N-Acetylcysteine in the Treatment of Excoriation Disorder: A Randomized Clinical Trial

Importance  Excoriation (skin-picking) disorder (SPD) is a disabling, underrecognized condition in which individuals repeatedly pick at their skin, leading to noticeable tissue damage. To date, there has been no clearly effective pharmacologic or psychological treatment for SPD.

Objective  To determine whether N-acetylcysteine, an amino acid that appears to restore extracellular glutamate concentration in the nucleus accumbens, will be more effective than placebo in reducing compulsive picking behavior.

Design, Setting, and Participants  A randomized, double-blind trial was conducted at ambulatory care centers at the University of Minnesota (September 12, 2011, to June 15, 2012) and the University of Chicago (December 17, 2012, to June 26, 2015) and included 66 adults with SPD. Data analysis was performed from July 16 to September 9, 2015.

Interventions  N-acetylcysteine (dosing range, 1200-3000 mg/d) or placebo was administered for 12 weeks.

Main Outcomes and Measures  Participants were assessed using measures of skin-picking severity, including the modified Yale-Brown Obsessive Compulsive Scale (NE-YBOCS); total scores range from 0 to 40, with higher scores reflective of greater symptom severity. Another measure of skin-picking severity was the Clinical Global Impression-Severity Scale; total scores range from 1 (normal) to 7 (among the most extremely ill patients), and improvement ratings range from 7 (very much worse) to 1 (very much improved). Selected cognitive tasks included the Intra-dimensional/Extra-dimensional Shift Task to examine cognitive flexibility, with the key outcome measures being the number of errors, and Stop-Signal Reaction Time task, which evaluates motor inhibition. Outcomes were examined using a linear mixed-effects model.

Results  Of the 66 participants (31 randomized to placebo and 35 to N-acetylcysteine) included in the analysis, 59 (89%) were women; mean (SD) age was 34.8 (11.0) years. Compared with placebo, N-acetylcysteine treatment was associated with significant improvements in the NE-YBOCS. At baseline, NE-YBOCS scores were 18.9 and 17.9 for the treatment and placebo groups, respectively, and at 12 weeks, the scores were 11.5 and 14.1 for the treatment and placebo groups, respectively (P = .048). For the Clinical Global Impression-Severity scale, baseline scores were 3.5 and 4.0 and 12-week scores were 3.0 and 4.2, respectively (P = .003). These effects were significant both in terms of treatment by time interactions and post hoc tests at 1 or more individual time points. At the study’s end point, of the 53 participants who completed the study, 15 of the 32 participants (47%) receiving N-acetylcysteine were much or very much improved compared with 4 of the 21 participants (19%) receiving placebo (P = .03). There were no significant differences between the active and placebo arms in terms of psychosocial functioning.

Conclusions and Relevance  N-acetylcysteine treatment resulted in significant reductions in skin-picking symptoms and was well tolerated. The glutamate system may prove a beneficial target in treating SPD and other compulsive behaviors.

Trial Registration  clinicaltrials.gov Identifier: NCT01063348

Excoriation (skin-picking) disorder (SPD) is a disabling, underrecognized condition in which individuals repeatedly pick at their skin, leading to noticeable tissue damage.¹ Psychosocial problems, reduced quality of life, and medical problems (eg, infections) are common among individuals with SPD.^2,3

Although apparently fairly common (estimated prevalence rate, 1.4%-5.4%),^4-6 SPD remains poorly understood, with limited data regarding underlying pathophysiology and treatment.⁷ There is currently no medication treatment for SPD approved by the US Food and Drug Administration. In fact, to our knowledge, there have been only 4 small placebo-controlled trials published on the treatment of SPD. Three of the studies^8-10 examining a selective serotonin reuptake inhibitor produced mixed effects, and 1 trial¹¹ investigating lamotrigine found no significant differences compared with placebo. Given the lack of clearly effective pharmacologic treatments for SPD, more research is needed.

Glutamatergic dysfunction has been implicated in the pathophysiology of compulsive or habitual behaviors.^12,13 In preclinical studies, levels of glutamate within the nucleus accumbens appear to mediate reward-seeking behavior.¹⁴ Furthermore, glutamate manipulations alter extradimensional set shifting (a form of cognitive flexibility) and appear to exert these effects via the nucleus accumbens.^15,16

N-acetylcysteine, an antioxidant and cysteine prodrug, increases extracellular levels of glutamate in the nucleus accumbens, possibly through a mechanism whereby inhibitory metabotropic glutamate receptors are stimulated, and this then reduces synaptic release of glutamate.¹⁷ Restoring extracellular glutamate concentration in the nucleus accumbens may therefore block reinstitution of compulsive behaviors.¹⁸ A nonmutually exclusive, hypothesized mechanism of action of N-acetylcysteine involves its antioxidant properties. Cysteine provided by N-acetylcysteine increases cellular production of glutathione and thereby may have neuroprotective effects on the brain.¹⁹

Early research into trichotillomania, a disorder with familial, clinical, and biological links to SPD,²⁰ found that N-acetylcysteine was a potentially effective treatment option for adults.²¹ Based on those data, we conducted a randomized, double-blind, placebo-controlled study to examine the tolerability and efficacy of N-acetylcysteine in the treatment of adults with SPD. Our key hypothesis was that N-acetylcysteine would be more effective than placebo in ameliorating the symptoms of SPD.

Box Section Ref ID

Men and women aged 18 to 65 years with a primary, current DSM-5 diagnosis of excoriation (skin-picking) disorder were recruited by newspaper advertisements and referrals. Exclusion criteria included (1) unstable medical illness; (2) history of seizures; (3) lifetime bipolar disorder, dementia, or psychotic disorder; (4) current (past 3 months) substance use disorder; (5) current suicide risk; (6) previous treatment with N-acetylcysteine; (7) diagnosis of asthma (owing to the alleged potential of N-acetylcysteine to worsen asthma); (8) current pregnancy or inadequate contraception in women of childbearing potential; and (9) initiation of pharmacotherapy or psychotherapy within 3 months prior to study entry (Figure 1). Current use of psychotropic medications was allowed if the dose had been stable for 3 months and there were no plans to modify the dose. Participants who changed dosage of medication or initiated therapy were to be discontinued from the study; however, no one was withdrawn for this reason.

Data were collected from September 12, 2011, to June 15, 2012, at the University of Minnesota and then from December 17, 2012, to June 26, 2015, at the University of Chicago. The same protocol was used at both locations, and the same investigator (J.E.G.) conducted the research. The institutional review boards for both universities approved this randomized clinical trial (the trial protocol is available in Supplement 1) and the informed consent procedures. One investigator (J.E.G.) discussed the potential risks of the study, as well as alternative treatments for SPD, with participants. After a complete description of the study and an opportunity to ask questions, participants provided written informed consent; they received financial compensation. This study was carried out in accordance with the principles of the Declaration of Helsinki.²² Investigational new drug application number 108195 was assigned by the US Food and Drug Administration for the use of N-acetylcysteine.

Eligible participants were assigned to 12 weeks of double-blind N-acetylcysteine or placebo treatment. Each university’s investigational pharmacy randomized all participants (block sizes of 8, using computer-generated randomization with no clinical information) to either the N-acetylcysteine or matching placebo group in a 1:1 fashion. All participants were seen every 3 weeks during the 12-week period. The initial dose of N-acetylcysteine was 1200 mg/d and was increased to 2400 mg/d by week 3. At week 6, the dose was increased to 3000 mg/d for the remaining 6 weeks of the study. If clinically necessary, the dose was raised more slowly or the target dose of 3000 mg/d was not reached. Because N-acetylcysteine often emits an odor of sulfur, to protect the blinding, all participants were told at study entry that their medication may or may not have an odor and that the effectiveness of N-acetylcysteine was independent of this odor. No participant ever mentioned the odor.

The dose range was based on safety and efficacy data from studies using N-acetylcysteine, although the dose in the present study was increased to 3000 mg/d instead of the 2400-mg/d dose used in the previous trichotillomania study.²¹ We selected this higher maximal dose because the dose of 2400 mg/d used in the trichotillomania trial was very well tolerated, suggesting that higher doses merited exploration in subsequent trials. No participants were withdrawn from the study because of nonadherence to the study medication (ie, failing to take medication for ≥3 consecutive days).

Demographics and clinical features of SPD were assessed with an unpublished, semistructured interview (eMethods in Supplement 2). The semistructured interview included proposed diagnostic criteria for SPD as well as questions regarding SPD’s phenomenology. After publication of the DSM-5, all participants were retrospectively assessed based on their earlier answers to the semistructured interview to determine whether they met full diagnostic criteria, and all did meet those criteria. Psychiatric comorbidity was assessed using the Structured Clinical Interview for DSM-IV.²³

The primary outcome measure was the change from baseline using the Yale-Brown Obsessive Compulsive Scale modified for Neurotic Excoriation (NE-YBOCS).²⁴ The NE-YBOCS is a 10-item scale that assesses picking symptoms during the past 7 days. The first 5 items compose the urge/thought subscale, and items 6 to 10 make up the behavior subscale. The NE-YBOCS has demonstrated good preliminary psychometric properties. The NE-YBOCS total scores range from 0 to 40, with higher scores reflective of greater symptom severity.²⁵

Both NE-YBOCS subscales were evaluated as secondary efficacy measures. Other secondary outcome measures included

1. Skin Picking Symptom Assessment Scale, which is a self-report scale that has satisfactory test-retest reliability and satisfactory change over time.²⁵ The Skin Picking Symptom Assessment Scale is scored from 0 to 24, with higher scores indicative of greater symptom severity.²⁴

2. Clinical Global Impression (CGI)–Improvement and CGI-Severity scales,²⁶ which were used to rate changes in the symptoms of SPD based on observations of the excoriated lesions as well as the NE-YBOC and self-report measures. The CGI uses a 7-point scale to evaluate global severity of illness and improvement or worsening of symptoms. Severity scores range from 1 (normal) to 7 (among the most extremely ill patients), and improvement ratings range from 7 (very much worse) to 1 (very much improved).

3. Sheehan Disability Scale,²⁷ which is a 3-item, self-report scale that assesses functioning in work, social or leisure activities, and home and family life. The total score for the Sheehan Disability Scale ranges from 0 to 30, with higher scores equating to greater psychosocial dysfunction.

4. Hamilton Anxiety Rating Scale²⁸ and Hamilton Depression Rating Scale,²⁹ which were used to assess anxiety and depressive symptoms. Hamilton Anxiety Rating Scale scores range from 0 to 56, with higher scores reflective of greater anxiety symptoms. The scores for the Hamilton Depression Rating Scale range from 0 to 56, and higher scores reflect greater depressive symptoms.

5. Quality of Life Inventory,³⁰ which is a self-administered rating scale that assesses 16 life domains, such as health, work, and relationships; this measure was used only at baseline and the study end point. Scores on the Quality of Life Inventory range from 0 (high quality of life) to 77 (very low quality of life).

Cognitive assessments consisted of 2 previously validated tests taken from CANTABeclipse, version 4.0 (Cambridge Cognition). The choice of cognitive challenges was based on the clinical features of SPD, knowledge from animal literature regarding the likely role of the glutamatergic system in cognition, and previous research demonstrating that individuals with SPD often exhibit significant deficits of motor inhibition and cognitive flexibility.^15,16,31 The order of the tasks was fixed, and participants completed the tasks at first visit (baseline) and end point.

Cognitive flexibility (ie, set shifting) was measured using the Intra-dimensional/Extra-dimensional Shift Task.³² Participants were presented with 2 stimuli on a computer screen for each trial and attempted to learn an underlying rule about which stimulus was correct. Key outcome measures were the number of errors made at the intradimensional shift and at the extradimensional shift stage.

The Stop-Signal Reaction Time task was used to assess motor inhibition.³³ On this task, participants respond to a left- or right-facing arrow that appears on a computer screen in a rapid fashion. Corresponding motor responses were measured, as were the participants’ ability to inhibit responses when an auditory beep (stop signal) occurred on a subset of trials. The time taken to internally suppress prepotent motor responses was measured (ie, stop-signal reaction times).

Safety assessments at each visit included evaluations of sitting blood pressure, heart rate, and weight. Adverse events were documented at each visit and included time of onset and resolution, severity, action taken, and outcome.

Demographic and baseline visit characteristics for the N-acetylcysteine and placebo groups were compared using χ² and 2-tailed, unpaired t tests to determine whether group differences occurred at randomization. All randomized participants were included in the analyses of baseline demographics and safety using an intention-to-treat principle.

Efficacy analyses included 60 individuals who had returned for at least 1 postrandomization visit (N-acetylcysteine, 34 [57%]; placebo, 26 [43%]). The statistical model was a linear mixed-effects (LME) regression model that included terms for treatment group, time, and treatment-by-time interaction. The analyses were run using the nlme package on R for Windows, version 3.2.2 (R Foundation; http://www.r-project.org). Previous studies³⁴ suggest that LME without imputation may provide more accurate and stable results than LME models using fixed and multiple imputation methods for handling missing data; therefore, no imputation was undertaken for missing data. All tests of hypotheses were performed using a 2-sided significance level at P ≤ .05. Rates of treatment response, defined by a CGI-Improvement Scale rating at end point of much or very much improved, were calculated for the 53 (80%) participants who completed the study.

With limited data regarding effect sizes for the primary outcome measure in SPD treatment studies, we determined sample size based on the previous study of N-acetylcysteine in trichotillomania.²¹ In that study, 25 individuals per group provided a sufficient sample to determine significant differences. Data analysis was conducted from July 16 to September 9, 2015.

Demographic and clinical characteristics of the participants at baseline are presented (Table 1 and Table 2). There were no statistically significant imbalances regarding demographics or baseline symptoms between treatment groups. Baseline skin-picking scores were reflective of moderate severity, overall baseline psychosocial dysfunction was mild, and quality of life was in the low-average range.

Of 71 individuals screened, 66 (59 women [89%]; mean [SD] age, 34.8 [11.0] years) with SPD were randomized: 31 to receive placebo and 35 to receive N-acetylcysteine (Figure 1). The 66 randomized participants reported a mean age at the onset of SPD of 12.2 (6.5) years. Most (50 [76%]) participants picked skin from multiple sites, and most (56 [85%]) had never sought mental health treatment for SPD.

No participant was currently undergoing psychotherapy, and 22 (33%) were receiving a psychotropic medication. The rates of psychotropic medication use did not differ significantly between treatment groups.

A total of 32 of 35 participants (91%) assigned to treatment with N-acetylcysteine and 21 of 31 individuals (68%) assigned to treatment with placebo completed the 12-week trial (Figure 1). The rate of study completion differed significantly between treatment groups (Fisher exact test, P = .03), although there were no statistically significant pretreatment differences between completers and noncompleters on any measure. Of the 13 participants who failed to complete the study, all withdrew owing to an inability to adhere to the study schedule.

As indicated in Table 2 and Figure 2, there were statistically significant treatment type-by-time interactions for the NE-YBOCS total, NE-YBOCS urge/thought subscale, and CGI-Severity Scale scores. For NE-YBOCS total and urge/thought subscale scores, significant benefits for N-acetylcysteine over placebo were seen at all nonbaseline time points (weeks 3, 6, 9, and 12). For the CGI-Severity Scale, significant benefits were observed at weeks 6, 9, and 12. The CGI-Severity Scale scores for the N-acetylcysteine group improved from 3.5 at baseline to 3.0 at 12 weeks, while scores for the placebo group were 4.0 at baseline and 4.2 at 12 weeks. During the course of the 12 weeks, the NE-YBOCS total score demonstrated a 38.3% reduction in skin-picking symptoms for the N-acetylcysteine group (from 18.9 at baseline to 11.5 at 12 weeks) compared with 19.3% for placebo (from 17.9 at baseline to 14.1 at 12 weeks). The observed means (SDs) for each outcome for each time period are included in the eTable in Supplement 2.

The CGI-Improvement Scale score demonstrated significant improvement by study end point: of the 32 participants who completed the study, 15 (47%) participants assigned to treatment with N-acetylcysteine were much or very much improved compared with 4 (19%) of the 21 in the placebo group who completed the study (Fisher exact test, P = .03). However, participants assigned to treatment with N-acetylcysteine did not demonstrate greater improvement in psychosocial functioning. Although functioning improved to a greater numeric extent for those assigned to the N-acetylcysteine group, the Sheehan Disability Scale total score did not differ significantly between groups. In addition, quality of life did not change significantly for either group (Table 2).

There were no serious adverse experiences in participants assigned to treatment with N-acetylcysteine or placebo. The few adverse experiences were of mild intensity and did not differ significantly between groups. The only reported adverse event in the placebo group was nausea (1 participant [3%]); those in the N-acetylcysteine group included nausea (5 participants [14%]), dry mouth (1 [3%]), constipation (2 [6%]), and dizziness (1 [3%]).

There was a treatment type-by-time interaction regarding extradimensional set-shifting performance owing to the beneficial effects of placebo over N-acetylcysteine therapy (Table 2). However, because we examined 3 cognitive measures, the finding of change in set shifting is not significant when using a Bonferroni correction for multiple comparisons. No such treatment type–by-time interaction was observed for intradimensional set shifting or for Stop-Signal Reaction Times.

This randomized, double-blind clinical trial indicates that N-acetylcysteine is safe and more effective than placebo for treatment of SPD. The beneficial effects of N-acetylcysteine are consistent with those of previous controlled research²¹ finding the drug to be effective for adults with trichotillomania, case reports³⁵ demonstrating that N-acetylcysteine is beneficial for SPD, and a recent case series³⁶ on use of N-acetylcysteine in the treatment of skin picking in 35 individuals with Prader-Willi syndrome. N-acetylcysteine is an amino acid that is available without prescription and generally well tolerated.²¹ The agent could be an effective treatment option for people with SPD.

The benefits achieved with N-acetylcysteine treatment in the present study appear to be primarily in the reduction of urges or cravings to pick (reflected by the NE-YBOCS subscale score) rather than the actual behavior. Because urges to pick are perhaps somewhat analogous to drug cravings, this finding is consistent with research^14,17 using N-acetylcysteine in the treatment of cocaine addiction. This finding may also suggest that a treatment other than N-acetylcysteine might be more effective in people who pick automatically or with little conscious awareness.

The efficacy of N-acetylcysteine in this study lends further support to the hypothesis that pharmacologic manipulation of the glutamate system may target core symptoms of compulsive behaviors.¹² This narrative accords well with animal data on compulsive grooming. One mouse model of grooming behaviors is the Sapap3 knockout mouse.³⁷ These mice exhibit excessive self-grooming behaviors that appear to be somewhat analogous to skin picking. Sapaps are proteins that act between glutamate receptor–binding proteins and the cytoskeleton, and Sapap3 is highly expressed in glutamatergic synapses of the striatum. Genetic variants of Sapap3 have been associated³⁸ with skin picking in humans. Therefore, the beneficial effects seen in the present study may stem from the ability of N-acetylcysteine to modulate prefrontal glutamatergic drive to the nucleus accumbens.

Of course, there are alternative and nonmutually exclusive theories about how N-acetylcysteine may be helping in compulsive behaviors. For example, N-acetylcysteine increases glutathione in glial cells at the doses we used in the present study.³⁹ Recent animal research⁴⁰ suggests that the glutathione hypothesis of N-acetylcysteine is also consistent with the Sapap3 genetic research. This research has shown that Sapap3 knockout transgenic mice have lower striatal glutathione levels, suggestive of striatal hypermetabolism accompanied by oxidative stress. Treatment with N-acetylcysteine therefore may be beneficial via this potential antioxidant mechanism.

Treatment with N-acetylcysteine failed to produce statistically significant improvement in measures of psychosocial functioning or quality of life. One possible explanation might be associated with the baseline scores. Participants began with only mild functional impairment and a low-average quality of life. Given the limited range for improvement on these scales, it may not have been possible to observe statistically significant improvement with the sample size. Another explanation may refer to the long-term consequences of picking (eg, scars) rather than with the picking itself. Because symptom improvement may occur many weeks before scars heal, measures of functional impairment and quality of life might be more relevant if determined after several months of symptom improvement rather than over a short (12-week) period.

These results with N-acetylcysteine are, of course, preliminary and based on one small study that does not allow for complete evaluation of the benefit/risk balance; therefore, further research is needed. Although this study represents a potentially beneficial approach to treating SPD, there are several limitations. First, our formulation of N-acetylcysteine had a sulfurous odor, which may have jeopardized blinding in some cases. Although we told people that the medication may or may not have an odor, some may have searched this issue online and found that odor was associated with active N-acetylcysteine. The integrity of the blinding was never formally tested by asking participants to guess their assignment. Second, SPD appears to be a chronic disease⁷ that is likely to require treatment beyond 12 weeks. Long-term effects of N-acetylcysteine treatment thus require evaluation. Third, this study did not include psychotherapy. Cognitive behavioral therapy has shown benefit for SPD⁴¹ and should be considered in conjunction with medication. Fourth, the study was underpowered to look at adverse events. Based on our small sample, we could have only detected adverse events that occurred at the 10% or higher level with any certainty. Other research⁴² suggests that N-acetylcysteine may, in rare cases, produce adverse effects that are often intolerable (eg, nausea, vomiting, and cutaneous systemic hypersensitivity reaction, all in the context of intravenous infusions of N-acetylcysteine); we did not expect to see such effects in this study, and they did not occur. A larger study may provide more information on the tolerability of N-acetylcysteine treatment in this population. Finally, we did not examine the optimal N-acetylcysteine dose; whether some individuals would have responded to higher doses needs to be examined.

This investigation suggests that N-acetylcysteine appears to be effective and well tolerated in the acute treatment of SPD. As effective treatments for skin picking emerge, it becomes increasingly important that physicians and other mental health care professionals screen for the disorder to provide timely treatment.

Corresponding Author: Jon E. Grant, JD, MD, MPH, Department of Psychiatry and Behavioral Neuroscience, Pritzker School of Medicine, University of Chicago, 5841 S Maryland Ave, MC 3077, Chicago, IL 60637 (jongrant@uchicago.edu).

Submitted for Publication: December 8, 2015; final revision received January 11, 2016; accepted January 13, 2016.

Published Online: March 23, 2016. doi:10.1001/jamapsychiatry.2016.0060.

Author Contributions: Dr Grant had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Grant, Odlaug, Kim.

Acquisition, analysis, or interpretation of data: Grant, Chamberlain, Redden, Leppink, Odlaug.

Drafting of the manuscript: Grant, Chamberlain.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Grant, Chamberlain, Odlaug.

Administrative, technical, or material support: Grant, Chamberlain, Redden, Leppink, Odlaug.

Study supervision: Grant, Odlaug.

Conflict of Interest Disclosures: Dr Grant reported receiving research grants from the National Institute of Mental Health (NIMH U01MH076179), National Center for Responsible Gaming, and Forest and Roche Pharmaceuticals; yearly compensation from Springer Publishing for acting as editor-in-chief of the Journal of Gambling Studies; and royalties from Oxford University Press, American Psychiatric Publishing, Inc, Norton Press, and McGraw Hill. Dr Chamberlain reported being a paid consultant for Cambridge Cognition, and his involvement in the present study was facilitated by a grant from the Academy of Medical Sciences. Dr Odlaug reported receiving a research grant from the Trichotillomania Learning Center, previously consulting for and currently (after the study) being employed by H. Lundbeck A/S, and receiving honoraria and royalties from Oxford University Press. No other disclosures were reported.

Funding/Support:Great American Health provided the N-acetylcysteine to the investigational pharmacies at the Universities of Minnesota and Chicago and provided certificates of analysis and conformance to ensure compliance with the standards of Good Manufacturing Practice.

Role of the Funder/Sponsor: Great American Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.