[Predictive factors of suicidal behaviour recurrence in borderline personality disorder patients].

OBJECTIVES: The high prevalence of suicidal behavior (SB) in patients with borderline personality disorder (BPD) raises clinical questions in terms of screening and prevention, particularly for an emergency psychiatric department. The purpose of this prospective study was to determine the risk factors of the recurrence of SB during a one year follow-up in BPD patients consulting the emergency unit of the University Hospitals of Geneva (HUG) following a suicide attempt.

METHODOLOGY: All subjects included in the study (age 18-65) had been diagnosed with BPD according to DSM IV criteria. Furthermore, they all consulted the emergency psychiatric unit after a suicide attempt. The exclusion criteria were the presence of cognitive, bipolar or psychotic disorders. Almost all SB patients from the Canton of Geneva (350 000 inhabitants) are directed to the HUG emergency department. After one year, 95 subjects were included in the study, while the total number of emergency psychiatric consultations was of about 10 000. During the emergency consultations, the clinicians checked the DSM IV criteria for BPD and current Major Depressive Episode, following the usual guidelines, independently of the study. The clinicians were specifically trained to set up the diagnosis of BPD by means of the International Personality Disorder Examination (IPDE). The gravity of depressive disorders was assessed with the Hamilton Depression Rating Scale (HDRS). The recurrence of SB was recorded for every patient during one year.

RESULTS: Among the 95 patients included in the study, 34 patients (36%) were re-admitted to the emergency unit for one or several SB during the first year after inclusion. The recurrence of the SB was significantly higher in women (OR=9.8), in patients with past history of SB (OR=8.9) and in patients living alone (OR=2.5). Interestingly, the presence of a farewell letter seems to be a protective factor (OR=0.1) for SB. Furthermore, low economic status appears to be associated with a higher recurrence risk, but the trend is not statistically significant. Recurrence and intensity (HDRS) of the major depressive episode, drug addiction, and other disorders on axis I of DSM IV did not differ statistically in patients with or without SB recurrence.

CONCLUSION: In this preliminary study, we tried to identify patients at risk for SB, relating to early secondary prevention, starting from the first assessment at the emergency unit.