Serotonergic dysfunction in women with pure premenstrual dysphoric disorder: is the fenfluramine challenge test still relevant?

Psychiatry Research 87 Ž1999. 107]115

Serotonergic dysfunction in women with pure premenstrual dysphoric disorder: is the fenfluramine challenge test still relevant? Meir Steiner a,U , Lakshmi N. Yathamb, Margaret Coote c , Annette Wilkins c , Patricia Lepage a a

Departments of Psychiatry and Beha¨ ioural Neurosciences, St. Joseph’s Hospital, McMaster Uni¨ ersity, Hamilton, Ontario, Canada, L8N 4A6 b Department of Psychiatry, Uni¨ ersity of British Columbia, Vancou¨ er, British Columbia, Canada c Department of Psychiatry, St. Joseph’s Hospital, Hamilton, Ontario, Canada Received 1 April 1998; received in revised form 21 June 1999; accepted 22 June 1999

Abstract The fenfluramine ŽFEN. neuroendocrine challenge paradigm, which involves measuring the response of prolactin ŽPRL. release to an oral challenge dose of FEN, provides a means of assessing serotonin Ž5-HT. function. The purpose of this study was to ascertain the role of 5-HT in premenstrual dysphoric disorder ŽPMDD. by measuring: Ž1. PRL and cortisol ŽCORT. responses to FEN; and Ž2. platelet 3 H-imipramine binding levels, in females with pure PMDD Žwithout a past or present comorbid mood disorder. in comparison to healthy controls. FEN challenge tests were administered to nine female patients with pure PMDD and nine healthy female controls during the follicular and late luteal phases of a menstrual cycle. There were no differences in the PRL response to FEN for women with PMDD compared to healthy controls. However, the trend toward a delayed response to FEN and a significant negative correlation between Dmax PRL and basal CORT in patients but not in controls during both phases of the menstrual cycle suggest an underlying 5-HT dysfunction in patients as compared to controls. This is further supported by the finding of significantly lower Bmax 3 H-imipramine binding levels in the patients during the late luteal phase. Q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Premenstrual dysphoria; Serotonin; Fenfluramine challenge

U

Corresponding author. Tel.: q1-905-522-1155, ext. 3605; fax: q1-905-521-6098. E-mail address: [email protected] ŽM. Steiner.

0165-1781r99r$ - see front matter Q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 9 9 . 0 0 0 6 2 - 1

108

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115

1. Introduction Premenstrual syndrome is characterized by a multitude of physical and mood symptoms that appear during the week prior to menstruation and resolve within a week after onset of menses. Most women in their reproductive years experience some premenstrual symptoms. However, 3]8% of these women experience extremely distressing premenstrual mood symptoms that interfere with their lifestyle, relationships and occupational functioning Žfor review, see Steiner, 1997.. This more severe presentation has now been recognized by and included as research diagnostic criteria in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition ŽDSM-IV. as premenstrual dysphoric disorder ŽPMDD; American Psychiatric Association, 1994.. The essential features of PMDD, in particular irritability, depressed mood, marked anxiety, affective lability and anger, are phenomenologically consistent with the serotonin Ž5-HT. hypothesis of depression Žreviewed in Lepage and Steiner, 1991., but to date the only convincing evidence to support this hypothesis is the treatment literature which has demonstrated consistently the efficacy of serotonin re-uptake inhibitors ŽSRIs. in the treatment of this disorder ŽSteiner et al., 1997.. Two studies have compared serotonergic and non-serotonergic antidepressants head-to-head in women with this disorder further confirming that the response to serotonergic agents was not only significantly superior to placebo but also to the other non-serotonergic drug ŽEriksson et al., 1995; Pearlstein et al., 1997.. The fenfluramine neuroendocrine paradigm ŽFEN. involves measuring the magnitude of prolactin ŽPRL. release to an oral challenge dose of fenfluramine as a means of assessing central 5-HT function. Fixed and weight-dependent doses of d-fenfluramine and d,l-fenfluramine have been used to assess 5-HT function in a variety of psychiatric disorders. Blunted PRL responses to FEN have been consistently associated with episodes of major depression, with impulsive aggression in a variety of personality disorders and with suicidality in depressed patients Žfor reviews,

see Yatham and Steiner, 1993; Newman et al., 1998.. Only two studies have used FEN as a probe in women with premenstrual syndrome. Of these, one reported no difference in PRL release to d-FEN ŽBancroft and Cook, 1995., whereas the other found a blunted PRL response to d,l-FEN ŽFitzGerald et al., 1997. when patients were compared to controls. PRL responses induced by both d- and d,l-FEN are attenuated by 5-HT2 antagonists ŽNewman et al., 1998. and d,l-fenfluramine has been found to evoke a similar PRL response to that evoked by d-fenfluramine in healthy human subjects ŽCoccaro et al., 1996a.. Hence, differences in pharmacology of the two compounds are unlikely to explain the discrepancy in PRL release between the two studies. These data are also difficult to interpret because the patients in the former study experienced premenstrual depression, but may not have met criteria for PMDD, whereas the nine patients in the latter study met criteria not only for PMDD but also had a past history of major depression or dysthymia. Platelets have also been used as a model for central nervous system 5-HT function in studies of major depression. Platelets have 5-HT transporter and 5-HT2 receptor sites similar to central 5-HT neurons. Ashby et al. Ž1988. reported that the Vmax of 5-HT uptake and 5-HT content in platelets of women with premenstrual symptoms were significantly decreased during the premenstrual phase compared to findings in control subjects. Rojansky et al. Ž1991. reported that subjects with dysphoric premenstrual changes demonstrated lower 3 H-imipramine receptor binding compared to controls during the early luteal phase only, whereas Steege et al. Ž1992. showed that both follicular and luteal phase levels were lower in patients but reached statistical significance only in the follicular phase. While far from conclusive, these reports support the hypothesis that fluctuations in central serotonergic function may contribute to premenstrual dysphoric symptoms. The purpose of this study was to further ascertain the role of 5-HT in PMDD by measuring: Ž1.

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115

PRL and cortisol ŽCORT. responses to FEN; and Ž2. platelet 3 H-imipramine binding levels in women with pure PMDD Žwithout a past or present comorbid mood disorder. in comparison to healthy controls.

2. Methods 2.1. Subjects The nine participating patients were adult women who had been referred to our universityaffiliated Women’s Health Concerns Clinic and had screened positive for a diagnosis of PMDD ŽAmerican Psychiatric Association, 1994.. To qualify for this diagnosis, women were required to report at least a 1-year history of symptomatology, identify five or more attributable symptoms Žone or more being irritability or mood-related. and demonstrate recurrent severe symptoms and social or occupational impairment specific to the premenstruum. These criteria were confirmed prospectively through the use of a daily symptoms diary and self-rating scales that were completed during at least two consecutive symptomatic menstrual cycles. Psychiatric history was assessed using the Schedule for Affective Disorders and Schizophrenia, current and lifetime versions ŽSADS-C and -L. to establish Research Diagnostic Criteria ŽRDC. ŽSpitzer et al., 1978.. Patients who were diagnosed with another mood disorder or with PMDD and a history of mood disorders were excluded from this study. Nine healthy female volunteers who responded to notices placed in the medical school and on the hospital campus served as control subjects. All controls underwent a structured psychiatric interview to ensure the absence of current Axis I or II diagnoses and also completed the daily symptom diary and self-rating scales for two menstrual cycles to confirm that they did not meet criteria for premenstrual syndrome or PMDD. All subjects were required to have menstrual cycle lengths of no shorter than 24 days and no longer than 35 days in order to be included in the study. Menstrual cycle length was tracked on the daily calendar, which captured data for a com-

109

plete menstrual cycle Žfirst day of menses until day before first day of menses of next cycle.. Coexisting medical problems were ruled out by physical examination and routine blood and urine tests. Women who had required any prescription medications within 2 weeks of screening were not included in the study. All subjects were within 25% of their normal body weight and none had a history of recent weight loss. The protocol was approved by the institution’s ethics committee, and all participants gave written informed consent prior to screening. 2.2. Design All subjects were studied on two occasions. Testing had to coincide with both the follicular Ždays 7]11. and late luteal Ždays y6 to y2. phases of a menstrual cycle. Ratings from these cycle days are used by researchers in this area to confirm the ‘on]offness’ of premenstrual symptoms and are used to confirm a diagnosis. On the night before the FEN challenge, subjects stayed fasting beginning at midnight. At 08.00 h an indwelling IV catheter was inserted into the antecubital vein, and 3.3% dextrose in 0.3% saline was administered throughout the challenge period. All subjects remained supine and fasting during the duration of this protocol. At 08.45 h, approximately 120 ml of whole blood was collected. Platelets were isolated and membrane suspensions prepared in triplicate. Aliquots were frozen at y808C until assayed for 3 H-imipramine binding. The binding data, Bmax Žthe maximum number of binding sites. and K d Ždissociation constant. were calculated by Scatchard analysis using the computerized ligand program BDATA ŽE.M.F. Software, Knoxville, TN, USA.. Protein concentration was determined by the method of Lowry et al. Ž1951.. Baseline plasma PRL and CORT were collected at y15 and y5 min. An oral challenge of FEN Žd,l-fenfluramine hydrochloride, 60 mg. was administered at 09.00 h. Post-FEN challenge samples for plasma PRL and CORT were collected hourly thereafter for 6 h. CORT and PRL samples were kept on ice until centrifugation and then frozen at y208C until assayed by radioim-

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115

110

munoassay. PRL levels were determined using the Hybritech immunoradiometric kit. The Diagnostic Products Corp. radioimmunoassay kit was used to measure CORT. The 3 H-imipramine binding assays in platelets were all ligand receptor binding techniques. All samples at all times were assayed in triplicate.

ŽPhase., time point ŽTime. and Phase = Time within groups. MANOVA was conducted for delta PRL, PRL and CORT. Response values were also compared at each time point using Student’s ttests. Paired t-tests were used when baseline variables were compared between the two phases within subjects. All tests were two-tailed and a P-value of - 0.05 was considered significant.

2.3. Analysis Primary outcome was defined as the delta PRL response over time which was calculated by averaging the two baseline PRL measures and subtracting the mean baseline value from the levels recorded for each time point during the challenge. Delta PRL responses were compared using two methods: repeated measure analysis of variance ŽMANOVA. and Student’s t-tests of response levels at each time point during the test. Baseline data were compared using Student’s t-tests. Challenge response analysis was conducted using MANOVA with study group ŽGroup., time point ŽTime. and Time= Group terms. MANOVA was also run to compare menstrual cycle phases Žfollicular and late luteal.

3. Results Nine female patients and nine female controls participated in the study. The mean age ŽS.D.. was 37.5 Ž5.3. years. Progesterone levels were similar between groups and demonstrated expected increases from the follicular to luteal phases ŽTable 1.. There were no differences between the groups in baseline PRL or CORT at either the follicular or the late luteal phased associated with ovulatory cycles ŽTable 1.. 3 H-Imipramine binding levels Ž Bmax imipramine. were similar between the groups during the follicular phase but were significantly lower in the patients

Table 1 Baseline characteristics Patients Ž n s 9. Mean Age Žyears. Follicular phase Progesterone Žnmolrl. Prolactin Žngrml. Cortisol Žnmolrl. 3 H-Imipramine, Bma x Žfmolrmg protein. 3 H-Imipramine, Kd ŽnM. Late luteal phase Progesterone Žnmolrl. Prolactin Žngrml. Cortisol Žnmolrl. 3 H-Imipramine, Bma x Žfmolrmg protein. 3 H-Imipramine, Kd ŽnM. a

Controls Ž n s 9. S.D.

Mean

S.D.

36.3

6.0

38.7

4.5

2.0 6.9 308.8

1.3 3.4 89.8

1.4 5.7 265.0

0.5 3.4 82.8

1036.4 1.5

329.0 1.0

1023.0 1.6

444.7 0.7

23.7 7.6 286.2

22.0 4.0 107.2

23.2 7.8 319.3

15.5 4.4 97.2

688.8 1.5

260.5a 1.5

1150.7 1.8

372.3 1.0

Patients vs. controls during the luteal phase Ž Ps 0.008.; luteal vs. follicular phase in patients only Ž Ps 0.03..

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115

111

Table 2 PRL response to the FEN challenge: results of repeated measure analysis of variance Ža. Within group Controls Delta PRL PRL CORT Patients Delta PRL PRL CORT

Time

Phase

Time= Phase

F5 s 8.96, Ps 0.00 F6 s 12.56, P s 0.00 F6 s 1.28, Ps 0.28

F1 s 0.21, Ps 0.66 F1 s 1.60, Ps 0.24 F1 s 4.93, Ps 0.06

F5 s 2.30, Ps 0.06 F5 s 1.48, Ps 0.20 F6 s 0.90, Ps 0.51

F5 s 9.83, Ps 0.00 F6 s 10.14, P s 0.00 F6 s 2.4, P s 0.04

F1 s 0.75, Ps 0.41 F1 s 3.99, Ps 0.08 F1 s 0.67, Ps 0.44

F5 s 1.87, Ps 0.12 F6 s 1.75, Ps 0.13 F6 s 0.47, Ps 0.83

Time

Group= Time

F5 s 11.93, Ps 0.00 F5 s 15.98, Ps 0.00 F6 s 15.04, Ps 0.00 F6 s 18.14, Ps 0.00 F6 s 1.82, Ps 0.10 F6 s 1.86, Ps 0.10

F5 s 0.43, F5 s 0.47, F6 s 0.43, F6 s 0.30, F6 s 2.07, F6 s 0.21,

Žb. Between groups

Cycle phase

Group

Delta PRL

F L F L F L

F1 s 0.63, F1 s 0.24, F1 s 0.00, F1 s 0.20, F1 s 0.00, F1 s 1.60,

PRL CORT

Ps 0.44 Ps 0.63 Ps 0.98 Ps 0.67 Ps 0.99 Ps 0.23

Ps 0.83 Ps 0.80 Ps 0.86 Ps 0.94 Ps 0.06 Ps 0.97

FEN, fenfluramine; PRL, prolactin; CORT, cortisol; F, follicular; L, luteal.

as compared to the controls during the late luteal phase Ž t s 3.05, 16 d.f., Ps 0.008.. The difference between the follicular and late luteal Bmax imipramine was significant within the patient group Ž t s 2.64, 8 d.f., Ps 0.03. but not within the control group. The results of within-group analyses of responses to FEN by menstrual cycle phase are listed in Table 2a. Significant time effects were identified within both groups for delta PRL and PRL but not CORT. There were no statistically significant Phase or Time= Phase results, although the control demonstrated a trend toward a phase effect in CORT response, and a Time= Phase effect for delta PRL response. Results of the analyses for delta PRL, PRL and CORT between-group differences are listed in Table 2b. Significant time effects occurred during both the follicular and late luteal phases for delta PRL and PRL response to FEN. No other statistically significant effects were identified, although there was a trend toward a Group= Time effect in CORT response for the follicular phase measures. Analysis of delta PRL response at each time

point revealed a trend towards an initial delay in PRL response to the FEN challenge in patients as compared to controls, and while the delta PRL values were similar at q60 and q180 min, the patient values were lower at q120 min during both the follicular Ž t s 1.89, 16 d.f., P s 0.077. and late luteal Ž t s 1.91, 16 d.f., Ps 0.075. phases ŽFig. 1.. CORT responses in the patients were decreased for both the follicular and late luteal phase challenges but did not reach statistical significance. There were no Time= Phase effects for CORT within the patient or control groups. A significant inverse relationship was identified for baseline CORT concentrations and the maximal delta PRL response for patients during the follicular Ž r s y0.79, Ps 0.011. and late luteal phases Ž r s y0.89, Ps 0.001., with the combined phases showing a significant inverse correlation Ž Ps 0.001. ŽFig. 2.. Thus, higher baseline CORT values were associated with decreased PRL responses in patients with PMDD whereas for the controls CORT was positively correlated with Dmax PRL in the follicular and luteal phases and the correlations were not significant.

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115

112

Fig. 1. Delta PRL response to FEN.

4. Discussion This is the first study to report on results of FEN challenge in women who meet DSM-IV criteria for pure PMDD Žwithout a past or present comorbid mood disorder.. Patients with PMDD demonstrated no differences in delta PRL response to FEN when compared to healthy controls. Secondary findings of this study include: Ž1. a trend toward a delayed PRL response to FEN demonstrated by the patients and not the controls at q120 min during the follicular and the luteal phases of the menstrual cycle; Ž2. an inverse correlation between basal CORT levels and Dmax

Fig. 2. Correlation between maximal delta PRL response and basal CORT Žpatients only, both phases combined..

PRL response to FEN in both phases; and Ž3. decreased 3 H-imipramine binding levels Ž Bmax . during the late luteal phase only, when compared to controls. The response curves for each group and phase ŽFig. 1. are similar to findings of O’Keane et al. Ž1991., who reported that the response to d-FEN in healthy controls was lowest in the early follicular phase, cycle days 2]4. While our classification of the follicular phase was days 7]11, this may still explain the trend towards a lower follicular phase response curve compared to the late luteal phase response curve in the controls. Acute administration of FEN in a challenge paradigm leads to an increase in the amount of 5-HT in the synaptic cleft through stimulation of presynaptic 5-HT neurons and inhibition of 5-HT re-uptake. This increased 5-HT activates postsynaptic 5-HT receptors leading to an increase in PRL release. The amount of PRL released appears to depend on the dose of FEN administered, with higher doses leading to release of larger amounts of PRL compared to lower doses ŽQuattrone et al., 1983; Muldoon et al., 1996.. PRL release begins to rise at 60 min and reaches a peak between 180 and 240 min and returns to normal levels between 300 and 420 min, and the rise and fall in PRL appears to correspond with rise and fall in serum levels of FEN ŽGarattini and Caccia, 1979; Quattrone et al., 1983; McBride et al., 1990.. A diminished PRL response at 120 min but not at 180 min could be the result of delayed absorption of FEN in the PMDD subjects. Since we did not measure serum FEN levels, we cannot exclude this possibility, although it seems unlikely as none of the previous studies in women with premenstrual syndrome reported any changes in drug absorption. An alternative explanation might be that the presynaptic 5-HT neuron in PMDD subjects would require higher levels of the drug for the system to respond with release of 5-HT. This is consistent with our findings indicating that women with PMDD secreted normal amounts of PRL Žsimilar to levels of controls. at 180 min when brain FEN concentrations were expected to be highest but had little or no PRL release at 60 and 120 min when brain FEN concentrations were lower. This hypothesis

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115

could be tested in future studies using different doses Ži.e. 30, 45, and 60 mg. of FEN in a challenge paradigm. PRL release induced by FEN challenge in humans appears to be mediated by postsynaptic 5-HT2 receptors as such release is attenuated by pretreatment with the 5-HT2 blockers ritanserin ŽGoodall et al., 1993. and amesergide ŽCoccaro et al., 1996c.. There are conflicting or negative findings for pindolol ŽPark and Cowen, 1995; Palazidou et al., 1995., a beta-blocker with 5-HT1A antagonistic properties and with the 5-HT3 blocker ondansetron ŽCoccaro et al., 1996b.. The fact that PRL release for patients in this study was similar to that of controls would suggest that 5-HT2 receptor sensitivity is unaltered in women with PMDD. This is also consistent with findings of a previous study which showed that 5-hydroxytryptophan Ž5-HTP. induced CORT release, a response shown to be mediated by 5-HT2 receptors ŽLee et al., 1991., was unaltered in women with late luteal phase dysphoric disorder compared to controls ŽVeeninga and Westenberg, 1992.. Bancroft and Cook Ž1995. did not identify differences in PRL response to FEN between women with premenstrual depression and controls. However, at least some of these women would not have met criteria for PMDD and may constitute a different population compared to our study subjects. Bancroft and Cook Ž1995. also did not report whether any of the index patients had a positive history for mood disorders. On the other hand, FitzGerald et al. Ž1997. reported significant blunting in PRL response to FEN challenge which is perhaps not surprising given that all their patients had a history of a mood disorder in addition to the current diagnosis of PMDD. In a recent study, Dyck-Flory et al. Ž1998. reported that individuals who met criteria for a major depressive episode in the past demonstrated a blunted PRL response to FEN during a challenge performed at least 1 year into remission, suggesting that the blunting of PRL response to FEN may be a trait marker for major depression. Our finding of a strong inverse correlation between basal CORT and Dmax PRL has also been demonstrated in depressed patients

113

ŽMitchell and Smythe, 1990; O’Keane et al., 1991; Lichtenberg et al., 1992; Malone et al., 1993.; however, there is increasing evidence of this same correlation in healthy subjects ŽCleare and Bond 1997.. Imipramine binds to the 5-HT transporter in platelets and the characteristics of the 5-HT transporter in platelets have been reported to be similar to those of the transporter in 5-HT neurons ŽAshby et al., 1988.. The PMDD patients in our study had significantly lower late luteal phase Bmax imipramine levels compared to controls, and these lower levels were also significantly decreased as compared to their own follicular phase levels. This is consistent with the findings of previous studies which also reported lower Bmax imipramine binding in women with premenstrual depression compared to controls ŽAshby et al., 1988; Rojansky et al., 1991; Steege et al., 1992. and with two reviews which have concluded that most investigators have found significantly lower levels of Bmax imipramine in subjects with depression compared to controls ŽEllis and Salmond, 1994; Owens and Nemeroff, 1994.. Limitations of this study include the small sample size, the use of fixed dose D,L-FEN, and the inability to report serum fenfluramine levels. Future studies of larger samples of women should clearly define current and past psychiatric history Žincluding familial loading. in addition to biological confirmation of menstrual cycle phase and serum FEN levels. Whether such a study will actually be performed is questionable, since the therapeutic administration of fenfluramine has been suspended as a result of reports of cardiac side effects. If indeed this is the end of the FEN challenge test ‘era’, more direct approaches to assess central 5-HT function, including imaging techniques such as PET, can be expected to supplant this neuroendocrine approach ŽNewman et al., 1998.. In conclusion, we found no differences in the PRL response to FEN for women with pure PMDD compared to healthy controls. However, the trend toward a delayed response to FEN, as well as a significant negative correlation between Dmax PRL and basal cortisol in patients but not in controls during both phases of the menstrual

114

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115

cycle, suggests an underlying 5-HT dysfunction in patients as compared to controls. This is further supported by the finding of significantly lower Bmax imipramine levels in the patients during the late luteal phase.

Acknowledgement This study was supported by a grant-in-aid from SmithKline Beecham Canada Inc. The authors wish to thank Janice Rogers and Carol Ballantyne for their assistance in the preparation of the manuscript. References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. American Psychiatric Association, Washington, DC, pp. 717]718. Ashby, C.R., Jr., Carr, L.A., Cook, C.L., Steptoe, M.M., Franks, D.D., 1988. Alteration of platelet serotonergic mechanisms and monoamine oxidase activity in premenstrual syndrome. Biological Psychiatry 24, 225]233. Bancroft, J., Cook, A., 1995. The neuroendocrine response to D-fenfluramine in women with premenstrual depression. Journal of Affective Disorders 36, 57]64. Cleare, A.J., Bond, A.J., 1997. Does central serotonergic function correlate inversely with aggression? A study using D-fenfluramine in healthy subjects. Psychiatry Research 69, 89]95. Coccaro, E.F., Kavoussi, R.J., Cooper, T.B., Hauger, R.L., 1996a. Hormonal responses to D- and D,L-fenfluramine in healthy human subjects. Neuropsychopharmacology 15, 595]607. Coccaro, E.F., Kavoussi, R.J., Cooper, T.B., Hauger, R., 1996b. 5-HT3 receptor antagonism by ondansetron does not attenuate prolactin response to D-fenfluramine challenge in healthy human subjects. Psychopharmacology 127, 108]112. Coccaro, E.F., Kavoussi, R.J., Oakes, M., Cooper, T.B., Hauger, R., 1996c. 5-HT2ar2c receptor blockade by amesergide fully attenuates prolactin response to D-fenfluramine challenge in physically healthy human subjects. Psychopharmacology 126, 24]30. Dyck-Flory, J.D., Mann, J.J., Manuck, S.B., Muldoon, M.F., 1998. Recovery from major depression is not associated with normalization of serotonergic function. Biological Psychiatry 43, 320]326. Ellis, P.M., Salmond, C., 1994. Is platelet imipramine binding reduced in depression? A meta-analysis. Biological Psychiatry 36, 292]299. Eriksson, E., Hedberg, M.A., Andersch, B., Sundblad, C., 1995. The serotonin reuptake inhibitor paroxetine is superior to the noradrenaline reuptake inhibitor maprotiline in

the treatment of premenstrual syndrome. Neuropsychopharmacology 12, 167]176. FitzGerald, M., Malone, K.M., Li, S., Harrison, W.M., McBride, P.A., Endicott, J., Cooper, T., Mann, J.J., 1997. Blunted serotonin response to fenfluramine challenge in premenstrual dysphoric disorder. American Journal of Psychiatry 154, 556]558. Garattini, S., Caccia, S., 1979. Comparison of the plasma levels of fenfluramine in rats after a toxic dose and in man after a maximal therapeutic dose. Toxicology Letters 3, 285]290. Goodall, E.M., Cowen, P.J., Franklin, M., Silverstone, T., 1993. Ritanserin attenuates anorectic, endocrine and thermic responses to D-fenfluramine in human volunteers. Psychopharmacology 112, 461]466. Lee, M.A., Nash, J.F., Barnes, M., Meltzer, H.Y., 1991. Inhibitory effect of ritanserin on the 5-hydroxytryptophan-mediated cortisol, ACTH and prolactin secretion in humans. Psychopharmacology 103, 258]264. Lepage, P., Steiner, M., 1991. Gender and serotonergic dysregulation: implications for late luteal phase dysphoric disorder. In: Cassano, G.B., Akiskal, H.S. ŽEds.., SerotoninRelated Psychiatric Syndromes: Clinical and Therapeutic Links, International Congress and Symposium Series Number 165. Royal Society of Medicine Services Limited, London, UK, pp. 131]143. Lichtenberg, P., Shapira, B., Gillon, D., Kindler, S., Cooper, T.B., Newman, M.E., Lerer, B., 1992. Hormone responses to fenfluramine and placebo challenge in endogenous depression. Psychiatry Research 43, 137]146. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, 265]271. Malone, K.M., Thase, M.E., Mieczkowski, T., Myers, J.E., Stull, S.D., Cooper, T.B., Mann, J.J., 1993. Fenfluramine challenge test as a predictor of outcome in major depression. Psychopharmacology Bulletin 29, 155]161. McBride, P.A., Tierney, H., DeMeo, M., Chen, J-S., Mann, J.J., 1990. Effects of age and gender on CNS serotonergic responsivity in normal adults. Biological Psychiatry 27, 1143]1155. Mitchell, P., Smythe, G., 1990. Hormonal responses to fenfluramine in depressed and control subjects. Journal of Affective Disorders 19, 43]51. Muldoon, M.F., Manuck, S.B., Jansma, C.L., Moore, A.L., Perel, J., Mann, J.J., 1996. D,L-Fenfluramine challenge test: experience in nonpatient sample. Biological Psychiatry 39, 761]768. Newman, M.E., Shapira, B., Lerer, B., 1998. Evaluation of central serotonergic function in affective and related disorders by the fenfluramine challenge test: a critical review. International Journal of Neuropsychopharmacology 1, 49]69. O’Keane, V., O’Hanlon, M., Webb, M., Dinan, T., 1991. D-Fenfluraminerprolactin response throughout the menstrual cycle: evidence for an oestrogen-induced alteration. Clinical Endocrinology 34, 289]292.

M. Steiner et al. r Psychiatry Research 87 (1999) 107]115 Owens, M.J., Nemeroff, C.B., 1994. Role of serotonin in the pathophysiology of depression: focus on the serotonin transporter. Clinical Chemistry 40, 288]295. Park, S.B., Cowen, P.J., 1995. Effect of pindolol on the prolactin response to D-fenfluramine. Psychopharmacology 118, 471]474. Palazidou, E., Stephenson, J., Butler, J., Coskeran, P., Chambers, S., McGregor, A.M., 1995. Evidence for 5-hydroxytryptamine 1A receptor involvement in the control of prolactin secretion in man. Psychopharmacology 119, 311]314. Pearlstein, T.B., Stone, A.B., Lund, S.A., Scheft, H., Zlotnick, C., Brown, W.A., 1997. Comparison of fluoxetine, bupropion, and placebo in the treatment of premenstrual dysphoric disorder. Journal of Clinical Psychopharmacology 17, 261]266. Quattrone, A., Tedeschi, G., Aguglia, U., Scopacasa, F., Di Renzo, G.F., Annunziato, L., 1983. Prolactin secretion in man: a useful tool to evaluate the activity of drugs on central 5-hydroxytryptaminergic neurones. Studies with fenfluramine. British Journal of Clinical Pharmacology 16, 471]475. Rojansky, N., Halbreich, U., Zander, K., Barkai, A., Goldstein,

115

S., 1991. Imipramine receptor binding and serotonin uptake in platelets of women with premenstrual changes. Gynecologic and Obstetric Investigation 31, 146]152. Spitzer, R.L., Endicott, J., Robins, E., 1978. Research Diagnostic Criteria: rationale and reliability. Archives of General Psychiatry 35, 773]782. Steege, J.F., Stout, A.L., Knight, D.L., Nemeroff, C.B., 1992. Reduced platelet tritium-labelled imipramine binding sites in women with premenstrual syndrome. American Journal of Obstetrics and Gynecology 167, 168]172. Steiner, M., 1997. Premenstrual syndromes. Annual Review of Medicine 48, 447]455. Steiner, M., Judge, R., Kumar, R., 1997. Serotonin re-uptake inhibitors in the treatment of premenstrual dysphoria: current state of knowledge. International Journal of Psychiatry in Clinical Practice 1, 241]247. Veeninga, A.T., Westenberg, H.G.M., 1992. Serotonergic function and late luteal phase dysphoric disorder. Psychopharmacology 108, 153]158. Yatham, L.N., Steiner, M., 1993. Neuroendocrine probes of serotonergic function: a critical review. Life Sciences 53, 447]463.