No eligible patients could be recruited.

About 50% of all stroke patients develop post-stroke depression (PSD). A meta-analysis has shown that rTMS treatment can reduce depressive symptoms in PSD patients. In addition to rTMS alone for the improvement of depression, the question arises as to whether a combination therapy of rTMS plus antidepressant medication can achieve a stronger or longer-term effect in PSD patients. Unfortunately, there are currently no trials of combination therapy with rTMS and drug therapy in PSD patients. Therefore, this study will investigate whether combination therapy of antidepressant and rTMS can provide additional relief of depressive symptoms compared to antidepressant and sham rTMS therapy. It is assumed that the additional active rTMS achieves a faster normalization of affect and drive than with a sham rTMS, so that the patients benefit from neurorehabilitation measures earlier and more sustainably.

Depression is one of the most common forms of mental illness. According to studies by the World Health Organization (WHO), the World Bank and the European Brain Council [1], depression is the leading disease in Europe and Germany since the early 1990s.

Besides drug or psychotherapeutic treatment, repetitive transcranial magnetic stimulation (rTMS) is currently being used as a new non-invasive therapy for depression. The rTMS applies an electromagnetic coil to the patient's head, creating a magnetic field. Impulses emanating from the coil trigger a multitude of reactions at the point of stimulation which, for example, can alter the metabolism, lead to a release of neurotransmitters and a change in gene expression [2-3]. Pulses with a frequency ≤1Hz lead to a reduction of the excitability of the neurons and to an inhibition of cortical activity. In contrast, frequencies ≥5 Hz increase the excitability of neurons and increase cortical activity [4-5].

A large number of studies has already shown that rTMS in depressive patients leads to an improvement in depressive symptoms and has been shown to have an antidepressant effect [6]. In the United States, rTMS has been approved by the Food and Drug Administration (FDA) since 2008 as a treatment for patients with depression who do not respond to antidepressant drug therapy. The FDA recommends a high-frequency (10Hz) rTMS on the left dorsolateral prefrontal cortex (DLPFC) five days a week for four to six weeks [7]. The stimulation of the DLPFC is based on the valence hypothesis that the right hemisphere specializes in the processing of negative emotions and the left hemisphere is specialized in the processing of positive emotions [8] and the DLPFC controls emotional processing [9-10]. Activation of the left DLPFC is therefore associated with the processing of positive emotions [11].

About 50% of all stroke patients develop post-stroke depression (PSD) [12]. A meta-analysis has shown that rTMS treatment can reduce depressive symptoms in PSD patients [13]. In addition to rTMS alone, it is unkown if a combination therapy of rTMS plus antidepressant medication can achieve a stronger or longer-term antidepressive effect in PSD patients. Unfortunately, there are currently no trials of combination therapy with rTMS and drug therapy in PSD patients. Previous studies with depressive patients provide both results that suggest an additional effect of combination therapy [14-19] and results that found no difference between drug-only therapy and combination with rTMS [20-24]. The comparability of the studies is difficult due to the heterogeneity of the study designs. However, it is noticeable that a younger age (<50 years), an intervention duration of rTMS of four weeks, a higher dose of the antidepressant, an inter-train interval (interval between the trains) of <30 seconds and a total number of pulses of <1250 per day, associated with positive effects. However, further studies are needed that address the issue of an additional effect of combination therapy. In addition, a neurological disease was considered to be an exclusion criterion in some of the studies performed [14-15; 20; 23]. It is therefore questionable whether the study results can be transferred to PSD patients.

Therefore, this study will investigate whether combination therapy of antidepressant and rTMS can provide additional relief of depressive symptoms compared to antidepressant and sham rTMS therapy. It is assumed that the additional active rTMS achieves a faster normalization of affect and drive than with a sham rTMS, so that the patients benefit from neurorehabilitation measures earlier and more sustainably.

Inclusion Criteria:

• first insult
• Post-stroke Depression (17 item version of the Hamilton Depression Rating Scale [HAM-D]> 18 points)
• capacity to consent

Exclusion Criteria:

• insufficient cardiorespiratory stability
• previous depression or previous use of antidepressants
• pre-stroke psychological illnesses (eg psychosis, bipolar disorder)
• severe cognitive impairment
• aphasia
• lefthanded
• decreased seizure threshold or history of epileptic seizures
• taking medicines that lower the seizure threshold (local anesthetics, cortisone, alcohol, neuroleptics)
• hemorrhages and cerebral edema (e.g., subarachnoid haemorrhage, intracerebral hemorrhage, subdural hematoma, epidural hematoma)
• fresh and healed head wounds near the area to be stimulated
• missing bone cover (relief spread)
• colonization with a germ requiring isolation (e.g., MRSA, 3MRGN, 4MRGN)
• recent myocardial infarction or higher grade cardiac arrhythmias
• contraindications to rTMS: Metallic or magnetic implants containing iron, cobalt or nickel (e.g., pacemakers, brain pacemakers, automatic insulin pumps, electrodes, plates, clips, implanted hearing aids, dental implants, metal endoprostheses, metal parts, or metal fragments in the body).
• pregnancy
• no consent for study participation by the patient

The randomization of the patients is carried out by a non-medical employee. This co-worker maintains a randomly generated randomization list, indicating to the particular study code (e.g., TMS-PSD-001; TMS-PSD-002), whether to perform an active or sham treatment. The employee pulls a number from an envelope before beginning the intervention of each enrolled patient and then looks at the list to identify the particular group assignment. There will always be randomization packages of ten patients each. Once these are used up, a new list and a new envelope with study codes are prepared.

In addition, the "real" and the placebo coils are coded before starting studies (eg "A" and "B"). Only the uninvolved employee is the assignment to the real - or placebo coil known. The uninvolved employee randomly assigns the study code containing the information on which coil to use ("A" or "B"). In this way a blinding of performing physician, patient and the evaluating scientist is guaranteed.