), Chap. 19, pp. 397-427. New York: Marcel Dekker, 1982. PA~KAI P, JOHANNSON G, POX B. Mood, alertness and sympathetic-adrenal medullary activity during the menstrual cycle. Psychosom Med 1974; 36: 503-512.
Hormones,
behaviour
and the menstrual
cycle
233
46. KASHIWAGI T, MCCLURE JN, JR, WETZEL KD. Premenstrual affective syndrome and psychiatric disorder. Dis Nerv Sys 1976; 37: 116-l 19. 47. DIAMOND SB, RUBINSTEIN AA, DUNNER DL, FIEVE RR. Menstrual problems in women with primary affective illness. Camp Psychiut 1976; 17: 541-548. 48. CHECKLEY S. Endocrine changes in psychiatric illness. In Ctinical Neuroendocrinology (Edited by BESSER GM, MARTINI L), Chap. 6, pp. 266-282. London: Academic Press, 1982. 49. KLAIBER EL, BROVEPMAN DM, V&EL W, KENNEDY JA, NADEAU CJL. Estrogens and central nervous system function: electroencephalography, cognition and depression. In Behuvior ond the Menstrual Cycfe(Edited by FRIEDMAN RC), Chap. 13, pp. 267-289. New York: Marcel Dekker, 1982. 50. WINSTON F. Oral contraceptives, pyridoxine and depression. Am J Psychiut 1973; 130: 1217-1221. 51. GREEN J. Recent trends in the treatment of premenstrual syndrome: a critical review. In Behavior and the Menstrual Cycle (Edited by FRIEDMAN RC), Chap. 18, pp. 367-395. New York: Marcel Dekker, 1982. 52. ELLIS DP, AUSTIN P. Menstruation and aggressive behavior in a correction center for women.
JCriminul Law Criminal PoliceSci 1971; 62: 388-395. 53. D’ORBAN PT, DALTON J. Violent crime and the menstrual cycle. PsycholMed 1980; 10: 353-359. 54. DALTON K. Cyclical criminal acts in premenstrual syndrome. Luncet 1980; 11: 1070-1071. 55. BERLINS M, SMITH T. Should PMT be a woman’s all-purpose excuse? The Times, 12 November 1981, p. 12. 56. BRAHAMSD. Premenstrual syndrome: a disease of the mind? Lancet 1981; 11: 1238-1240. 57. Lancet Editorial: Premenstrual syndrome, 1981; 11: 1393-1394. 58. RICHARDS JR. PMT-an obstacle in the fight for female equality? The Listener, 29 April 1982, pp. 9-10. 59. DALTON K. Violence and the premenstrual syndrome. Police Surgeon 1982; 21: S-15. 60. WHITEHEAD RE. Women pilots. JAviution Med 1934; 5: 47-49. 61. DALTON K. Menstruation and accidents. BrMed J 1960; 11: 1425-1426. 62. WILSON EW, RENNIE PIC. TheMenstrual Cycle. London: Lloyd-Luke, 1976. 63. Royal Society for the Prevention of Accidents (1977) PMT-Premenstrual Tension. Pamphlet. 64. TIGER L. Male dominance? Yes, alas. A sexist plot? No. Times Mug 25 October 1970, p. 132. 65. DALTON K. Menstruation and examinations. Luncet 1986; 11: 1386-1388. 66. WALSH RN, BUDTZ-OLSEN B, LEADER C, CUMMINS RA. The menstrual cycle, personality and academic performance. Arch Gen Psychiut 1981; 38: 219-221. 67. WICKHAM M. The effects of the menstrual cycle on test performance. Br J Psycho/ 1958; 49: 39-41. 68. SUMMERB. Menstrual cycle changes and intellectual performance. Psychosom Med 1972; 35: 515-534. 69. GOLUB S. The effect of premenstrual anxiety and depression on cognitive function. JPersonulify Sot Psycho1 1976; 34: 99-104. 70. SLADE P, JENNER FA. Results from performance tests in different phases of the menstrual cycle.
JPsychosom Res 1980; 24: 6-8. 71. DALTON K, WILLIAMSJGP. Women in sport. In Sports Medicine (Edited by WILLIAMSJGP, SPERRYN PN), pp. 210-215. London: Edward Arnold, 1976. 72. WEARING MP, YOHOSZ MD, CAMPBELL R, LOVE EJ. The effect of the menstrual cycle on tests of physical fitness. J Sports Med Physicul Fitness 1972; 12: 38-41. 73. BAUSENSWEINI. Sportarztl. Praxis 1960; 3: 12. 74. ANDERSCH B. Epidemiological, hormonal and water balance studies on premenstrual tension. Thesis, University of Goteborg, 1980. 75. LAMB WM, ULETT GA, MASTERS WH, ROBINSON DW. Premenstrual tension, EEG, hormonal and psychiatric evaluation. Am JPsychiut 1953; 109: 840-848. 76. STEIN G. The maternity blues. In Motherhood and Mental Illness (Edited by BROCK~NGTON I, KUMAR R), pp. 119-154. London: Academic Press, 1982. 77. VAN KEEP PA, LEHERT P. The premenstrual syndrome-an epidemiological and statistical exercise. In The Premenstrual Syndrome (Edited by VAN KEEP PA, UTIAN WH), pp. 31-42. Lancaster: MTP Press, 1981.