453-460,463_0602cns_phillips

453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 453 Feature Article
Pharmacologic Treatment of
Body Dysmorphic Disorder:
Review of the Evidence and a
Recommended Treatment Approach
ABSTRACT
and occupational/academic functioning.6-8 Many are house- Research on effective pharmacotherapy for body dysmorphic bound, require psychiatric hospitalization, and attempt sui- disorder (BDD) has rapidly increased in recent years, with cide.6-8 Quality of life is notably poor: one study found that emerging data consistently indicating that serotonin reuptake mental health-related quality of life for BDD patients was inhibitors (SRIs) are often efficacious for this disorder. poorer than for the United States’ population as a whole as Although data are limited, it appears that higher SRI doses well as for patients with depression or a chronic or acute and longer treatment trials than those used for many other psy- medical condition (eg, diabetes or acute myocardial infarc- chiatric disorders are often needed to treat BDD effectively. tion).9 Furthermore, although it is relatively common,10 BDD Approaches to treatment-resistant BDD have received little usually goes undiagnosed in clinical settings.11,12 investigation, but available data indicate that switching toanother SRI and several SRI-augmentation strategies may be WHY TREAT BODY-IMAGE DISTURBANCE
helpful. This article reviews the empirical literature on BDD WITH MEDICATION?
and offers a recommended approach to the pharmacotherapy of There are a number of reasons to think that medication this distressing and often disabling disorder. might be effective for BDD. Obsessional preoccupation with CNS Spectrums 2002;7(6):453-463 perceived appearance flaws and compulsive, repetitivebehaviors, such as mirror-checking, excessive grooming, INTRODUCTION
and skin-picking, are core features of the disorder.6 In some The pharmacotherapy of body dysmorphic disorder (BDD) ways, BDD resembles obsessive-compulsive disorder involves several apparent paradoxes. One paradox is that (OCD)13 and this suggests a role for serotonin reuptake although the empirical literature on this disorder’s pharma- inhibitors (SRIs). Many patients have notable depressive cologic treatment is limited, there appears to be considerable features,6,14 suggesting a role for antidepressants. Insight is consensus on how clinicians should initially treat these usually poor or absent, and nearly half of patients are delu- patients.1 Perhaps a more fundamental paradox involves the sional7 (ie, completely convinced that their view of their very nature of the disorder: BDD is a disorder of disturbed appearance “defect” is accurate), raising the question of body image, which almost certainly has sociological, cul- whether antipsychotics may be useful. In addition, although tural, and psychological roots,2 yet many patients respond to the development of BDD almost certainly involves psycho- logical and sociocultural factors, it also likely involves neu- As recently as the 1980s, patients with BDD were robiological disturbances,2 with very preliminary data described as “extremely difficult” to treat,3 and a noted der- suggesting a role for abnormalities in serotonergic func- matologist stated, “The author knows of no more difficult tion15,16 and corticostriatal circuitry.17 Moreover, other types patients to treat than those with body dysmorphic disorder.”4 of body-image disturbance, such as autotopagnosia (disori- However, as emerging empirical evidence and clinical entation with respect to body surface) or unilateral neglect experience indicate that most patients with BDD may bene- (involving a decreased awareness of body parts) are well fit from pharmacotherapy, this pessimism is gradually being known to have a neurologic basis. While these disturbances do not have direct implications for pharmacotherapy, they It should first be underscored that BDD warrants aggres- highlight the neurobiologic basis of some types of body- sive treatment. Patients with this disorder are markedly dis- tressed, with scores on measures of perceived stress Although most patients with BDD seek often-costly exceeding those reported for most psychiatric patients.5 nonpsychiatric treatment, such as dermatologic treatment Most patients experience significant impairment in social and cosmetic surgery,19 available data indicate that these Dr. Phillips is professor of psychiatry at Butler Hospital and in the Department of Psychiatry at Brown University School of Medicine in Providence,Rhode Island. Please direct all correspondence to: Dr. Phillips, Butler Hospital, 345 Blackstone Blvd., Providence, RI 02906. Tel: 401-455-6490; Fax: 401-455-6539. Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 454 Feature Article
approaches are usually ineffective. In a study of 250 adults improvement in BDD symptoms, whereas 18 non-SRI tri- with BDD seen in a psychiatric setting, who received a total cyclic trials led to no overall improvement in BDD symptoms.
of 484 nonpsychiatric treatments, only 7.3% of all treat- Prospective Clinical Series
ments led to both a decrease in concern with the treated In a recent chart-review study of 90 patients treated by body part and overall improvement in BDD.19 Although the author in her clinical practice, clinically significant prospective data are lacking, these findings suggest that improvement occurred with 63.2% (n=55) of adequate SRI nonpsychiatric treatment should not be recommended for trials, with similar response rates for each type of SRI.34 In these patients. Reports of BDD patients being violent this study, 17.5% (n=10) of adequate SRI trials resulted in toward surgeons20 or committing suicide in a dermatology full remission and 31.7% (n=18) in partial remission.34 setting21,22 underscore the importance of implementing more In a study of 33 children and adolescents with BDD,35 10 of 19 (53%) subjects treated with an SRI had an improve-ment in BDD symptoms on the CGI, and 10 (45%) of 22 EVIDENCE ON THE EFFICACY
SRI trials led to an improvement. Considering only the 13 OF MEDICATION FOR BDD
SRI trials conducted by the authors, which tended to be Serotonin Reuptake Inhibitors
more aggressive trials, eight (62%) resulted in improvement The SRIs (clomipramine and the selective SRIs [SSRIs]) in BDD. No non-SRI medications (n=8 trials) were effective have received the lion’s share of pharmacologic research.
Data from clinical series, open-label studies, and controlled Open-Label Studies
studies consistently indicate that these agents are often, and Three systematic open-label studies of SRIs have been perhaps selectively, effective as single agents for BDD. SRIs done, two with fluvoxamine and one with citalopram. In a are widely used and recommended for BDD.1 Response to 16-week study of 30 subjects who met criteria for the an SRI usually results in decreased distress and time preoc- Diagnostic and Statistical Manual of Mental Disorders, cupied with the “defect,” improved functioning (eg, return Fourth Edition, (DSM-IV) BDD,23 19 (63%) patients to work or school), and improvement in depressive symp- responded (in an intent-to-treat analysis) based on ≥30% toms.23 Repetitive behaviors, such as mirror-checking and improvement on the Yale-Brown Obsessive-Compulsive Scale Modified for BDD (BDD-YBOCS)36 and the CGI.
Case Reports
BDD-YBOCS scores significantly decreased, from 31.1±5.4 Several early case reports suggested efficacy for fluoxe- at baseline to 16.9±11.8 at termination. Six responders dis- tine and clomipramine,24-26 in some cases after failure to continued fluvoxamine, all of whom relapsed, with BDD respond to a variety of psychotropic agents. Efficacy of symptoms significantly improving in the four patients who SSRIs in children and adolescents was subsequently restarted an SRI. In a 10-week open-label study of fluvox- reported,27-30 with relapses occurring after medication dis- amine up to 300 mg/day,37 10 of 15 patients were much or continuation.27 Intravenous, pulse-loaded clomipramine was very much improved on the CGI. Ten of 12 patients who also noted to be efficacious in two cases.31 completed the study were responders. In a recent open- Retrospective Studies
label study of citalopram in 15 BDD patients, BDD-YBOCS Largely retrospective data from larger patient series sug- scores decreased from 30.7±4.9 at baseline to 15.3±10.6 at gest that SRIs, but not other medications, are often effective termination (P<.001), with 73.3% (n=11) of subjects rated for BDD. In an early series of 30 patients, 58% responded as responders on the BDD-YBOCS and the CGI (KA to SRIs, whereas only 5% responded to other medications.6 Phillips, MD, and F Najjar MD, unpublished data, 2001). In an expansion of this series, consisting of 130 patients Controlled Studies
who had received a total of 316 medication trials, 42% of 65 Two controlled-treatment trials of BDD have been com- SRI trials led to improvement on the Clinical Global pleted. In the first, a double-blind crossover study of Impressions (CGI) scale, compared to 30% of 23 trials with clomipramine versus desipramine conducted by Hollander monoamine oxidase inhibitors (MAOIs), 15% of 48 trials and colleagues,38 subjects were treated for 8 weeks with with non-SRI tricyclics, 3% with neuroleptics, 6% with a each medication. Forty patients entered the study and 29 variety of other medications (eg, benzodiazepines and mood were randomized. Clomipramine was superior to stabilizers), and 0% of electroconvulsive therapy (ECT) tri- desipramine in improving BDD symptoms and functional als.32 (Although in this series the SRI response rate was disability. Treatment efficacy was independent of the pres- higher than for the other medications, it was relatively low ence or severity of comorbid diagnoses of OCD, depression, compared with subsequent rates reported later in this arti- or social phobia. This study is important because, consistent cle. This probably reflects the fact that at this early stage of with the aforementioned retrospective data, it suggests that BDD research, it was not known what an adequate SRI trial BDD should be treated with an SRI rather than a non-SRI consisted of, and many patients received SRI doses and trial durations that were probably inadequate for treating BDD).
The first placebo-controlled study was recently com- Another retrospective study of 50 patients by Hollander and pleted. It found that fluoxetine was more effective than colleagues33 similarly reported that 35 SRI trials resulted in placebo for BDD.39 Seventy-four patients with DSM-IV BDD Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 455 Feature Article
criteria were enrolled and 67 were randomized to 12 weeks research on this medication’s efficacy is clearly warranted.
of treatment in this double-blind, parallel-group study.
MAOIs, too, may have a role in treating BDD. In the author’s Fluoxetine was significantly more effective than placebo for aforementioned retrospective series, MAOIs were effective BDD beginning at week 8 and continuing at weeks 10 and in 30% of 23 cases, in some cases after patients failed an 12 (<.001). The response rate was 53% (18 of 34) to flu- SRI.7 Several case reports have noted a good response to oxetine and 18% (6 of 33) to placebo (P=.003). As in the MAOIs, one with tranylcypromine and one with a combina- clomipramine/desipramine study, treatment efficacy was tion of phenelzine, trimipramine, and perphenazine.42 independent of the presence of major depression or OCD.
Tricyclic antidepressants other than clomipramine Efficacy of SRIs for Delusional BDD
appear generally ineffective for BDD. The Perhaps the most intriguing finding in the pharmacother- clomipramine/desipramine trial described above found that apy literature is that patients with delusional BDD appear to desipramine was significantly less effective than respond to SRIs.7,34,35,38-40 This finding is counterintuitive, clomipramine for BDD.38 Because the study did not include given that these patients are diagnosed not only with BDD a placebo control, it cannot be stated with certainty that but also with a psychotic disorder (delusional disorder of desipramine is ineffective for BDD (it is possible that the somatic type), a group of disorders usually treated placebo would have produced an even lower response rate).
with antipsychotics. In the previously noted placebo-con- However, other published data are consistent with a com- trolled fluoxetine study, a similar percentage of delusional parative lack of efficacy for tricyclics other than and nondelusional patients had response of BDD symp- clomipramine. A previously noted retrospective study of toms to fluoxetine (50% and 55%, respectively).39 In the 50 patients found that non-SRI tricyclic trials led to no desipramine/clomipramine crossover study, clomipramine overall improvement in BDD symptoms,33 and the author’s was more effective than desipramine regardless of whether previously noted retrospective series found that only 15% of patients had insight or held their dysmorphic misperception 48 trials with non-SRI tricyclics improved BDD symptoms.32 with delusional intensity.38 Of interest, clomipramine was There are no data available on the efficacy of mirtazapine even more effective for delusional patients than for nondelu- and bupropion for BDD, and only very limited data on nefa- sional patients. The results from these controlled studies are zodone.34 In the author’s experience, these medications are consistent with earlier findings from case reports,24,25,29 clini- generally ineffective as single agents for BDD.
cal series,7,35 and open-label trials,40 which also indicatedthat delusional patients had a high rate of response to SRIs.
Antipsychotics as Single Agents
Several of these studies39,40 had the methodologic advantage Given that a high percentage of patients with BDD are of using a reliable and valid scale41 to classify subjects prior delusional, an important question is whether antipsychotics to treatment as delusional or nondelusional.
are effective as single agents for BDD, including its delu- Several studies also examined the different question of sional form. This question has received surprisingly little whether insight improves with treatment. In other words, do empirical investigation. In early case reports, antipsychotics treated patients develop more accurate beliefs about their were generally reported to be ineffective, with negative appearance, realizing that they do not look abnormal? In an results for loxitane, trifluoperazine, thioridazine, flu- open-label fluvoxamine study, insight significantly penthixol, pimozide, and unspecified agents.42 The author’s improved.40 In the blinded crossover study, clomipramine previously noted retrospectively assessed series, in which but not desipramine resulted in significant improvement in 52% of patients had been delusional, showed a response in insight.38 In the fluoxetine study, insight did not improve only 1 of 49 trials.7 A recent case report, however, noted more with fluoxetine than with placebo, although insight improvement in BDD symptoms with olanzapine.43 improved significantly more in treatment responders (to either Pimozide has had the reputation of being effective for fluoxetine or placebo) than in treatment nonresponders.39 monosymptomatic hypochondriacal psychosis, a broad cate- These intriguing findings indicate that SRIs are effective gory currently known as delusional disorder of the somatic for BDD symptoms in delusional patients. They also suggest type, which includes delusional BDD.3 However, data from that insight may improve with SRI treatment, although the only a small number of case reports are available for data are somewhat contradictory. These data further suggest patients clearly diagnosed with delusional BDD as opposed that certain types of psychosis (perhaps the delusional vari- to another type of somatic delusional disorder. These reports ants of putative OCD-spectrum disorders) may respond to suggest that pimozide may be effective for BDD, although the author has found pimozide alone to be ineffective in asmall number of cases (n=8).32 The efficacy of this medica- Other Antidepressants
tion warrants further, more rigorous investigation.
Venlafaxine, a mixed serotonin-norepinephrine reuptake inhibitor, may have a role in treating BDD. Although there Other Medications as Single Agents
are no published reports on this medication’s efficacy, in the Data are very limited but generally suggest that other author’s clinical experience, BDD may respond to venlafax- medications are generally ineffective as single agents for ine, even after the failure of numerous SRIs. More systematic BDD (although they may have a role as SRI-augmentation Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 456 Feature Article
agents, as discussed in a later section). Early case reports none of which showed ECT efficacy for BDD, although sev- noted a lack of response for lithium, alprazolam, diazepam, eral patients had a transient improvement, primarily in and unspecified benzodiazepines.42 The author’s previously noted retrospective study similarly found that a variety ofother medications (eg, mood stabilizers, benzodiazepines, Psychosurgery
stimulants), used alone, were usually ineffective for BDD.32 One published case report noted improvement in BDD In the previously described series of 33 children and ado- symptoms with a modified leucotomy.45 In another, signifi- lescents with BDD,35 none of the eight treatments with non- cant improvement occurred with a capsulotomy (P Mindus, SRI medications significantly decreased BDD symptoms.
MD, oral communication, 2001), and in a third case, with abilateral anterior cingulotomy and subcaudate tractotomy (E OTHER SOMATIC TREATMENTS
Cassem, MD, personal communication, 2001). However, intwo cases, an anterior capsulotomy was ineffective for BDD Electroconvulsive Therapy
(SA Rasmussen, oral communication, 2001).
There are only limited data on the efficacy of ECT. A 1991 review found seven published case reports, six of SEROTONIN REUPTAKE INHIBITOR
which reported an unsuccessful outcome with ECT.42 One AUGMENTATION STRATEGIES
subsequent report noted a good response to ECT.44 In the SRI augmentation with pharmacotherapy has consider- author’s series of 130 cases, none of eight ECT trials was able face validity. Augmentation strategies are widely used successful, although the data were largely retrospective.32 in other disorders, including OCD and depression, disorders The author is aware of approximately 10 additional cases, with similiarites to BDD. Augmentation has a particular TABLE. SUMMARY OF CONTROLLED AND OPEN-LABEL PHARMACOTHERAPY STUDIES IN BDD*
BDD-YBOCS‡); effect size: d=0.70 disability (response rate of 65% vs 35% on BDD-YBOCS§) quality of life and functioning also significantly improved BDD-YBOCS=the Yale-Brown Obsessive-Compulsive Scale Modified for Body Dysmorphic Disorder; CMI=clomipramine; DMI=desipramine; CGI=Clinical GlobalImpressions scale.
*Case reports, case series, and retrospective studies are not included in the table but are described in the text.
†Results are reported for an intent-to-treat analysis for all studies except for the clomipramine/desipramine trial, which used a minimum treatment analysis.
‡Response was defined as 30% or greater decrease in total BDD-YBOCS score; the BDD-YBOCS assesses BDD severity during the past week based on preoc-
cupation with the perceived defect (time occupied, interference with functioning due to the preoccupation, distress, resistance, and control), associated repet-
itive behaviors, such as mirror-checking (time spent, interference with functioning, distress if the behaviors are prevented, resistance, and control), insight,
and avoidance.
§Response was defined as ≥25% decrease in total BDD-YBOCS score.
\\ KA Phillips KA,MD, and F Najjar, MD, unpublished data, 2001.
Phillips KA. CNS Spectrums. Vol 7, No 6. 2002.
Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 457 Feature Article
appeal for patients who have partially responded to an SRI co-occurring depressive symptoms, but less effective for and for whom relapse—which could occur with a switch to BDD symptoms. Venlafaxine augmentation of selective another SRI—could be risky (eg, leading to increased sui- SRIs (SSRIs) also has some promise. These approaches cide risk). SRI-augmentation strategies may in fact be more effective for patients who have had a partial SRI response,as opposed to no SRI response, with a chart-review study A RECOMMENDED PHARMACOTHERAPEUTIC
reporting augmentation response rates of 40.5% (n=15) for APPROACH TO THE PATIENT WITH BDD
partial SRI responders versus only 18.2% for SRI nonre- The following recommendations are based on available data and are also informed by the author’s clinical experi-ence. Empirical evidence on effective pharmacotherapy for Buspirone
BDD, while dramatically increasing, is still limited. This is A rationale for using buspirone is that it is a partial ago- especially the case for augmentation strategies and treat- receptors, and it has been shown to effec- ment-resistant BDD. It should be kept in mind that the fol- tively augment SRIs in depression46 and in some studies of lowing recommendations are general suggestions that may OCD.47 In a chart-review study of BDD patients in which require modification to appropriately treat an individual buspirone was added to an SRI after an adequate trial, patient. Comorbidity, for example, may indicate the need for 33.3% (n=12) of trials were successful, with a large effect additional treatment, as in the case of type I bipolar disorder size.34 The mean buspirone dose was 56.5±15.2 mg/day.
requiring a concomitant mood stabilizer. For certain Buspirone augmentation of SRIs appears as effective for patients with substance abuse or dependence, it may be unwise to use a benzodiazepine or a stimulant. A proposedalgorithm for the pharmacotherapy of BDD, which is consis- Clomipramine Plus a Selective Serotonin
tent with the following recommendations, has been pub- Reuptake Inhibitor
In the previously noted chart-review study, the response 1) Recognize and diagnose BDD.
rate to clomipramine augmentation of an SSRI or vice versa This is the first essential step to effective treatment.
was 44.4% (n=4). This was higher than for the other aug- While this point may appear self-evident, studies indicate mentation strategies.34 The effect size was small to medium, that although BDD is a relatively common disorder,10,11 it is which was smaller than for the other augmentation strate- nearly always missed in clinical practice, in both inpatient gies, although the power of the analysis was limited. and outpatient settings.11,12 The diagnosis can be made usingrelatively straightforward questions which establish that the Antipsychotics
patient is preoccupied with nonexistent or minimal appear- While antipsychotics alone do not appear effective for ance flaws, which cause clinically significant distress or BDD (although data are limited to retrospective findings), antipsychotic augmentation of SRIs has considerable face 2) Provide psychoeducation about BDD.
validity, given that many patients have prominent delusions Providing education about BDD is an important next step of reference and delusional conviction about the perceived that often facilitates engagement of the patient in treatment.
appearance defect. In addition, this approach has been Web sites51 and books2 for the layperson may be helpful, shown to be efficacious in OCD.49 The only data on this especially because these patients may be reluctant to accept strategy come from the aforementioned chart-review study, psychiatric care (because of their poor insight, desire for in which the response rate to antipsychotic augmentation nonpsychiatric treatment such as surgery, or for other rea- was only 15.4% (n=2); however, the effect size for atypical sons). Other general aspects of treatment, such as the chal- lenge of engaging delusional patients in treatment, involvingfamily members in treatment, and collaborating with derma- Other Agents
tologists or surgeons, are reviewed elsewhere.52,55 The previously noted chart-review study provides the 3) Use an SRI as a first-line approach, including
only data on the use of other agents as SRI augmenters, in for delusional patients.
which the response rate to lithium augmentation of SRIs The aforementioned data indicate that SRIs are effective was 20% (n=1), with a large effect size, and the response for a majority of patients, including those who are delusional.
rate to methylphenidate augmentation was 16.7% (n=1), Fluoxetine, clomipramine, fluvoxamine, and citalopram have also with a large effect size.34 In addition to the notable received the most investigation, but sertraline and paroxetine improvement in BDD symptoms occasionally seen with appear effective. Although no controlled studies have these approaches, symptoms of depression or anergia may directly compared one SRI with another, very preliminary improve. This may in turn make BDD symptoms easier to data suggest that all SRIs have approximately equivalent tolerate. There are no published reports on the efficacy efficacy.34 Most clinicians would begin treatment with an of bupropion or mirtazapine augmentation of SRIs, but in SSRI rather than clomipramine, given the latter’s greater the author’s experience they may sometimes be helpful for propensity to cause side effects and toxicity in overdose. Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 458 Feature Article
4) Use the maximum SRI dose recommended or
have an even higher success rate, longer durations have not tolerated unless a lower dose is effective.
been studied. In the author’s clinical experience, however, Although dose-finding studies have not been done, BDD onset of efficacy after 16 weeks appears unlikely. It is there- appears to often require higher SRI doses than those typi- fore recommended that the medication be changed at 12–16 cally used for depression. Doses reported in most published weeks if an adequate dose has been reached and the studies have been fairly high (eg, 77.7±8.0 mg/day in the response is inadequate at that time.
fluoxetine study39 and 238.3±85.8 mg/day in a fluvoxamine 6) Continue an effective SRI for at least 1 year, if not
study23), but these studies used a forced-titration schedule with a relatively high target dose. Nonetheless, relatively No continuation, maintenance, or discontinuation studies high mean doses have also been used in the author’s clinical have been done in BDD, so empirical data that might guide practice, despite the fact that an attempt was often made to longer-term treatment are lacking. Nonetheless, in find the lowest effective dose: 66.7±23.5 mg/day for fluoxe- the author’s clinical experience, relapse with continued tine; 308.3±49.2 mg/day for fluvoxamine; 55.0±12.9 treatment is rare, and many patients who respond by mg/day for paroxetine; 202.1±45.8 mg/day for sertraline; 12–16 weeks appear to experience further gradual improve- and 203.3±52.5 mg/day for clomipramine.34 ment with continued pharmacotherapy. It is not clear Clinical experience indicates that some patients warrant whether further improvement is due to a direct effect of the and benefit from SSRI doses that exceed the maximum rec- medication, or to behavioral changes (eg, less social avoid- ommended dose—for example, 400 mg/day of fluvoxamine, ance) made possible by medication response, which in turn 80–100 mg/day of citalopram, or 100 mg/day of paroxetine.
facilitate further improvement. This phenomenon appears to These higher doses are best suited to patients with a partial be a continuation of an earlier response, rather than appear- response to the highest recommended dose who are tolerat- ing the medication well, and patients who have failed many The only data on relapse (from a chart-review study) SSRIs. Clomipramine doses, however, should not exceed indicate that relapse is likely with SRI discontinuation, occurring in 84% of patients.34 It is unknown whether the The rapidity of dose titration will depend on a number of relapse rate differs following different treatment durations.
factors, including illness severity (with quicker titration The recommendation to continue an effective SRI for at generally advisable for sicker patients, especially those who least 1 year is somewhat arbitrary, especially because the are suicidal), medication tolerability, and patient prefer- optimal treatment duration has not been systematically ence. Generally, however, it is advisable to reach the maxi- studied. It would seem wise to tailor the duration of treat- mum recommended medication dose by weeks 5–9 of ment to the individual patient. For severely ill patients (eg, treatment, if tolerated. For example, in the fluoxetine study, those with numerous, potentially lethal suicide attempts due patients received 20 mg/day for 2 weeks, with the dose to BDD), it is probably best to continue the medication for raised by 20 mg/day every 10 days to a maximum dose of life. Patients who have had several relapses when discontin- 80 mg/day, if tolerated (no patients dropped out of this study uing the medication may also need longer-term pharma- because of medication side effects).39 In the citalopram cotherapy. In the author’s experience, discontinued study, 20 mg/day was initially prescribed, with the dose medication that was previously effective is usually effective raised to 40 mg/day after 2 weeks and 60 mg/day after with reinitiation, although occasional patients do not have 4 weeks (KA Phillips, MD, and F Najjar, MD, unpublished as robust a response as with its initial use.
data, 2001). This approach may result in treating with a 7) If an effective SRI is discontinued, discon-
higher SRI dose than is necessary, but it has the advantage tinue it carefully.
of not missing a response because of undertreatment, and it Clinical experience suggests that if a decision is made to also prevents an unnecessarily protracted-treatment trial discontinue effective medication, it should be slowly that could result from slow titration.
tapered (eg, over months) rather than suddenly discontin- 5) Treat with an SRI for 12–16 weeks before
ued. Because the risk of relapse appears high, discontinua- making a final determination of efficacy.
tion should ideally be done at a time when the patient is not Studies that have assessed time to SRI response have highly stressed and is able to tolerate a relapse if it occurs.
reported a mean time of 6–9 weeks and as long as 16 Whether concomitant cognitive-behavior therapy reduces weeks.23,33,35,39 The only exception is a recently completed the risk of relapse with medication discontinuation is study of citalopram that reported a mean time to response of only 4.6±2.6 weeks (KA Phillips, MD, and F Najjar, MD, 8) Do not simply treat depression.
unpublished data, 2001). In a fluvoxamine study, 6 (32%) of A majority of patients with BDD have major depression, the 19 eventual responders had not responded by week 8, and many additional patients have depressive symptoms although most (94%) responded by 12 weeks.23 Most studies that do not qualify for a diagnosis of major depression.14 A used a fairly rapid titration schedule, so an even longer time depressive disorder diagnosis and depressive symptoms to response might be expected with a slower dose increase.
often appear secondary to BDD (although this does not While it is possible that trials longer than 16 weeks would always seem to be the case). In the author’s experience, Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 459 Feature Article
focusing treatment on depression but not BDD is a common appealing, however, because the medication is usually so clinical error. An effective regimen for depression will not well tolerated and relatively easy to prescribe. It may also necessarily effectively treat BDD; however, an effective reg- improve concomitant anxiety and possibly depressive symp- imen for BDD will usually effectively treat depression.23,38,39 toms. Doses of 30–60 mg/day are recommended, and some As previously noted, although data are very limited, non- patients benefit from an even higher dose (eg, 90 mg/day).48 SRI antidepressants (with the possible exception of ven- As previously noted, clomipramine may also be effective lafaxine) appear ineffective as single agents for BDD.
and may be the preferred strategy for patients with severe Non-SRI antidepressants also appear relatively ineffective depression comorbid with severe BDD. However, SSRI aug- for accompanying depressive symptoms, as indicated by a mentation with clomipramine is riskier than with other aug- study by Hollander and colleagues38 in which depressive mentation approaches because SSRIs have the potential to symptoms were reduced more by clomipramine than by unpredictably and sometimes dramatically increase blood desipramine. This was also the case for subjects with a levels of clomipramine, which has a low therapeutic index.
comorbid diagnosis of depression. This finding suggests that Although this approach is generally well tolerated, patients BDD is not simply a symptom of depression; if it were, should be followed closely for signs of toxicity. Clomipramine desipramine should have been as effective as clomipramine.
levels should be monitored (the author’s practice is to check Another reason to not simply treat depression is that longer a clomipramine level at a dose of 25 mg/day or 50 mg/day treatment trials and higher SRI doses than those usually after adding it to an SSRI). Clomipramine and an SSRI gen- used for depression are often needed to treat BDD symp- erally should not be combined without first attempting to toms. In addition, several23,37 (although not all39), studies have found that relatively long trials may also be needed to Antipsychotic augmentation of an SRI appears less often treat depressive symptoms in BDD patients (7.0±4.6 weeks effective than buspirone or clomipramine augmentation, but in one study53 and more than 6 weeks in another.37 the evidence is too limited at this time to draw any firm con- 9) Try sequential SRIs if necessary.
clusions about the comparative efficacy of these Although all SRIs appear effective, one may be more approaches. This approach would seem reasonable for delu- effective than another for an individual patient. If one SRI sional patients who fail an SRI or for severely ill delusional fails, another should be tried, as some patients will respond.
patients as initial treatment in combination with an SRI. It The author has treated a number of patients who responded is unknown whether atypical antipsychotics are more effec- to an SRI after failing all other SRIs, venlafaxine, and tive than typical antipsychotics, including pimozide, but numerous augmentation strategies. In the only study to they are likely to be better tolerated. It is possible that neu- examine this question, of those subjects who failed an initial roleptic augmentation is more effective for patients who per- adequate SRI trial, 42.9% (n=6) responded to at least one form tic-like behaviors, such as skin-picking or hair- subsequent adequate SRI trial, and 43.5% (n=10) of subse- plucking, though this question has not been investigated.
quent adequate SRI trials received by these subjects were Adding lithium or a stimulant appears more effective for effective.34 Among responders to an initial SRI who were depressive than BDD symptoms, but diminishing depres- subsequently treated with a different SRI, 92.3% (n=12) of sive symptoms is clearly beneficial and may also help subsequent trials also led to improvement.34 patients better tolerate BDD symptoms. Although it is theo- 10) Venlafaxine or an MAOI may be worth trying.
retically possible that skin-picking, a tic-like behavior, In the author’s experience, venlafaxine may be effective might worsen with stimulant treatment, the author has not for BDD. Although data on MAOIs are very limited, these agents appear less effective than SRIs but perhaps more What constitutes an adequate augmentation trial effective than other antidepressants. Because of the risk of is unclear, but a duration of 6–8 weeks is probably ade- side effects and hypertensive crisis, it would seem best to quate to determine whether the approach will be effec- restrict MAOI use to patients who have failed adequate tri- tive. It is probably best to use a 12-week trial for als of all of the SRIs and venlafaxine.
11) Avoid using antipsychotics as monotherapy,
13) Is it better to augment an SRI or switch to
even for delusional patients.
another SRI?
As discussed in an earlier section, antipsychotics appear The answer to this question is unclear. The only study to ineffective as single agents, even for delusional BDD date that sheds any light on this issue found that in patients patients. SRIs alone appear effective for these patients. This who had failed an adequate SRI trial, 43.5% (n=10) of sub- important issue, however, requires further investigation.
sequent adequate SRI trials were effective versus 33.3% 12) Consider buspirone, clomipramine, an
(n=8) of various SRI-augmentation strategies.34 However, antipsychotic, or other SRI-augmentation
40.5% (n=15) of augmentation trials were successful in par- strategies.
tial SRI responders.34 The relatively small number of trials Regarding the choice of an augmentation agent, no and the study’s methodological limitations preclude drawing methodologically rigorous studies have compared one aug- a firm conclusion about which approach is more effective.
mentation agent to another. Buspirone augmentation is From a clinical perspective, if a patient has failed three SRI Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 460 Feature Article
trials without any attempt at augmentation, it would seem ment response can be optimized. Equally compelling is the that augmentation should be tried. Conversely, if a patient need to evaluate cognitive-behavior therapy as an SRI “aug- has failed several augmentation strategies with one SRI, a mentation” strategy. Another important avenue of research switch to another SRI should be considered. Numerous fac- is to compare the efficacy of cognitive-behavior therapy ver- tors need to be considered, however. For example, the better sus SRIs versus their combination to determine which treat- the SRI response, the less appealing it would be to discon- ments are most effective for which patients.
tinue it and try another SRI. Augmentation may also be The longest systematic pharmacotherapy study to date more appealing if there is a high risk of serious conse- was only 16 weeks in duration, making continuation and quences due to relapse, which could occur when switching maintenance studies necessary to confirm clinical impres- sions that SRI response is generally maintained, and may 14) Consider benzodiazepine use for agitated or
even further improve, with longer-term treatment. Because anxious patients.
BDD can be so severely distressing and impairing, the like- Because patients with BDD can be extremely distressed, lihood of relapse with medication discontinuation is a par- anxious, and agitated, adjunctive benzodiazepine use ticularly important question; the only published data on this should be considered. This approach may be used for a issue come from a single chart-review study, underscoring short duration (eg, until an SRI begins to work) or for the the need for rigorous placebo-controlled, relapse-prevention longer term if clinically indicated. Benzodiazepines may studies. Once studies such as the above more firmly estab- enable treatment with an SRI if an SRI causes agitation. A lish which pharmacotherapeutic approaches are efficacious benzodiazepine, by alleviating anxiety and insomnia, can for BDD, effectiveness studies are needed that evaluate the make BDD symptoms more tolerable. It is also possible that usefulness of medications under “real world” conditions (eg, benzodiazepines may have a more direct effect on BDD for patients in community settings who are not excluded symptoms, although in the author’s experience, this effect, if because of comorbid conditions, such as substance abuse).
it occurs at all, tends to be minimal. Clonazepam, which Finally, elucidation of BDD’s underlying neurobiology, may have serotonergic properties54 and possibly be effective which to date has received virtually no investigation, may for OCD,1 may be preferable to other benzodiazepines, provide fruitful leads for new pharmacologic approaches to although this question has not been studied. The potential for substance abuse or dependence must be considered,although in the author’s clinical experience, relatively few CONCLUSION
As recently as a decade ago, virtually nothing was known 15) Consider cognitive-behavior therapy.
about effective pharmacotherapy for BDD. Pharmacologic Although the efficacy of cognitive-behavior therapy is approaches for BDD have received far less investigation beyond the scope of this review, it is worth noting that this than for many other common and severe mental disorders, treatment approach should also be considered.55 There are and much remains to be learned. At the same time, our no data available to address the question of whether phar- knowledge of effective pharmacologic treatment for BDD macotherapy or cognitive-behavior therapy are more effec- has dramatically increased to the point where a majority of tive, or whether combining these treatments is more patients with this relatively common disorder may obtain effective than using either one alone. In the author’s experi- relief from their distressing and often disabling illness. CNS
ence, these approaches are complementary. For example,partial response to an SRI can make it possible for a REFERENCES
severely depressed, delusional, or suicidal patient to engage 1. March JS, Frances A, Carpenter D, et al. The expert consensus guideline series: in and tolerate cognitive-behavior therapy. In the author’s treatment of obsessive-compulsive disorder. J Clin Psychiatry. 1997;58(suppl opinion, medication is always indicated for severely ill, 2. Phillips KA. The Broken Mirror: Understanding and Treating Body Dysmorphic severely depressed, or highly suicidal patients.
Disorder. New York, NY: Oxford University Press; 1996.
3. Munro A, Chmara J. Monosymptomatic hypochondriacal psychosis: a diagnostic FUTURE PHARMACOTHERAPY
checklist based on 50 cases of the disorder. Can J Psychiatry. 1982;27:374-376.
RESEARCH NEEDS
4. Cotterill JA. Body dysmorphic disorder. Dermatol Clin. 1996;14:457-463.
Virtually all aspects of pharmacotherapy for BDD need to 5. DeMarco LM, Li LC, Phillips KA, et al. Perceived stress in body dysmorphic be studied, as research in this area is still in its early stages.
disorder. J Nerv Ment Dis. 1998;186:724-726.
One of the most pressing needs is for additional placebo- 6. Phillips KA, McElroy SL, Keck PE Jr, et al. Body dysmorphic disorder: 30 cases of imagined ugliness. Am J Psychiatry. 1993;150:302-308.
controlled SRI trials, including trials that compare SRIs to 7. Phillips KA, McElroy SL, Keck PE Jr, et al. A comparison of delusional and other medication classes, such as non-SRI antidepressants, nondelusional body dysmorphic disorder in 100 cases. Psychopharmacol Bull. to determine whether SRIs are truly preferentially effica- cious for BDD. Because patients typically experience a par- 8. Phillips KA. Body dysmorphic disorder. In: Phillips KA, ed. Somatoform and tial response, rather than remission, with SRIs, it is also Factitious Disorders. Washington DC: American Psychiatric Publishing; 2001.
critically important to evaluate pharmacologic augmentation Oldham JM, Riba MB, series eds. Review of Psychiatry Series, Vol 20, No. 3. strategies in a methodologically rigorous way, so that treat- 9. Phillips KA. Quality of life for patients with body dysmorphic disorder. J Nerv Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S
453-460,463_0602CNS_PHILLIPS 4/20/09 4:19 PM Page 463 Feature Article
10. Bienvenu OJ, Samuels JF, Riddle MA, et al. The relationship of obsessive-com- 37. Perugi G, Giannotti D, Di Vaio S, et al. Fluvoxamine in the treatment of body pulsive disorder to possible spectrum disorders: results from a family study. Biol dysmorphic disorder (dysmorphophobia). Int Clin Psychopharmacol. 11. Phillips KA, Nierenberg AA, Brendel, G, et al. Prevalence and clinical features of 38. Hollander E, Allen A, Kwon J, et al. Clomipramine vs desipramine crossover body dysmorphic disorder in atypical major depression. J Nerv Ment Dis. trial in body dysmorphic disorder: selective efficacy of a serotonin reuptake inhibitor in imagined ugliness. Arch Gen Psychiatry. 1999;56:1033-1039.
12. Grant JE, Won Kim S, Crow SJ. Prevalence and clinical features of body 39. Phillips KA, Albertini RS, Rasmussen SA. A randomized placebo-controlled dysmorphic disorder in adolescent and adult psychiatric inpatients. J Clin trial of fluoxetine in body dysmorphic disorder. Arch Gen Psychiatry. 13. McElroy SL, Phillips KA, Keck PE Jr. Obsessive-compulsive spectrum disor- 40. Phillips KA, McElroy SL, Dwight MM, Eisen JL, et al. Delusionality and ders. J Clin Psychiatry. 1994;55(suppl):33-51.
response to open-label fluvoxamine in body dysmorphic disorder. J Clin 14. Phillips KA. Body dysmorphic disorder and depression: theoretical considera- tions and treatment strategies. Psychiatr Q. 1999;70:313-331.
41. Eisen JL, Phillips KA, Baer L, et al. The Brown Assessment of Beliefs Scale: 15. Barr LC, Goodman WK, Price LH. Acute exacerbation of body dysmorphic disor- reliability and validity. Am J Psychiatry. 1998;155:102-108.
der during tryptophan depletion [letter]. Am J Psychiatry. 1992;149:1406-1407.
42. Phillips KA. Body dysmorphic disorder: the distress of imagined ugliness. Am J 16. Craven JL, Rodin GM. Cyproheptadine dependence associated with an atypical Psychiatry. 1991;148:1138-1149.
somatoform disorder. Can J Psychiatry. 1987;32:143-145.
43. Grant JE. Successful treatment of nondelusional body dysmorphic disorder with 17. Deckersbach T, Savage CR, Phillips KA, et al. Characteristics of memory olanzapine: a case report. J Clin Psychiatry. 2001;62:297-298.
dysfunction in body dysmorphic disorder. J Int Neuropsychol Soc. 44. Carroll BJ. Response of major depression with psychosis and body dysmorphic disorder to ECT [letter]. Am J Psychiatry. 1994;151:288-289.
18. Ames FR. The Neurological Basis of Body Image. In: Castle DJ, Phillips 45. Hay GG. Dysmorphophobia. Br J Psychiatry. 1970;116:399-406.
KA, eds. Disorders of Body Image. Hampshire, England: Wrightson 46. Joffe RT, Schuller, DR. An open study of buspirone augmentation of sero- tonin reuptake inhibitors in refractory depression. J Clin Psychiatry. 19. Phillips KA, Grant JD, Siniscalchi J, et al. Surgical and nonpsychiatric medical treatment of patients with body dysmorphic disorder. Psychosomatics, 47. Jenike MA, Baer L, Buttolph L. Buspirone augmentation of fluoxetine in patients with obsessive-compulsive disorder. J Clin Psychiatry. 1991;52:13-16.
20. Phillips KA, McElroy SL, Lion JR. Body dysmorphic disorder in cosmetic 48. Phillips KA. An open study of buspirone augmentation of serotonin-reup- surgery patients [letter]. J Plastic Reconst Surgery. 1992;90:333-334.
take inhibitors in body dysmorphic disorder. Psychopharmacol Bull. 21. Cotterill JA. Dermatological non-disease: a common and potentially fatal distur- bance of cutaneous body image. Br J Dermatol. 1981;104:611-619.
49. McDougle CJ, Goodman WK, Price LH, et al. Neuroleptic addition in flu- 22. Cotterill JA, Cunliffe WJ. Suicide in dermatological patients. Br J Dermatol. voxamine-refractory obsessive-compulsive disorder. Am J Psychiatry. 23. Phillips KA, Dwight MM, McElroy SL. Efficacy and safety of fluvoxamine in 50. Phillips KA. Pharmacologic treatment of body dysmorphic disorder: a review of body dysmorphic disorder. J Clin Psychiatry. 1998;59:165-171.
empirical data and a proposed treatment algorithm. Psychiatr Clin North Am. 24. Sondheimer A. Clomipramine treatment of delusional disorder, somatic type. J Am Acad Child Adolesc Psychiatry. 1988;27:188-192.
51. Body Dysmorphic Disorder (BDD) and the body image program. Butler Hospital 25. Hollander E, Liebowitz MR, Winchel R, et al. Treatment of body-dysmorphic Web site. Available at: http://www.butler.org/bdd. Accessed June 1, 1999.
disorder with serotonin reuptake blockers. Am J Psychiatry. 1989;146:768-770.
52. Phillips KA. Body dysmorphic disorder: diagnostic controversies and treatment 26. Brady KT, Austin L, Lydiard RB. Body dysmorphic disorder: the relationship to challenges. Bull Menninger Clin. 2000;64:18-35.
obsessive-compulsive disorder. J Nerv Ment Dis. 1990;178:538-540.
53. Phillips KA, McElroy SL. Treatment response of depression in patients with 27. Phillips KA, Atala KD, Albertini RS. Body dysmorphic disorder in adolescents.
body dysmorphic disorder. Abstract presented at: New Clinical Drug Evaluation J Am Acad Child Adolesc Psychiatry. 1995;34:1216-122.
Unit Program (NCDEU) 38th Annual Meeting Abstracts of Poster Presentations; 28. Albertini R, Phillips KA, Guvremont D. Body dysmorphic disorder in a young Boca Raton, Fl: National Institute of Mental Health, June 1998:146.
child [letter]. J Am Acad Child Adolesc Psychiatry. 1996;35:1425-1426.
54. Wagner HR, Reches A, Yablonskaya E, et al. Clonazepam-induced up- 29. El-Khatib HE, Dickey TO. Sertraline for body dysmorphic disorder [letter]. J Am regulation of serotonin-1 and serotonin-2 binding sites in rat frontal cor- Acad Child Adolesc Psychiatry. 1995;34:1404-1405.
tex. Adv Neurol. 1986;43:645-651.
30. Heimann SW. SSRI for body dysmorphic disorder [letter]. J Am Acad Child 55. Phillips KA, Castle DJ. Body dysmorphic disorder. In: Castle DJ, Phillips Adolesc Psychiatry. 1997;36:868.
KA, eds. Disorders of Body Image. Hampshire, England: Wrightson 31. Pallanti S, Koran LM. Intravenous, pulse-loaded clomipramine in body dysmor- phic disorder: two case reports. CNS Spectrums. 1996;1:54-57.
32. Phillips KA. Pharmacologic treatment of body dysmorphic disorder.
Psychopharmacol Bull. 1996;32:597-605.
33. Hollander E, Cohen L, Simeon D, et al. Fluvoxamine treatment of body dysmor- phic disorder [letter]. J Clin Psychopharmacol. 1994;14:75-77.
34. Phillips KA, Albertini RS, Siniscalchi J, et al. Effectiveness of pharmacotherapy for body dysmorphic disorder: a chart-review study. J Clin Psychiatry.
2001;62:721-727.
35. Albertini RS, Phillips KA. 33 cases of body dysmorphic disorder in children and adolescents. J Am Acad Child Adolesc Psychiatry. 1999;38:453-459.
36. Phillips KA, Hollander E, Rasmussen SA, et al. A severity rating scale for body dysmorphic disorder: development, reliability, and validity of a modified versionof the Yale-Brown Obsessive-Compulsive Scale. Psychopharmacol Bull.
1997;33:17-22.
Volume 7 – Number 6 • June 2002 MBL Communications Inc.
C N S S P E C T R U M S

Source: http://www.diariodeumadismorfia.com.br/blog/wp-content/uploads/2012/10/Pharmacologic-Treatment-of-Body-Dysmorphic-Disorder.pdf