February 2010  << Back  

  Can't open the PDF? Click here for help.



Augmentation with naltrexone to treat compulsive sexual behavior: A case series

Nancy C. Raymond, MD

Department of Psychiatry, University of Minnesota Medical School, Minneapolis, MN, USA

Program in Human Sexuality, Family Medicine and Community Health, University of Minnesota Medical School, Minneapolis, MN, USA

Jon E. Grant, JD, MD

Department of Psychiatry, University of Minnesota Medical School, Minneapolis, MN, USA

Program in Human Sexuality, Family Medicine and Community Health, University of Minnesota Medical School, Minneapolis, MN, USA

Eli Coleman, PhD

Program in Human Sexuality, Family Medicine and Community Health, University of Minnesota Medical School, Minneapolis, MN, USA

BACKGROUND: Compulsive sexual behavior (CSB) is generally characterized by recurrent and intense sexually arousing fantasies, sexual urges, and behaviors, which cause individuals distress or impair daily functioning. Descriptive studies of individuals with paraphilic and nonparaphilic CSB indicate that they experience urges to engage in problematic sexual behavior. The opiate antagonist naltrexone has been successfully used to treat a number of disorders in which urges to engage in problematic behavior are a central feature, such as alcoholism. We hypothesized that naltrexone would reduce the urges and behaviors associated with CSB.

METHODS: Records of 19 male patients with CSB who were treated with naltrexone at an outpatient adult sexual health clinic were retrospectively reviewed.

RESULTS: Nearly all patients were already taking other psychotropic medications when naltrexone was initiated. Seventeen (89%) of the 19 patients reported a reduction in CSB symptoms when taking naltrexone for a period ranging from 2 months to 2.3 years, as judged by Clinical Global Impression scores of 1 or 2, indicating “very much improved” or “much improved.” Five (26%) of the 19 patients chose to discontinue the medication.

CONCLUSIONS: Naltrexone may be a useful adjunctive treatment for CSB.

KEYWORDS: impulse control disorder, sexual disorder NOS, compulsive sexual behavior, sexual addiction, naltrexone, paraphilia



Compulsive sexual behavior (CSB) is generally characterized by recurrent and intense sexually arousing fantasies, sexual urges, and behaviors, which cause individuals distress or impair daily functioning.1-3 The main CSB types are paraphilic and nonparaphilic, the main distinction being that paraphilic CSB involves sexual arousal associated with unconventional and sometimes illegal sexual behaviors, whereas nonparaphilic CSB involves conventional or normative sexual behavior.2,4 In paraphilic CSB, the preoccupation is with behavior as described in the paraphilia section of the DSM-IV-TR.5 Nonparaphilic CSB is not described in the DSM-IV-TR, although an example of a type of CSB appears under the Sexual Disorder, NOS category: “Distress about a pattern of repeated sexual relationships involving a succession of lovers who are experienced by the individual only as things to be used.” Examples of nonparaphilic CSB from the literature include: (1) having sex with multiple partners, (2) excessive masturbation, (3) a constant need for sexual activity in a relationship, (4) excessive use of pornography (via the Internet, videos, magazines, etc.), or (5) problematic use of telephone sex or interactive sexual activity via the Internet (using technologies such as video streaming).3,4 To meet the proposed criteria for either type of CSB, the fantasies, sexual urges, and behaviors must cause the patient clinically significant distress.

To our knowledge, no epidemiologic studies have been conducted on the prevalence and incidence of nonparaphilic CSB. A recent study of college students, however, found that 3.7% meet criteria for CSB.6 Based on clinical experience and an extrapolation from survey data, Coleman7 estimated that about 5% of the population experiences this disorder. Similarly, Schaffer and Zimmerman8, citing an unpublished estimate, suggested that 6% of the population is affected. Grant et al9 found a 4.9% lifetime prevalence of nonparaphilic CSB in a series of 204 patients; these patients were consecutively admitted to an inpatient psychiatry unit and interviewed using a semistructured clinical interview designed to identify impulse control disorders.

CSB is often associated with significant personal and interpersonal morbidity. For instance, there is evidence that CSB is associated with unsafe sexual behavior and that individuals with CSB are at risk of contracting and spreading human immunodeficiency virus (HIV)10,11 and other sexually transmitted infections. CSB often results in dissolution of marriages or other significant relationships, loss of employment because of activities such as viewing of pornography Web sites while at work, or unwanted pregnancy.1,4 Patients who present for treatment are often despondent, even suicidal, because they feel they cannot control their sexual behavior despite the consequences.1,4 Those with CSB also report increasing tension or arousal prior to engaging in the problematic sexual behavior and often report experiencing regret or guilt after performing the sexual act in an attempt to resolve that tension or arousal.1,12

Descriptive studies have shown that individuals with paraphilic and nonparaphilic CSB experience urges to engage in their problematic sexual behavior. In one study of 25 individuals with nonparaphilic CSB, all reported urges to engage in the problematic sexual behavior, and 24 of the 25 spent time trying to resist the urges, which they characterized as unwelcome.12 In another study, 23 of 34 individuals in treatment for CSB for varying lengths of time reported having urges to engage in CSB.13 In a third study, 67% of participants (n = 34) reported that they “had sexual urges that they felt were out of control or caused subjective distress.”1

Data regarding pharmacologic treatment of CSB are sparse. Case studies and case series have indicated that selective serotonin reuptake inhibitors (SSRIs) and other antidepressants may offer some benefit in treating CSB symptoms.14-23 In the only controlled pharmacologic study of nonparaphilic CSB, 28 men were assigned to citalopram or placebo.24 Those on citalopram demonstrated significant reductions in sexual desire, frequency of masturbation, and use of pornography. Four participants, however, did not report a significant decrease in high-risk sexual behavior.

Treatment of paraphilic CSB may also include the use of antiandrogens; however, side effects of these medications can lead to significant morbidity (for a review of this topic, see Reilly et al25).

Although study results are promising, data on treatment response to pharmacotherapy are limited. Because medications currently in use may not be effective for all individuals with CSB, other approaches are needed.

The opiate antagonist naltrexone has shown promise in treating addictive behaviors that have phenomenologic links to CSB. Naltrexone has been successfully used to treat a number of disorders in which urges to engage in problematic behavior are a central feature, such as alcoholism,26-29 borderline personality disorder with self-injurious behavior,30 and bulimia nervosa and binge-eating disorder.31,32 In addition, research supports the effectiveness of opiate antagonists in treating pathological gambling and kleptomania.33-35 It has been proposed that naltrexone works in these settings by dampening urges through inhibition of dopamine release within the ventral striatum region.36

Published case reports in adults37,38 and a case series in adolescents indicate that naltrexone is effective in the treatment of CSB.39 In these publications, patients indicated that naltrexone decreased urges to engage in problematic sexual behaviors and decreased the frequency of those behaviors.

Based on these encouraging preliminary reports of naltrexone in the treatment of CSB, we conducted a retrospective chart review to examine the efficacy of naltrexone in the treatment of CSB. We hypothesized that naltrexone would reduce the urges associated with CSB, as well as the behavior resulting from those urges.



Records of 19 patients with paraphilic or nonparaphilic CSB who were treated with naltrexone in an outpatient sexual health clinic at the University of Minnesota, Minneapolis from 2002 through 2006 were retrospectively reviewed. All patients were in individual and group psychotherapy that was primarily based on cognitive-behavioral approaches and also incorporated elements of psychodynamic, family systems, and object relations theories.40 All patients were male. Baseline information included age at the time of treatment at the clinic, comorbid psychiatric diagnosis based on clinical interviews, and concurrent medications. Charts were reviewed by one of the physicians in a group of 4 practitioners, which included 2 psychiatrists, a psychiatry resident, and a psychiatric clinical nurse specialist.


All patients were given a baseline Clinical Global Impression (CGI) score based on the retrospective review of the severity of CSB symptoms prior to initiating naltrexone. The CGI is a reliable and valid scale assessing the severity of psychiatric symptoms using a 7-point scale that ranges from “not ill at all” to “among the most extremely ill.”41 After reviewing the documentation of each subsequent medication management appointment, the rater assigned an improvement score ranging from 1 to 7, in which 1 meant “very much improved” and 7 meant “very much worse.” To arrive at the scores, the clinician reviewed the physician notes and, if necessary, notes from either the group or individual psychotherapy appointment that occurred within 2 weeks of the psychiatric appointment.

The Sexual Symptom Assessment Scale (S-SAS) was also given at multiple visits to 13 (79%) of the 19 patients.13 The S-SAS is a 12-item, self-rated scale with each item scored from 0 to 8 (for a discussion of the scale psychometric, see Raymond et al13). The scale includes items such as the severity, frequency, and duration of sexual urges and behaviors.13 The S-SAS was not in use during the time some of the patients were being treated in the clinic. When available, these scores were reviewed and taken into consideration when the visit rating was assigned.

Medication administration procedure

In general, patients were started on naltrexone, 25 to 50 mg/d, and were instructed to increase the medication to 100 mg/d after 1 to 2 weeks, unless they thought they had a good clinical response to the lower dose. On average, patients were seen 2 to 3 weeks after initiating the medication and then monthly thereafter until they were stable. Naltrexone could be increased to a maximum of 200 mg/d.

Liver function tests (LFTs) were generally performed before naltrexone was initiated, at 2 to 4 weeks after initiating the medication, at 2 to 4 weeks after every dosage change, at 3 months after the last dosage change, and then every 6 months thereafter. Patients were instructed not to use over-the-counter analgesics while taking naltrexone at doses higher than 50 mg/d because this drug combination can result in elevated liver enzymes.42 All patients were informed that naltrexone was being used “off-label” to treat CSB and that there was no FDA-approved medication for the treatment of CSB; the hypothesized scientific basis for the potential efficacy of naltrexone in the treatment of CSB was also described.36,38 In addition, patients were informed that the FDA-approved dose of naltrexone was 50 mg/d for relapse prevention of alcoholism and that naltexone had a “black box” warning regarding hepatotoxicity associated with doses over 50 mg/d. The Institutional Review Board of the University of Minnesota, Minneapolis approved the chart review protocol.



The mean patient age at the time of naltrexone initiation was 44.1 (SD ± 9.4) years (range, 28 to 62 years). On average, patients had been in treatment at the clinic for 2.1 (SD ± 1.9) years (range, 0.8 to 7.8 years) before starting naltrexone. During this time, all patients were participating in individual therapy for CSB, and 16 of the 19 (84.2%) were also in group therapy for CSB. Fourteen (73.7%) patients endorsed being primarily or exclusively heterosexual, and 5 (26.3%) endorsed being primarily or exclusively homosexual. Two patients (10.5%) were HIV positive. Sixteen (84.2%) of the 19 patients were on medications to treat concurrent disorders and in an attempt to reduce their CSB symptoms.

Compulsive sexual behavior diagnoses

Eight (42%) of the patients had at least one primary paraphilic diagnosis (see TABLE 1 for diagnoses). Eleven (58%) of the patients had primary nonparaphilic CSB subtypes: 5 described their primary problematic sexual behavior as having sex with multiple partners; 6 stated that their primary problem was the compulsive use of pornography on the Internet while masturbating or use of other forms of sexual interaction via the Internet while masturbating, such as use of chat rooms or live video streaming. Of the 19 patients, 8 appeared to have only one type of CSB, according to the retrospective chart review. The frequencies of each of the CSB subtypes are listed in TABLE 1.


Retrospective CSB subtypes as diagnosed during initial clinical interview

Paraphilic CSB (n = 8)
   Exhibitionism 2
   Pedophilia 2
   Sexual masochism 2
   Sexual sadism 1
   Transvestic fetishism 2
   Voyeurism 2
Nonparaphilic CSB (n = 11)
   Compulsive masturbation (only) 1
   Internet pornography (cybersex) 8
   Pornography and/or pornographic bookstores 3
   Sex with multiple partners 5
   Telephone sex 1
CSB: compulsive sexual behavior.
Comorbid psychiatric diagnoses

Axis I and II diagnoses were based on the initial clinical interview by the psychologist and psychiatrist when the patient first entered the clinic. The frequencies of Axis I and Axis II diagnoses appear in TABLE 2. Only one patient did not have a concurrent psychiatric diagnosis during the course of treatment for CSB.


Retrospective Axis I and Axis II comorbidity as diagnosed during initial clinical interview

Axis I and II comorbidities (n = 18)a
Affective disorders
   Major depressive disorder 13
   Dysthymia 1
   Depression NOS 1
   Bipolar II disorder 2
   Cyclothymic disorder 1
Anxiety disorders
   Generalized anxiety disorder 5
Substance use disorders
   Alcohol abuse 2
   Cannabis abuse 1
Adjustment disorder 1
   Cluster B traits 2
CSB: compulsive sexual behavior; NOS: not otherwise specified.
aOne of the 19 patients had no Axis I or II comorbidities during the course of treatment for CSB.
Concurrent medications

Fifteen patients (79%) were taking either an SSRI or the serotonin-norepinephrine reuptake inhibitor (SNRI) venlafaxine. Two patients (11%) were on bupropion alone, and 3 (16%) were on bupropion with an SSRI. Seven patients (37%) were on an additional medication—gabapentin or buspirone—to treat anxiety. Three patients (16%) were on mood stabilizers, and 2 (11%) were on quetiapine to either treat anxiety or help with sleep. One patient (5%) was on N-acetyl cysteine to augment the naltrexone and help protect against liver damage.

In order to attribute improvement in CSB symptoms to the addition of naltrexone, it is important to know whether other medications have been changed at the same time. All but 4 patients had been on the same medication regimen for at least 6 months at the time naltrexone was initiated. One patient started sertraline 1 month before naltrexone was initiated. Three patients had stopped taking a medication—dextroamphetamine, lamotrigine, and nefazodone—the day naltrexone was started. Three patients (16%) were taking no psychiatric medications other than naltrexone.

Liver function tests

None of the patients had elevated aspartate aminotransferase (AST) or alanine aminotransferase (ALT) levels at baseline, and none developed elevated AST (32.2 ± 13.2) or ALT (20.6 ± 5.4) levels during the course of treatment with naltrexone.

Treatment response

Seventeen of the 19 (89%) patients taking naltrexone had a positive response, as judged by a score of 1 or 2 on the CGI scale, indicating “very much improved” or “much improved,” respectively.

  • Seven (37%) of the 19 patients had a response within 1 to 2 months and were stabilized on a dose of naltrexone with good response throughout the follow-up period. Of these 7 patients, 2 were maintained on 50 mg/d, 2 were maintained on 75 mg/d, and 3 were maintained on 100 mg/d of naltrexone.

  • Five (26%) of the 19 patients had a good initial response for an average of 8.4 (SD ± 4.8) months and then needed 1 or 2 dosage increases due to worsening symptoms. After the dosage was increased, they again had a positive response that was maintained for the remainder of the follow-up period (2 to 18 months). Three of the 5 patients were maintained on a dose of 150 mg/d, 1 was maintained on 200 mg/d, and 1 was maintained on 100 mg/d.

  • Five (26%) additional patients who had a positive response had a trial of naltrexone that lasted 4 months or less. Of these patients, 2 (11%) had a good response for 3 months and were then lost to follow-up; 2 (11%) took the medication for a brief period (1 and 3 months) and then discontinued it, despite initial improvement, because of sedation; and 1 (5%) patient stopped taking naltrexone because it was interfering with his sleep and causing numbness and tingling in his extremities. All of the patients who discontinued naltrexone due to side effects experienced worsening of CSB symptoms after discontinuation.

Of these 17 patients, 2 (11%) successfully lowered or eliminated their dose. One patient had been on a maximum dose of 150 mg/d that was then reduced to 50 mg/d, and he was still in remission at the time of this writing. The second patient was maintained on 100 mg/d for 2 years, then discontinued the medication and has not been engaging in CSB for 1 additional year.

The mean effective dose was 104 (SD ± 41) mg/d for the 17 patients who had a positive response to naltrexone. We compared the clinical characteristics of the patients who required higher doses of naltrexone with those who did not and could not find any distinguishing characteristics. On average, patients who responded were treated with naltrexone for 1.0 (SD ± 1.0) year, with a range of 2 to 28 months.

Finally, 2 (11%) of the 19 patients did not have a positive response to medication. Both men stopped the medication because of little perceived benefit, and one also mentioned sedation as a contributing factor.


The majority of the patients—17 (89%) of 19—had a reduction in CSB symptoms when taking naltrexone, as judged by CGI scores of 1 or 2 indicating “very much improved” or “much improved,” respectively. This improvement was seen despite the fact that many of these patients had received therapy for CSB for 9 months to 8 years; however, they were still symptomatic at the time naltrexone was started. Five (26%) of the 19 patients chose to discontinue naltrexone, including 3 responders who stopped taking the medication because of side effects and 2 nonresponders who stopped because of lack of efficacy. In summary, 12 (63%) of 19 patients had a good response to naltrexone and were able to tolerate the medication.

Seven of the responders (37% of the 19 patients) had a good long-term outcome on the initial treatment dose of between 50 mg/d and 100 mg/d of naltrexone. However, 3 of these patients took naltrexone for less than 8.4 months—the average length of time before patients who required a dosage increase obtained an increase. Five of the patients did well on doses of 50 to 100 mg/d for a number of months and then required a dosage increase in order to maintain remission of symptoms. Taken together, these data indicate that for a significant subset of patients, the effect of the initial dosage wears off after a period of time and a dosage increase is needed to maintain remission.

After a period of about 2 years of taking naltrexone, 2 (11%) of the 19 patients were able to either discontinue or markedly reduce their use of the medication and still maintain remission. At the time of this writing, none of the other patients have attempted to discontinue the medication; however, this raises the question of whether naltrexone could be successfully discontinued after a significant period of stability.

Limitations of this study include its retrospective nature, the lack of a placebo or control sample, the relatively small number of patients, and the fact that all but 3 of the patients were on concurrent medications. The clinicians who conducted the retrospective assessments were not blind to the patient’s treatment condition. The CGI has been validated for psychiatric conditions but has not been used before for assessing severity and improvement for CSB symptoms. The Axis I and Axis II diagnoses as well as the CSB subtypes were based on a clinical rather than a structured interview. Additionally, the patient follow-up period was quite variable.

This is the first published retrospective case series reporting on the use of naltrexone in combination with other psychotropic medications to treat adults with CSB. These patients are representative of other primarily heterosexual treatment samples of CSB patients described in the literature; however, not all treatment programs offer the extensive psychotherapy programs offered by our clinic.

The results are consistent with 2 other reports in the literature. One report included 2 cases of CSB involving unsafe sexual behavior with multiple partners.38 Both study participants stopped engaging in problematic sexual behavior when treated with naltrexone and reported marked decreases in urges to engage in CSB. The residual passing thoughts about the behavior were no longer compelling and did not lead to acting on the urges. A second case series reported that naltrexone was effective in treating 21 adolescent sex offenders (mean age 15.2) who had offended against younger children.39 Study participants were placed on naltrexone if they: (1) masturbated excessively (>2 times per day), (2) felt they could not control their arousal (spent more than 30% of their awake time in sexual fantasies), or (3) had sexual fantasies or behavior that regularly intruded into and interfered with their functioning in the treatment program. Given these criteria, all participants would likely meet criteria for paraphilic CSB. Twenty-one out of a cohort of 30 consecutively admitted patients met these criteria. The maintenance dosages of naltrexone ranged from 100 to 200 mg/d, with an average dosage of 170 mg/d. The author noted that there was no clinical benefit at dosages of less than 100 mg/d. Dosages above 200 mg/d added no additional benefit. All participants initially responded to the medication. Fifteen of the 21 patients (71%) had a positive result for at least 4 months, as judged by a decrease in sexual fantasies and masturbation. The average frequency of masturbation in the group decreased from an average of 2 times per day to 2 times per week with naltrexone. Very few of the patients in our case series could be maintained on a 50 mg/d dose, and our clinical experience also indicates no additional benefit for dosages over 200 mg/d.

Reports of liver toxicity and increases in LFTs with the use of naltrexone doses above 50 mg/d led to an FDA black box warning to not exceed 50 mg/d when prescribing naltrexone. There is evidence that the elevated liver enzymes in patients taking high doses of naltrexone are related to concurrent administration of a common group of over-the-counter medications, the nonsteroidal anti-inflammatory drugs (NSAIDs).43,44 One theory regarding the mechanism of this interaction is that the toxicity is due to NSAIDs causing a change in the metabolic pathway from the major metabolic pathways of reduction and glucuronidation to a reactive metabolite pathway (catalyzed by cytochrome P-450). This could explain the elevated transaminase levels observed in patients taking NSAIDs when on naltrexone. In a review of 41 consecutive cases of patients told not to take NSAIDs when they were on naltrexone, only 2 participants developed mild elevations in liver enzymes, and they later admitted to taking NSAIDs.45


Results of our study suggest that naltrexone may be a useful adjunctive treatment for CSB. This case review and a review of the other case series in juvenile patients indicate that many patients need doses >50 mg/d to reduce CSB symptoms but that doses >200 mg/d seldom, if ever, add additional benefit, while potentially increasing the risk of side effects. LFTs must be monitored if dosages >50 mg/d are required. We recommend obtaining baseline LFTs and repeating LFTs 1 month after initiating the medication, 1 month after each dosage change, 3 months after a maintenance dosage is determined, and every year thereafter. Based on the evidence cited above, it is prudent to advise against concurrent use of NSAIDs when naltrexone is being given at dosages >50 mg/d. If a patient disregards this advice, check LFTs more frequently and consider discontinuing naltrexone. If LFTs are elevated during routine testing, inquire about the use of NSAIDs before deciding whether to change or discontinue the naltrexone dose.

Most of our patients were taking concurrent medications for other psychiatric disorders. SSRIs were the most common medications given with naltrexone. An important question for further research is whether a single medication or combinations of medications are more effective to treat CSB. As with most of the disorders in which behavioral excess is a major symptom, such as eating disorders, pathological gambling, and borderline personality disorder, psychotherapy may be an important addition to medication. All of our patients received concurrent psychotherapy. Efficacy of psychotherapy vs medications alone vs combination treatment is another empirical question that needs to be addressed. Additional research is needed to determine the best course of medication and psychotherapy treatment for this group of patients.

DISCLOSURES: Dr. Grant receives grant/research support from Forest Pharmaceuticals. Drs. Raymond and Coleman report no financial relationship with any company whose products are mentioned in this article, or with manufacturers of competing products.


  1. Black DW, Kehrberg LL, Flumerfelt DL, et al. Characteristics of 36 subjects reporting compulsive sexual behavior. Am J Psychiatry. 1997;154:243–249.
  2. Coleman E. Compulsive sexual behavior: new concepts and treatments. J Psychol Human Sex. 1991;4:37–52.
  3. Kafka MP, Hennen J. The paraphilia-related disorders: an empirical investigation of nonparaphilic hyper-sexuality disorders in outpatient males. J Sex Marital Ther. 1999;25:305–319.
  4. Coleman E, Raymond N, McBean A. Assessment and treatment of compulsive sexual behavior. Minn Med. 2003;86:42–47.
  5.  Diagnostic and statistical manual of mental disorders. 4th ed, text revision. Washington, DC: American Psychiatric Association; 2000.
  6. Odlaug B, Grant JE. Prevalence of impulse control disorders in a college sample. Presented at: Annual Meeting of the American Psychiatric Association; May 3-8, 2008; Washington, DC.
  7. Coleman E.  Is your patient suffering from compulsive sexual behavior? Psychiatr Ann. 1992;22:320–325.
  8. Schaffer SD, Zimmerman ML. The sexual addict: a challenge for the primary care provider. Nurse Pract. 1990;15:25–26,28,33.
  9. Grant JE, Levine L, Kim D, et al. Impulse control disorders in adult psychiatric inpatients. Am J Psychiatry. 2005;162:2184–2188.
  10. Emmons CA, Joseph JG, Kessler RC, et al. Psychosocial predictors of reported behavior change in homosexual men at risk for AIDS. Health Educ Q. 1986;13:331–345.
  11. Rompa D, Difranceisco W, Kelly J. Predictors of high-risk sexual behaviors in a sample of HIV-positive men who engage in sex with men. Abstracts of the 12th World AIDS Conference; June 28-July 3, 1998; Geneva, Switzerland.
  12. Raymond NC, Coleman E, Miner MH. Psychiatric comorbidity and compulsive/impulsive traits in compulsive sexual behavior. Compr Psychiatry. 2003;44:370–380.
  13. Raymond NC, Lloyd MD, Miner MH, et al. Preliminary report on the development and validation of the Sexual Symptom Assessment Scale. Sexual Addiction and Compulsivity. 2007;14:119–129.
  14. Coleman E, Gratzer T, Nesvacil L, et al. Nefazodone and the treatment of nonparaphilic compulsive sexual behavior: a retrospective study. J Clin Psychiatry. 2000;61:282–284.
  15. Emmanuel NP, Lydiard RB, Ballenger JC. Fluoxetine treatment of voyeurism. Am J Psychiatry. 1991;148:950.
  16. Fedoroff JP. Serotonergic drug treatment of deviant sexual interests. Annals of Sex Research. 1993;6:105–121.
  17. Kafka MP. Paraphilia-related disorders: common, neglected, and misunderstood. Harv Rev Psychiatry. 1994;2:39–40.
  18. Kafka MP. Successful antidepressant treatment of nonparaphilic sexual addictions and paraphilias in men. J Clin Psychiatry. 1991;52:60–65.
  19. Kafka MP, Coleman E. Serotonin and paraphilias: the convergence of mood, impulse and compulsive disorders. J Clin Psychopharmacol. 1991;11:223–224.
  20. Kafka MP, Prentky R. Fluoxetine treatment of nonparaphilic sexual addictions and paraphilias in men. J Clin Psychiatry. 1992;53:351–358.
  21. Pearson HJ. Paraphilias, impulse control, and serotonin. J Clin Psychopharmacol. 1990;10:323.
  22. Perilstein RD, Lipper S, Friedman LJ. Three cases of paraphilias responsive to fluoxetine treatment. J Clin Psychiatry. 1991;52:169–170.
  23. Stein DJ, Hollander E, Anthony DT, et al. Serotonergic medications for sexual obsessions, sexual addiction, and paraphilias. J Clin Psychiatry. 1992;53:267–271.
  24. Wainberg ML, Muench F, Morgenstern J, et al. A double-blind study of citalopram versus placebo in the treatment of compulsive sexual behaviors in gay and bisexual men. J Clin Psychiatry. 2006;67:1968–1973.
  25. Reilly DR, Delva NJ, Hudson RW. Protocols for the use of cyproterone, medroxyprogesterone, and leuprolide in the treatment of paraphilia. Can J Psychiatry. 2000;45:559–563.
  26. Anton RF, Moak DH, Waid LR, et al. Naltrexone and cognitive behavioral therapy for the treatment of outpatient alcoholics: results of a placebo-controlled trial. Am J Psychiatry. 1999;156:1758–1764.
  27. Jaffe AJ, Rounsaville B, Chang G, et al. Naltrexone, relapse prevention, and supportive therapy with alcoholics: an analysis of patient treatment matching. J Consult Clin Psychol. 1996;64:1044–1053.
  28. Monterosso JR, Flannery BA, Pettinati HM, et al. Predicting treatment response to naltrexone: the influence of craving and family history. Am J Addict. 2001;10:258–268.
  29. O’Malley SS, Jaffe AJ, Chang G, et al. Naltrexone and coping skills therapy for alcohol dependence. A controlled study. Arch Gen Psychiatry. 1992;49:881–887.
  30. Roth AS, Ostroff RB, Hoffman RE. Naltrexone as a treatment for repetitive self-injurious behaviour: an open-label trial. J Clin Psychiatry. 1996;57:233–237.
  31. Marrazzi MA, Bacon JP, Kinzie J, et al. Naltrexone use in the treatment of anorexia nervosa and bulimia nervosa. Int Clin Psychopharmacol. 1995;10:163–172.
  32. Mitchell JE, Morley JE, Levine AS, et al. High-dose naltrexone therapy and dietary counseling for obesity. Biol Psychiatry. 1987;22:35–42.
  33. Grant JE, Kim SW. An open-label study of naltrexone in the treatment of kleptomania. J Clin Psychiatry. 2002;63:349–356.
  34. Grant JE, Potenza MN, Hollander E, et al. Multi-center investigation of the opioid antagonist nalmefene in the treatment of pathological gambling. Am J Psychiatry. 2006;163:303–312.
  35. Kim SW, Grant JE, Adson DE, et al. Double-blind naltrexone and placebo comparison study in the treatment of pathological gambling. Biol Psychiatry. 2001;49:914–921.
  36. Kim SW. Opioid antagonists in the treatment of impulse-control disorders. J Clin Psychiatry. 1998;59:159–164.
  37. Grant JE, Kim SW. A case of kleptomania and compulsive sexual behavior treated with naltrexone. Ann Clin Psychiatry. 2001;13:229–231.
  38. Raymond NC, Grant JE, Kim SW, et al. Treatment of compulsive sexual behaviour with naltrexone and serotonin reuptake inhibitors: two case studies. Int Clin Psychopharmacol. 2002;17:201–205.
  39. Ryback RS. Naltrexone in the treatment of adolescent sexual offenders. J Clin Psychiatry. 2004;65:982–986.
  40. Coleman E. Treatment of compulsive sexual behavior. In: Rosen RC, Leinblum SR, eds. Case studies in sex therapy. New York, NY: Guilford Press; 1995:333–349.
  41. Guy W. ECDEU assessment manual for psychopharmacology. Rockville, MD: National Institute of Mental Health; 1976:218-222. US Department of Health, Education and Welfare publication (ADM) 76-338.
  42. Kim SW, Grant JE, Adson DE, et al. A preliminary report on possible naltrexone and nonsteroidal analgesic interactions. J Clin Psychopharmacol. 2001;21:632–634.
  43. Kalyanaraman N, Remmel RP. Identification of naltrexone-glutathione conjugates. Drug Metab Rev. 2004;36:170.
  44. Kalyanaraman N, Remmel RP. Inhibition of naltrexone glucuronidation by nonsteroidal anti-inflammatory drugs (NSAIDs). Drug Metab Rev. 2003;35:57.
  45. Kim SW, Grant JE, Yoon G, et al. Safety of high-dose naltrexone treatment: hepatic transaminase profiles among outpatients. Clin Neuropharmacol. 2006;29:77–79.

CORRESPONDENCE Nancy C. Raymond, MD Department of Psychiatry University of Minnesota Medical School F282/2AW 2450 Riverside Avenue Minneapolis, MN 55454 USA E-MAIL: raymo002@umn.edu