Detection of comorbidity with Borderline Personality Disorder in patients with Bipolar Disorders

Detection of comorbidity with Borderline Personality Disorder in patients with Bipolar Disorders

Riconoscimento della comorbidità con il disturbo borderline di personalità nei pazienti affetti da disturbo bipolare

  1. Lai, S. Pirarba, F. Pinna, B. Carpiniello

Department of Public Health, Section of Psychiatry and Psychiatric Clinic, University of Cagliari, Italy





Difficulties are frequently encountered in distinguishing between Bipolar Disorder and Borderline Personality Disorder, with dif- ferential diagnosis being complicated by the presence of comor- bidity. The present study aims to evaluate the utility of the Mil- lon Clinical Multiaxial Inventory-III (MCMI-III) in discriminating patients affected by “pure” Bipolar Disorder from those affected by Bipolar Disorder with Borderline Personality or Other Person- ality Disorder.



57 patients (M = 20, F = 37; mean age 47.9 ± 10.8 yrs) af- fected by BD (BD-I 51%; BD-II 49%) in clinically stable remis- sion were recruited; 28 patients were affected by BD (49.1%), 18 by BD and BPD (31.6%), 11 by BD plus Other Personality Disorders (OPD) (19.3%). Subjects were submitted to SCID-I and SCID-II and rated by the CGI-severity and GAF scales, and MCMI-IIII.



MCMI-III scales focusing on “clinical syndromes” and  “se- vere clinical syndromes” revealed significantly higher mean scores for comorbid patients on all scales, with the exception    of somatization and posttraumatic stress scales. In particular, BD + BPD scored highest on Anxiety, Bipolar-manic, Alcohol dependence, Drug dependence and Thought Disorder scales, while BD + OPD scored highest only on the Dysthymia scale. With regard to “clinical personality patterns”, highly significant increases in mean scores were obtained for depressive, narcis- sistic, antisocial, sadistic-aggressive, passive-aggressive scales among BD + BPD patients, who conversely displayed the lowest scores on the obsessive-compulsive scale. Moreover, the highest scores on Avoidant, Dependent and Self-Defeating Scales were obtained by BD + OPD patients, who likewise scored lowest on the Histrionic Scales; no difference in mean scores was detected for the Schizoid scale between BD, BD + BPD, and BD + OPD patients. On taking into account “severe pathology scales”, the highest mean scores for the Borderline scale were detected among BD + BPD, and among BD + OPD for Schizotypal Scale; no inter-group differences emerged with regard to the Paranoid scale. Cluster B and C scales discriminated respectively between BD + BPD and BD + OPD patients (Table II).



MCMI-III may prove to be useful in identifying Bipolar patients with comorbid BPD in routine clinical practice.




La distinzione fra disturbo borderline di personalità e distur-   bo bipolare dell’umore comporta frequenti difficoltà, tenendo conto del fatto che la diagnosi differenziale è non raramente complicata dalla presenza di una comorbidità. Il presente studio ha l’obiettivo di valutare l’utilità del MCMI-III nel discriminare pazienti affetti da un disturbo bipolare “puro” rispetto a pazienti affetti da disturbo bipolare in comorbidità con disturbo border- line o con altri disturbi di personalità.



Sono stati reclutati 57 pazienti (M = 20, F  =  37,  età  media 47,9 ± 10,8 anni) affetti da disturbo bipolare (BD-I 51%; BD-II 49%) in condizioni di stabilizzazione  clinico-sintomatologica;  28 pazienti erano affetti da solo disturbo bipolare (BD) (49,1%),

18 da disturbo bipolare e disturbo borderline  di  personalità (BD + BPD) (31,6%), 11 (19,3%) da disturbo bipolare e altri disturbi di personalità (BD + OPD). Tutti i soggetti sono stati sottoposti alla SCID-I e alla SCID-II e ad una valutazione me- diante le scale CGI-gravità, GAF ed il Millon Clinical Multiaxial Inventory (MCMI-III).



Le scale MCMI-III riguardanti le “sindromi cliniche” e le “sin- dromi cliniche severe” hanno posto in luce punteggi significa- tivamente superiori nei pazienti con comorbidità fatta eccezio- ne per quanto riguarda le scale relative alla “somatizzazione” e allo “stress postraumatico”. In particolare, nei pazienti affetti da BD + BPD sono emersi i punteggi più alti alle scale di ansietà, bi- polare-maniacale, dipendenza da alcol, dipendenza da sostanze, e disordini del pensiero, mentre i pazienti affetti da DB + OPD hanno dimostrato i punteggi più alti alla scala della distimia. Per quanto riguarda le scale dei “Patterns clinici di personalità” so- no emersi punteggi medi significativamente più elevati alle scale relative ai patterns depressivo, narcisistico, antisociale, sadico- aggressivo, passivo-aggressivo tra i pazienti affetti da BD + BPD, i quali al contrario hanno dimostrato i punteggi più bassi alla scala ossessivo-compulsiva. Viceversa, i pazienti affetti da BD + OPD hanno dimostrato i punteggi più alti alle scale dei patterns evi- tante, dipendente, e autofrustrante e quelli più bassi alla scala istrionica. Nessuna differenza è emersa fra i tre gruppi di pazienti per quanto riguarda la scala schizoide. Prendendo in considera- zione le scale relative alle “patologie gravi”, i punteggi in assoluto più elevati alla scala Borderline sono emersi fra pazienti affetti da BD + BPD mentre quelli relativi alla scala schizotipica sono emersi fra i soggetti affetti da BD + OPD; nessuna differenza fra i tre gruppi per quanto concerne la scala paranoide. Infine le scale dei Cluster B e C discriminavano rispettivamente i soggetti affetti da BD + BPD e da BD + OPD (Tab. II).



Il MCMI-III si dimostra strumento utile nella pratica clinica rou- tinaria al fine di individuare pazienti bipolari affetti da comor- bidità con disturbo borderline di personalità.


Parole chiave

Disturbo bipolare • Disturbo di personalità • Comorbidità • Dimensio- ni di personalità • Millon Clinical Multiaxial Inventory-III




Based on a series of evidences including high rates of co- morbidity, frequent overlapping of symptom features, risk factors and response patterns to pharmacological treat- ments between borderline personality (BPD) and major mood disorders, particularly Bipolar Disorders, some au- thors have raised the question as to whether BPD is an independent disease or might possibly be more appro- priately classified as belonging to the spectrum of mood disorders 1-3. Accordingly, three possible hypotheses have been put forward: 1) BPD is a variant of affective disor- ders; 2) borderline personality predisposes to mood dis- orders; 3) the two disorders may have etiological features in common 4 5. These hypotheses have lead to a contro- versy which remains unresolved 6-11.

From a clinical point of view, differential diagnosis be- tween these two disorders may prove exceedingly diffi- cult 12. Affective instability, explicitly included as a criteri- on for borderline personality (DSMIVTR), may character- ize even bipolar disorders 13, whilst impulsivity, another recognized feature of BDP, may be found among bipolar patients 14 15. Thus, a major issue from a clinical perspec- tive is represented by the difficulty of diagnosing patients presenting with both affective instability and impulsivity, clinical features commonly shared both by Bipolar pa- tients and Borderline patients 13 16. Diagnostic problems are further complicated by the high rate of comorbidity between bipolar disorder and cluster B and C personality disorders 17 18. In particular, comorbid borderline personal- ity disorders are detected in a percentage of cases ranging from 7% to 41% in bipolar patients 17-21. These diagnostic difficulties have been confirmed by several recent stud- ies, revealing not only a problem of underdiagnosis of bi- polar disorder, but also an equally marked occurrence of overdiagnosis 22. Moreover, a very recent study regarding 82 outpatients previously diagnosed as having a bipolar disorder that was not confirmed when they were inter- viewed by means of the Structured Clinical Interview for

DSMIV, revealed that these patients overdiagnosed with bipolar disorder were significantly more likely to be diag- nosed with other disorders, and in particular borderline personality disorder 23. This finding confirms the intrinsic difficulties encountered in differential diagnosis between BDP and BD.

Thus, the distinguishing of bipolar patients with or with- out BPD comorbidity is an outstanding problem. In view of the well known difficulties in adopting structured clini- cal interviews in common clinical practice, self-admin- istered personality evaluation instruments might be of use in discriminating between bipolar disorders with and without BPD; in particular, the Millon Clinical Multiaxial Inventory-III (MCMI-III) 24 seems to be a highly suitable candidate for this purpose, in view of its dimensional ap- proach covering a variety of clinical personality patterns, severe personality pathology and clinical syndromes of- ten missed in the categorical evaluation 25. Starting from these premises, the aim of the present study, as part of  a ongoing study project on comorbidity between bipo- lar and personality disorders, was to test the usefulness of MCMI-III in discriminating bipolar disorders with and without comorbid borderline B personality disorder.


Materials and Methods


Criteria applied for inclusion in the study were: age 18-65 years; lifetime diagnosis of bipolar I or bipolar II disorder according to DSMIVTR criteria 26; absence of current de- pressive, manic/hypomanic or mixed episode according to DSMIV criteria, together with stable clinical remission over the last month and providing of informed consent to take part in the study. Exclusion criteria were: patients with a past or current schizophrenic, schizoaffective or other psychotic disorder; patients with a past or current mental disorder due to a medical condition; current men- tal retardation or other significant cognitive disturbances; current severe physical illness; concurrent alcohol and/or other substance abuse/dependence. All consecutive out- patients attending a university community mental health centre who fulfilled the above mentioned criteria were enrolled in the study. Following routine protocols, pa- tients were diagnosed by a senior psychiatrist on the ba- sis of a non-structured clinical interview. They were also submitted to SCID I 27 and SCID II 28 by a fellow in psychi- atry (LL) trained in conducting the interviews, to confirm the diagnosis of bipolar disorder type I or II and evaluate the presence of a concurrent personality disorder. Clini- cal history and demographical data were collected from clinical records. Severity of psychopathology was evalu- ated by means of the Clinical Global Impression sever-  ity scale (CGI-s) 29 and Global Assessment Functioning (GAF) scale 30. Personality characteristics were evaluated using the Italian Version 31 of the MCMI III 24, a 175 item self-administered questionnaire which takes into consid- eration 25 scales: 11 focusing on “moderate personality disorders” (schizoid, avoidant, depressive, dependent, histrionic, narcissistic, antisocial, sadistic-aggressive; ob- sessive-compulsive; passive-aggressive; masochistic), 3 assessing severe personality disorders (schizotypic, bor- derline, paranoid); 6 moderate clinical syndromes (anxi- ety, somatization, mania, disthymia, alcohol depend- ence, substance dependence,posttraumatic stress disor- der); and 3 scales evaluating severe clinical syndromes (thought disorder, major depression, delusional disorder), as well as 4 control scales. Ratings at each MCMI-III scale are expressed as Base Rate (BR) scores. Patients were also assessed by other self evaluation tools such as the Bar- ratt Impulsiveness Scale 32 and the Aggression Question- naire 33 to ascertain additional personality traits (results not reported here).


Statistical analysis

Statistical analysis was performed by means of statistical package SPSS-11. Pearson’s Chi square test and Fisher Exact Test were used for non continuous variables. T test for independent samples and One-Way Analysis of Vari- ance with post hoc Bonferroni’s Test were used for con- tinuous variables.



The sample selected for the present study on the ba-    sis of the above mentioned inclusion/exclusion criteria was originally made up of 60 subjects. 3 subjects re- fused to take part in the study. The final sample com- prised  57  bipolar  patients  (29  bipolar  I  51%,  28 bipolar II, 49%), 20 males (35.1%) and 37 females (64.9%); mean age was 47.9 ± 10.8 yrs (males: 45.2 ± 10.15; females: 49.03 ± 11.21, t = -1.269, df = 55, p = 0.210); mean  years  of  education  were  10.77  ±  3.89  (males: 10.0 ± 3.34; females: 11.84 ± 4.11, t = -1.714, df = 55, p  =  0.09);  22  patients  (38%)  were  employed  (males: n = 8,40%; females: n = 14, 37.8%), 35 (62%) were unemployed (males: n = 12, 60%; females: n = 23, 62.2%; chi square test = 0.016, df = 1, p = 0.901); 21 (36.8%) patients were married (males: n = 9, 45%; females:     n = 12, 32.4), 36 were singles (males: n = 11, 55%; females: n = 25, 67.6%, chi square test = 0.424, df = 1,   p = 0.515). An Axis II comorbidity was found in 50.8% of the sample. 28 patients were affected by bipolar disorder (49.1%), 18 by bipolar disorder and borderline personal- ity disorder (31.6%), 11 (19.3%) by bipolar disorder and other personality disorders (obsessive-compulsive n = 2; obsessive-compulsive+schizoid N = 1; avoidant N = 2; paranoid n = 2; histrionic n = 1; dependent n = 1, not otherwise specified n = 3). No difference was detected among BP, BP/BPD and BP/OPD patients with respect to education, marital status, and occupation. All patients were submitted to routine treatment (clinical monitoring psychopharmacological treatment, supportive psycho- therapy).




Clinical Variables

Clinical variables of patients with BD, BD/BPD and BD/ OPD are reported in Table I. No difference was found be- tween groups for age at onset of Bipolar Disorder and du- ration of illness. GAF mean score was significantly higher among BD patients, CGI mean score was significantly higher in BP/OPD patients. No significant difference in mean number of drugs taken per patient was detected. Mean number of attempted suicides was significantly higher among BP/BPD patients both respect to BD and BD/OPD patients.



Mean scores at MCMI-III are reported in Table II. As- sessment of clinical personality patterns revealed highly significant increased mean scores for depressive, narcis- sistic, antisocial, sadistic-aggressive, passive-aggressive scales among BD + BPD patients, who conversely dis- played the lowest scores in obsessive-compulsive scale; BD + OPD patients showed the highest scores in Avoid- ant, Dependent and Self-Defeating Scale and the lowest scores in Histrionic Scales; no difference in mean scores for schizoid scale was found between BD,  BD  +  BPD  and BD + OPD patients. When Severe Pathology Scales were taken into account, the highest means scores in Borderline scale was detected  among  BD  +  BPD  and  in Schizotypal Scale among BD + OPD scale; no inter- group differences were found at Paranoid scale. With re- gard to “clinical syndromes” scales, BD + BPD patients were characterized by significantly higher mean scores at anxiety, bipolar-manic, alcohol dependence, and Drug- dependence scale, while BD + OPD patients achieved the highest scores in dysthymia scale; no difference in somatization and Posttraumatic stress disorder scale was found between groups. In scales assessing “severe clinical syndromes” BD + BPD patients rated significantly higher at Thought Disorder scale, while no difference between groups were found at Major Depression and Delusional Disorder scales. Finally, when scales of “Clusters” were examined, BD + BPD patients were significantly higher on “B” and BD + OPD patients on “C” subscale. No difference was found between groups was found for cluster “A” scale


BD BD + BPD BD + OPD Total Statistics
Mean Age (± sd) at onset of Bipolar Illness 27.79 ± 11.18 26.28 ± 10.87 31.91 ± 17.16 28.11 ± 12.35 F = -0.720

df = 56

p = 0.491

Mean duration (± sd) of illness (yrs) 22.54 ± 10.08 18.17 ± 11.35 14.36 ± 13.14 19.58/-11.37 F = 2.349

df = 56

p = 0.105

Mean number of drugs prescribed 2.38 ± 1.09 2.56 ± 0.96 2.40 ± 1.07 2.44 ± 1.04 F = 0.151

Df = 51

P = 0.860

Mean score (± sd) at CGI 3.21 ± 0.68 3.61 ± 0.60 3.82 ± 0.6o 3.46 ± 0.68 F = 4.190

df = 56

p = 0.02

Mean score (± sd) at GAF 69.64 ± 5.64 65.17 ± 9.18 63.36 ± 3.82 67.02/-7.67 F = 3.744

df = 2,54 p = 0.03*

Mean number of attempted suicides 0.54 ± 1.07 1.67 ± 1.68 0.85 ± 0.44 0.85 ± 1.35 F = 5.914

Df = 54

p = 0.005**

BD: Bipolar Disorder; BD + BPD: Bipolar Disorder + Borderline Personality Disorder; BD + OPD: Bipolar Disorder + other Personality Disorder;

* post-hoc test = BD > BD + OPD, p = 0.034; ** post-hoc test = BD + BPD > BD, p = 0.012; BD + BPD > BD + OPD, p = 0.012.



Prior to discussion of the results obtained, several limita- tions of the present study should be acknowledged. First, the sample examined was made up of patients judged to be in “stable” remission, mainly to avoid the possible in- fluence of a highly symptomatic status on personality as- sessment 17. The authors did however include “remitted” patients who no longer met criteria for depressive, manic, mixed or hypomanic episodes according to DSMIVTR in the study, thus not excluding the possibility of persisting subsyndromal status which may have influenced, at least in part, personality evaluation. Secondly, the sample in-

vestigated was relatively small and comprised a mixed sample of bipolar patients, thus limiting the possibility of achieving a separate evaluation of the impact of axis II co- morbidity on bipolar I and bipolar II patients. Moreover, the overall high level of axis II comorbidity found in the present study is likely to have been influenced by the re- ferral patterns employed in our unit, which often receives secondary and tertiary referrals from other centres, thus re- sulting in the sample including numerous complex cases. Conversely, the exclusion of patients affected by comorbid alcohol and/or drug abuse/dependence may have reduced the generalizability of findings obtained. A further limita- tion is represented by the lack of a control group of patients with BPD alone, a difficulty encountered in similar studies, due to the relative scarcity of such individuals, with 80%  or more of BPD patients being affected by comorbid mood disorders 34. Fifty-one percent of bipolar patients observed in the present study were affected by a comorbid personal- ity disorder, a finding exceeding figures obtained in clinical studies reported in literature: a previous study conducted in the same country by Rossi et al. 18 found a 42% preva- lence rate of comorbid personality disorder among bipolar patients, while Brieger et al. 20 reporting pooled data from seven studies estimated a comorbidity rate of 45.2%. This finding appears to be of considerable importance, particu- larly in view of the fact that in routine clinical practice the use of standardized diagnostic techniques is infrequent and detection of comorbidity may be overlooked 35; indeed, examination of clinical records of cases included in this study revealed how an axis II diagnosis was present only  in a minority of cases (7/29, 24.1%). With regard to the main sociodemographic and clinical characteristics of the samples, “pure” bipolar patients do not differ significantly from patients with borderline or other comorbid person- ality disorders. However, mean scores obtained at CGI-s were significantly higher in comorbid cases than in “pure” bipolars, although in both sub-samples the degree of sever- ity was low (mean scores of approx. 3.7 in the presence of comorbidity and 3.2 in non-comorbid patients), indicating the presence of residual symptoms, as expected in patients in stable remission. The presence of less pronounced mean GAF scores in comorbid patients (approx. 65 compared to approx. 70 non-comorbid patients) demonstrated a poor functional status in these subjects. Taken together, these results are largely convergent with findings emerging from other clinical studies 17 36 37. In particular, in line with the findings of the present study, George et al. 17 reported more severe symptoms and psychosocial adjustment in comor- bid subjects than in bipolar patients in remission. In the present study the rate of attempted suicides was approx. three times higher in bipolar patients with comorbid BPD respect to “pure” BP and 7.6 times higher than in bipo-  lar patients with other comorbid personality disorders, a difference that may not be linked to significant differenc- es in duration or severity of illness, type of treatments or sociodemographic variables of subsamples examined. As shown in literature, suicidality characterizes both bipolar disorders 38 39 and personality disorders 40, particularly clus- ter B personality disorders 41, more specifically borderline personality disorder 42. The findings of this study, therefore, indicate that comorbidity with BPD considerably increases the risk of self-harm in bipolar patients. The finding is con- sistent with results reported by Moran et al. 43 and Ucok   et al. 44 who showed a significant major risk of attempted or complete suicide among psychotic patients with comorbid PDs compared to patients lacking comorbidity. Results are also in line with those obtained by Garno et al. 45, demon- strating how lifetime suicide attempts in bipolars are signifi- cantly associated with cluster B comorbidity.


Scales BD BD + BPD BD + OPD Statistics
Clinical Personality Patterns
Schizoid 56.43 (22.82) 58.00 (25.70) 73.36 (27.57) F = 1.964, N.S.
Avoidant 40.04 (31.61) 44.61 (27.65) 78.26 (28.45) F = 6.670, p = 0.003

BD + OPD > BD p = 0.002

BD + OPD > BD + BPD, p = 0.015

Depressive 54.00 (32.51) 83.39 (19.06) 80.82 (18.88) F = 8.143, p = 0.001

BD + BPD > BD, p = 0.002 BD + OPD > BD, p = 0.02

Dependent 50.43 (30.34) 62.33 (25.46) 76.73 (18.09) F = 3.997, p = 0.025

BD + OPD > BD, p = 0.025

Histrionic 59.96 (17.89) 56.39 (16.04) 41.18 (21.09) F = 4.258, p = 0.019

BD > BD + OPD, p = 0.016

Narcissistic 62.96 (18.43) 70.78 (17.97) 48.36 (24.10) F = 4.530, p = 0.015

BD + BPD > BD + OPD, p = 0.012

Antisocial 43.00 (22.58) 69.94 (14.05) 58.64 (24.58) F = 9.540, p = 0.000

BD + BPD > BD, p = 0.000

Aggressive 51.64 (24.13) 69.17 (11.90) 61.73 (20.93) F = 4.393, p = 0.017

BD + BPD > BD, p = 0.015

Compulsive 60.61 (12.76) 39.06 (15.07) 50.09 (12.73) F = 14.01, p = 0.000

BD > BD + BPD, p = 0.000

Passive-Aggressive 62.50 (29.90) 83.78 (18.65) 76.36 (26.69) F = 3.800, p = 0.029

BD + BPD > BD, p = 0.029

Self-defeating 47.43 (33.21) 66.22 (24.16) 76.55 (23.46) F = 4.821, p = 0.012

BD + OPD > BD, p = 0.02

Severe Personality Pathology
Schizotypal 29.36 (31.96) 60.22 (23.67) 66.36 (23.89) F = 10.060, p = 0.000

BD + BPD > BD, p = 0.002 BD + OPD > BD, p = 0.002

Borderline 44.32 (30.35) 82.22 (18.51) 67.55 (22.11) F = 12.189, p = 000

BD + BPD > BD, p = 0.000 BD + OPD > BD, p = 0.044

Paranoid 54.07 (23.62) 65.56 (26.00) 57.55 (36.58) F = 0.983, NS
Clinical Syndromes
Anxiety 56.71 (28.89) 75.39 (24.11) 71.82 (22.08) F = 3.156, p = 0.051
Somatoform 41.71 (24.59) 57.83 (30.15) 50.82 (24.19) F = 2.092, NS
Bipolar-manic 53.14 (26.40) 73.39 (12.71) 60.91 (22.85) F = 5.883, p = 0.005

BD + BPD > BD, p = 0.012

Dysthymia 46.07 (28.05) 67.89 (18.58) 69.27 (23.62) F = 4.526, p = 0.015

BD + BPD > BD, p = 0.015 BD + OPD > BD, p = 0.032

Alcohol-dependence 45.36 (26.37) 61.89 (14.18) 59.64 (15.57) F = 3.877, p = 0.027

BD + BPD > BD, p = 0.040

Drug dependence 45.21 (22.99) 62.06 (15.36) 51.18 (22.55) F = 3.539, p = 0.034

BD + BPD > BD, p = 0.029

Post-traumatic 44.80 (21.90) 47.88 (19.87) 50.88 (18.90) F = 0.37 NS
Scales BD BD + BPD BD + OPD Statistics
Severe Clinical Syndromes
Thought disorder 42.79 (27.83) 66.22 (16.44) 63.55 (16.83) F = 6.866, p = 0.002

BD + BPD > BD, p = 0.004 BD + OPD > BD, p = 0.041

Major depression 44.43 (29.68) 62.56 (31.72) 63.91 (34.25) F = 2.546, NS
Delusional disorder 40.68 (27.44) 60.28 (26.73) 52.64 (33.55) F = 2.689, NS
Cluster A 43.36 (21.84) 60.72 (19.68) 65.18 (24.42) F = 4.043,p = 0.023
Cluster B 52.46 (12.75) 69.50 (11.62) 53.45 (13.29) F = 11.063, p = 0.000

BD + BPD > BD, p = 0.000

BD + BPD > BP + OPD, p = 0.004

Cluster C 53.04 (19.66) 57.00 (14.86) 70.00 (15.67) F = 3.699, p = 0.031

BD + OPD > BD, p = 0.027

BD: Bipolar Disorder; BD + BPD: Bipolar Disorder + Borderline Personality Disorder; BD + OPD: Bipolar Disorder + Other Personality Disor- der.


Several studies conducted previously applied the Millon Clinical Multiaxial Inventory in the psychological assess- ment of patients affected by bipolar mood disorder: Choca et al. 46 used MCMI-I to evaluate patients with major af- fective disorders, Wetzler et al. 47 compared unipolar and bipolar patients by means of MCMI-II and Turley et al. 48 used MCMI-II in recent onset bipolar disorder. However, the present study was the first to apply MCMI-III in the comparison of bipolar patients with and without comor- bid personality disorder. Results obtained at “clinical syn- dromes” and “severe clinical syndromes” scales of MCMI- III revealed significantly higher mean scores for comorbid patients at all scales, excluding somatization and posttrau- matic stress scales; in particular BD + BPD achieved the highest scores in several scales such as Anxiety, Bipolar- manic, Alcohol dependence and Drug dependence and Thought Disorder scales, whilst BD + OPD showed the highest scores only at Dysthymia scale. However, the most significant increase in rating from a clinical point of view (BR scores ≥ 74) was detected in BD + BPD for anxiety, bipolar-manic scales. These results suggested a more pro- nounced presence of anxiety and mood-related symptoms in BD patients with comorbid BPD, even when judged in clinical remission, a finding which may be interpreted as expression of the intrinsic affective component of BPDs.

As expected, comorbid bipolar patients rated statistically significant different scores for all “personality patterns” scales (with the sole exception of the schizoid scale) and “severe pathology of personality” scales (excluding the “paranoid” scale). BD + BDP displayed the highest scores in the majority of scales including depressive, narcissis- tic, antisocial, sadistic-aggressive, passive-aggressive and Borderline scales and the lowest for obsessive-compulsive scales while BD + OPD rated higher scores at Avoidant, Dependent, Self-Defeating and Schizotypal scales; on taking into account only clinically relevant scores (≥ 74) BD + BPD achieved exceedingly high scores at “depres- sive”, “self-defeating” and “borderline” scales of personal- ity patterns while BD + OPD showed very high scores at “Avoidant” scale. With regard to the BD + BPD sample the results obtained seem to reflect the intrinsic clinical characteristics of borderline personality disorders, whilst for BD + OPD they likely reflect the composition of this subsample, mainly constituted by cluster C disorders (ap- prox. 70% pts of this group). The latter hypothesis seems to be confirmed by the higher scores obtained at “clus- ter B” scales and “cluster C” scales respectively by BD + BPD and by BD + OPD. Interestingly, the very low scores achieved by BP + BDP patients at obsessive-compulsive scale may be interpreted as a confirmation of the hypothe- sis that obsessive-compulsive dimension should no longer be considered a trait of the anxiety domain but rather as an extreme of a personality trait ranging from excessive self-control to impulsivity, as indicated by studies showing an inverse correlation between impulsivity measures and obsessive-compulsive disorders 49 50. As a consequence, the finding in bipolar patients deemed in clinical remission of higher than expected scores at clinical scales regarding mood and anxiety dimensions, of exceedingly high scores at scales intrinsically linked to BPD (borderline) or evalu- ating affective dimensions of personality (depressive, self- defeating), and very low scores at obsessive-compulsive scale possibly indicating marked impulsivity traits, may orient the clinician to suspect a comorbidity with border- line personality disorder, which might have been missed on the basis of clinical evaluation alone. Thus, in the light of the importance of this comorbidity in terms of course, outcome and therapeutic management of bipolar patients, the possibility of improving diagnostic accuracy by means of a user-friendly self-evaluation instrument such as MC- MI-III may be of particular relevance in routine clinical practice.

To conclude, the results obtained in the present study un- derline the general utility of MCMI-III in distinguishing between patients affected by “pure” bipolar disorder and bipolar patients with comorbid personality disorders; in particular, the tool clearly differentiates bipolar patients with comorbid borderline personality disorder from pa- tients with other personality disorders



1 Atre-Vaida N, Hussain S.  Borderline personality disorder and bipolar mood disorders: two distinct disorders or a con- tinuum? J Nerv Ment Dis 1999;57:313-5.

2 Deltito J, Martin L, Riefkhol J, et al. Do patients with border- line personality disorder belong to the bipolar spectrum? J Affect Disord 2001;67:221-8.

3 Akiskal HS. Subaffective disorders: dysthymic, cyclothymic andf bipolar II disorders in the borderline realm. Psychiatr Clin North Am 1981;4:25-46.

4 Gunderson JG, Philiphs Ka. A current view of the interface between borderline personality disorder and depression. Am J Psychiatry 1991;148:967-75.

5 Koenisberg HW, Anwunah I, News AS, et al. Relationship between depression and borderline personality disorder. Depress Anxiety 1999;10:158-67.

6 Smith DJ, Muir WJ, Blackwood DKR. Is Borderline  Dis- order part of the Bipolar Spectrum? Harv Rev Psychiatry 2004;12:133-9.

7 Berrocal C, Ruiz Moreno MA, Rando MA, et al. Borderline personality disorder and mood spectrum. Psychiatry Re- search 2008;159:300-7.

8 Paris J. Borderline or bipolar? Distinguishing borderline per- sonality disorder from bipolar spectrum disorder. Harv Rev Psychiatry 2004;12:140-5.

9 Paris J, Gunderson J, Weinberg I. The interface between bor- derline personality disorder and bipolar spectrum disorder. Compr Psychiatry 2007;48:145-54.

10 Benazzi F. Borderline personality disorder and bipolar II disorder in private practice depressed outpatients. Compr Psychiatry 2000;42:106-10.

11 Blacker D, Tsuang MT. Contested boundairies of bipolar disorder and the limits of categorical diagnosis in psychiatry. Am J Psychiatry 1992;149:1473-83.

12 Bolton S, Gunderson JG. Distinguishing borderline personal- ity disorder from bipolar disorder: differential diagnosis and implications, Am J Psychiatry 1996;153:1202-7.

13 Henry C, Mitropoulou V, News A, et al. Affective instability and impuilsivity in borderline personality disordes:similarities and differences. J Psychiatr Res 2001;35:307-12.

14 Swann A, Dougherty DM, Pazzaglia PJ, et al. Increased impulsivity associated with severity of suicide attempts history in patients with bipolar disorder. Am J Psychiatry 2005;162:1680-7.

15 Lewis M, Scott J, Fragou S. Impulsivity, personality and bipo- lar disorder. European Psychiatry, 2009;24:464-9.

16 Magill CA. The boundary between borderline personality disorder and bipolar disorder: current concepts and chal- lenges. Can J Psychiatry 2004;49:551-6.

17 George EL,  Miklowitz  DJ,  Richards  JA,  et al.  The comorbidity of bipolar disorder and Axis II personality disorders: prevalence and clinical correlates. Bipolar Disorders 2003;5:115-22.

18 Rossi A, Marinangeli MG, Butti G, et al. Personality disorders in bipolar and depressive disorders. J Affect Disord 2001;65:3-8.

19 Bellino S, Allasia L, Bogetto F. Borderline personality dis- order and bipolar disorder II: comparison of personality di- mensions. Giorn Ital Psicopat 2007;13:293-9.

20 Brieger P, Ehrt U, Marneros A. Frequency of comorbid per- sonality disorders in bipolar and unipolar affective disorders. Compr Psychiatry 2003;44:28-34.

21 Schiavone P, Dorz S, Conforti D, et al. Comorbidity of DSM- IV Personality Disorders in unipolar and bipolar affective dis- orders: a comparative study. Psychol Rep 2004;95:121-8.

22 Zimmermann M, Ruggero CJ, Chelminski I, et al. Is Bipolar Disorder overdiagnosed? J Clin Psychiatry 2008;69:935-40.

23 Zimmermann M, Ruggero CJ, Chelminski I, et al. Psychiatric diagnoses in patients previously overdiagnosed with Bipolar disorder. J Clin Psychiatry 2010;71:26-31.

24 Millon T, Millon CM, Davis M, MCMI-III:Millon Clinical Multiaxial Inventory-III, Minneapolis, MN: National Com- puter Systems 1994.

25 Krueger RF, Piasecki TW. Toward a dimensional and psy- chometrically-informed approach to conceptualizing psy- chopathology. Behav Res Ther 2001;40:485-9.

26 American Psychiatric Association. Diagnostic and Statistical Manual fof Mental Disorders. Fourth edition Text Revised (DSM-IV,TR). Washington, DC: American Psychiatric Asso- ciation 2000.

27 First MB, Spitzer RL, Williams JBW, et al. Structured Clini- cal Interview for DSM IV Axis I Disorders- Research Version (SCID I, version 2.0). New York: Biometrics Research De- partment, New York State Psychiatric Institute 1996.

28 First MB, Gibbon M, Spitzer RL. Structured Clinical Interview for DSM IV Axis II-Personality Disorders (SCID II,Version 2.0). New York: Biometrics Research Department, New  York State Psychiatric Institute 1996.

29 Guy W. Clinical Global Impression (CGI). In: ECDEU As- sessment Manual for Psychopharmacology, Revised. NIMH Publ No (Adm), 76-338.  Rockville,  MD:  NIMH  1976,  pp. 217-22.

30 American Psychiatric Association. Global Assessment of functioning scale (GAF). In: Diagnostic and Statistical Man- ual fof Mental Disorders. Fourth edition (DSM-IV). Washing- ton, DC: American Psychiatric Association 1994, pp. 30-2.

31 Zennaro A, Ferracuti S, Lang M, et al. MCMI-III, Millon Multiaxial Clinical Inventory. Ed. italiana. Firenze: Giunti OS 2008.

32 Patton JH, Stanford MS, Barratt ES. Factor structure of the Barratt Impulsivness scale. J Clin Psychol 1995;51:768-74.

33   Buss A, Perry M. The Aggression Questionnaire. J Pers Soc Psychol 1992;63:452-9.

34 Wilson ST, Stanley B, Oquendo MA, et al. Comparing im- pulsiveness, hostility, and depression in borderline disorder and bipolar II disorder. J Pers Soc Psychol 2007;68:1533-9.

35 Zimmermann M, Mattia JI. Psychiatric diagnosis in clini- cal practice: is comorbidity missed? Compr Psychiatry 1999;40:182-91.

36 Krishnan K. Psychiatric and medical comorbidities of bi- polar disorder. Psychosom Med 2005;67:1-8.

37 Preston GA, Marchant BK, Reimherr FW, et al. Borderline personality disorder in patients with bipolar disorder and re- sponse to Lamotrigine. J Affect Disord 2004;79:297-303.

38 Jamison KR. Suicide and bipolar disorder. J Clin Psychiatry 2000;61(Suppl 9):47-51.

39 Hoyer EH, Olesen AV, Mortensen PB. Suicide risk in pa- tients hospitalised because of an affective disorder: a follow up study, 1973-1993. J Clin Psychiatry 2004;78:209-17.

40 Krysinska K, Heller TS, De Leo D. Suicide and deliberate self-harm in personality disorders. Curr Opin Psychiatry 2006;19:95-101.

41 Pompili M, Ruberto A, Giraldi P, et al. Suicidality in DSMIV cluster B personality disorders. An overview. Ann Ist Super Sanita 2004;40:475-83.

42 Oldham JM. Borderline personality disorder and suicidality.

Am J Psichiatry 2006;163:20-6.

43 Moran P, Walsh E, Tyrer P, et al. Does comorbid personality disorder increase the risk of suicidal behaviour in psychosis? Acta Psychiatr Scand 2003;107:441-8.

44 Ucok A, Karaveli D, Kundakci E, et al. Comorbidity of per- sonality disorders with bipolar mood disorders. Compr Psy- chiatry 1998;39:72-4.

45 Garno JL, Goldberg JF, Ramirez PM, et al. Bipolar Disorders with Comorbid Cluster B Personality Disorder features; im- pact on suicidality. J Clin Psychiatry 2005;66:339-45.

46 Choca J, Bresolin L, Okonek A, et al. Validity of the Millon Clinical Multiaxial Inventory in the assessment of affective disorders. J Pers Assess 1988;52:96-105.

47 Wetzler S, Khadivi A, Oppenheim S. The psychological as- sessment of depression: unipolars versus bipolars. J Pers As- sess 1995;65:557-66.

48 Turley B, Bates  GW,  Edwards  J,  et  al.  MCMI-II  diag- nosis in recent onset bipolar disorders. J Clin Psychol 1992;48:320-9.

49 Fossati A, Barratt ES, Borroni A, et al. Impulsivity, aggres- siveness and DSMIV Personality Disorders. Psychiatry Res 2007;149:157-67.

50 Tomassini A, Riccardi I, Cerroni G, et al. The different mean- ing of impulsiveness related to the Millon Clinical Multiaxial Inventory (MCMI-III) model in anxiety disorders. Giorn Ital Psicopat 2009;15:375-83.



Dr. Bernardo Carpiniello, Department of  Public  Health-Section  of  Psychiatry,  University  of  Cagliari,  via  Liguria  13,  09127  Cagliari,  Italy  • Tel. +39 070 41518 • Fax: +39 070 480083 • E-mail:


Leave a Reply

%d bloggers like this: