depression and women across the reproductive...

Co-sponsored by

September 26 – 28, 2013 | Westin Tampa Harbour Island

Depression and Women Across the Reproductive Lifecycle

David R. Rubinow, MD University of North Carolina at Chapel Hill School of Medicine Chapel Hill, NC

Learning Objective

Identify factors that increase the risk of mood disturbance in women at key time points when hormonally-related triggers are present (e.g., menses, pregnancy, postpartum, perimenopause)

1

Learning Objective

Design and implement treatment plans to manage the mood symptoms in women at various life stages

2

Audience Response

A. Yes, I discourage patients from taking estrogen B.  It has made no difference C. Data is mixed, so I am cautious D.  I ignore it and recommend estrogen when

appropriate

Has the Women’s Health Initiative impacted the way you treat your peri-menopausal and menopausal patients

Gonadal Steroids, Like All Steroid Hormones, are Derivatives of Cholesterol

Hu J, et al. Nutr Metab (Lond). 2010;7:47. PMID: 20515451.

Principles

●  Reproductive hormones wire and rewire the brain ●  Reproductive hormones regulate brain circuitry ●  Changes in reproductive hormones alter brain states of

clinical relevance ●  Reproductive endocrine-related mood disorders account

for a substantial amount of morbidity and mortality ●  These disorders are NOT hormone deficiencies ●  Reproductive steroids are both context-dependent and

context-determining

Rubinow DR et al. Hormones, Brain and Behavior, v4, 2395-2428, 2009. Rubinow, D.R., and Schmidt, P.J.: Gonadal steroids, brain, and behavior: role of context. Dial. Clin. Neurosci., 4:123-138, 2002. Schmidt PJ, et al. N Engl J Med. 1998;338(4):209-16. PMID:9435325.

Reproductive Hormones Wire and Rewire the Brain

Reproductive Hormones Regulate Brain Circuitry

PFC = prefrontal cortex K Berman et al. Unpublished data. Rolls ET. Front Hum Neurosci. 2013;7:74. PMID: 23508210. Ochsner KN, et al. Neuroimage. 2004;23(2):483-499. PMID: 15488398.

Estrogen Add-Back

Leuprolide Alone

Progesterone Add-Back

Steroids Regulate PFC Functional Correlations With Left Hippocampal Formation Activity

OFC Activity in Response to Emotional Stimuli in the Context of Behavioural Inhibition Across the Menstrual Cycle

OFC = orbitofrontal cortex Protopopescu X, et al. Proc Natl Acad Sci USA. 2005;102(44):16060-5. PMID: 16247013.

Periods of Reproductive Endocrine Change are Associated With Altered Mood States and Reproductive Endocrine-Related Mood Disorders Account for Substantial Morbidity and Mortality

● Puberty ● Menstrual Cycle ● Puerperium ● Perimenopause

Images Courtesy of PJ Schmidt; DHEA = dehydroepiandrosterone Grumbach MM. Horm Res. 2002;57 Suppl 2:2-14. PMID: 12065920.

Gene Network

Ojeda SR, et al. Mol Cell Endocrinol. 2010;324(1-2):3-11. PMID: 20005919.

Ojeda SR, et al. Mol Cell Endocrinol. 2010;324(1-2):3-11. PMID: 20005919.

Gene network regulating puberty

Reproductive Endocrine-related Mood Disorders

● Premenstrual dysphoria

● Perinatal depression

● Perimenopausal depression

Reproductive Endocrine-related Mood Disorders

● Premenstrual dysphoria ! Prevalence – 5% ! Cause of 14.5 million Disability Associated

Live Events annually in US ! Requires prospective confirmation of

restriction of symptoms to the luteal phase

Meltzer-Brody S, Stuebe, A. Best Pract Res Clin Obstet Gynecol. September 2013. [ePub ahead of print]. doi:10.1016/j.bpobgyn.2013.08.009.

Efficacy of GnRH-A in the Symptoms of PMDD

GnRH-A = gonadotropin-releasing hormone agonist; PMDD premenstrual dysphoric disorder. Schmidt PJ, et al. N Engl J Med. 1998;338(4):209-16. PMID:9435325.

(Weekly Means + SEM)

Most ANOVA-R Phase X Week X Group: F = 2.8 p < 0.05

Phase X Group: F = 42.0, p < 0.001

Anx

iety

Least Baseline

Baseline

Leuprolide

Placebo

5

4

3

2

1

5

4

3

2

1

Most

Anx

iety

Least

Central Nervous System Effects of GnRH Agonist & Gonadal Steroid Replacement

GnRH-A = gonadotropin-releasing hormone agonist; IM = intramuscular; QM = every month; BID = twice a day. Schmidt PJ, et al. N Engl J Med. 1998;338(4):209-16. PMID:9435325.

OR Estradiol 0.1 mg/d Progesterone 200 mg BID Placebo

Time (weeks) • 0 • 12 • 24

Leuprolide acetate (3.75 mg IM QM)

Procedures

Steroid Precipitation of PMDD Symptoms

PMDD = premenstrual dysphoric disorder Schmidt PJ, et al. N Engl J Med. 1998;338(4):209-16. PMID:9435325.

Most

Sad

ness

Leuprolide Alone P4 Replacement

3

2

1 Least

Sad

ness

Controls

E2 Replacement

PMDD ANOVA-R Phase X Group: F1,23 = 19.6 p < .001 3

2

1

Differential Brain Regional Activation by Ovarian Steroids in PMDD

PMDD = premenstrual dysphoric disorder Schmidt PJ, unpublished data.

Subgenual Cingulate

(-6, 20, -10) p = .0002

Controls (n = 32) R

elat

ive

regi

onal

Cer

ebra

l Blo

od F

low

(-m

m/1

00g/

min

)

74

72

70

68

66

64

62

60

Estradiol Leuprolide Progesterone

Patients with PMDD

(n = 16)

Differential Brain Regional Activation by Ovarian Steroids in PMDD

PMDD = premenstrual dysphoric disorder Schmidt PJ, unpublished data.

Subgenual Cingulate

Rel

ativ

e re

gion

al C

ereb

ral F

low

(-m

m/1

00g/

min

)

Estradiol Leuprolide Progesterone

Controls Patients

70

65

60

Medial Orbitofrontal Cortex

55

57

59

61

63

Rel

ativ

e re

gion

al C

ereb

ral F

low

(-m

m/1

00g/

min

)

Controls Patients Medial Orbitofrontal Cortex (-12, 42, 12), p = .000015

Subgenual Cingulate (-6, 20, -10), p = .0002

PMDD: Is it Triggered by the Level of Steroid or the

Change in Level of Steroid?

Reproductive Endocrine-related Mood Disorders

● Premenstrual Dysphoria

● Perinatal Depression

● Perimenopausal Depression

Perinatal Depression

● Prevalence – 10% - 15% (MDE = 7.5%; mDE = 6.5%)

● Suicide is the leading cause of maternal death

MDE = major depressive episode; mDE = minor depressive episode Meltzer-Brody S, Stuebe, A. Best Pract Res Clin Obstet Gynecol. September 2013. [ePub ahead of print]. doi:10.1016/j.bpobgyn.2013.08.009.

Background: Perinatal Depression ●  COMMON

!  10%-15% prevalence !  4 million women give birth annually in U.S.; one-half million with perinatal

depressive disorder !  Most common, unrecognized complication of perinatal period

!  Compare to the prevalence rate of gestational diabetes at 2%-5% ●  MORBID

!  Devastating consequences for patient and family !  Low maternal weight gain, preterm birth !  Impaired bonding between mother and infant !  Increased risk of suicide and infanticide

●  MISSED !  No practice guidelines or routine screening !  Symptoms often different from “classic DSM-IV depression”

Meltzer-Brody S, et al. Arch Womens Ment Health. 2013 Aug 1. [Epub ahead of print] PMID: 23904137. Meltzer-Brody S, Stuebe, A. Best Pract Res Clin Obstet Gynecol. September 2013. [ePub ahead of print]. doi:10.1016/j.bpobgyn.2013.08.009.

Distinguishing Characteristics of Perinatal Mood Symptoms

● Anxiety or agitation ● Depressed mood ● Sadness, weepiness ●  Irritability ● Hypervigilance about the baby ● OR lack of interest in the newborn ●  Impaired concentration or feeling overwhelmed ● Feelings of dependency or guilt Austin MP, et al. Acta Psychiatr Scand. 2005;112(2):97-104. PMID: 15992390.

Other Facts and Dilemmas

● There is no free lunch ● Inadequacy of database ● During pregnancy, there are medications to

avoid, but…. ● Pregnancy affects medication levels

! Increased volume of distribution, hepatic blood flow ! Decreased plasma binding proteins ! Effects of reproductive steroid on metabolic enzymes

Koren G. J Popul Ther Clin Pharmacol. 2011;18(3):e523-527. PMID: 22113390. Bodén R, et al. Arch Gen Psychiatry. 2012;69(7):715-21. PMID: 22752236.

Profound Neuroendocrine Changes at Time of Birth

Rel

ativ

e H

orm

one

Con

cent

ratio

n

Russell JA, et al. Brain preparations for maternity⎯adaptive changes in behavioral and neuroendocrine systems during pregnancy and lactation. An overview. In: Progress in Brain Research; 2001.

E2+ PROG PLACEBO PLACEBO

BASELINE HYPO- GONADAL

ADDBACK WITHDRAWAL 0 4 8 12 16 20 24

Leuprolide acetate

What Is the Hormone Trigger in Perinatal Depression?

E2 = estradiol; PROG = progesterone Bloch M, et al. Am J Psychiatry. 2000;157(6):924-930. PMID: 10831472.

Weekly Cornell Depression Scale Scores (One Subject)

Bloch M, et al. Am J Psychiatry. 2000;157(6):924-930. PMID: 10831472.

0

5

10

15

20

25

30

35

40

45

Baseline Week 6 Week 8 Week 2 Week 4 Week 6 Week 8 Week 2 Week 4

Add back

Sco

re

Early Late Follow-up

Withdrawal

PPD Study: Cornell Depression Score

(Mean ±SEM) PPD = Postpartum depression Bloch M, et al. Am J Psychiatry. 2000;157(6):924-930. PMID: 10831472.

BASELINE ADDBACK WITHDRAWL FOLLOW UP

Cor

nell

D

epre

ssio

n S

core

0

5

10

15

20

25

PPD Normal

These Disorders Are NOT Hormone Deficiencies

Reproductive Endocrine-related Mood Disorders

● Premenstrual Dysphoria

● Perinatal Depression

● Perimenopausal Depression

Perimenopausal Depression

● Prevalence – 20%

● Is associated with a 50% increase in cardiovascular mortality

Rubinow DR, Girdler SS. Depress Anxiety. 2011;28(6):E1-E15. PMID: 21648024.

E2 Withdrawal Precipitates Depressive Symptoms in Asymptomatic Women With a Past Perimenopausal Depression

E2 = estradiol; PMD = perimenopausal depression; CESD = Center for Epidemiologic Studies Depression Scale; DB = double. Schmidt PJ, et al. unpublished data

• 15

• 10

• 5

•  CESD

•  PMD -

•  PMD + (n = 26) •  PMD - (n = 30)

• E2 Continuous

• Placebo

• E2 DB Blind

• E2 Open

•  p < 0.01 •  p < 0.01

•  p < 0.01

• 0

•  PMD + • E2 Withdrawal • E2 Withdrawal

Therapeutic Trial of Estradiol, SERM, and Phytoestrogen in Perimenopausal Depression (n = 38)

SERM = selective estrogen-receptor modulator; HDRS = Hamilton Rating Scale for Depression; Tx*Time = treatment by time. Schmidt PJ. Unpublished data.

0 5

10

15

20

25

HDRS-17

Baseline Week 8

Estradiol

p < .01

*Week 8 E2 vs. Placebo: p < .05

ANOVA-R: Tx*Time F1,34 = 8.8 p < .001

* 0 5

10

15

20

25

Baseline Week 8

p < .05

Raloxifene

0 5

10

15

20

25

Baseline Week 8

p = NS Placebo

0 5

10

15

20

25

HDRS-17

Rimostil p = NS

Baseline Week 8

Trials and Hormone Regimens (E+P or E vs. Placebo) Characteristic

(Hormone Therapy; Placebo)

WHI E+P WHI E WHIMS E+P WHIMS E WHISCA E+P WHISCA E HERS E+P ESPRIT E EMS E+P WISDOM E+P ULTRA E

Number 8506;8102 5310;5429 2229;2303 1464;1483 690;726 434;452 1380;1383 513;504 70;72 2196;2189 191;185

Mean age, y 63.2;63.3 63.6;63.6 63.2;63.3 63.6;63.6 73.69;73.86 74.01;74.02 67;67 62.3;62.9 75;74.5 63.3;63.3 66.8;66.7

Nonwhite race,% 16.1;16.0 24.5;24.9 - 17.3;16.4 8.4;7.0 14.09;13.08 12;10 3;3 4.3;9.7 1;1.4 7.2;8.1

Previous or current HT, % 26.1;25.6 47.8;48.9 21.8;22.4 45.8;44.7 21.2;22.6 49.54;46.24 1.7;1.7 12;10 31.4;23.6 55;54.3 -

Hysterectomy age <40 y, % - 39.8;39.8 - - - - - - - - -

Hysterectomy age 40–49 y, % - 43.2;42.2 - - - - - - - - -

Bilateral oophorectom, % - 39.5;42.0 - - - - - - - - -

Never pregnant, % 10.1;10.3 9.3;8.5 - - - - - - - - -

Baseline Characteristics of Participants in Randomized, Controlled Trials

Marjoribanks J, et al. Cochrane Database Syst Rev. 2012;7:CD004143. PMID: 22786488. Nelson HD, et al. Ann Intern Med. 2012;157(2):104-113. PMID: 22786830. Schierbeck LL, et al. BMJ. 2012;345:e6409. PMID: 23048011.

Outcome E+P vs. Placebo E vs. Placebo

HR (95% CI) Differences in events per 10,000 women-years (95% CI)* HR (95% CI)

Differences in events per 10,000 women-years (95%

CI)* Cancer

Invasive breast 1.25 (1.07–1.46) 8 (3–14) more 0.77 (0.62–0.95) 8 (1–14) less

Colorectal 0.75 (0.57–1.00) Not significant 1.11 (0.82–1.50) Not significant

Lung 1.23 (0.92–1.63) Not significant 1.17 (0.81–1.69) Not significant

Endometrial 0.78 (0.52–1.16) Not significant Not reported Not reported

Ovarian 1.58 (0.77–3.24 Not significant Not reported Not reported

Cervical 1.44 (0.47–4.42) Not significant Not reported Not reported

Cardiovascular events

Coronary heart disease (CHD death and total MI) 1.22 (0.99–1.51) Not significant 0.95 (0.78–1.15) Not significant

Stroke 1.34 (1.05–1.71) 9 (2–15) more 1.36 (1.08–1.71) 11 (2–20) more

Thromboembolic events

Deep vein thrombosis 1.88 (1.38–2.55) 12 (6–17) more 1.47 (1.06–2.05) 7 (1–14) more

Pulmonary embolism 1.98 (1.36–2.87) 9 (4–14) more 1.37 (0.90–2.07) Not significant

Diabetes

Self-reported new diagnosis requiring treatment with drugs 0.79 (0.67–0.93) 15 (4–26) less 0.88 (0.77–1.01) Not significant

Results of Hormone Therapy Trials

Marjoribanks J, et al. Cochrane Database Syst Rev. 2012;7:CD004143. PMID: 22786488. Nelson HD, et al. Ann Intern Med. 2012;157(2):104-113. PMID: 22786830. Schierbeck LL, et al. BMJ. 2012;345:e6409. PMID: 23048011.

E+P vs. Placebo E vs. Placebo

Outcome HR (95% CI) Differences in events per 10,000 women-years (95% CI)* HR (95% CI)

Differences in events per 10,000 women-years (95% CI)

* Fractures

Hip 0.67 (0.47–0.95) 6 (1–10) less 0.61 (0.41–0.91) 7 (1–12) less Vertebral 0.68 (0.48–0.96) 6 (1–11) less 0.62 (0.42–0.93) 6 (1–12) less Total fractures 0.76 (0.69–0.83) 46 (29–63) less 0.70 (0.63–0.79) 56 (37–75) less

Mortality All-cause mortality 1.04 (0.91–1.18) Not significant 1.02 (0.91–1.15) Not significant Breast cancer mortality 1.96 (1.00–4.04) Not significant 0.37 (0.13–0.91) 2 (1–3) less Lung cancer mortality 1.71 (1.16–2.52) 5 (1–8) more Not reported Not reported

Gallbladder Gallbladder disease (cholecystitis and cholelithiasis)

1.61 (1.30–2.00) 20 (11–29) more 1.79 (1.44–2.22) 33 (20–45) more

Cognitive function Probable dementia 2.05 (1.21–3.48) 22 (5–39) more 1.49 (0.83–2.66) Not significant Mild cognitive impairment 1.07 (0.74–1.55) Not significant 1.34 (0.95–1.89) Not significant

Urinary incontinence Overall urinary incontinence (stress, urge, or mixed)

1.39 (1.27–1.52) 872 (591–1153) more 1.53 (1.37–1.71) 1271 (883–1660) more

Results of Hormone Therapy Trials

Marjoribanks J, et al. Cochrane Database Syst Rev. 2012;7:CD004143. PMID: 22786488. Nelson HD, et al. Ann Intern Med. 2012;157(2):104-113. PMID: 22786830. Schierbeck LL, et al. BMJ. 2012;345:e6409. PMID: 23048011.

Reproductive Steroids Are Both Context-Dependent and Context-Determining ● Susceptibility

● What do we mean by gene : environment interactions?

Rubinow DR, et al. Dialogues Clin Neurosci. 2002;4(2):123-37. PMID: 22033695.

Estradiol Effects in the Forced Swim Test: Strain Differences

Wistar Hannover rat and Long Evans rat. Koss WA, et al. Physiol Behav. 2012;106(2):81-6. PMID: 22266677.

Strain-dependent Effects of Estradiol in the Forced Swim Test

Koss WA, et al. Physiol Behav. 2012;106(2):81-6. PMID: 22266677.

Immobility

Vehicle Estradiol Withdrawal

Wistar Long Evans

Sec

onds

300

250

200

150

100

50 0

*

*

N = 12 N = 12 N = 12 N = 6 N = 7 N = 7

Biology and Environment Meet at the Level of the Dance Partners ● SMRT, high-fat diet, and obesity/insulin

resistance

SMRT = silencing mediator of retinoid and thyroid hormone receptors Fang S, et al. Proc Natl Acad Sci USA. 2011;108(8):3412-7. PMID: 21300871.

High-Fat Diet Induced Severe Obesity in SMRTmRID1 Mice

SMRT = silencing mediator of retinoid and thyroid hormone receptors; WT = wild type NC = normal control; HFD = high-fat diet. Fang S, et al. Proc Natl Acad Sci USA. 2011;108(8):3412-7. PMID: 21300871.

Effects of Genetic Differences are Unmasked by Environment

● SMRT mutation, high-fat diet, and obesity/insulin resistance

● GABA-A receptor mutation, pregnancy, and depression

SMRT = silencing mediator of retinoid and thyroid hormone receptors Fang S, et al. Proc Natl Acad Sci USA. 2011;108(8):3412-7. PMID: 21300871.

Deficient NS Signaling: Postpartum “Depression” and “Infanticide”

NS = neurosteroid ; WT = wild type Wistar; THIP = 4,5,6,7-tetrahydroisoxazolo-[5,4-c]pyridin-3-ol Maguire J, Mody I. Neuron. 2008;59(2):207-13. PMID: 18667149.

Embrace the Uncertainty!

Abbreviations EMS = Estrogen Memory Study ESPRIT = Oestrogen in the Prevention of Reinfarction Trial HERS = Heart and Estrogen/Progestin Replacement Study HT = hormone therapy MI = myocardial infarction P = progestin PE = pulmonary embolism SBP = systolic blood pressure ULTRA = Ultra-Low-Dose Transdermal Estrogen Assessment WHI = Women’s Health Initiative WHIMS = Women’s Health Initiative Memory Study WHISCA = Women’s Health Initiative Study of Cognitive Aging WISDOM = Women’s Intl Study of Long Duration Oestrogen After Menopause CHD = coronary heart disease HR = hazards ratio CI = confidence intervals E + P; estrogen plus progestin E = estrogen only RCT = randomized controlled trial

Questions & Answers

Co-sponsored by