N -Acetylcysteine as Treatment for Self-Injurious Behavior in a Child with Autism

To The Editor:

Self-injurious behaviors (SIB), including skin picking/scratching, self-biting, and head banging have been reported to occur in as many as 50% of children with autism spectrum disorders (ASD) (Rojahn 1994; Richards et al. 2012). More than 14% reportedly exhibit “severe” SIB, which indicates that the behavior results in functional impairment or life-threatening injury (Baghdadli et al. 2003; Duerden et al. 2012). An inverse relationship exists between the presence SIB and the severity of intellectual impairment (Berkson 1983; Rojahn and Esbensen 2002; Symons et al. 2005; Hillery and Dodd 2007), and SIB frequently co-occurs in many syndromes often associated with intellectual/developmental disabilities, such as Lesch–Nyhan syndrome (Nyhan and Wong 1996), Rett syndrome (Deb 1998), Prader–Willi syndrome (Dykens and Kasari 1997), and Fragile X syndrome (Symons et al. 2003). Numerous hypotheses regarding the etiology of SIB have emerged over the years, including (but not limited to) SIB as a symptom of physical discomfort/organic illness, the result of a neurochemical imbalance, a learned behavior, a form of communication, or a form of self-stimulation (Rana et al. 2013).

SIB is often approached and treated as a discrete behavior, independent from other maladaptive behaviors (Bodfish and Lewis 2002); however, self-injury in ASD is generally associated with other impairing symptoms, including high levels of impulsivity, irritability, and stereotypies, paired with lower intellectual functioning (Matson et al. 2008; Richman et al. 2013). SIB in children with ASD are commonly targeted for treatment by means of pharmacology (Soorya et al. 2008), communication strategies (Johnson and Rodriguez 2013), and behavioral modification (Minshawi 2008 Devlin et al. 2011). To date, no single intervention strategy for SIB has emerged as consistently effective (King 2000; Lang et al. 2010; Oliver and Richards 2010).

Limited data support the use of any medication class specifically for SIB, whether in autism or other developmental disabilities (King 2000; Rana et al. 2013). A number of medications are sometimes used to try to address self-injury, including serotonin reuptake inhibitors (SRIs), atypical antipsychotics, α agonists, and opioid antagonists. The most frequently used medications overall for individuals with ASD have been SRIs (Aman et al. 2005), although these medications have shown ambiguous efficacy in children with ASD (McDougle et al 2000; King et al. 2009; Williams et al. 2013), as well as in pathological skin picking (Simeon et al 1997; Bloch et al 2001; Arbabi et al. 2008). Atypical antipsychotics have been the second most frequently used pharmacological intervention for children with ASD (Aman et al. 2005). Although efficacy in the treatment of irritability, and compulsive and disruptive behaviors has been demonstrated in large-scale, randomized controlled trials (RCTs), little evidence exists for specific benefit for SIB (McCracken et al. 2002; Shea et al. 2004; McDougle et al. 2005; Marcus et al. 2009; Owen et al. 2009). Atypical antipsychotics have also not demonstrated efficacy in reducing SIB in adults with intellectual disability, with or without ASD (Ruedrich et al 2008). Further, a systematic review of the literature demonstrated serious adverse effects, the most common of which is significant weight gain (Politte and McDougle 2013). Although efficacy of opioid antagonists (predominately naltrexone) had been suggested for self-injurious/skin-picking behaviors in small controlled trials (King 2000), larger controlled trials have not shown an advantage of naltrexone over placebo (Campbell et al. 1993; Willemsen-Swinkels et al. 1995). Case reports in ASD and in nonhuman primates suggest that α agonists, including clonidine and guanfacine, could provide benefit for SIB (Macy et al. 2000; Symons et al. 2004), but larger open label guanfacine data and a small, randomized crossover trial of clonidine in ASD were primarily focused on hyperactivity symptoms (Jaselskis et al. 1992; Posey et al. 2004; Scahill et al. 2006).

N-acetylcysteine (NAC) has recently received attention in the treatment of grooming disorders as well as in ASD. NAC is a United States Food and Drug Administration (FDA) approved antioxidant, most commonly used for the treatment of acetaminophen/paracetamol overdose. NAC is also used to restore hepatic concentrations of cysteine, which lead to glutathione synthesis (Atkuri et al. 2007). In addition to these uses, NAC is also considered an emergent treatment for trichotillomania (repetitive hair pulling), with one positive RCT of a dissolving preparation (PharmaNAC) in adults (Grant et al. 2009) and one negative trial of a different NAC preparation in children (Bloch et al. 2013). There is also one case report of NAC 1800 mg/day showing benefit for skin picking (Odlaug and Grant 2007). NAC has also recently been considered for symptoms associated with ASD. Hardan and colleagues (2012) reported an RCT demonstrating significant decrease in irritability/agitation (including SIB) and repetitive behaviors in children with ASD. Based upon the lack of severe associated adverse events and the data in trichotillomania, skin picking, and ASD, NAC seemed like a reasonable option for a young boy with ASD and severe, disfiguring skin gouging behavior.

Case Report

VJ, a 4-year-old boy with a history of ASD, was referred to the Vanderbilt Treatment Resistant Autism Consultation Clinic for severe SIB that had shown minimal response to prior treatments. VJ's cognitive functioning was reportedly in the average range, with demonstrated strength in spatial reasoning, which fell into the above average range. His SIB presented as disfiguring face scratching in the form of a characteristic, stereotyped twisting motion, and reportedly began spontaneously when he was 6 months of age. This behavior, which occurred throughout the day, but especially upon waking or potentially during sleep, was largely concentrated on or around the nose and had resulted in bleeding, erosions, excoriations, and scarring. There was no history of aggression or harm directed toward others. VJ also had a history of significant hyperactivity, impulsivity, and irritability resulting in daily tantrums. His medication regimen at the time of evaluation included gabapentin (250 mg/5 mL) 2 mL twice daily, clonidine 0.25 mg every morning and 0.5 mg every evening, sertraline 5 mg once daily (QD), cetirizine 10 mL QD, and hydroxyzine 10 mg three times daily. VJ was also on a no-wheat diet, because of a reported increase in scratching behavior following consumption of wheat.

Some improvements in the self-injury had been noted since beginning hydroxyzine, with reduced scratching and picking during the day; however, self-injury persisted when he was upset or at night. VJ's irritability and tantrums had reportedly improved with the addition of sertraline; although no improvement in SIB was noted. Gabapentin, which was started for possible neuropathy, had resulted in little improvement in the scratching behavior. Previous attempts at treating the SIB included plaster casts on the arms for several months, sewing his sleeves closed, and straight-arm splints. He was able to escape from most attempts to block the SIB except for the plaster casts, which allowed his excoriations to fully heal. Unfortunately, he returned to the gouging behavior soon after the casts were removed. The family also reported that a previous steroid injection led to a dramatic decrease in the SIB; although subsequent trials of oral steroids resulted in no improvement.

After a discussion of various behavioral and medication options, NAC (PharmaNAC formulation) was initiated at 450 mg QD and titrated up by 450 mg each week to 1800 mg. VJ showed significant decrease in his self-injurious symptoms after reaching the 1800 mg dose. His dose of gabapentin was also increased to three times daily shortly before this titration period; therefore, it was initially unclear which medication led to the benefit. He ran out of NAC and stopped taking it for 10 days while awaiting shipment. The cessation of NAC resulted in a rapid resumption of his self-injury. When the NAC was resumed at 1800 mg QD, he again showed a significant decrease in SIB, so that all of his excoriations had completely healed over the course of 2 months. Although he was no longer gouging/scratching his face, his parents reported that some mild rubbing behavior persisted in the same pattern, primarily when he was sleepy.

Discussion

Although it is difficult to draw broad conclusions from a single case, the relationship between NAC and the patient's cessation of severe SIB is certainly suggestive. The rapid re-occurrence of VJ's SIB during the interruption of NAC, followed by decrease in the behavior after resuming the drug, seems to support the efficacy of NAC for this child. This outcome is also supported by another case report, in which NAC was used in successfully reducing problematic symptoms in an 8-year-old child with ASD, including aggression, hyperactivity, and repetitive nail biting (Ghanizadeh and Derakhshan 2012).

Two potential mechanisms of action have been proposed for NAC in psychiatric disorders. First, NAC can modulate glutamatergic function. It is metabolized to cystine, a substrate for the glial glutamate-cystine antiporter. The glial uptake of cystine by this antiporter leads to the reverse transport of glutamate into the extracellular space. In turn, stimulation of the type 2/3 metabotropic glutamate receptors inhibits the neuronal release of glutamate, which results in an overall reduction in synaptic glutamate (Baker et al. 2002; Moran et al. 2005; LaFleur et al. 2006). Second, the cysteine provided by NAC is also a precursor in the synthesis of glutathione, the predominant antioxidant in the brain (Dean et al. 2011). It is not yet clear how these mechanisms of action may account for improvements in compulsive behavior.

Further research is needed to understand the benefits and potential risks associated with NAC for individuals with ASD and SIB. Ideally, this would be achieved in a randomized, placebo-controlled study. One potential challenge is the contrast between VJ's self-injurious behavior, which is similar to skin picking or hair-pulling in its stereotyped, repetitive motion, and other forms of SIB such as head banging. The short-term return in SIB symptoms after NAC withdrawal suggests that a crossover design with a short washout period may allow subjects to serve as their own controls, increasing power to detect a specific drug benefit in the context of diverse forms of SIB. Results of such a study could then be used to identify a subpopulation most likely to respond to NAC versus placebo in a parallel group design. If NAC is found to be effective for SIB in ASD, its lack of significant adverse events in previous RCTs suggest that it will be better tolerated than current medications, such as risperidone or aripiprazole, that are approved for treatment of irritability/agitation.