This study focuses on possible mechanism mediating duloxetine effects on painful physical symptoms in patients suffering from MDD. Our hypothesis is based on the assump¬tion of dual impairment of the somatosensory system in these patients. Hypalgesia to phasic experimental pain may be due to diminished spinal and brainstem transmission. Hyperalgesia may be at¬tributed to increased interoceptive perception (somatic complaints, especially those consist¬ing in pain) due to sensitisation or lack of inhibition of the interceptive perception. These ef¬fects seem to be mediated by specific brain regions (e.g. the right insula). The investigators intent to test if duloxetine effects on these somatic complaints, especially pain complaints are due to a nor¬malization of these interceptive alterations which have been reported to be associated with depressive symptoms. The investigators hypothesize that treatment with duloxetine will normalize "patho¬logical" activation patterns (as assessed by fMRI) associated with increased interoceptive perception.

Inclusion Criteria:

1. Male and female (female are enrolled only under the condition of proof of using a safe and medially accepted contraceptive. According to the "Note for guidance on non-clinical safety studies for the conduct of human clinical trials for pharmaceuticals" (CPMP/ICH/286/95, modification) contraceptives with a pearl-index<1% are considered safe and effective. Among these only hormonal contraceptives, micro-pill and tubal ligantions will be accepted in this study)
2. Between the ages of 21 and 65 years; and
3. A diagnosis of major depression (HDRS>=15) and
4. A minimum of 25 mm on a VAS measure of painful symptoms (ranging from 0-100 mm), but no need for regular pain medication and HAMD-17 item 13 minimum rated as 1,
5. Medically and neurologically healthy on the basis of history, physical examination, EKG, screening laboratories (CBC w/ differential, TSH, Free-T4, ASAT, ALAT, GGT, BUN, creatinine, calcium, phosphorous, magnesium, total protein, albumin, electrolytes) and
6. Absence of substance abuse on the basis of history and urine toxicology at screening.

Exclusion Criteria:

1. DSM-IV psychiatric and substance abuse diagnosis (excluding nicotine dependence) by history or psychiatric evaluation that includes a structured diagnostic interview for non-patient populations (SCID-NP)

2. Current suicide risk sufficient to preclude treatment on an outpatient base: Higher than "2" on the "suicide" item of HAMD-17, or history of suicide attempt(s) in the past 12 months, or who, in the investigator's judgment, poses a current suicidal or homicidal risk.

3. Clinical indications of organic brain disease, dementia, or cognitive impairment.

4. History of substance dependence other than nicotine and consumption within the last year

5. Any medical condition that would preclude the use of duloxetine treatment as

   • A known hypersensitivity to duloxetine or any of the inactive ingredients.
   • Intake of monoamine oxidase inhibitors (MAOIs) or suffering from uncontrolled narrow-angle glaucoma.
   • Signs of any hepatic insufficiency
   • Intake of inhibitors of CYP1A2 (as fluvoxamine, ciprofloxacin, enoxacin)
   • End-stage renal disease (requiring dialysis) or severe renal impairment (estimated creatinine clearance [CrCl] <30 mL/min)

6. Subnormal intellectual potential as assessed by the WST-IQ [Metzler. und Schmidt, 1992]

   • Healthy controls will meet exclusion and inclusion criteria, but will have no current or prior diagnosis of major depression.