Summary
Objectives
Current empirical studies show renewed interest in defence mechanisms and demonstrate the role of defences in development, psychopathology and social functioning. Traditionally, defence mechanisms were considered as unconscious processes to protect individuals from the awareness of unacceptable thoughts, emotions, impulses or wishes. More recent theories have suggested that defence mechanisms maintain self-esteem and control negative emotions in order to restore a more functional state of mind, like the immune system. Defences can be ordered on a continuum, differing in degree of maturity: the less adaptive or immature are associated with personality disorders. Cluster B personality disorders are very difficult to differentiate for their high comorbidity and low discriminant validity. Some authors have tried to describe the defence profile of borderline personality disorder (BPD), indicating a more frequent use of action and image distorting defences. However, comparisons of defences of different Cluster B categories are still limited. The aim of the present study is to search for the differences in defence mechanisms among three Cluster B personality disorders (borderline, narcissistic, and histrionic personality disorder). The second aim is to examine, in a group of BPD patients, the defensive characteristics that are significantly related to clinical response to standard treatment at our centre (mood stabilizers and/or new antipsychotics combined with interpersonal or supportive psychotherapy).
Methods
The hundred forty-four consecutive outpatients who received a DSM-IV-TR diagnosis of borderline (BPD), or narcissistic (NPD), or histrionic personality disorder (HPD) were recruited (55 males and 89 females). Patients were between 18 and 60 years old. All subjects were tested with a semi-structured interview for clinical variables, the Structural Clinical Interview for DSM-IV Axis II Disorders (SCID II) and the response evaluation measure- 71 (REM-71) to evaluate 21 defence mechanisms. Analysis of variance (ANOVA) was used for statistical analysis (p ? 0.05). The subgroup of 66 patients with BPD were tested with a semistructured interview and the Borderline Personality Disorder Severity Index-IV (BPDSI-IV) at the time of the first visit (T0) and one year later (T1) to assess the difference in severity of clinical symptoms. The 21 defences measured with the REM-71 were included in a regression model with the change of BPDSI score (T0-T1) as a dependant variable. An intention to treat (ITT) analysis was performed.
Results
Patients with NPD showed more differences in the use of defences compared with the other two Cluster B PDs. Acting out (vs. BPD, p = 0.009), dissociation (vs. BPD, p = 0.008; vs. HPD, p = 0.011) and reaction formation (vs. BPD, p = 0.027; vs. HPD, p = 0.038) were found to be less frequently used by narcissistic patients. In contrast, patients with BPD and HPD shared most defences: only humour was more expressed by borderline patients than by histrionic patients (p = 0.023) (Table I). Multiple regression analysis showed that the change in BDPSI score (T0-T1), which indicates the change of symptom severity, was inversely related to omnipotence (p = 0.0005) and withdrawal (p = 0.013), and directly related with suppression (p = 0.032), sublimation (p = 0.024), displacement (p = 0.0005) and passive aggression (p = 0.001) (Table II).
Discussion
The present study found a noticeable, but not complete overlap of the defences used by patients with different Cluster B personality disorders. A possible explanation is that borderline, narcissistic and histrionic disorders are different clinical pictures derived from the same psychopathology, namely the borderline personality organization described by Otto Kernberg. With regard to the relationship between specific PDs and single defence mechanisms, three defences scored higher in the BPD group than in NPD patients: acting out, reaction formation and dissociation. Consistent with our results, previous studies observed that BPD is related to maladaptive action defences. Significantly higher scores of reaction formation in the borderline group could be imputable to the presence of obsessive-compulsive symptoms (27.3%) in BPD patients. The higher score of dissociation in HPD compared with NPD is a somewhat predictable result, because dissociation is usually considered phenomenon related to hysteric/histrionic personality. Also, reactive formation was more frequently used by histrionic than narcissistic patients. Although they are considered two PDs with a different degree of psychopathological abnormality and clinical severity, BPD and HPD shared almost the same defence profile. Actually, in our sample the two disorders differed only for the defence humour, which was more frequently used by BPD patients. Considering the second aim of the study, some adaptive defences in the BPD subgroup (suppression, sublimation, displacement and passive aggression) can be considered as predictors of positive response to standard treatment. However, immature defences like omnipotence and withdrawal appear to be predictive factors of a negative outcome. These results merit further investigation.