| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Psychiatry (S.S.G., J.L., P.A.S., C.A.P., K.C.L.), Cardiology (A.L.H.), and Dentistry (N.L.C.), University of North Carolina at Chapel Hill and Psychiatry (A.S.), Duke University Medical Center, Durham, North Carolina.
Address reprint requests to: Susan S. Girdler, Ph.D., CB #7175, Medical Research Bldg. A, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7175. Email: susan_girdler{at}med.unc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: Twenty-eight women with prospectively confirmed PMDD were compared with 28 non-PMDD women for cardiovascular and neuroendocrine measures at rest and in response to mental stressors, and for ß-adrenergic receptor responsivity, during both the follicular and luteal phase of the menstrual cycle. Structured interview was used to assess psychiatric history and prior sexual and physical abuse experiences. All subjects were free of current psychiatric comorbidity and medication use.
RESULTS: More PMDD women had prior sexual and physical abuse experiences than controls (20 vs. 10, respectively). Relative to nonabused PMDD women, PMDD women with prior abuse (sexual or physical) exhibited significantly lower resting norepinephrine (NE) levels and significantly greater ß1- and ß2-adrenoceptor responsivity and greater luteal phase NE reactivity to mental stress. For non-PMDD control women, abuse was associated with blunted cortisol, cardiac output, and heart rate reactivity to mental stress relative to nonabused controls.
CONCLUSIONS: The results of this initial study suggest that a history of prior abuse is associated with alterations in physiological reactivity to subsequent mental stress in women, but that the biological correlates of abuse may be different for PMDD vs. non-PMDD women.
Key Words: premenstrual dysphoric disorder, • abuse, • stress, • norepinephrine, • cortisol.
Abbreviations: BMI = body mass index;; CO = cardiac output;; DBP = diastolic blood pressure;; HPA = hypothalamic-pituitary-adrenal;; HR = heart rate;; MAP = mean arterial pressure;; PMDD = premenstrual dysphoric disorder;; SCID = structured clinical interview;; SBP = systolic blood pressure;; SV = stroke volume;; TPR = total peripheral resistance.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Although there have been no studies of biological correlates of abuse in PMDD, there does exist a handful of studies that have examined autonomic or HPA-axis measures in non-PMDD women with abuse histories, with mixed results. For example, although some studies report lower plasma cortisol (9) or a blunted cortisol response to dexamethasone in sexually abused women (10), it is difficult to assess the independent effects of abuse history on HPA-axis function in these studies because many of the abused women had comorbid posttraumatic stress disorder (PTSD) (eg, 10). Indeed, Lemieux and Coe (11) found that only women with both sexual abuse histories and current PTSD exhibited greater cortisol and norepinephrine levels relative to healthy controls, whereas abused women with no PTSD showed normal neuroendocrine levels.
Many of the existing studies are also confounded by the fact that a large proportion of the abused cohort exhibit current major depression and/or use psychotropic agents (eg, 11–14). The important interactive effects of current depression and histories of abuse were recently documented by Heim et al. (12). In that study, only women with both a history of abuse and current major depression showed heightened cortisol and heart rate responses to stress relative to abused women with no current depression, nonabused depressed women, and healthy controls. Similar effects involving depression and abuse on HPA-axis function in children have been reported (15).
Consequently, this study was designed to explore biological correlates of abuse both in women meeting strict diagnostic criteria for PMDD and in healthy, non-PMDD women. Unlike women in the vast majority of prior research on abuse and neuroendocrine factors, all women in the present study were free from current depression and the use of psychotropic agents. Thus, this study sought to provide initial evidence on histories of abuse and biological measures in women, independent of psychiatric comorbidity and medication use.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Procedures
Assessment of PMDD
The Prospective Record of the Impact and Severity of Menstrual Symptoms (PRISM) calendar (16) was used to classify both PMDD women and controls. In addition to symptom severity ratings, the PRISM calendar also incorporates measures of life-style impact, life events, and the use of medications. Calendars were completed daily for two to three menstrual cycles.
Criteria for PMDD were based on those of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) (17) which include 1) rating of symptoms as moderate or severe (as opposed to mild) premenstrually, 2) moderate to severe symptoms on at least 3 of the 6 premenstrual days, 3) a total of five or more symptoms premenstrually, 4) at least one moderate to severe emotional symptom, 5) evidence that symptoms impacted function, 6) a symptom-free period in the follicular phase, and 7) criteria 1 to 5 met on two consecutive menstrual cycles. Non-PMDD control women met the following criteria: 1) not completely asymptomatic during the premenstrual week (ie, in order to exclude women biased to nonreporting); 2) having only mild emotional symptoms premenstrually; 3) moderate physical symptoms on fewer than 3 days premenstrually; and 4) no evidence that symptoms impacted normal functioning.
Structured Clinical Interviews
Psychiatric histories
Structured clinical interviews (SCID) based on DSM-III-R criteria for Axis I disorders were conducted. All diagnoses were based on a consensus diagnostic session with a psychiatrist (CP). For past major depressive disorder, 7 months in full remission was required before testing. For other Axis I disorders, 3 years in full remission was required.
Sexual and physical abuse histories
At the end of the SCID, subjects were asked about sexual and physical abuse histories using a modified version of a validated interview (17). To meet criteria for sexual abuse incidents as an adult, there had to be clear threat of harm or force (pressure for sexual activity was not sufficient). To meet criteria as a child, the threat of force did not have to be as clearly established if it was implied by the age differential between perpetrator and victim. Sexual abuse was defined as any of two types of sexual experiences: those involving forced sexual touching (including oral sex and vaginal penetration with objects), and those involving intercourse. Physical abuse was divided into two categories: beat, hit or kicked, and life threat. Physical abuse was only counted if the incident occurred separately from sexual abuse.
Physiological Recording Procedures
Blood pressures (BP) were recorded noninvasively using the auscultatory technique. A semi-automated blood pressure monitor was used to operate the blood pressure cuff, with cuff pressure and Korotkoff sounds (K-sounds) recorded in analogue form and displayed on a computer screen. Systolic blood pressure corresponded to the onset of K-sounds and diastolic blood pressure corresponded to the disappearance of K-sounds. Manual stethoscopic readings were taken initially using a sphygmomanometer in order to ensure correct placement of the microphone.
Impedance cardiography was used to permit noninvasive monitoring of cardiac performance (18), including stroke volume (SV), heart rate (HR) and preejection period. A custom-designed impedance cardiograph (HIC-100, Bioimpedance Technology, Inc., Model 100, Chapel Hill, NC) was used in conjunction with a tetrapolar band electrode configuration to record impedance dZ/dt and Zo signals. Impedance and electrocardiogram signals were processed on-line by specialized computer software (BIT, Chapel Hill, NC) with subsequent manual editing to improve accuracy. For each minute of interest, a 30-second continuous sample of waveforms (obtained concurrently with BP) was processed to generate an ensemble-averaged cardiac cycle, from which SV was determined by means of the Kubicek et al. (19) equation, and HR was determined by the mean interbeat interval. Cardiac output (CO) and total peripheral resistance (TPR) for these same minutes were then calculated using standard formulae (18).
Experimental Procedures
Each subject was tested twice—once during her follicular phase (days 4–8) and once 8 to 12 days after home urine testing (Clearplan Easy) revealed the luteinizing hormone surge. All cycles were later confirmed to be ovulatory using serum progesterone. Cycle phase at first testing was counterbalanced within groups. Before testing, subjects were instructed to refrain from all over-the-counter medications for 24 hours, caffeine for 8 hours, and nicotine for 1 hour. Although time of day of testing for the 2 sessions was held constant for each subject, it was allowed to vary between subjects in order to reduce the barriers that exist to participating in clinical research studies for women (20). Thus, 5 controls and 14 PMDD women began testing in the morning (700–1130 hours), 13 controls and 4 PMDD women began testing in the early to mid-afternoon (1200–1600 hours) and 10 controls and 10 PMDD women began testing in the late afternoon (1630–1800 hours). More PMDD women were tested in the morning, whereas more controls were tested in the early- to mid-afternoon (
2 (2)= 9.0, p < .05). However, time of day did not differ as a function of abuse status, with 11 abused and 8 nonabused women tested in the morning, 9 abused and 8 nonabused women tested in the early-to-mid-afternoon, and 10 abused and 10 nonabused women tested in the late afternoon. Analyses for all physiological measures controlled for time of day (see Design and Data Analysis).
Immediately on arriving at the laboratory, subjects were instrumented for cardiovascular monitoring. Next, an intravenous (IV) line was established in an arm vein and once in place, a curtain was drawn that prevented the subject from viewing the IV. A minimum of 15 minutes elapsed between establishing the IV and beginning baseline rest. Subjects were exposed to the following tasks in fixed order, and a 5-minute recovery period separated the stressors.
Baseline rest
Subjects rested quietly for 10 minutes, during which time cardiovascular measures were collected during minutes 1, 3, 5, 8, and 10. Blood was sampled for baseline levels of norepinephrine (NE) and cortisol and for progesterone during minute 10. Speech stressor: Subjects were presented with a hypothetical situation involving an interpersonal conflict and were given 2 minutes in which to prepare (Speech Preparation) to give a 3-minute, tape-recorded talk (Speech) describing what her actions and emotional responses would be in the situation. Cardiovascular measures were taken during minute 2 of Speech Preparation and during minutes 1 and 3 of the Speech. Blood was sampled at the end of minute 1 of Speech for NE. Paced auditory serial addition test: Subjects were instructed to add each number presented on a tape to the immediately preceding number, stating the answer aloud. This 9.5-minute test is composed of 4 series, with progressively shorter interdigit intervals. Cardiovascular measures were taken once each series, and during series 3 (at 6.5 minutes), blood was sampled for NE responses to math stress and for the delayed cortisol response to speech stress (21).
ß-Adrenergic receptor responsivity testing
Mental stress testing was followed by a 20-minute recovery period, during which the subject rested quietly in the supine position. BP was measured continuously using the Finapres (Ohmeda, Madison, WI) noninvasive blood pressure monitor. The standardized isoproterenol sensitivity test was used to evaluate ß-adrenergic receptor responsiveness in terms of the chronotropic dose of isoproterenol required to increase HR by 25 beats/min (CD25) (22). Progressively increasing bolus doses of isoproterenol (0.125, 0.25, 0.5, 1.0, 2.0, and 4.0 µg) were injected until an increase in HR of at least 25 beats/min was observed. HR responses following each dose were computed as the shortest of three successive electrocardiogram R-R intervals after drug injection, compared with the shortest three R-R intervals at rest (preinjection). The linear regression model of log dose/HR response for each subject was used to determine CD25 exactly by interpolation. The CD25 measure provides an index of cardiac ß1- receptor responsiveness (23). A vascular ß2-receptor responsiveness index was also derived by determining the vasodilatory dose required to decrease TPR by 40% (VD40), using log dose/TPR response interpolation (24). Both the CD25 and VD40 indices are inversely related to receptor responsiveness.
Hormone and Neuroendocrine Assays
Blood for cortisol was collected into EDTA-treated tubes and blood for NE was collected into heparin-treated tubes. Tubes were placed immediately on ice and cold centrifuged to separate plasma within a few minutes after collection. Plasma was pipetted into aliquot tubes, rapidly frozen, and maintained at -80°C until assayed. Plasma levels of cortisol were determined by radioimmunoassay (RIA) using commercial kits from ICN Pharmaceuticals. The sensitivity of the assay is excellent at 0.07 µg/dl. The specificity of the RIA for cortisol is high, showing only 0.05% to 2.2% cross-reactivity with most similarly structured compounds. Plasma levels of NE were determined using the high-performance liquid chromatography technique. The lower limit of quantification with this system is 25 pg/ml, and the intra- and interday coefficients of variation are less than 10%. Serum levels of progesterone were determined a using radioimmunoassay (RIA) kits from ICN Pharmaceuticals. The specificity of the antiserum for progesterone is very high, showing only 0.01% to 2.5% cross-reactivity with other steroid compounds. Progesterone levels less than 3 ng/ml in the "luteal" phase of the cycle were considered reflective of an anovulatory cycle. Based on this criterion, all women included in this report exhibited ovulatory cycles.
Psychosocial Questionnaires
The Beck Depression Inventory (BDI) (25) and the Spielberger State Anxiety Inventory (Form Y-2) (26) were completed during both phases of the menstrual cycle. Because the assessment of Post-Traumatic Stress Disorder (PTSD) was not included as a standard module in the SCID interview based on DSM-III-R criteria, we assessed PTSD-like symptoms using the supplementary PTSD (PS) scale of the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) (27).
Design and Data Analysis
Because so few controls had histories of sexual abuse (see Results), for analytical purposes women were coded as having an abuse history if they experienced either sexual or physical abuse in their lifetime. Group differences in psychosocial measures were analyzed using a 2(Group: PMDD vs. controls) x 2(Abuse) x 2(Phase) repeated-measures analysis of variance with cycle phase as the repeated factor.
Next, we examined differences in resting cardiovascular and neuroendocrine baseline measures. For each dependent measure, a 2(Group) x 2(Abuse) x 2(Phase) repeated-measures analysis of covariance (ANCOVA) was used, with phase as the repeated factor and time of day of testing as the covariate.1 A similar 2(Group) x 2(Abuse) x 2(Phase) repeated measures ANCOVA was employed for measures of CD25 and VD40. Reactivity to stress was analyzed using change scores (stress level –baseline level), employing 2(Group) x 2(Abuse) x 2(Phase) x 2(Task) repeated-measures ANCOVAs. Where significant interactions emerged, subsequent simple effects analyses were conducted in order to examine the source of the effect.
We were unable to obtain blood samples from one control subject, yielding a sample of 27 controls and 28 PMDD women for neuroendocrine analyses. Technical difficulties resulted in our inability to perform the ß-receptor responsivity testing in 6 women (3 controls and 3 PMDD), resulting in a sample of 25 controls and 25 PMDD women for CD25 and VD40 analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
2 (1) = 4.9, p < .05) and physical abuse (17 vs. 8 women; 2 (1) = 5.9, p < .05). Thus, more PMDD women had abuse histories (either sexual or physical) relative to controls (20 vs. 10 women; 2 (1) = 7.2, p < .01). Three (30%) of the abused control women reported that their abuse occurred repeatedly (as opposed to a single episode), whereas 12 (60%) of the abused PMDD women reported repeated abuse experiences. Regarding age at first abuse experience, for the controls, 1 woman experienced her first abuse as a child (
13 years of age), 5 as an adolescent (14–17 years of age), 3 as an adult, whereas the age was unknown for 1. For the PMDD women, 10 experienced their first abuse as a child, 6 women experienced their first abuse as an adolescent, and 4 women experienced their first abuse as an adult. Regarding proximity of last abuse experience, for the controls, on average 12.4 years had elapsed since the last abuse episode (range 9–20 years). For the PMDD women, 14.9 years had elapsed since the last abuse experience (range 3–32 years).
Psychiatric, Psychosocial, and Demographic Factors
There were no significant differences as a function of either PMDD diagnosis or abuse histories in age, body mass index, or number of cigarette smokers (Table 1). There were also no differences in the percentage of women meeting DSM-III-R criteria for prior histories of major depression, anxiety disorders, eating disorders, or substance abuse/dependence disorder. PMDD women, regardless of abuse history, scored higher on the PTSD scale of the MMPI (F(1,51) = 21.2, p = .0001). There were no significant effects involving abuse histories for the PTSD scale.
|
Baseline Neuroendocrine and Cardiovascular Levels
Norepinephrine
A Group x Phase interaction emerged (F(1,50) = 4.6, p < .05), reflecting the menstrual cycle influence on baseline NE levels in PMDD women only (F(1,25) = 5.5, p <.05), because they exhibited greater luteal phase levels relative to their own follicular phase levels (Table 2). This phase effect in PMDD women yielded greater baseline NE in PMDD women vs. controls during the luteal phase only (p = .01). We also obtained a Group x Abuse interaction for baseline NE (F(1,50) = 4.6, p < .05). Only for PMDD women did abuse history influence baseline NE (F(1,25) = 6.1, p < .05), since abused PMDD women exhibited lower baseline NE during both cycle phases relative to nonabused PMDD women. Because of the differences in baseline NE, mean baseline NE was included as a covariate for all NE reactivity analyses.
|
Cardiovascular measures
There were no significant effects (Table 2).
Neuroendocrine and Cardiovascular Reactivity to Mental Stress
Norepinephrine reactivity
Phase x Task x Abuse (F(1,49) = 5.6, p < .05), Phase x Task x Group (F(1,49) = 4.0, p < .05) and Task x Group (F(1.49) = 4.3, p < .05) interactions were obtained (Fig. 1). Simple effects analyses were then conducted separately by group. For PMDD women, a Phase x Abuse interaction followed (F(1,24) = 3.9, p = .05), reflecting the phase-specific effects of abuse. During the luteal phase, abused PMDD women displayed greater NE reactivity to speech stress than nonabused PMDD women (p < .05), whereas during the follicular phase there was no effect of abuse. For controls, differences in NE reactivity to stress as a function of abuse history were nonsignificant.
|
|
|
ß-Adrenoceptor Responsivity
Histories of abuse modulated ß1-adrenoceptor responsivity in PMDD women only (Group x Abuse: F(1,45) = 9.7, p < .01). Abused PMDD women exhibited greater ß1-adrenoceptor responsivity (ie, lower CD25 values) than did nonabused PMDD women in both the follicular (1.4 vs. 2.7 µg, respectively, p < .01) and luteal phases (1.4 vs. 2.8 µg, respectively, p < .01). No differences were seen for ß1-adrenoceptor responsivity between abused and nonabused controls in either the follicular (2.2 vs. 2.1 µg, respectively) or luteal phases (2.0 vs. 1.8 µg, respectively). Similarly, for ß2-adrenoceptor responsivity, PMDD women with abuse histories had greater responsivity (ie, lower VD40 values) than did nonabused PMDD women (Group x Abuse: F(1,45) = 4.3, p < .05), in both the follicular (0.8 vs. 1.2 µg, respectively, p = .05) and luteal phases (0.7 vs. 1.4 µg, respectively, p = .01). No differences were seen in ß2-receptor responsivity between abused and nonabused controls in either the follicular (1.1 and 1.0 µg, respectively) or luteal phase (1.3 and 1.0 µg, respectively).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Thus, our results for blunted cortisol and myocardial reactivity to stressors in abused, non-PMDD women suggest that even for healthy women, with no current psychopathology, the stress of even one traumatic abuse incident occurring at any point in the developmental lifespan may be associated with alterations in sympathetic and HPA-axis responsivity to subsequent stress. This interpretation is consistent with the results of Resnick et al. (9), who reported that adult women with at least one prior assault incident (as a child or adult) exhibited lower plasma cortisol levels after a subsequent rape stress than women with no prior assault history. Whether there are long-term physical and/or psychological repercussions of such stress-induced neuroendocrine dysregulation in women who are currently healthy is of clinical interest. This is especially true in light of the recent emerging evidence that blunted HPA-axis function is associated with a number of stress-related bodily disorders (29).
In contrast to the non-PMDD women, abused PMDD women showed alterations in measures reflecting adrenergic function only, specifically heightened NE reactivity to speech stress in the luteal phase, and greater ß1- and ß2-adrenoceptor responsivity coupled with lower baseline levels of plasma NE relative to nonabused PMDD women in both cycle phases. The lower baseline NE levels and greater adrenergic receptor responsivity may be directly related because the adrenergic receptors would be expected to upregulate in response to diminished agonist levels. Alterations in peripheral ß-adrenergic receptor function may have pathophysiological relevance because increased ß-adrenoceptor density, which is linearly related to ß1-adrenoceptor responsiveness to the isoproterenol procedure (30), has been documented in clinical depression (31). Moreover, in the only other study of which we are aware that has measured ß-adrenoceptor function in PMDD, Gurguis and colleagues (32), who did not assess abuse, found that PMDD women displayed greater ß-adrenoceptor density than controls, and also that ß-adrenoceptor density was correlated with premenstrual anxiety in the PMDD group. Thus, alterations in adrenergic physiology that we observed only in the abused PMDD women support the conceptualization that there may be different subgroups of PMDD women with differing pathophysiological mechanisms.
Regardless of prior abuse, our study also found a number of notable differences between PMDD women as a group and healthy controls. For example, despite the fact that more PMDD women than controls were tested in the morning, a time when cortisol levels were at their highest, the PMDD women exhibited lower baseline cortisol levels in both cycle phases. This observation, in the face of an opposing circadian influence, combined with the results from other studies documenting diminished HPA-axis activation in PMDD (33, 34, 35–37), suggests that blunted HPA-axis function may be a robust characteristic of PMDD. We also observed that only for PMDD women did the menstrual cycle exert an effect on basal NE levels, resulting in PMDD women having greater NE levels than controls during the luteal, symptomatic phase only. A similar phase-related shift for cerebrospinal fluid levels of the NE metabolite, MHPG, has been previously reported in PMDD (38), although no comparison with controls was made in that study. A menstrual cycle influence on NE levels is in contrast to the lack of a cycle effect for most other neuroendocrine measures examined in PMDD women (2). This leaves open the possibility that NE plays a pathophysiological role in this disorder.
Despite the numerous strengths to our study, including the use of strict diagnostic procedures to diagnose PMDD, the use of a validated interview to establish abuse histories, and the careful exclusion of subjects with comorbid depression or anxiety or who used psychotropic agents, it is not without its limitations. The primary limitation of our study concerns the small sample sizes involved, especially in the abused, non-PMDD (N = 10) and the nonabused, PMDD (N = 8) groups. Although these percentages are consistent with reported rates of abuse in community samples (39) and PMDD populations (6–8), indicating that our study did not suffer from a selection bias, the small cell sizes involved leaves open the possibility that some of our findings may be spurious. Although the robustness of our results are bolstered by the consistent pattern of effects observed, especially the consistent evidence for a blunted stress reactivity pattern in the abused, non-PMDD women, the small cell sizes render our findings preliminary. A related limitation to our study stems from the fact that we statistically analyzed a large number of dependent variables in a relatively small sample of women, raising concerns regarding Type I error rates. Thus, the results of this initial examination of stress-induced alterations in HPA-axis and adrenergic measures in women who have experienced sexual or physical abuse are intended to serve a heuristic value for larger scale investigations.
Second, given the important focus of our study on cortisol and cortisol reactivity to stress, a clear drawback to our study design was that we tested subjects at different times of the day. Although time of day of testing was controlled for in all analyses, and although there were no differences in the proportion of abused vs. nonabused women tested as a function of time, nor was there any significant diurnal influence on cortisol reactivity, the possibility remains that diurnal factors could have contributed to some of our effects involving cortisol.
Despite the study limitations, the temporal gap between traumatic events and dysregulation in measures reflecting the HPA-axis and adrenergic physiology later in life is a remarkable feature of our preliminary findings because many years, if not decades, had elapsed since some of the women in the present study had experienced abuse. This is consistent, however, with the work of Wang and Mason (40) showing persistent disturbance in hypothalamic–pituitary–thyroid axis function in male combat veterans 40 to 50 years after combat. This is also consistent with the observation for altered ACTH responses to stress in adult women with histories of childhood sexual abuse (12). Thus, it is possible that dysregulation of neuroendocrine factors in women with abuse histories represents an adaptation gone awry if the stress axes continue to prepare the individual for a severe stressor that no longer exists but somehow remains encoded in the brain (41).
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
The authors thank Dot Faulkner for manuscript preparation, Sara Benjamin and Catherine Stanwyck for their roles as study coordinator, and Unipath Diagnostics for their generous donation of Clearplan Easy Ovulation prediction kits.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
2Baseline cortisol levels and cortisol reactivity to stress from a subset of these same PMDD (N=24) and control women (N=12), although not analyzed as a function of prior abuse, were previously reported (28).
3Because of the diurnal influence on plasma cortisol levels, additional analyses were conducted to examine the influence of time of day of testing on mean cortisol levels and cortisol reactivity to stress. For absolute cortisol levels, although women tested in the morning had the highest levels followed by early- and then late-afternoon testing, the effect of time of day was not significant (F(2,49) = 2.2, p = .12). There was no evidence of a diurnal influence on cortisol reactivity to stress.
Received for publication May 16, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. Leeners, R. Stiller, E. Block, G. Gorres, B. Imthurn, and W. Rath Effect of Childhood Sexual Abuse on Gynecologic Care as an Adult Psychosomatics, October 1, 2007; 48(5): 385 - 393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Girdler, K. S. Thompson, K. C. Light, J. Leserman, C. A. Pedersen, and A. J. Prange Jr. Historical Sexual Abuse and Current Thyroid Axis Profiles in Women With Premenstrual Dysphoric Disorder Psychosom Med, May 1, 2004; 66(3): 403 - 410. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
