The purpose of this study is to identify, prior to prescribing, which neuropathic pain patients will benefit from duloxetine more specific the investigators aims are to:

• Verify whether presence of chronic pain alters the pain modulation mechanisms, such as DNIC (diffuse noxious inhibitory control) and TS (temporal summation).
• Investigate whether anti-neuropathic medications such as duloxetine indeed change the pain modulation profile, and whether this profile change is associated with a reduction of clinical pain.

There is no accepted practice for selecting among recommended medications for the individual neuropathic pain patient. Guidelines published to date provided the evidence for their efficacy, however, data is not available on how to choose the right medication for the right patient in order to avoid long 'trial and error's. We hypothesize that medications affecting specific process of pain modulation will be more efficacious in patients expressing dysfunction of that specific process. Therefore, medications that enhance descending inhibition such as SSNRI will be more efficacious in patients with less-efficient pain inhibition. The latter is assessed by the conditioned pain modulation (CPM) paradigm. Accordingly, the aim of this study is to examine this hypothesis in painful diabetic neuropathy patients, using duloxetine, an SSNRI agent assumed to augment descending pain inhibition by reuptake inhibition of noradrenalin and serotonin in the spinal cord dorsal horn synapses. We expect to find better effect of duloxetine in those patients whose pain inhibition capability is less efficient, as expressed by their baseline CPM. Further, we aim to evaluate whether pro-nocieptive pattern of pain modulation indeed reverses in response to treatment. This will be explored by comparing the CPM responses before and after treatment, and by correlating pain alleviation with the possible changes in CPM.

Inclusion Criteria:

• Patients diagnosed as having painful diabetic neuropathy.
• Pain is experienced for more than 3 months.
• Pain severity is ≥ 4 on a 0-10 scale (last month average).

Exclusion Criteria:

• Patient already receiving duloxetine or another SNRI/SSRI.
• Known hypersensitivity to duloxetine or any of the inactive ingredients.
• Treatment with a monoamine oxidase inhibitor (MAOI) within 14 days of randomization or potential need to use an MAOI during the study or within 5 days of discontinuation of study drug.
• Uncontrolled narrow-angle glaucoma
• Because of the risk of serious ventricular arrhythmias and sudden death potentially associated with elevated plasma levels of thioridazine (Mellaril), Cymbalta and thioridazine should not be co-administered
• Inability to perform psychophysical testing, due to language or perceptual barriers.