We reviewed case reports submitted to FDA, conducted an observational study using data from FDA’s Sentinel System, and reviewed observational and animal studies in the published literature.
We evaluated reporting trends for all neuropsychiatric adverse events associated with montelukast use reported in the FDA Adverse Event Reporting System (FAERS) database from the date of FDA approval in February 1998 through May 2019. There was an increase in reporting of neuropsychiatric events around the time of the initial communications from FDA in 2008. Other increases in reporting were due to duplicate reports or foreign reports. Despite outside influences on reporting patterns, we continue to receive reports of serious neuropsychiatric events with montelukast.
We also performed a focused evaluation of completed suicides. Our analysis included only reports submitted to FDA, so there may be additional cases about which we are unaware. We identified 82 cases of completed suicide associated with montelukast, with many reporting the development of concomitant neuropsychiatric symptoms prior to the event. Forty-five cases were reported in patients older than 17 years, 19 cases were reported in those 17 years and younger, and 18 cases did not provide the age of the patient. The majority of the cases were reported by a family member or on social media. Most cases (48/82) did not contain sufficient information to evaluate the relationship between montelukast and the adverse events. The cases did not include key information such as the time to onset of the event, the use of concomitant medications, the presence of past or current comorbidities, including psychiatric illness; the degree of asthma control; and the presence of other risk factors for the events. Of the remaining 34 cases that were better documented, many contained additional risk factors that may have contributed to the suicide such as the use of medications or presence of comorbidities associated with increased risk for self-harm or behavioral disturbances. Six cases specifically reported concerns about not receiving education from a health care professional regarding the potential for neuropsychiatric side effects.
Using data from the FDA’s Sentinel System from January 1, 2010, to September 30, 2015, we investigated if there is an increased risk of hospitalized and treated outpatient depressive disorders, self-harm, and suicides associated with montelukast use for asthma compared to inhaled corticosteroids (ICS). We also evaluated if the risk of neuropsychiatric adverse events with montelukast compared to ICS was modified by the 2008 FDA Early Communication and changes to the montelukast prescribing information, or was affected by age, sex, and/or psychiatric history. Patients (n=457,377) 6 years and older diagnosed with asthma and exposed to montelukast or ICS were matched 1:1 on propensity scores. The risk of inpatient depressive disorder associated with montelukast use compared to ICS was not significant (overall HR: 1.06; 95% CI: 0.90-1.24). There were no significant risks among males, females, patients 12 years and older, patients with a psychiatric history, or after the 2008 FDA communication and prescribing information changes. Exposure to montelukast was also not associated with self-harm (HR:0.92; 95% CI: 0.69-1.21) or modified self-harm (HR: 0.81; 95% CI: 0.63-1.05). Four suicides occurred (two exposed to montelukast, two exposed to ICS), all in patients older than 18 years with a psychiatric history. Exposure to montelukast was significantly associated with a decreased risk of treated outpatient depressive disorder (overall hazard ratio [HR]: 0.91; 95% confidence interval [CI]: 0.89-0.93). Decreased risks were seen among patients with a history of a psychiatric disorder, in patients 12 to 17 years as well as 18 years and older, and in both females and males.
The Sentinel study had limitations. The study relied on outcomes for which patients sought medical attention that were recorded in health care claims. Thus, it was unable to evaluate either the entire spectrum of neuropsychiatric events or events that did not result in a billed encounter. Some neuropsychiatric events may have been handled by discontinuation of the drug without a health care encounter. Most of the usage occurred after the 2008 FDA communication and prescribing information changes about the risk of neuropsychiatric events, so montelukast patients may have been informed to cease treatment should depressive symptoms develop, resulting in a decreased risk among montelukast users. Lastly, the study was unable to adjust for socioeconomic status. However, a literature review did not reveal evidence that montelukast and ICS are prescribed disproportionally to patients of varying socioeconomic status. Patients with higher socioeconomic status may be more likely to seek asthma management through outpatient visits, resulting in increased surveillance for neuropsychiatric adverse events.
We also reviewed evidence from animal studies, which suggest montelukast could act directly on cells in the brain. Orally administered montelukast (10 mg/kg/day for 7 days) was detectable in brain tissue and cerebrospinal fluid in rats, providing evidence of its ability to cross the blood-brain barrier.1
