Training of Neural Responding in BPD (IP5n)

Emotion-related brain activation is made visible for patients via neurofeedback with the aim to improve discriminability of emotional arousal and emotion regulation. With functional magnetic resonance imaging (fMRI), information of current brain activation is imaged and fed back to the patient via a visual display. Patients with borderline personality disorder (BPD) usually hyper-activate brain regions associated with emotion. In this study, BPD patients will be provided with neurofeedback from the amygdala, which is crucial for the processing of emotions. The aim of the study is to observe, whether amygdala-neurofeedback would help BPD patients to improve emotion regulation. Compared to a control condition, improved brain self-regulation and emotion regulation is expected with three neurofeedback training sessions.

Patients with BPD show increased emotional reactivity, slow return to baseline, and severe emotion dysregulation symptoms. On the neural level, BPD patients hyper-activate the amygdala and hypo-activate the prefrontal cortex in response to emotional stimuli. The prefrontal cortex and the amygdala are crucial nodes of the brain's emotion regulation network and thus it is assumed, that dysregulation within this network is key to BPD symptoms. Psychotherapy treatments specialized for BPD teach patients to monitor emotional arousal and to develop emotion regulation skills. However in the long run and despite of important therapeutic advances, the majority of BPD patients keep reporting significant impairments in functioning after psychotherapy. To explore new types of therapy in BPD, the investigators have applied real-time fMRI neurofeedback, where patients are provided with their brain activation via a visual display. In previous work they found that BPD patients and healthy participants can down-regulate amygdala activation with real-time fMRI neurofeedback, and increase connectivity between the amygdala and the prefrontal cortex. Yet, we do not yet fully understand the potential effects of amygdala neurofeedback on emotion. BPD patients (n=25) participate in a three-session fMRI neurofeedback training with 2-7 days between sessions (within 2 weeks). The effect of the training will be measured before and after training. Primarily, the investigators expect an improvement in emotion regulation, secondarily, reductions in BPD symptoms are expected. Hypotheses: With fMRI neurofeedback, BPD patients improve significantly in self-report and psychophysiological measures of emotion regulation with fMRI neurofeedback training. BPD patients show significantly reduced symptom severity in self-report measures with neurofeedback training.

25 patients with BPD. In a diagnostic session, diagnostics of psychiatric disorders are conducted. For BPD diagnosis, the International Personality Disorder Examination (IPDE) is used and symptom severity is assessed with the Borderline Symptom List. The Treatment group will receive fMRI amygdala neurofeedback training (3 sessions within 2 weeks). Patients in regular psychotherapeutic treatment (treatment-as-usual) will not be excluded.

The Blood Oxygenation Level Dependent (BOLD) signal from the amygdala, recorded with functional magnetic resonance imaging, is utilized as a feedback signal to patients.

T0: max 7 days before first training session (depends on patient's availability), T1: max 7 days after third training session, T2 (Follow up): 6 weeks after T1

Fear-potentiated startle with instructed emotion regulation vs. natural responding to emotional pictures

T0: max 7 days before first training session (depends on patient's availability), T1: max 7 days after third training session, T2 (Follow up): 6 weeks after T1

Peripheral physiologic measure: resting heart rate variability (relation of high vs. low frequencies in spectrum)

T0: max 7 days before first training session (depends on patient's availability), T1: max 7 days after third training session, T2 (Follow up): 6 weeks after T1

T0: max 7 days before first training session (depends on patient's availability), T1: max 7 days after third training session, T2 (Follow up): 6 weeks after T1

Central nervous system measures: amygdala BOLD response in Sternberg-Working Memory test with emotional vs. neutral distractor images

T0: max 7 days before first training session (depends on patient's availability), T1: max 7 days after third training session, T2 (Follow up): 6 weeks after T1

ZAN-BPD structured interview (acquisition in T0 and T2), BSL-23 self-report questionnaire (acquisition in T0, T1, T2; time lag matched to treatment group).

T0: max 7 days before first training session (depends on patient's availability), T1: max 7 days after third training session, T2 (Follow up): 6 weeks after T1

Inclusion Criteria: Current BPD (≥ 5 DSM-V criteria), female, informed consent for study participation Exclusion Criteria: Psychotropic medication 2 weeks before start (SSRIs excluded) Lifetime diagnosis of schizophrenia or bipolar I Substance dependence in the preceding year Current substance use Pregnancy Epilepsy Antecedent cranial or brain injuries Organic brain diseases Severe medical or neurological condition BMI<16.5 Metallic non-removable items in or on the body which are not MR compatible, Permanent make-up Claustrophobia, left-handedness

Pseudonymized data will be shared with projects from the BrainTrain-network, http://www.braintrainproject.eu

Paret C, Kluetsch R, Zaehringer J, Ruf M, Demirakca T, Bohus M, Ende G, Schmahl C. Alterations of amygdala-prefrontal connectivity with real-time fMRI neurofeedback in BPD patients. Soc Cogn Affect Neurosci. 2016 Jun;11(6):952-60. doi: 10.1093/scan/nsw016. Epub 2016 Feb 1.

Paret C, Ruf M, Gerchen MF, Kluetsch R, Demirakca T, Jungkunz M, Bertsch K, Schmahl C, Ende G. fMRI neurofeedback of amygdala response to aversive stimuli enhances prefrontal-limbic brain connectivity. Neuroimage. 2016 Jan 15;125:182-188. doi: 10.1016/j.neuroimage.2015.10.027. Epub 2015 Oct 16.

Paret C, Kluetsch R, Ruf M, Demirakca T, Hoesterey S, Ende G, Schmahl C. Down-regulation of amygdala activation with real-time fMRI neurofeedback in a healthy female sample. Front Behav Neurosci. 2014 Sep 18;8:299. doi: 10.3389/fnbeh.2014.00299. eCollection 2014.