Potential Risks of Poorly Monitored Ketamine Use in Depression Treatment

At the time of his initial presentation to a tertiary medical center in 2012, “Mr. A” was a 52-year-old divorced man with a 30-year history of recurrent major depressive disorder, persistent depressive disorder, and a remote history of outpatient treatment for alcohol use disorder as a young adult. His first episode of major depression occurred at age 22 in association with a suicide attempt (he jumped off a four-story building) and subsequent hospitalization. Mr. A’s medical history was significant for corrected hypothyroidism. The prospective course of illness encompassed four subsequent psychiatric admissions at our medical center, annually from 2012 to 2014, illustrating clear treatment-resistant depression and a reemerging pattern of substance misuse (alcohol, benzodiazepines, and ketamine).

Mr. A’s admission in the summer of 2012 focused on consultation for depression and consideration of ECT. Prior medication trials of optimal dose and duration included trazodone, escitalopram, bupropion, and mirtazapine with lamotrigine and aripiprazole augmentation strategies. Mr. A reported negligible alcohol consumption (a single drink on rare occasions). He received seven bilateral ECT treatments, and his symptoms improved; his score on the Hamilton Depression Rating Scale declined from 37 on admission to 8 at discharge. He was discharged after 18 days on mirtazapine monotherapy with recommendations to maximize his dosing of mirtazapine and to add lithium for augmentation following his ECT course. He received three additional ECT treatments prior to self-discontinuing his ECT course and being lost to follow-up.

Mr. A’s 2013 admission, 15 months later, was for major depression with suicidal ideation with intent to overdose on prescribed medications. Mr. A reported drinking 1–2 beers a day and misuse of lorazepam (he was taking 5 mg daily, although it was prescribed at 1 mg twice daily). However, on admission, his urine drug abuse test was negative, including absence of alcohol and benzodiazepines. During hospitalization, Mr. A reported that earlier in the year, he participated in an out-of-state research study in which he received a single intravenous ketamine infusion, which he reported provided antidepressant benefit for 4–5 days. Because of his perceived mood improvement, he left the study and sought out a ketamine prescriber. A neurologist in another state prescribed intranasal ketamine for depression, and Mr. A reported taking the prescribed ketamine intranasally with continued benefit. In contrast to the previous admission, ECT was not instituted in 2013. Discharge recommendations after this 5-day admission included follow-up with outpatient prescribers, changing lorazepam to clonazepam at 1 mg twice daily, and abstinence from alcohol. He did not receive ketamine while hospitalized.

Mr. A’s next admission for major depression, 3 months later in early 2014, differed from previous admissions based on a reemergence of increasing alcohol use (he reported consuming four drinks daily). On admission, his urine drug abuse test was negative, including absence of alcohol. He admitted to using ketamine beyond the prescribed recommendations, although ketamine screening was not included in the urine drug abuse test. We contacted the out-of-state prescriber, who verified prescribing ketamine, 150 mg/mL, 0.5–1 mL intranasally every 4 hours as needed for depression. However, Mr. A reported using higher than prescribed doses of ketamine and more frequently (10–12 times a day) before admission, which he said he needed to relieve his depressive symptoms. He reported that the antidepressant benefit of ketamine lasted 2–3 hours after administration, but noted transient dissociative side effects lasting up to 20 minutes (“feeling like I am in a dreamy state” and “trippy effect … can be kind of intense”). He mentioned using intranasal ketamine in his car and stated that his mother had concerns about his driving during those times (“She thought I was swerving”). His family reported that ketamine was “ripping his life apart” and that he had “no control over it [ketamine use].” Mr. A met criteria for ketamine dependence based on his use of increased amounts of and increased tolerance to ketamine, driving while using despite concern from family, and loss of employment due to ketamine use. We discussed our concerns with the patient and his ketamine prescriber, who agreed to cease prescribing. Mr. A insisted that ketamine was the only prescribed substance that had helped his depression, and he reported his intention to seek more ketamine despite his treatment team’s recommendations. A petition for civil commitment was not supported by the court. During the hospitalization, we discussed alternative medication recommendations to consider, including ECT, which Mr. A. declined because of memory difficulties associated with his previous treatment. After he was discharged from a 7-day admission, Mr. A participated in an outpatient program, but he often appeared to be intoxicated in group sessions, and he acknowledged continued use of prescribed ketamine. He was occasionally asked to leave group sessions because of an inability to maintain sobriety from ketamine.

Mr. A’s fourth admission for depression, a month after the previous admission, was associated with a suicide attempt by alcohol and overdose of prescribed eszopiclone. A urine drug test was negative, including for alcohol. Mr. A reported that he had discontinued ketamine because he was unable to obtain the drug. He reported moderate alcohol consumption (1–2 beers two or three nights a week) but adamantly denied alcohol abuse or dependence. The treatment team again petitioned for civil commitment. A stay of commitment was granted, which meant that the court would not enforce commitment as long as Mr. A participated in recommended voluntary treatment, including outpatient chemical dependency evaluation and medication management.

We were notified of his death from a single-car crash 3 months later. Autopsy findings included a blood alcohol concentration of 0.133% (well over the 0.08% legal limit), tetrahydrocannabinol, and bupropion (prescribed); tests for ketamine and norketamine were negative. The family indicated the belief that Mr. A’s death was a suicide facilitated by alcohol.