Naltrexone as an Adjunctive Treatment for Adolescents With Deliberate Self-Harm Behavior and Cannabis Use: A Case Series

AFFILIATIONS:
¹Medical Student, Physician Assistant Online Program, Yale School of Medicine, New Haven, CT
²Psychiatry Department, Larkin Community Hospital, South Miami, FL
³Child and Adolescent Psychiatry Department, Family Center for Recovery, Lantana, FL

CITATION:
Wei A, Navarro M, Yeager V, Goubran B, Moran RA. Naltrexone as an adjunctive treatment for adolescents with deliberate self-harm behavior and cannabis use. Consultant. 2023;63(2):e1. doi:10.25270/con.2022.11.000006.

Received March 28, 2021; Accepted April 26, 2021; Published online November 29, 2022.

DISCLOSURES:
The authors report no relevant financial relationships.

CORRESPONDENCE:
Annie Wei, PA-S3, Yale School of Medicine, PO Box 208083, New Haven, CT 06520-8083 (annie.wei@yale.edu)

ABSTRACT: Nonsuicidal self-injury (NSSI) or deliberate self-harm is defined as the volitional, direct injury of the body tissue without suicidal intent. Substance abuse complicates the presentation because it masks the precipitating psychiatric underpinning, and some data show an association between NSSI and cannabis use. Antidepressants and antianxiety medications have been favored for tackling the primary psychiatric disease underlying NSSI; however, there are limited data for the utility of naltrexone as adjunctive therapy for self-harming behavior with co-occurring substance abuse, particularly in genetically female adolescent patients. Three genetically female adolescent patients were referred for self-harming behavior all with concurrent cannabis use. Each patient had a history of refractory self-harming behavior and substance abuse with nonadherence to nonpharmacological and pharmacological treatments. Naltrexone was added to each patient's treatment regimen. On follow-up, all 3 patients had maintained compliance throughout inpatient and outpatient treatments, resulting in remission of both substance use and self-harming behaviors. The case series highlights the potential benefits of naltrexone as an adjunctive therapy for genetically female adolescent patients with self-harming behaviors and co-occurring substance abuse disorders. Further research is warranted to establish the benefits of naltrexone.

Key words: naltrexone, nonsuicidal self-injury, NSSI, self-harm, adolescents, addiction, craving, substance use disorders, self-injury, risk factor

Introduction

Self-harm is a nonfatal act that involves the deliberate destruction of body tissue or the deliberate ingestion of a noningestible substance, illicit drug, or substance in an attempt to harm one’s self.^1,2  Self-harm, without the intent of suicide, is the most frequent reason for psychiatric emergency room visits.³ However, the exact rates are variable due to the inconsistent definitions used to describe deliberate self-harm (DSH), self-harm, self-mutilation, self-injury, parasuicide, attempted suicide, and nonsuicidal self-injury (NSSI).⁴ There are many reasons why people self-harm, including temporary alleviation of negative emotions,^4,5 expression of self-directed anger or self-punishment, a desire to influence others, or a compulsion to produce a physical sign of emotional distress.⁵

Both suicidal and NSSI are becoming increasingly prevalent in adolescents.^4,5 Studies have estimated an overall lifetime prevalence of self-harm behaviors to be around 16.9%, with most starting during adolescence; these behaviors are most often associated with comorbid psychiatric diagnoses, such as mood disorders and substance use disorders.^2,4-7 Genetically female adolescents are more likely to engage in these behaviors than their male counterparts, with a risk ratio of 1.72 (95% confidence interval [Cl], 1.57-1.88) and a higher incidence of long-lasting self-injury behavior.⁸ Self-harm typically begins around the age of 12 to 13 years, peaks around age 15 to 16 years, and decreases by adulthood.^2,4,9 This behavior is also the most prominent predictor of suicide in adolescents as it increases both the risk for suicidal ideation and the ability to act on the ideation.^2,5,7,10 Adolescent self-harm, both sporadic and repetitive, is associated with anxiety, depression, and substance use during adolescence and into adulthood.^2,8,10,11 Attempted suicide tends to happen during or immediately after substance use.^2,12

Studies have evaluated the efficacy of various psychotherapeutic interventions for NSSI, with adolescent dialectical behavioral therapy (DBT) and family-centered therapy showing the most promise.^9,13 Some have suggested that psychosocial treatment as usual may be just as effective as DBT and family-centered therapy, and  front-loading treatments together with long-term care may also be beneficial.^9,12 Although persistence of self-harm reduction varied among studies of these interventions, it was typically sustained for 6 to 12 months.¹⁴ The efficacy of psychotherapy for adolescents with self-harm and co-occurring mood disorders or substance use has been limited.⁹

There is currently no consensus on first-line pharmacological treatment for patients exhibiting self-harm behaviors. Small studies have suggested there may be some benefit to using atypical antipsychotics, such as olanzapine in intellectually disabled patients, aripiprazole in patients with borderline personality disorder (BPD), and ziprasidone in adolescents.^14,15 Selective serotonin reuptake inhibitors (SSRIs), such as fluoxetine, as well as serotonin-norepinephrine reuptake Inhibitors (SSNRIs) and N-acetylcysteine have also been shown to be beneficial in small studies in adolescents with clinical depression.^14,15

Naltrexone has the highest affinity for mu-opioid receptors and is a known nonselective pure opioid antagonist.¹⁶ It has been approved by the Food and Drug Administration (FDA) in the management of alcoholism and opioid use disorders, particularly due to its role on the brain’s reward system through the nucleus accumbens.¹⁶ Furthermore, naltrexone has been reported in multiple small studies to be an effective pharmacologic option in reducing self-injurious behavior, particularly in patients with bipolar disorder and intellectual/developmental disabilities.^17,18

CASE REVIEWS

This case series reports on three genetically female adolescents who were referred for evaluation of repetitive episodes of self-harm and concomitant substance use. In each case, adjunctive naltrexone was added to residential and step-down pharmacological and nonpharmacological treatment.

Case 1. A 16-year-old white transgender (female-to-male) patient presented to a long-term adolescent residential treatment facility for management of bipolar II disorder, gender dysphoria, cannabis use, attention deficit hyperactivity disorder (ADHD), and suicidal ideation. The patient had a 5-year history of repetitive self-harm with scissors and utility knives, and he reported triggers related to his self-harm as “relationships, being misgendered, and [guilt over his] father’s suicide.” The patient had been hospitalized three times before the current admission for self-harm and suicidal ideation while on a home medication regimen that included clonidine 0.15 mg, trazodone 50 mg, testosterone 50 mg, and lamotrigine 75 mg. Reported substance use on admission included daily cannabis use, approximately 1 "dime bag" (approximately 1 gram) daily for 10 months, “laughing gas” twice when he was 13 years old, and sporadic use of alprazolam 3 to 5 mg for the last 4 years. He had also been consuming alcohol occasionally since childhood; he reported drinking approximately four shots of hard liquor or one to two beers every two weeks in recent months.

At admission, the patient was started on 50 mg of naltrexone once daily. All of the patient’s home medications were continued. Over the next month, nonpharmacological therapies during the treatment course included cognitive behavioral therapy (CBT), DBT, 12-step facilitation, contingency management, family systems therapy, motivational enhancement therapy, psychodynamic psychotherapy, and motivational interviewing. The patient reported no substance use cravings or urges to self-harm throughout the treatment course. No side effects of naltrexone were noted. The patient with adhered to the naltrexone regimen throughout his treatment course.

After 16 days of residential treatment, the patient was discharged to the care of a community psychiatrist for management. He went on to receive 11 days of intensive outpatient treatment and 45 days of outpatient treatment. The total follow-up time on naltrexone was 72 days.

Case 2. A white, genetically female 16-year-old patient presented to a long-term adolescent residential treatment facility for recurrent major depressive disorder, severe cannabis use disorder, and ADHD (predominantly inattentive type). The patient had a 3-year history of self-harm behavior in which she used razors to cut her upper arm and occasionally her leg. The patient reported no suicidal ideation on admission but stated that she had completed a questionnaire the previous day on which she marked “suicidal thoughts for most days out of a month”; this disclosure was the reason for her current admission. On further questioning, the patient revealed that if she ever wanted to die, she had a plan to use a gun at her aunt’s house. Triggers and stressors included her “parents taking everything away,” school, family, and romantic relationships. The patient reported three involuntarily admissions to the hospital for approximately 3 days duration per admission. The last admission was within 3 months of current treatment.  In one self-harming event prior to admission, 3 days after she cut her arm,  the patient reported that she “was not aware it was the sharper end of the razor and cut too deep,” when describing the incident. Per her mother, her psychotropic medication regimen at the time of admission included fluoxetine, buspirone, and oxcarbazepine, but the dosages were unknown. Reported substance use on admission included cannabis use for 2 years, 20 puffs daily; casual tobacco cigarette use for 2 years, five puffs socially; and alcohol consumption, 2 beers socially.

After admission, the patient reported 5 days of cannabis cravings and 6 days of reported self-harm cravings, culminating in one event of self-harm inflicted with a pencil to her thigh, wrist, and ankles. At this time, 20 days after admission, the patient was started on naltrexone 50 mg and venlafaxine 75 mg once daily for depression. The naltrexone dose was increased to 100 mg 43 days later due to 9 reported days of self-harm cravings, 6 reported days of cannabis cravings, and one self-harm episode in which she cut her thighs with a razor. Immediately following the dose increase, the patient reported 12 days of self-harm cravings and 12 days of reported cannabis cravings; during this time, she had two self-harm episodes in which she again cut her thighs with razors, and she also had one episode in which she violently punched the wall.

The patient remained on the 100 mg dose of naltrexone for approximately 7 months. Additionally, per report, after approximately 3 months aripiprazole 2 mg and lisdexamfetamine 30 mg (for ADHD) were added to her regimen, and the venlafaxine dose was increased to 300 mg. On this regimen, the patient's coping mechanisms improved, and reported self-harm and cannabis cravings decreased. After 136 days with no self-harm events and 68 days with no episodes of explosive behavior, the patient's naltrexone dose was decreased to 50 mg/day. Twenty-five days later, the patient stopped lisdexamfetamine due to side effects of appetite loss and insomnia, which resolved after discontinuation. The only adverse reaction to naltrexone reported by the patient was dizziness on the first day.

Non-pharmacological therapies during the patient's treatment course included CBT, DBT, 12-step facilitation, contingency management, family systems therapy, motivational enhancement therapy, psychodynamic psychotherapy, and motivational interviewing. The patient with adhered to naltrexone therapy throughout her treatment course. The total follow-up time on naltrexone was 223 days. Self-harm and cannabis cravings were well controlled at time of discharge. The patient completed 129 days of residential treatment, 91 days of intensive outpatient treatment, and 98 days of outpatient treatment.

Case 3. A white, genetically female 15-year-old patient presented to a long-term adolescent residential treatment facility because of “self-harm, suicide, that kind of stuff,”severe major depressive disorder without psychotic features, severe cannabis use disorder, and social anxiety disorder. Upon admission, the patient reported a 4 to 5year history of self-harm behaviors using razors to cut her arms or legs, "if there was no space available [in her arms].” She reported using cannabis for the past 3 years, taking two hits off a bowl once or twice a month; she also reported social tobacco use, one cigarette at a time, and one isolated episode of alcohol consumption.

On physical examination, bilateral healed abrasions were noted on the upper extremities. The patient reported cutting herself nearly every day, usually one time per day, to “project mental pain into physical pain” and stop herself from crying. The last known cutting event occurred approximately 2 weeks ago, and consequently her mother discarded all razors from her home. The patient reported her stressors and triggers included family conflict and a history of physical, mental, verbal, and emotional abuse by her father, who no longer resides in the home. In addition to self-harm behavior, the patient reported multiple attempted suicide events using “pills” and one attempt by hanging. The patient self-reported that they had previously engaged in individual therapy which was unsuccessful in stopping self-harming behavior. The type of therapy and duration of therapy was unknown. No previous hospitalizations were reported. Two days ago, the patient was prescribed fluoxetine by a primary care physician but the medication was never started. The patient did not want to take the medications at that time.

Five days after admission, the patient was started on sertraline 25 mg for depression and anxiety. This dosage was titrated to 100 mg over 37 days. During this time, the patient reported 17 days of self-harm cravings, 1 event of NSSI without trigger or stressor, and 2 substance cravings. On day 25 of treatment, naltrexone 50 mg was initiated; this dose was increased 17 days later to 100 mg because of 8 days of self-harm cravings and one unwitnessed self-harm event. There were self-harm craving reports on 2 other days shortly after the naltrexone dose change but no self-harm events. No substance cravings were reported after the naltrexone dose was increased. The patient had no adverse reactions and adhered to treatment for the duration of her stay. Nonpharmacological therapies throughout the treatment course included CBT, DBT, 12-step facilitation, contingency management, family systems therapy, motivational enhancement therapy, psychodynamic psychotherapy, and motivational interviewing.

At the time of discharge following 28 days of residential treatment, the patient's self-harm and substance cravings were well controlled. The total follow-up time on naltrexone was 150 days. The patient was discharged to a community psychiatrist and received 21 days of intensive outpatient treatment followed by 127 days of outpatient treatment.

Discussion

Recent studies have suggested that self-harm is an addictive behavior, similar to gambling or video gaming, that has both neurobiological and psychological mechanisms.¹⁹ As in substance use disorders, the neurobiological addictive pathways of self-harm have been thought to be mediated by the endogenous opioid reward pathways, particularly the delta-receptor–mediated Met-enkephalins and mu-receptor–mediated beta-endorphins that increase dopaminergic activity and euphoria.^3,9,19,20 Like substance use, self-harm has been conceptualized as an addictive behavior with positive reinforcements, such as endogenous opioid release.^5,21 Repeated self-harm causing endogenous opioid release may affect the brain’s mesocortical dopamine reward system, leading to tolerance and addiction, and studies have suggested that those with self-harm report meeting addictive criteria including loss of control over self-harm behaviors, tolerance, and craving.^19,21 Several reports have found persistent changes in opioid receptor binding sites in patients with a history of chronic, severe childhood stress, trauma, and self-harm. This opioid deficiency may lead to self-harm behaviors, and these patients may require increased levels of endogenous opioids to restore homeostasis.³

Psychological addictive mechanisms of self-harm, or the “sensitizing hypothesis,” involve automatic positive reinforcement by relieving emptiness and ameliorating emotional tension and pain.¹⁹ Elevated endogenous opioid levels in response to self-harm may affect addictive behaviors by relieving the psychological pain that triggers self-harm. Similar to environmental triggers that precipitate drug use, it is thought that the return of life stressors and emotions trigger self-harm, increasing the risk of relapse.¹⁹

Besides the addiction model, the “pain hypothesis” suggests that people with self-harm are desensitized to pain, which perpetuates self-harm behaviors.^22,23 This may be related to an altered hypothalamic–pituitary–adrenal (HPA) axis, increased endogenous opioid, and decreased cortisol and adrenocorticotropic hormone (ACTH) levels, which leads to increased pain thresholds and a blunted stress response.^3,19,23 In addition to physiological mechanisms, psychological mechanisms and emotional dysregulation may play a role in increased pain tolerance as well.²⁴

Naltrexone, a pure opioid antagonist, is currently only FDA-approved to treat alcohol and opioid use disorder but has also shown efficacy in decreasing self-harm, cravings, and cannabis use.^17,25 Naltrexone has a high affinity for mu-opioid binding sites and a lesser affinity for kappa- and delta-opioid sites, which are thought to also modify the HPA axis. Naltrexone also affects the cannabinoid system and is effective in decreasing the frequency of cannabis cravings.²⁵ By inactivating the mu-opioid receptor, naltrexone is hypothesized to impair the alcohol- and opioid-induced dopamine release in the striatum (more specifically the nucleus accumbens) and reduce cravings.²⁶ Its role in attenuating addictions may also be partially explained by its ability to block the opioid-mediated inhibition of the HPA axis, thereby increasing cortisol levels and decreasing alcohol consumption through reduced cravings.^19,20 By normalizing the HPA axis and diminishing opioid- and cannabinoid-mediated reward pathways and pain tolerance, naltrexone may act on both the addiction and pain hypotheses of self-harm pathophysiology and therefore be an effective treatment for self-harm as well as cannabinoid use.

Although a number of case reports and small studies have suggested that  naltrexone may be efficacious for treating self-harm,^17,18 these studies have primarily involved individuals with neurodevelopmental disorders (NDD) and BPD. In patients with NDD, self-harm is associated with higher degrees of intellectual disability and may be related to environmental impoverishment, behavioral maladaptation, hypersensitivity to stimuli, altered neurotransmitters, and opioid system disruptions.²⁷ Patients with BPD, meanwhile, may demonstrate self-harm due to impulsivity and mood instability.¹⁸ The proposed etiologies of impulsivity and self-harm in patients with BPD include decreased serotonin levels, dysregulated dopaminergic activity, and altered opioid systems with instability and either increased or excessive activity.²⁸

Studies on the efficacy of naltrexone for self-harm in patients with NDD have yielded conflicting data. A few randomized control studies reported that naltrexone was not effective in reducing self-harm in adults with self-harm. Conversely, there were other small studies and case reports of naltrexone responders.²⁷ It has been suggested that the underlying neurobiological mechanisms causing behavioral symptoms (including self-harm) in patients with NDD may differ from those found in individuals without NDD, thus explaining why naltrexone appears to be less effective in patients with NDD.²⁸

The current FDA-approved pharmacologic treatment for patients with NDD, specifically autism, and problematic behaviors (eg, irritability, aggression, tantrums, self-harm) are the second-generation antipsychotics (SGAs) risperidone and aripiprazole.²⁷ However, side effects, such as akathisia, which has been associated with increased risk of suicidal ideation, and significant drug-drug interactions limit the clinical use of these SGAs. Other medications, such as clonidine, N-acetylcysteine, riluzole, mirtazapine, may be efficacious, but studies have been limited to case reports and small controlled studies; most of these reports were not focused on self-harm alone. Naltrexone has been recommended to be a last-line, treatment-refractory option for patients with NDD.²⁷

Evidence for naltrexone for self-harm in patients with BPD is also limited to small studies and case reports. A small open-label trial in female adults with BPD and history of DSH reported termination of self-harm behaviors in six out of seven patients, and all patients reported the loss of analgesia and dysphoria during self-harm after starting naltrexone.²⁹ Open-label studies have used antidepressants, such as fluoxetine, and case reports have suggested the efficacy of SGAs, mood stabilizers, and alpha-2 agonists to treat BPD-associated DSH.²⁸ There are no FDA-approved treatments, but SSRIs continue to be suggested as first-line treatment, followed by SGAs.²⁸

To the authors' knowledge, limited research has been reported on the efficacy of naltrexone on genetically female adolescents with history of self-harming behavior co-occurring with substance use disorders who are treated with long-term residential and step-down treatment and nonpharmacological treatments. The current study population consisted of patients with histories of both substance use and self-harm, which may suggest that their self-harm behaviors could be driven by addictive mechanisms and the reward pathway, particularly through the release of endogenous opioids. As substance use and self-harm may have similar neurobiological and psychological mechanisms, it follows that adolescents with substance use may be more likely to engage in self-harm. Therefore, treatments for one may also affect the other.

Past research has suggested that substance use disorders and self-harm may be related to dysregulated pro-opiomelanocortin (POMC), the precursor to both beta-endorphins and ACTH.^30,31 Patients with certain POMC polymorphisms may respond to naltrexone for self-harm as naltrexone addresses both the opioid and HPA axis pathways of self-harm.^30,31 Although the presence of the POMC gene was not identified in this patient population, this case series may support the efficacy of naltrexone to treat opioid- and HPA axis–mediated addictive self-harm.

The patients in this case series demonstrated that they were refractory to other forms of nonpharmacological and pharmacological treatment before initiation of naltrexone. Two of the three had previous psychiatric hospitalizations and were receiving SSRIs, SGAs, mood stabilizers, and/or clonidine, and all had participated in nonpharmacological therapy modalities, such as CBT and DBT, before initiation of naltrexone for self-harm. Non-pharmacological treatments, including DBT, have demonstrated efficacy in adolescents exhibiting self-harm behaviors; therefore, naltrexone may have provided a synergistic effect with these extensive residential and intensive outpatient nonpharmacological treatments, influencing their outcomes. The effects of naltrexone on these three patients may have also provided an adjunctive therapy effect with other pharmacological treatments.

Study Limitations

Although this case series may suggest the efficacy of naltrexone to treat self-harm and supports opioid- and HPA axis–mediated mechanisms of self-harm, it is limited by its study design and multiple concomitant treatment regimens. Due to the study design and concurrent therapies, this paper does not isolate the contributing factors in remission of the self-harming behaviors. Further research is needed to understand the self-harming behavior within adolescents’ females with co-occurring substance use disorders. 

Conclusions

This case series described three genetically female adolescents with histories of substance use and self-harm who responded to naltrexone as an adjunctive treatment to synergistic pharmacological and nonpharmacological therapeutics. Although there is no consensus on the physiological and psychological mechanisms of self-harm, patients with concomitant substance use may engage in an addictive form of self-harm and thus benefit from naltrexone. As self-harm is a prevalent issue among adolescents, these cases highlight the need for further studies on the mechanisms of self-harm as well as the efficacy of naltrexone to treat addictive self-harm behaviors