Bipolar Disorders 2000: 2: 71–74Printed in Ireland. All rights reser6ed

Brief Report

Comorbidity of obsessive-compulsivedisorder in recovered inpatients with bipolardisorder

Kruger S, Braunig P, Cooke RG. Comorbidity of obsessive-compulsivedisorder in recovered inpatients with bipolar disorder.Bipolar Disord 2000: 2: 71–74. © Munksgaard, 2000

Objecti7e: To determine the frequency of obsessive-compulsive disorder(OCD) in inpatient subjects with bipolar disorder (BD) and to examinethe clinical characteristics of BD subjects with OCD.

Method: The sample consisted of 143 inpatient subjects with DSM-III-R BD-I and BD-NOS (BD-II), recovered from a current episode ofeither depression or mania. Demographic and clinical variables wereobtained on the day of admission. Current comorbid conditions includ-ing OCD were determined by the Structured Clinical Interview forDSM-III-R following recovery from the acute affective episode.

Results: The frequency of current OCD was 7% (N=10). All BD sub-jects with OCD were BD-II, were male, and had a diagnosis of currentdysthymia. They had fewer episodes and a higher incidence of priorsuicide attempts than bipolar subjects without OCD. None of the bipo-lar subjects with OCD fulfilled criteria for cyclothymia.

Conclusions: Our findings suggest that BD-II, OCD, dysthymia, andsuicidality cluster together in some subjects with BD. We discuss theclinical implications of our findings.

Stephanie Krugera,Peter Braunigb,1 and RobertG Cookea

a Centre for Addiction and Mental Health,Clarke Institute of Psychiatry, Mood andAnxiety Disorders Division, University ofToronto, Ontario, Canada; b Zentrum furPsychiatrie und Psychotherapie, RuhrUniversitat Bochum, Germany

Key words: bipolar-II disorder –comorbidity – dysthymia – OCD –suicidality

Received 18 May 1999, revised andaccepted for publication 28 July 1999

Corresponding author: Dr StephanieKruger, Centre for Addiction and MentalHealth, Clarke Institute of Psychiatry, Moodand Anxiety Disorders Division, 250 CollegeSt., M5T 1R8, Toronto, Ontario, Canada.Fax: +1 416 979 6864; e-mail:stephaniekrueger@compuserve.com

Until recently, the comorbidity of obsessive-com-pulsive disorder (OCD) with bipolar disorder (BD)was considered to be rare and was thus not system-atically investigated. The few case reports on thiscomorbidity found OCD to occur during bipolardepression and during the euthymic interval of BDand to remit during mania (1–5).

The first strong evidence of a frequent link be-tween OCD and BD came from the EpidemiologicCatchment Area studies, where Boyd et al. (6)found OCD to occur much more frequently ineuthymic subjects with BD than in the generalpopulation (odds ratio=18). Subsequent system-

atic studies, despite methodologic differences, haveconfirmed the finding of a frequent comorbidity ofthe two disorders, with rates of OCD in BD rang-ing from 8 to 35% (7–16).

Despite this increasing evidence for a strongcomorbidity between BD and OCD, data on thecharacteristics of bipolar subjects with OCD arestill scarce. In the present study, we further investi-gate the complex relationship between OCD andBD by examining the frequency of OCD in asample of recovered inpatients with BD, the clini-cal characteristics of the subjects with comorbidOCD and BD, and the association of coexistingOCD and BD with other psychiatric diagnoses.

Methods

Subjects were drawn from a total of 1214 succes-sive admissions to the General Psychiatry Division,

1 Klinik fur Psychiatrie, Psychotherapie und Psychosomatikam Klinikum Chemnitz, University affiliated Hospital of theUniversity of Leipzig, Germany.

71

Kruger et al.

Westfalisches Zentrum fur Psychiatrie, Universityof Bochum, Germany, over a 2-year period. Sub-jects eligible for inclusion in the study had to meetDSM-III-R (17) criteria either for BD or for BD-NOS depressed phase (equivalent to DSM-IV cate-gories of BD-I and BD-II), they had to be free ofacute medical illness, and had to give informedconsent to participate in the current study. Of 150eligible patients, 145 agreed to participate and 143completed the study. The diagnosis of BD wasmade clinically on admission by the intake psychi-atrist and later confirmed by one of the authors(P.B. or S.K.). Demographic and clinical variablesincluding the number and types of previousepisodes (depression, hypomania, and mania) wereobtained on admission from medical records andpatient interviews. Rapid cycling was defined asthe presence of four or more affective episodes peryear. Diagnoses of current OCD and other currentcomorbid conditions were made using the Struc-tured Clinical Interview for DSM-III-R (SCID)(18) following recovery from the acute affectiveepisode and within 1 week before discharge. Tofulfil criteria for recovery, affective symptoms hadto be sufficiently improved to enable the patient tosatisfactorily function in social and vocational ar-eas. Recovery was clinically determined by one ofthe authors (P.B.) and was substantiated with ascore of B7 on the 17-item Hamilton DepressionRating Scale (19) or of B4 on the Young ManiaRating Scale (20). The comorbid diagnoses wereassessed after recovery and not during the acuteepisode to increase reliability. All SCID interviews

were performed by practicing psychiatrists (P.B.and S.K.) with good interrater reliability for prin-cipal (k=0.096) and comorbid (k=0.94) diag-noses established from joint ratings.

Psychiatric comorbidity including OCD wasdefined as the presence of any DSM-III-R psychi-atric diagnosis within the past month in additionto the principal BD diagnosis (21). Dysthymia andcyclothymia were not considered comorbid unlessit could be determined that subjects had alreadyfulfilled the full criteria for these diagnoses at least2 years before the onset of the affective episode.Mean hospitalization was 4296 days for all sub-jects, which is an average length of stay for sub-jects with BD in this hospital.

To evaluate differences between subjects withand without OCD, x2 analysis was adopted forcategorical variables and t-tests were performedfor continuous variables. A Bonferroni correctionfor multiple comparisons was applied using SPSS,for Windows (version 7.0, licensed to InformationCommons, University of Toronto).

Results

On admission, 109 subjects were diagnosed withBD-I (31 manic and 78 depressed) and 34 withBD-II (all depressed). Table 1 outlines the demo-graphic and clinical characteristics and the differ-ences between the two groups. The frequency rateof current OCD in our subjects with BD was 7%(N=10). All BD subjects with OCD were BD-II,had been hospitalized for depression, were male,

Table 1. Clinical characteristics and current comorbidity in bipolar subjects with and without OCD (N=143)

Non-OCDOCDdfChi2N=133 (93%) pN=10 (7%)Variables

Bipolar-I 0 (0) 108 (81) 34.1 1 0.000Bipolar-II 10 (100) 0.00024 (18) 114.7Rapid cycling 0.17611.820 (15)1 (10)Male 1 0.00034.743 (32)10 (100)

0.000179.17 (5)10 (100)Dysthymia47 (35) 5.2Cyclothymia 1 0.0220 (0)51 (38) 10.0Suicide attempts 1 0.0029 (90)

0.17411.921 (16)1 (10)Substance abusePanic disorder 0 (0) 10 ( 7) 8.1 1 0.369

0.21411.520 (15)3 (30)Phobic disorder117.620 (15) 0.6752 (20)Generalized anxiety disorder

6 (4) 4.7Somatization disorder 1 0.4930 (0)Binge eating 0 (0) 21 (16) 1.9 1 0.675

T df p

0.01544.03 (14.0)44.1 (12.7)Mean age (SD) 0.666141Age at onset of BD (SD) 31.1 (9.4) 30.5 (11.2) 0.151 141 0.880Number of episodes (SD) 5 (5) 11 (14) −1.3 141 0.006

72

Comorbidity of OCD in recovered inpatients with BD

and had a concurrent diagnosis of dysthymia thathad been present at least since adolescence. Theyhad fewer prior affective episodes but a higherincidence of prior suicide attempts than bipolarsubjects without OCD. None of the bipolar subjectswith OCD fulfilled criteria for cyclothymia. Nosignificant differences were found for the otherclinical variables examined (Table 1). After Bonfer-roni correction, all positive findings were still signifi-cant except cyclothymia. At the time of assessment,none of the bipolar subjects with OCD were receiv-ing selective serotonin reuptake inhibitors (SSRIs),all were receiving mood stabilizers, four requiredbenzodiazepines, and none received neuroleptics.All were treated with cognitive psychotherapy.

Discussion

The frequency of comorbid OCD and BD of 7% inthe sample is consistent with the findings in severalother recent prospective and controlled studies (7–10), which reported comorbidity rates from 8 to13%. However, three studies, including one from ourgroup, reported much higher frequency rates rang-ing from 21 (12, 13) to 35% (11). Possible explana-tions for these discrepant findings includedifferences in setting or methodology, such as thephase of BD illness during which subjects wereassessed for comorbidity, in the use of epidemiologicdata versus clinical data from inpatients or outpa-tients, in choice of diagnostic interview, and inwhether lifetime or current diagnoses of OCD weremade.

In the current report, all subjects with OCD andBD were male with BD-II. They had a comorbiddiagnosis of dysthymia more often than subjectswithout OCD, had more prior suicide attempts, andfewer affective episodes. These results are in keepingwith various previous reports of association amongthese clinical variables. Klein et al. (22) reported thata high proportion of their patients with depressionand dysthymia also experienced hypomania at fol-low up and had first degree relatives with BD-II.Chen and Dilsaver (12) reported an increased rateof suicidal and morbid thoughts and a higher fre-quency rate of suicide attempts in subjects withcomorbid BD-I and OCD. Dysthymia has also beenlinked to OCD (23). Perugi et al. (15) assessedlifetime comorbidity in 86 subjects with atypicaldepression and found a frequency of OCD of 20%in so-called ‘soft bipolars’, a diagnosis which in partoverlaps with the DSM-III-R diagnosis of BD-NOS(BD-II). The higher numbers of previous suicideattempts despite fewer episodes are consistent withreports in the literature that patients with OCD andpsychiatric comorbidity have higher suicide rates

than patients with OCD and no other comorbidity(24). Clinically, this finding is of relevance becausethe comorbidity of BD-II and OCD seems to putpatients at risk for suicidality, calling for closemonitoring of these patients.

Our findings suggest that the clinical featuresobserved in subjects with comorbid OCD and BD-IIreflect a shared underlying diathesis. Serotonin dys-function has been hypothesized in BD-II (25, 26),OCD (27), dysthymia (28–30), and suicidality (31).Furthermore, SSRIs are an effective treatment forboth OCD (32, 33) and dysthymia (34, 35), and arereported to provide a more favorable response in thetreatment of BD-II than tricyclic antidepressants(36). Thus a possible, albeit simplistic, hypothesiswould be that a common defect in serotonergicfunction underlies the combination of symptomsobserved in our subjects with comorbid OCD andBD.

We cannot explain the higher percentage of malesin the group of BD subjects with OCD because themajority of epidemiologic studies on both OCD andBD and on the comorbidity of OCD in BD havereported an even gender distribution (11, 12, 37).Further studies on larger samples of BD-II subjectsare needed to verify this finding.

None of our patients with OCD and BD-IIreceived SSRIs at the time of assessment and allreceived cognitive therapy because of reports thateven in low doses, SSRIs may trigger a switch intomania and hypomania in these cases (38, 39).

Limitations of our study include the relativelysmall sample size of bipolar patients and non-blindassessment procedures. However, we used system-atic rating instruments to minimize the influence ofrater bias. Also, retrospective assessment of datarelated to history of illness, such as number ofprevious episodes, by patient interview might impacton the reliability of the information. However, wehave no reason to believe that our patients wereunreliable informants, particularly as most informa-tion was confirmed by chart review.

Conclusions

The present study is an examination of the clinicalfeatures of recovered inpatients with BD. Amongthem we identified a subset with comorbid OCD.These subjects all had BD-II, had more dysthymia,significantly less cyclothymia, and a higher fre-quency of prior suicide attempts than subjects with-out OCD. Although these results are preliminary,we may tentatively conclude that subjects withBD-II should be routinely screened for current OCDand dysthymia and should be monitored for suiciderisk even more carefully than BD subjects withoutOCD. We have hypothesized that there may

73

Kruger et al.

be a common serotonergic linkage in these comorbidsubjects’ syndromes. Further study of the possiblecommon pathophysiology among these comorbidconditions may lead to new developments in theunderstanding and treatment of these disorders.

References

1. Morel BA. Traite des Maladies Mentales. Paris: LibrairieVictor Masson, 1866.

2. Stengel E. A study on some clinical aspects of the relation-ship between obsessional neurosis and psychotic reactiontypes. J Ment Sci 1945; 91: 166–187.

3. Baer L, Minichiello WE, Jenike MA. Behavioural treatmentof obsessive-compulsive disorder with concomitant bipolaraffective disorder. Am J Psychiatry 1985; 142: 358–360.

4. Gordon A, Rasmussen SA. Mood related obsessive-compul-sive symptoms in a patient with bipolar affective disorder.J Clin Psychiatry 1986; 49: 27–28.

5. Keck PE, Lipinsky LE, White K. An inverse relationshipbetween mania and obsessive-compulsive disorder: a casereport. J Clin Psychopharmacol 1986; 6: 123–124.

6. Boyd JH, Burke JD, Gruenberg E, Holzer CE, Rae DS,George LK, Karno M, Stolzman R, McEvoy I, Nesadt G.Exclusion criteria for DSM-III. A study of co-occurrence ofhierarchy-free syndromes. Arch Gen Psychiatry 1984; 41:983–989.

7. Strakowski S, Keck PE, McElroy SL, Lonczak HS, WestSA. Chronology of comorbid and principal syndromes infirst-episode psychosis. Compr Psychiatry 1995; 36 (2):106–112.

8. Strakowski SM, Tohen M, Stoll AC, Faedda GL, GoodwinDC. Comorbidity in mania at first hospitalisation. Am JPsychiatry 1992; 149 (4): 554–556.

9. Strakowski SM, Shelton RC, Kolbrener ML. The effects ofrace and comorbidity on clinical diagnosis in patients withpsychosis. J Clin Psychiatry 1993; 54: 96–102.

10. McElroy SL, Strakowski SM, Keck PE, Tugrul KL, WestSA, Lonczak HS. Differences and similarities in mixed andpure mania. Compr Psychiatry 1995; 36 (3): 187–194.

11. Kruger S, Cooke RG, Hasey GM, Jorna T, Persad E.Comorbidity of obsessive-compulsive disorder in bipolardisorder. J Affect Disord 1995; 34: 117–120.

12. Chen YW, Dilsaver SC. Comorbidity for obsessive-compul-sive disorder in bipolar and unipolar disorders. Psychol Res1995; 59: 57–64.

13. Russo A, Montgomoery SA. Frequency of anxiety disorderscomorbidity in bipolar depression, unipolar depression anddysthymia. J Affect Disord 1997; 42: 145–153.

14. Cosoff SJ, Hafner RJ. The frequency of comorbid anxietyin schizophrenia, schizoaffective disorder and bipolar disor-der. Aust N Z J Psychiatry 1998; 32: 67–72.

15. Perugi G, Akiskal HS, Lattanzi L, Cecconi D, MastrocinqueC, Patronelli A, Vignoli S, Bemi E. The high frequency of‘‘soft’’ bipolar (II) features in atypical depression. ComprPsychiatry 1998; 39 (2): 63–71.

16. Strakowksi SM, Sax KW, McElroy SL, Keck PE Jr,Hawkins JM, West SA. Course of psychiatric and substanceabuse syndromes co-occurring with bipolar disorder after afirst psychiatric hospitalization. J Clin Psychiatry 1998; 59(9): 465–471.

17. American Psychiatric Association. DSM-III-R. Diagnosticand Statistical Manual of Mental Disorders, 3rd edn,revised. Washington DC: American Psychiatric Press,1987.

18. Spitzer RLL, Williams JBW, Gibbon M. Structured ClinicalInterview for DSM-III-R-Patient Edition (SCID-P, Version1.0). Washington DC: American Psychiatric Press, 1990.

19. Hamilton M. A rating scale for depression. J NeurolNeurosurg Psychiatry 1960; 23: 56–62.

20. Young R, Biggs J, Meyer D. A rating scale for mania:reliability, validity and sensitivity. Br J Psychiatry 1978; 133:429–435.

21. Black DW, Winokur G, Bell S, Nasrallah A, Hulbert J.Complicated mania. Comorbidity and immediate outcomein the treatment of mania. Arch Gen Psychiatry 1988; 45:232–236.

22. Klein DN, Taylor EB, Harding K. Double depression andepisodic major depression: demographic, clinical, familial,personality and socioenvironmental characteristics andshort-term outcome. Am J Psychiatry 1988; 145: 1226–1231.

23. Angst J. Comorbidity of anxiety, phobia, compulsion anddepression. Int Clin Pharmacol 1993; 8 (Suppl. 1): 21–25.

24. Hollander E, Greenwald S, Neville D, Johnson J, HornigCD, Weissman MM. Uncomplicated and comorbid obses-sive-compulsive disorder in an epidemiologic sample. De-press Anxiety 1996; 4: 11–119.

25. DePaulo JR, Simpson SG. Therapeutic and geneticprospects of an atypical affective disorder. J Clin Psy-chopharmacol 1978; 7: 50S–54S.

26. Endicott J, Nee J, Andreasen NC. Bipolar II: combine orkeep separate? J Affect Disord 1985; 8: 17–28.

27. Insel TR, Mueller EA, Alterman I. Obsessive-compulsivedisorder and serotonin: is there a connection? Biol Psychi-atry 1985; 20: 1174–1188.

28. Ravindran AV, Chudzik J, Bialik RJ. Platelet serotoninmeasures in primary dysthymia. Am J Psychiatry 1994;1519: 1369–1371.

29. Lapierre YD, Ravindran A, Bakish D. Dysthymia andserotonin. Int Clin Psychopharmacol 1993; 8 (Suppl. 2):87–90.

30. Sechter D, Bonin B, Bizouard P. The potential role of5-HT-2 receptor antagonists in the treatment of dysthymia.Encephale 1993; 19: 753–754.

31. Brown GL, Goodwin FK. Cerebrospinal fluid correlates ofsuicide attempts and aggression. Ann NY Acad Sci 1986;487: 175–188.

32. Murphy DL, Pigott TA. OCD and its treatment withserotonin-selective agents. Progress Clin Biol Res 1990; 361:179–201.

33. Goodman WK, McDougle CJ, Price LH. Pharmacotherapyof obsessive-compulsive disorder. J Clin Psychiatry 1992; 53(Suppl. 4): 29–37.

34. Gay C. Biological treatment of dysthymia. Psychol Med1994; 26: 545–548.

35. Lapierre YD. Pharmacological therapy of dysthymia. ActaPsychiatry Scand 1994; 383 (Suppl. 89): 42–48.

36. Weiss MK, Tohen M, Zarate C. Diagnosis and managementof bipolar II disorder. Compr Ther 1994; 20: 121–124.

37. Perugi G, Akiskal HS, Pfanner C, Presta S, Gemignani A,Milanfranchi A, Lensi P, Ravagli S, Cassano GB. Theclinical impact of bipolar and unipolar affective comorbidityon obsessive-compulsive disorder. J Affect Disord 1997; 46:15–23.

38. Go FS, Malley EE, Birmaher B, Rosenberg DR. Manicbehaviors associated with fluoxetine in three 12 to 18 yearolds with obsesssive-compulsive disorder. J Child AdolescPsychopharmacol 1998; 8: 73–80.

39. Berk M, Koopowitz LF, Szabo CP. Antidepressant-inducedmania in obsessive-compulsive disorder. Eur Neuropsy-chopharmacol 1996; 6: 9–11.

74