Psychosomatics 42:314-319, August 2001
© 2001 The Academy of Psychosomatic Medicine
Fluoxetine in Pathologic Skin-Picking
Open-Label and Double-Blind Results
Mark R. Bloch, M.D.,
Michael Elliott, M.A.,
Holly Thompson, B.A., and
Lorrin M. Koran, M.D.
Received September 12, 2000; revised February 13, 2001; accepted February 23, 2001. From the Department of Psychiatry and Behavioral Sciences, Stanford University Medical Center, Stanford, CA. Address reprint requests to Dr. Bloch, 3433 Golden Gate Way, Suite 1A, Lafayette, CA 94549. E-mail: _rloch{at}excite.com
|
ABSTRACT
|
Various studies suggest that selective serotonin reuptake inhibitors (SSRIs) may be useful in treating pathologic skin-picking. The authors investigated the effectiveness of fluoxetine in treating this behavior. Fifteen subjects with clinically significant skin-picking were recruited by newspaper advertisement. They received 6 weeks of open-label treatment with fluoxetine. Responders were then randomized to 6 weeks of double-blind fluoxetine or placebo. Treatment effect was assessed with standardized rating scales. All 15 subjects completed open-label treatment, and 8 were responders. Of these eight, the four randomized to double-blind fluoxetine maintained clinically significant improvement. The four randomized to placebo returned to their baseline symptom level. Larger studies are needed to determine which individuals are likely to respond to fluoxetine and the relative effectiveness of fluoxetine, other SSRIs, and other forms of treatment.
Key Words: Fluoxetine • Skin-Picking
|
INTRODUCTION
|
Pathological skin-picking is not formally recognized in DSM-IV and has no widely accepted diagnostic definition. The phenomenology and comorbidity have been described in two series of modest size.1,2 Pathologic skin-picking begins as an urge to touch, scratch, squeeze, or dig at the skin, often in response to a minor flaw or mild acne. Tools, such as pins and tweezers are often used, and skin damage can range from mild to extreme; serious complications, such as scarring and cellulitis, may develop.1–3 The behavior often leads to disfigurement, shame, and social impairment.1–5
Pathologic skin-picking has been conceptualized as an obsessive–compulsive spectrum disorder (along with trichotillomania and nail-biting),6 as a self-mutilating behavior7 and as an impulse-control disorder.8 It may be accompanied commonly by mood disorders,1,2 obsessive–compulsive and other anxiety disorders,1,2 body dysmorphic disorder,1,9 and, rarely, by conditions such as Prader-Willi syndrome.4 In some cases, an underlying psychotic condition, such as a delusion of parasitosis, drives the skin-picking behavior. The differential diagnosis also includes factitious disorder, where self-inflicted skin wounds are termed "dermatitis artefacta."4
The epidemiology of pathologic skin-picking has not been studied in the general population, but one study reported a 2% prevalence in dermatology clinics,2 where it is known as "acne excorie,"4 "psychogenic excoriation,"1,10 or "neurotic excoriation."3,7 It seems to be more prevalent in women than in men, often starts in adolescence, and takes a chronic course.1–3,11
Pathologic skin-picking may respond to serotonergic agents. A double-blind study of fluoxetine found evidence of efficacy in two of three measures; improvement in skin-picking behaviors appeared to be independent of changes in depression and anxiety.11 An open-label study of sertraline found significant improvement in skin-picking behavior, with resultant reduction in lesions.12 A similar open-label study of fluvoxamine showed benefit across a variety of measures, and the effects appeared independent of mood.10 Case reports also have suggested preferential response to serotonergic agents.3,5,13
The purpose of this study is to investigate further the efficacy of fluoxetine in the treatment of pathologic skin-picking.
|
METHODS
|
Subjects
Subjects were recruited via newspaper advertisements and gave written informed consent before study entry. Eligible subjects were 18 years of age or older and able to participate in outpatient treatment. Inclusion criteria included symptoms of at least 6 months duration, lesions noticeable by the investigators, and clinically significant distress related to skin-picking. Exclusion criteria included mental disorders secondary to underlying medical illness, mental retardation, psychosis, obsessive–compulsive disorder, factitious disorders, dissociative disorders, active substance abuse or dependence, severe personality disorders, or a history of bipolar I or II disorder. We excluded subjects judged to present a serious suicide risk, those already taking psychiatric medications, and those with serious medical illnesses. Subjects with other dermatologic causes for skin-picking, for example, scabies, were also excluded. Women of childbearing potential were required to be using acceptable birth control methods and could not enter the study if already pregnant.
After telephone screening, all potential study subjects received a comprehensive psychiatric evaluation to determine eligibility. The evaluation included the Structured Clinical Interview for DSM-IV (SCID-I14); the Yale-Brown Obsessive Compulsive Scale (Y-BOCS), modified to rate skin-picking behaviors;15 the Skin Picking Treatment Scale (SPTS11); the 17-item Hamilton Rating Scale for Depression (Ham-D16); the self-rated Massachusetts General Hospital (MGH) Skin Picking Scale (MGH-SPS, unpublished; available on request); and the Clinical Global Impression-Severity scale (CGI-Severity). The modification of the Y-BOCS simply focuses each standard Y-BOCS probe on skin-picking, rather than on "obsessions" or "compulsions." The unpublished MGH-SPS is a seven-item questionnaire that asks the patient to rate, "on an average day," the frequency and intensity of urges to pick, degree of control, frequency of picking, attempts to resist, success in resisting, and distress associated with skin-picking. All evaluations were performed by one of the two study psychiatrists (MRB or LMK).
The screening medical work-up included complete medical history, physical examination, complete blood count, comprehensive survey panel, urinalysis, serum beta-HCG (women only), and an electrocardiogram.
Study Design
One week after the screening visit, subjects began fluoxetine 20 mg/day. They were seen again at the end of Weeks 1, 2, 4, and 6. Dosage was increased in nonresponders, as tolerated, at the end of Weeks 2 (up to 40 mg/day) and 4 (up to 60 mg/day).
At each visit, the following efficacy measures were obtained: modified Y-BOCS; the SSPTS; the MGH-SPS; and, the Ham-D. Ratings for each patient were performed by the same investigator across visits. The investigators met before starting the study to establish interrater reliability, but no formal interrater reliability studies were performed.
At Week 6, subjects were classified as "responders" or "nonresponders." "Response" was defined prospectively as a 30%-or-greater decrease in the modified Y-BOCS score. Nonresponders were excluded from the randomized continuation phase and referred for alternate treatments. Responders at Week 6 were randomized, double-blind, to either placebo or fluoxetine, to be continued through Week 12 at the Week 6 dose.
We hypothesized that subjects responding to fluoxetine in the open-label phase would relapse if randomized to placebo and maintain response if randomized to fluoxetine continuation. We chose this design for two reasons. First, we expected that offering open-label treatment would facilitate recruitment. We also believed that limiting the double-blind phase to open-label responders would be more likely to reveal a statistically significant response, given our small sample size. In this design, each subject randomized to placebo acts as her own control.
In the double-blind, placebo-controlled study phase, subjects were seen at Weeks 8 and 12, at which times the treatment efficacy rating scales were re-administered. At Week 12, subjects were again evaluated by an internist, and laboratory tests were repeated. Upon completing the study, subjects were offered continuation treatment with fluoxetine or alternative therapies.
Placebo and fluoxetine were dispensed from identical containers, and the capsules, as provided by Lilly Pharmaceuticals, appeared identical. Compliance was measured by pill counts performed by research assistants at study completion.
Statistical Analysis
Given the study's small size, and the study design in which each randomized patient served as her own control, we decided to report data on individuals. Also, pooled mean data for each group are reported across all visits for the MGH-SPS and the modified Y-BOCS. Spearman rank correlations were performed across all baseline measures, including the CGI-Severity and the SPTS. Results for latter two are not reported, since they were highly redundant with the MGH-SPS and the Y-BOCS results. We used the Mann-Whitney U test to compare the placebo and fluoxetine groups from baseline to the end of Week 12. Only nonparametric tests are reported since the data could not be assumed to be normally distributed.
|
RESULTS
|
We screened 21 potential subjects, of whom 6 were ineligible. The 15 subjects who entered the study were all women, one African American, the rest Caucasian. The subjects' mean age was 40.7 years (SD=11.5), and mean duration of symptoms was 25.1 years (SD: 9.1). Five subjects had current mood disorders: two had current dysthymia; one had current major depression and body dysmorphic disorder (BDD), with BDD symptoms unrelated to her skin-picking behavior; one had current major depression arising out of dysthymia; and, one, an adjustment disorder with mixed anxiety and depression. One subject had a history of drug abuse.
Study subjects described intense urges to pick, often exacerbated by anxiety or stress. Time spent picking divided the study group into thirds, with approximate picking times of 15–30 minutes per day, 1–2 hours per day, and 3–7 hours per day. Nine patients picked their faces; six their hands, cuticles and nails; four their legs; three their arms and shoulders; two their scalps; two their necks; and, one her back. Eight picked only one body area; seven picked more than one. Six used tools, such as pins, needles, tweezers, or cuticle tools in their picking. Subjects tended to feel for small skin defects or lesions, which then became sites of picking, usually to the point of bleeding. Most attempted to resist the behaviors and had varying degrees of success. The picking itself tended to be gratifying during the episode, painful in some but not all subjects, and usually relieved tension. The resulting lesions produced shame, self-criticism, guilt, and, in some cases, social withdrawal. Because the subjects failed to resist harmful impulses, experienced preceding tension and/or gratification in the act, and felt regret afterwards, their skin-picking behavior met DSM-IV diagnostic criteria for an Impulse Control Disorder, Not Otherwise Specified.
No subject stopped medication or left the study because of adverse drug reactions. Eight of the fifteen subjects were responders at Week 6 (i.e., achieved at least a 30% decrease in the modified Y-BOCS score). Four of the eight responded by Week 4 (Table 1). The other study measures, including the MGH-SPS, CGI-Severity and the SPTS, yielded similar results. The responders included 4 of the 5 subjects with current mood disorders (whose mood disorders remitted by randomization) and 3 of the 10 subjects without active comorbid conditions. In the one subject with comorbid major depression arising out of dysthymia, neither the mood disorder nor the skin-picking responded. All eight responders completed the double-blind, 6-week portion of the study.
For the 15 subjects in the 6-week open-label phase, the mean modified Y-BOCS score decreased from 22.9±4.6 at baseline to 13.1±7.6; the mean MGH-SPS score from 20.1±3.5 to 11.9±6.1; and, the Ham-D from 8.3±7.6 to 2.6±3.4. The eight responders accounted for the vast bulk of these changes. For example, the responders' mean modified Y-BOCS score fell from 24.0±5.1 to 7.3±4.4, whereas the non-responders' mean score fell only from 21.7±4.0 to 19.7±4.0. The responders' mean MGH-SPS scores fell 64%, whereas the nonresponders' mean decreased by only 10%. At baseline, the primary outcome measure, the modified Y-BOCS, correlated substantially with the SPTS (r=0.74) and CGI-Severity (r=0.65), modestly with the self-rated MGH-SPS (r=0.41), and minimally with the Ham-D (r=0.12).
There were no significant differences in skin-picking severity at baseline between the fluoxetine and placebo treatment groups, though there was a trend toward greater severity in the group randomized to double-blind fluoxetine. At Week 6, the four responders randomly assigned to placebo exhibited a trend toward greater severity, compared with the four assigned to fluoxetine, as measured by the modified Y-BOCS (Table 2), but not by the MGH-SPS (Table 1) or the SPTS.
After randomization, a clear divergence between the placebo and drug groups emerged across all measures. The placebo group experienced symptom-worsening between Weeks 6 and 12, whereas the fluoxetine group substantially maintained response through Week 12. At Week 12, two fluoxetine subjects reported practically no picking, and two had mild urges, occasional brief episodes, and little or no tissue damage. The placebo group had returned to the baseline level of symptoms by Week 12, with a mean modified Y-BOCS of 20.3±2.0 and mean MGH-SPS of 17.8±3.3. In contrast, the fluoxetine group exhibited an approximately 70% reduction in symptom score from baseline, with a mean Y-BOCS of 8.5±6.0 and mean MGH-SPS of 6.0±3.7. The Mann-Whitney U test statistics, calculated to compare the fluoxetine and placebo groups on percent change from baseline through Week 12, were identical for the MGH-SPS and the modified Y-BOCS (U=16, P=0.021; df=1).
We were able to contact six of the eight double-blind study subjects 21 to 30 months after they completed the study. One fluoxetine patient remained in remission on fluoxetine (27 months after the study); one had discontinued the drug because of sexual side effects and quickly relapsed. Two fluoxetine patients were lost to follow-up. One placebo patient resumed fluoxetine when the study ended and remains in remission on fluoxetine after 21 months. Three placebo patients did not resume fluoxetine, two because of side effects (nervousness and emotional "numbing"), and all have continued skin-picking behaviors during the follow-up period.
|
DISCUSSION
|
This study provides further evidence that fluoxetine may be useful in the treatment of pathologic skin-picking. Half the subjects responded favorably in a 6-week, open-label trial. Among these responders, those randomized to double-blind placebo lost the therapeutic effect, whereas the double-blind fluoxetine group largely maintained it, suggesting that the initial response was not a placebo effect. The study design allowed each subject to act as her own control, and both the individual and the pooled data support a drug effect. The long-term follow-up data also suggest drug effectiveness. The drug response in responders was robust, with a mean decrease in symptoms, as measured by observer and self-rated scales, of over 60% by Week 6. However, mild skin-picking urges and occasional skin-picking behaviors (although usually more benign) remained.
Given the half-lives of fluoxetine (T1/2=48–72 hours) and its major active metabolite, norfluoxetine (T1/2=96–384 hours),17 one might expect a gradual diminution in drug efficacy over several weeks in subjects switched from fluoxetine to placebo. Our data are consistent with this expectation. The placebo group experienced a partial return of symptoms 2 weeks after stopping the drug. After 6 weeks, their symptoms had returned to baseline levels.
Was the blind broken for subjects or investigators during the double-blind phase by changes in side effects? We believe it was not. In the double-blind study phase, neither the subjects nor investigators were confident about which subjects were assigned to active drug or placebo. Analyzing side-effect reports after the study ended, we found that three of four placebo subjects had no side effects at the time of randomization or afterward; one noted improved appetite after randomization. Of the four patients randomized to continued fluoxetine, two had no side effects when randomized or subsequently, one noted cessation of mild tremors and of increased sweating despite continued active medication, and one continued to report excessive yawning and sexual dysfunction.
Seven of our initial study group of 15, or 47% of the subjects, did not respond to the 6 weeks of open-label fluoxetine (i.e., achieve a  30% decrease in the modified Y-BOCS score). There are several possible explanations. First, one subject experienced some benefit but did not reach the response criterion. Second, a longer open-label phase might have produced more responders. Third, lack of response could reflect noncompliance, which we measured by physician inquiry at each visit (a notoriously poor means) and by counting pills remaining at the end of each study phase. We found no significant evidence of noncompliance. Since fluoxetine plasma levels vary widely among individuals receiving the same dose,18 measuring these levels would have provided limited information, unless a plasma level was zero. Finally, fluoxetine may truly be ineffective in some individuals with this disorder; that is, similarity in symptoms may not indicate pathophysiological homogeneity.
As in an earlier double-blind trial of fluoxetine treatment of skin-picking,11 we saw no clear way to predict fluoxetine responders. Individuals with and without comorbid mood symptoms, picking one vs. picking more than one body area, and using or not using tools in picking were included among the responders.
In summary, this study, the second randomized trial of fluoxetine in pathologic skin-picking, provides additional evidence of drug efficacy for some individuals with this impulse-control disorder. Improvement, when it occurs, usually occurs after about 1 month and progresses over 6 weeks or more. Furthermore, symptoms appear to reemerge with drug discontinuation along the time course of drug and metabolite elimination. Larger double-blind studies are needed to assess which individuals are likely to respond to fluoxetine; the relative effectiveness of fluoxetine, other serotonin reuptake inhibitors, and other treatment approaches; and the biological and psychological differences that separate responders from nonresponders to fluoxetine treatment.
|
ACKNOWLEDGMENTS
|
We would like to thank Sue Thiemann, M.A., for help with statistical analyses.
This study was supported, in part, by a grant from Eli Lilly Pharmaceuticals. None of the authors received direct compensation from Eli Lilly.
|
REFERENCES
|
-
Arnold LM, McElroy SL, Mutasim SF, et al: Characteristics of 34 adults with psychogenic excoriation. J Clin Psychiatry 1998; 59:509-514[Medline]
-
Wilhelm S, Keuthen NJ, Deckersbach T, et al: Self-injurious skin-picking: clinical characteristics and comorbidity. J Clin Psychiatry 1999; 60:454-459[Medline]
-
Gupta MA, Gupta AK, Haberman HF: Neurotic excoriations: a review and some new perspectives. Compr Psychiatry 1986; 27:381-386[CrossRef][Medline]
-
Koran LM: Obsessive-Compulsive and Related Disorders in Adults: A Comprehensive Clinical Guide. Cambridge, UK, Cambridge University Press, 1999, pp 202-207
-
Stout RJ: Fluoxetine for the treatment of compulsive facial picking (letter). Am J Psychiatry 1990; 147:370
-
Stein DJ, Hutt CS, Spitz JL, et al: Compulsive picking and obsessive-compulsive disorder. Psychosomatics 1993; 34:177-181[Free Full Text]
-
Fruensgaard K: Psychotherapy and neurotic excoriations. Int J Dermatol 1991; 30:262-265[Medline]
-
McElroy SL, Hudson JL, Phillips KA, et al: Clinical and theoretical implications of a possible link between obsessive-compulsive and impulse control disorders. Depression 1993; 1:121-132
-
Phillips KA, Taub SL: Skin-picking as a symptom of body dysmorphic disorder. Psychopharmacol Bull 1995; 31:279-288[Medline]
-
Arnold LM, Mutasim DF, Dwight MM, et al: An open trial of fluvoxamine treatment of psychogenic excoriation. J Clin Psychopharmacol 1999; 19:15-18[CrossRef][Medline]
-
Simeon D, Stein DJ, Gross S, et al: A double-blind trial of fluoxetine in pathologic skin-picking. J Clin Psychiatry 1997; 58:341-347[Medline]
-
Kalivas J, Kalivas L, Gilman D, et al: Sertraline in the treatment of neurotic excoriations and related disorders. Arch Dermatol 1996; 132:589-590
-
Gupta MA, Gupta AK: Fluoxetine is an effective treatment for neurotic excoriations: case report. Cutis 1993; 51:386-387[Medline]
-
Spitzer RL, Williams JBW, Gibbon M, et al: Structured Clinical Interview for DSM-III-R, Version 1.0. Washington, DC, American Psychiatric Press, 1990
-
Goodman WK, Price LH, Rasmussen SA, et al: The Yale-Brown Obsessive Compulsive Scale. Arch Gen Psychiatry 1989; 46:1006-1016[Abstract]
-
Hamilton M: A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23:56-62
-
Schatzberg AF, Cole JO, DeBattista C: Manual of Clinical Psychopharmacology, 3rd Edition. Washington, DC, American Psychiatric Press, Inc., 1997, p 72
-
Koran LM, Cain JW, Dominguez RA, et al: Fluoxetine serum levels are unrelated to outcome in obsessive-compulsive disorder. Am J Psychiatry, 1996; 153:1450-1454[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
R. J. Shaw, S. Dayal, J. Good, A. L. Bruckner, and S. V. Joshi
Psychiatric Medications for the Treatment of Pruritus
Psychosom Med,
November 1, 2007;
69(9):
970 - 978.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. A. Flessner and D. W. Woods
Phenomenological characteristics, social problems, and the economic impact associated with chronic skin picking.
Behav Modif,
November 1, 2006;
30(6):
944 - 963.
[Abstract]
[PDF]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 2001
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
