Interferon

Interferon is a biologic cytokine that carries antiviral and antiproliferative features, while simultaneously producing an inflammatory response within the central nervous system.⁴ Alfa and beta interferons have a place in therapy for hepatitis C virus (HCV), multiple sclerosis, and various oncologic conditions. Alfa interferons specifically carry a black box warning for neuropsychiatric side effects, including suicidal ideation, depression, aggressive behavior, and hallucinations.⁵

The prevalence of neuropsychiatric side effects in patients treated with interferon tends to vary depending on which interferon product is used, as well as which disease is being treated. Pegasys®, or peginterferon alfa-2a, is indicated for use in chronic HCV. The prevalence of neuropsychiatric adverse events in peginterferon clinical trials included 19% to 38% of patients who experienced irritability, anxiety, and nervousness, as well as 18% to 28% of patients who had depression.⁵ Betaseron®, or interferon beta-1b, is indicated for use in relapsing forms of multiple sclerosis. Of 1532 patients in the clinical trials for interferon beta-1b, 3 patients completed suicide and an additional 8 patients attempted suicide.⁶

The risk of psychiatric complications from interferon alfa therapy has been well documented throughout the years. In 1987, a study by Renault et al⁷ followed 58 patients with chronic HCV. Of these 58 patients, 17 patients developed psychiatric side effects, 3 patients had to discontinue therapy because of these side effects, and at least 2 patients attempted suicide. The researchers at that time classified the psychiatric side effects into three categories: organic personality syndrome, organic affective syndrome, and delirium. Organic personality syndrome included symptoms of irritability and hostility, whereas organic affective syndrome described symptoms of depression.

One case report by Sockalingam and Balderson⁸ chronicled a patient's major depressive episode with psychotic features secondary to treatment with pegylated interferon alfa-2b and ribavirin for chronic HCV. This 42-year-old male patient presented to an emergency department after a fight with a stranger and concurrent suicidal ideation. Two weeks prior to presentation, he had completed a total of 48 weeks of interferon alfa-2b therapy. He reported neuropsychiatric symptoms starting approximately 4 weeks before the event; these symptoms included suicidal ideation, insomnia, paranoia, ideas of reference, and religious preoccupation. This case demonstrates the risk for neuropsychiatric symptoms that may develop during interferon treatment as well as after completion of treatment.

When pegylated interferon was produced, many believed that patient quality of life would improve because of less frequent dosing of interferon treatment in chronic HCV. Kraus et al⁹ aimed to assess the prevalence of depression reported by patients who received pegylated interferon compared with conventional interferon therapy. Depression scores were measured for patients using the Hospital Anxiety and Depression Scale, and anger and hostility were assessed using the Symptom Checklist-90 Revised. When analyzing depression scores across both groups, depression increased significantly from baseline in both groups. No significant difference was detected in the depression scores between the two groups; therefore, the severity of depression remained the same regardless of interferon formulation. Of note, the authors did not specifically assess patient suicidality; however, depression increases the likelihood for suicidality.¹

There have been several proposed mechanisms that may explain how interferon induces neuropsychiatric side effects. The proposed mechanisms for interferon-induced depression include decreased serotonin (5-HT) concentrations, hypothalamic-pituitary-adrenocortical axis activation, increased nitric oxide concentrations, and inflammation.¹⁰ In general, interferon activates the immune system, having an impact on the metabolism of tryptophan. Tryptophan, a precursor to 5-HT, may get shunted away from the regular 5-HT pathway to another pathway, generating less serotonin and potentially having an impact on the central nervous system (CNS).¹¹ Decreased 5-HT concentrations may also result from the impact of chronic HCV on platelet 5-HT functions. With impaired platelet function, 5-HT reuptake is decreased.¹² Additionally, interferon may activate the hypothalamic-pituitary-adrenocortical axis, leading to increased adrenocorticotropin hormone and increased cortisol release. Increased cortisol has been found in patients with major depressive disorder.¹³ Similarly, interferon may release nitric oxide, which in turn activates the hypothalamic-pituitary-adrenocortical axis and monoamine neurotransmitter release, again resulting in increased cortisol concentrations.¹⁴ Lastly, the inflammatory theory describes how interferon increases cytokine release; cytokine release has been implicated in other studies of depression.¹⁵

As evidenced through case reports and clinical trials, interferon yields significant neuropsychiatric effects that could include suicidality. Patients prescribed interferon should be monitored for significant changes in behavior during treatment and upon discontinuation.

Efavirenz

Efavirenz is an antiretroviral agent, specifically a nonnucleoside reverse transcriptase inhibitor, indicated for use in human immunodeficiency virus (HIV) infection. Efavirenz is included in preferred antiretroviral treatment regimens.¹⁶ It is also, along with emtricitabine and tenofovir, an active ingredient in the combination product Atripla®. The package insert for efavirenz warns providers and patients of the risk for serious psychiatric symptoms.¹⁷

One particular study by Mollan et al¹⁸ compared the time to suicidality in patients treated with efavirenz versus those patients who did not receive efavirenz as part of their initial HIV treatment regimen. The hazard ratio of suicidality was significantly increased in the patients who received efavirenz compared with those who did not. A total of 9 deaths occurred during the study due to suicide: 8 patients were in the efavirenz group and 1 patient was not. The time to suicidality was decreased in patients receiving efavirenz compared with those who did not.

A cross-sectional study conducted by Fumaz et al¹⁹ assessed long-term neuropsychiatric side effects in 60 patients who received efavirenz for at least 1 year compared with 60 patients who did not receive said therapy. At the 1-year mark, 54% of patients in the efavirenz group reported a neuropsychiatric disorder, compared with 27% in the nonefavirenz group. Similarly, in another study by Lochet et al,²⁰ patients treated with efavirenz for at least 3 months were provided with a survey to assess neuropsychiatric effects prior to efavirenz treatment, during the first month of treatment, and 1 month prior to the survey. After being on efavirenz for 3 months, 13.2% of patients reported frequent suicidal ideation, of whom about 10% did not have suicidal ideation prior to starting efavirenz. In addition to suicidal ideation, other neuropsychiatric effects that persisted in patients beyond 3 months of therapy included anxiety, sadness, cognitive disorders, agitation, and aggressiveness. The researchers acknowledged that one limitation of their study was the inability to detect the impact that the disease progression of HIV and other life stressors may have had on the patients, which could have confounded the association believed to be from efavirenz alone.

One case report by Puzantian²¹ elucidates the neuropsychiatric complications believed to be due to efavirenz therapy. One 47-year-old man with acquired immune deficiency syndrome (AIDS) was hospitalized secondary to symptoms of severe depression that had started 4 weeks prior. His neuropsychiatric symptoms started around the time that he was initiated on efavirenz therapy. Efavirenz was discontinued when symptoms arose, but the patient remained depressed. During this patient's hospitalization, other biologic causes for depression were ruled out.

The neuropsychiatric effects of efavirenz may not only be detrimental to a patient's psychologic well-being, but may also compromise HIV treatment. Leutscher et al²² analyzed drug discontinuation rates among patients who started on an initial treatment regimen with efavirenz compared with patients who did not. Of the 168 patients who received efavirenz, 90 discontinued treatment; 60% of the patients who discontinued treatment did so secondary to neuropsychiatric effects, as opposed to only 3% of patients in the nonefavirenz treatment group. The most common reason for discontinuation in the nonefavirenz group was physician concern for cardiovascular effects.

A risk factor associated with efavirenz-induced neuropsychiatric side effects is thought to be elevated efavirenz plasma drug concentrations. A longitudinal study by Gutiérrez et al²³ measured the plasma efavirenz concentration at set time points for 18 months from the study onset in 17 patients with no psychiatric history. Of these 17 patients, 10 had neuropsychiatric side effects, and 4 of those patients had such significant neuropsychiatric side effects that efavirenz had to be discontinued. Plasma concentrations in patients experiencing neuropsychiatric effects were higher than in those who did not. A larger study by Marzolini et al²⁴ collected blood samples from 130 patients being treated with efavirenz to determine whether CNS toxicity could be predicted by plasma concentrations. CNS toxicity occurred three times more frequently in patients with plasma efavirenz concentrations greater than 4 μg/mL. Even with the potential relationship between efavirenz plasma concentration and CNS toxicity, modern-day practice does not measure efavirenz concentrations.

These case reports and clinical trials demonstrate an association between efavirenz and neuropsychiatric effects, including suicidality; however, the impact of the disease progression of HIV/AIDS cannot be excluded as a confounder.^1,20 In summary, patients and providers should monitor for increased suicidality throughout the course of HIV/AIDS and adjust antiretroviral therapy if necessary.

Varenicline

Varenicline (Chantix®, Pfizer, Mission, KS) is a partial nicotinic receptor agonist that is indicated for smoking cessation. In November 2007, the FDA released a warning reporting suicidal thoughts, aggressive/erratic behavior, and drowsiness from postmarketing reports of patients receiving varenicline therapy.²⁵ In 2009, the FDA required the addition of a black box warning to the varenicline product labeling for neuropsychiatric symptoms, including changes in behavior, hostility, agitation, depressed mood, suicidal thoughts and behavior, and attempted suicide.²⁶ In the clinical trials before FDA approval, the most common adverse effects identified included nausea, insomnia, abnormal dreams, and headaches.²⁷ There is debate about whether these adverse effects are caused by varenicline or related to the psychologic and psychiatric effects that are associated with smoking and/or smoking cessation.²⁸

Meyer et al²⁷ conducted a retrospective cohort study with data extracted from the Military Health System Data Repository. This study determined the rate ratio of neuropsychiatric hospital admissions of new varenicline users and new nicotine replacement therapy users within the Military Health System before the release of the FDA safety warning. After the two cohort groups were balanced to account for potential confounders, there was no statistically significant difference found in the rate of neuropsychiatric hospitalizations; however, looking at only hospitalizations could exclude suicide attempts limited to emergency department visits.

Mocking et al²⁸ conducted a randomized, double-blind, placebo-controlled study to compare the short-term effects of varenicline versus placebo on emotional processing (a biomarker of depressogenic effects), impulsivity (linked with suicidal behaviors), and cognitive performance in nonsmokers. After a 7-day course of varenicline or placebo administration, there was no suggestion of varenicline negatively influencing emotional processing or impulsivity, but varenicline instead enhanced cognitive performance. These findings contradict the evidence found from postmarketing case reports. Hong et al²⁹ also conducted a randomized, double-blind, parallel group, placebo-controlled study evaluating the effects of varenicline treatment on the neurobiologic and cognitive biomarkers in smokers and nonsmokers with a past medical history of schizophrenia or schizoaffective disorder. Evaluating neurobiologic and cognitive biomarkers was the primary outcome; however, clinical outcomes evaluating the severity of psychotic symptoms, negative symptoms, depression, and function were also recorded using the Brief Psychiatric Rating Scale, the Schedule for Assessment of Negative Symptoms, the Hamilton Rating Scale for Depression (HAM-D), and the Level of Function Scale and Global Assessment of Functioning, respectively. Suicidality was also assessed at each visit. The biomarkers were evaluated at 2 and 8 weeks; there was no evidence found at either end point that varenicline increased the psychiatric symptoms of depression, anxiety, or suicidality according to further review of depression/suicidality items on the rating scales listed above.

A follow-up review article by Gibbons and Mann³⁰ investigated all of the data obtained from the randomized, double-blind, placebo-controlled trials from phase 2 through phase 4 to determine whether varenicline is associated with a higher risk for neuropsychiatric adverse effects—specifically depression—suicidal thoughts/behaviors, and aggression/agitation. For each of these evaluated adverse effects, it was found that varenicline did not significantly contribute to adverse effects and that psychiatric illness did not moderate the effect of varenicline on adverse effects. Another small, open-label study by Philip et al³¹ was performed to evaluate varenicline as an adjunct agent in patients with depression who smoked cigarettes. Although this study's main objective was to evaluate varenicline treatment as adjunctive therapy for depression in smokers, a post hoc analysis was performed regarding the recent concerns about the potential increase in neuropsychiatric symptoms related to varenicline. A significant improvement was found in the core mood score from baseline to end point in the Quick Inventory of Depressive Symptomatology-Self-Rated-16 Item Scale, and no significant changes were found in the suicidality ratings.

There has been no pharmacologic explanation of how varenicline could potentially cause suicidal ideation. Although it works within the CNS on the nicotinic acetylcholine receptors, this is thought to reduce the cholinergic system hyperactivity that has been associated with depression symptoms.³² In 2011, the FDA released a safety review update following the studies focused on varenicline and neuropsychiatric effects sponsored by the FDA.³³ Neither sponsored study found an association between varenicline and increased hospitalizations, but the FDA recommendations remained the same. Recently, Pfizer requested the FDA revisit the black box warning in hopes of its removal following the results of observational studies and a Pfizer-sponsored meta-analysis. The Psychopharmacologic Drugs Advisory Committee and the Drug Safety and Risk Management Advisory Committee voted to retain the current black box warning statements and wait for the completion of the phase 4 trial specifically evaluating neuropsychiatric effects, which is anticipated to be completed in August 2016.³⁴

Isotretinoin

Isotretinoin is a retinoid used to treat nodular acne. Isotretinoin carries a Risk Evaluation and Mitigation Strategy because of its teratogenicity. The isotretinoin package insert also warns about the medication's risk of causing depression, psychosis, and suicidal ideation.³⁵

A United States–based analysis by Wysowski et al³⁶ was conducted to analyze the number of cases reported to the FDA of patients with depression and/or suicide attempts being treated with isotretinoin. The researchers looked at the FDA's Adverse Event Reporting System and selected adverse event reports for those patients on isotretinoin reporting depression, major depressive disorder, dysthymic disorder, suicidal depression, and aggravated depression from the years 1982 to 2000. From this time frame, the researchers found 431 reports. Of this number, 37 patients completed suicide—24 of these patients while using isotretinoin, and 13 after drug discontinuation. There were 110 patients who had received isotretinoin who were hospitalized for neuropsychiatric reasons. The remaining 284 patients reported a diagnosis of depression to the FDA. At the time of this report in 2001, isotretinoin ranked among the top 10 medications reported to the FDA for depression and suicide attempt. At that time, all medications ranking above isotretinoin had psychiatric indications.

Other studies have not found an association between isotretinoin and neuropsychiatric events. A population-based cohort study by Jick et al³⁷ looked at patients treated with isotretinoin versus those treated with oral antibiotics for acne compared with patients without exposure to either medication. Relative risk estimates for the development of depression or psychosis were around 1.0 in each exposure group. Another study by Rehn et al³⁸ followed Finnish male patients with acne who visited a dermatologist for treatment and were assessed at baseline, weeks 4 to 6, and weeks 10 to 12 using the Beck Depression Inventory. A total of 9 patients had a moderate level of depressive symptoms at baseline, but of that 9 at the start, 8 patients no longer reported significant depression by weeks 10 to 12. A total of 17 patients (13.5%) reported suicidal ideation at baseline. This decreased to only 7 patients reporting suicidal ideation by week 12. The question remains of whether some patients are more susceptible to mood effects from isotretinoin, or if the medication does not cause such effects.

One retrospective cohort study by Sundström et al³⁹ in Sweden aimed to analyze the risk of attempted suicide before, during, and after treatment with isotretinoin compared with the general population. The researchers linked discharge summaries and death records to patients entered in the isotretinoin patient registry. Patients included in this study were individuals prescribed isotretinoin between 1980 and 1990. These patients were then linked to hospital admissions for suicide attempts and accidents with unclear intent. A total of 5756 individuals comprised the final study cohort. Of this number, 128 individuals were discharged from the hospital for attempted suicide at least once, and 24 patients committed suicide. One year prior to isotretinoin treatment, the incidence ratio for attempted suicide increased compared with the general population but was not significant. The incidence ratio of first-time suicide attempts was greatest during and up to 6 months after treatment and was statistically significant. The risk of overall suicide attempts was also significantly increased during and up to 6 months after therapy. The incidence ratio decreased to that of the general population within 3 years after completion of treatment. The researchers concluded that an increased risk of attempted suicide was greatest during and up to 6 months after isotretinoin therapy; however, because of the increasing rates of suicide prior to isotretinoin therapy, risk from exposure to isotretinoin alone could not be established. The authors suggested that severe acne alone might contribute to increased suicidality.

The reason behind what could potentiate depressive symptoms in patients treated with isotretinoin was investigated by Bremner et al⁴⁰ through functional brain imaging. These researchers used positron emission tomography scans to image the brain prior to isotretinoin therapy and again 4 months after treatment, and these scans were compared with those of patients treated with antibiotics for acne. At baseline, patients in the isotretinoin group when compared with the antibiotic group demonstrated no difference in depressive symptoms measured by the HAM-D or in baseline positron emission tomography scans. After 4 months of therapy, positron emission tomography scans illustrated decreased brain metabolism in the orbitofrontal cortex in those who received isotretinoin compared with those who received antibiotics. Decreased orbitofrontal cortex metabolism has been associated with the onset of depressive symptoms.⁴¹

In 2010, the United Kingdom published guidelines for isotretinoin use with regard to its teratogenicity and neuropsychiatric effects.⁴² These guidelines suggested monitoring parameters for prescribers to follow. Based on the current evidence available, the United Kingdom recommended that providers inquire about previous psychiatric history in patients, inform patients and caregivers of the risk for neuropsychiatric effects, and encourage family and friends to speak to the patient and/or provider if they notice such changes occurring. Additionally, the United Kingdom recommended assessment of psychologic symptoms at each clinical appointment using standardized screening tools, such as The Beck Questionnaire or the Baer HANDS Questionnaire. Of a similar nature is the US FDA's iPLEDGE program. Although the main goal of the iPLEDGE program is to prevent pregnancy and birth defects from isotretinoin, the program also educates patients on the neuropsychiatric risks of the medication. A patient is told to stop isotretinoin and contact a provider right away if he or she develops any signs of mental illness.⁴³ In spite of the conflicting study results summarized above, governing bodies recommend caution with patients prescribed isotretinoin.

Montelukast

Montelukast (Singulair®) is a leukotriene-modifying agent (LTMA) that is FDA approved for both asthma and allergic rhinitis. In 2008, the FDA issued a statement warning the public about the possible association between the use of montelukast and behavior/mood changes, suicidality, and suicide.⁴⁴ This statement followed results from postmarketing surveillance that reported various neuropsychiatric side effects. The culminating event that prompted this warning and a subsequent systematic review of the literature was a media-reported suicide linked to the recent use of montelukast.⁴⁵ The FDA included other LTMAs, zafirlukast (Accolate®) and zileuton (Zyflo® and Zyflo CR®), in the systematic review.

The FDA released an update in January 2009 following the initial warning.⁴⁶ Of all the trials submitted to the FDA from all three drug companies, only 1 patient in the montelukast group and none in the other medication groups for zafirlukast and zileuton had suicidal ideation. The FDA noted that none of these clinical trials were designed to examine neuropsychiatric events, and therefore it maintained its previous warning while other clinical trial data were being assessed. In June 2009, the FDA announced that the manufacturers of the LTMAs would be required to add a precaution in the drug prescribing information because of postmarketing reports of neuropsychiatric events.⁴⁷

Since the release of the original FDA warning, many retrospective studies and articles have been published to evaluate the potential link of LTMAs, specifically montelukast, and suicidality. A population-based cohort study by Jick et al⁴⁸ was performed in the United Kingdom by using the General Practice Research Database. This study identified 23 500 patients who received 252 593 prescriptions for montelukast from 1998 to 2007. Among these patients, there was only 1 case of suicide, which involved a patient with a history of asthma and a 10-year history of depression who only received one 28-day supply of montelukast 2 years prior to her death. The authors concluded that if there is a suicidality risk associated with montelukast, it is extremely low.

A nested case-control study conducted by Schumock et al⁴⁹ looked at a cohort of asthmatic patients ages 5 to 24 years who were new to LTMAs or other asthma medications. The data were extracted through insurance claims for patients with a diagnosis of asthma. Ultimately, it was found that the current use of any LTMA was not associated with an increased risk for suicide attempts; paradoxically, although not statistically significant, the odds ratio was suggestive of a reduction in suicide attempts in those patients using LTMAs. A literature search performed by Schumock et al⁵⁰ evaluated publications associated with LTMAs and suicidality from 1995 to 2010 and used empirical data from the FDA's Adverse Event Reporting System for the review. The authors concluded that there were no well-conducted, comparative, observational studies of the association between LTMAs and suicide.

There is limited evidence of a specific pharmacologic mechanism linking LTMAs and suicidal behavior. It has remained controversial, but evidence has suggested a positive correlation between allergies and suicidality.⁴⁵ It has also been found that patients with asthma are at an increased risk for mental disorders.⁵⁰ There are few data describing the CNS effects on behavior mediated by the cysteinyl leukotriene type 1 receptor, which is where montelukast and zafirlukast bind; zileuton works further upstream, inhibiting 5-lipoxygenase.^48,50 Leukotrienes do exist in the brain and within the CNS, but LTMAs in particular work on inflammation in the airways. LTMAs can penetrate the blood-brain barrier and are found in the brain tissue in significant concentrations. This is why montelukast has been studied as a neuroprotective agent in traumatic brain injury, focal cerebral ischemia, and migraine headaches.⁵⁰

No clinical data or case reports have been able to definitively link LTMAs to the risk of suicidality. The American Academy of Allergy, Asthma, and Immunology (AAAAI) and the American College of Allergy, Asthma, & Immunology (ACAAI) released a joint statement shortly after the FDA released the initial warning statement for LTMAs. These committees believe the warning is based solely on case reports with no supporting clinical data and recommend continuing treatment with LTMAs unless suicidal thoughts are experienced.⁵¹ Most reviews and expert opinion articles conclude that although there is no definitive evidence linking the two, the FDA's precaution should not be ignored, and patients and clinicians should monitor for any changes in mood or increase in suicidal ideation.

Glucocorticoids

Glucocorticoids have countless roles in medication therapy, anywhere from treatment for chronic obstructive pulmonary disease exacerbation to treatment for Crohn disease. Overman et al⁵² aimed to estimate the prevalence of corticosteroid use in the United States. It was estimated that about 1.2% of the US population older than 20 years received an oral glucocorticoid between 1999 and 2008. Corticosteroids are not benign medications and can lead to adverse effects, such as osteoporosis, diabetes, Cushing syndrome, neuropsychiatric effects, and hypertension. This review will only focus on the neuropsychiatric effects.

Fardet et al.⁵³ analyzed the incidence rates of neuropsychiatric side effects in patients treated with glucocorticoids in the primary care setting and what risk factors might exist. Data was collected on patients from 1990 to 2008 who sought care at general practices in the UK that submitted information to The Health Improvement Network. Comparisons of neuropsychiatric effects were made between patients who received oral glucocorticoids to those who did not. A total of 786868 prescriptions for glucocorticoids were written for 372696 patients from 1990 to 2008. Of the patients who received glucocorticoids, 19 completed suicide, 90 attempted suicide, and over 6000 reported cases of depression. The hazard ratio for overall neuropsychiatric events in the glucocorticoid group compared to no treatment was significantly increased. The hazard ratio for suicide or suicide attempt was also significantly increased in the glucocorticoid-treated group. The authors found younger patients to be at greater risk for suicidal ideation and attempt compared to older patients. Patients with a history of psychiatric disorders were at the greatest risk for neuropsychiatric complications.

A cohort study by Fardet et al⁵⁴ assessed glucocorticoid-induced adverse effects, including patients' opinions on such side effects. Patients prescribed long-term corticosteroid regimens (duration >3 months) at high dosages (>20 mg prednisone per day) were included in the study. Adverse effects were assessed after 3 months of treatment. A total of 88 patients were included in the study results. The most commonly occurring adverse event was lipodystrophy, which was reported by 63% of patients. This was followed by neuropsychiatric effects, which were reported by 52.5% of patients, 5 of whom required hospitalization because of these effects. Two patients reported suicidal ideation. Per patient report, neuropsychiatric side effects were deemed the most distressing side effect in 17% of patients.

A Japanese case report by Matsumoto et al⁵⁵ was completed regarding a suicide associated with glucocorticoid use in a 68-year-old patient who received dexamethasone as an antiemetic for chemotherapy. Following severe nausea and vomiting after chemotherapy cycle one, the patient was admitted as an inpatient and the dose of dexamethasone was increased for the second cycle. The patient was also referred for psychiatric evaluation at that time. Her evaluation demonstrated increasing depressive symptoms without thoughts of death from days 7 through 20 of her first chemotherapy cycle, which had since resolved. The patient was discharged and scheduled for an outpatient visit 21 days later; however, the patient did not make it to that appointment because she had completed suicide on day 17. The patient had no previous psychiatric history, and the authors concluded none of the other medications could explain her suicide except for the glucocorticoid. It is important to note that dealing with the disease course itself may have been distressing and contributed to suicidality with or without medication involvement. Similarly, Uva⁵⁶ associated glucocorticoid use with suicidality. One patient described in the study was a 59-year-old woman who was prescribed a 7-day tapering course of prednisone. At her follow-up appointment the next day (day 2 of the steroid taper), she received an intramuscular injection of dexamethasone for continued symptoms. On day 4, she began experiencing neuropsychiatric symptoms, which by day 6 had progressed into an overwhelming depression. The woman then ingested 15 chlorzoxazone tablets and was found in an unconscious state, but would later deny suicidal ideation. She was hospitalized and discharged with a diagnosis of depression secondary to glucocorticoid use.

Bräunig et al⁵⁷ assessed 15 patients seen at the Bonn psychiatric clinic in Germany who developed glucocorticoid-induced psychosis between the years 1966 and 1988. A total of 5 of the 15 patients assessed were suicidal upon presentation, with 2 of these patients actually having attempted suicide. None of the patients who presented had any previous history of suicidal ideation or attempt. The authors further reviewed the literature for similar case reports and found that suicidal ideation was more common in patients presenting with glucocorticoid-induced psychosis compared with psychosis of organic origin.

A proposed mechanism of action for the neuropsychiatric side effects attributed to glucocorticoids involves the body's response to exogenous versus endogenous cortisol.⁵⁸ Normally, stress causes the hypothalamus to release corticotropin-releasing factor, and in turn the anterior pituitary releases corticotropin, which stimulates the adrenal gland to release cortisol. When a patient receives exogenous glucocorticoids, there is a selective activation of the glucocorticoid receptors and suppression of the adrenal cortisol secretion. The exogenous glucocorticoids continue to activate the glucocorticoid receptors in the brain while the mineralocorticoid receptors remain inactivated. This imbalance of receptor activation is one theory for the mood disturbances caused by glucocorticoids.⁵⁸

Overall, these findings, reinforced by other studies, demonstrate the neuropsychiatric effects associated with glucocorticoids.^52-61 Of those effects, suicidality (although seemingly rare) may arguably be the most severe.

Fluoroquinolones

Fluoroquinolones are a commonly prescribed class of antibiotics used for a range of infectious diseases, including urinary tract infections, respiratory infections, and sexually transmitted diseases. CNS effects, including convulsions, anxiety, confusion, depression, insomnia, hallucinations, and suicidal thoughts, are listed in the precautions section of the package inserts for such fluoroquinolones as levofloxacin (Levaquin®), ciprofloxacin (Cipro®), and moxifloxacin (Avelox™).^62-64

A nested case-control study was conducted using the UK General Practice Research Database following an influx of reports to the spontaneous reporting system in Germany concerning the relation of exposure to fluoroquinolone antibiotics and suicidal behavior.⁶⁵ The population of this study was restricted to patients who had used a quinolone antibiotic at some point between January 1, 1991, and April 30, 1995. The quinolones included were ciprofloxacin, enoxacin, acrosoxacin, and temafloxacin. Patients ages 15 to 84 years were separated into three groups: patients who completed suicide, patients who had a history of attempted suicide, and patients who had a diagnosis of suicidal ideation. Relative risk estimates showed no material differences in risk across the groups to suggest that fluoroquinolone antibiotics given alone or in combination with other antibiotics were associated with suicidal behavior. Potential confounding factors were accounted for in the adjusted relative risk estimates, and no statistically significant difference was noted.

Several case reports involving the commonly used fluoroquinolones, levofloxacin and ciprofloxacin, have been published associating these medications with suicidality. LaSalvia et al⁶⁶ reported a 75-year-old male patient with no past psychiatric history who experienced severe depression with symptoms including anhedonia, hopelessness, and suicidal ideation following levofloxacin therapy. Within 48 hours of discontinuing levofloxacin, his symptoms started to improve. Labay-Kamara et al⁶⁷ examined a similar case report involving a 26-year-old male patient who experienced depressed mood and suicidal thoughts following a 5-day inpatient course of levofloxacin and metronidazole and a subsequent outpatient course of ciprofloxacin. Suicide was attempted 2 days after discharge; shortly afterward, ciprofloxacin was discontinued and citalopram initiated. Per the patient's report, his symptoms started to greatly improve during a 24-hour period. Ciprofloxacin was implicated in a case report by Ahmed et al⁶⁸ following a patient with a year of recurrent depression culminating in a suicide attempt. It was discovered that each depressive episode was preceded by a 1-week treatment with ciprofloxacin. The result of the Naranjo causality scale was a score of 10 (definite being greater than 9). The Naranjo causality scale can be used to determine the probability that a medication caused an adverse effect.⁶⁹ The authors of these case reports admit there are many confounding factors, including age, current and past medical history, and polypharmacy, involved in these cases, which makes it difficult to determine a direct cause and effect correlation between fluoroquinolones and suicidality. However, the authors still considered the fluoroquinolone antibiotic the most likely offending factor in the onset of these patients' depression and suicidal ideation in each case.

There are several theories behind the possible mechanism of fluoroquinolones causing CNS adverse effects. The main theory focuses around the idea that the structural component of the fluoroquinolone molecules that is responsible for the activity against gram-positive organisms is also responsible for the CNS adverse effects because of their similarity to gamma-aminobutyric acid.⁷⁰ Fluoroquinolones inhibit gamma-aminobutyric acid binding at receptor sites within the CNS, which results in an excitatory effect.^66,68 Another possible theory includes the dopamine receptor and the N-methyl-d-aspartate gated ion channel, because fluoroquinolones are also thought to activate N-methyl-d-aspartate channels by chelating with magnesium and removing its channel-blocking effect. It has been thought that N-methyl-d-aspartate antagonists have an antidepressant effect. No clinical studies have been able to definitively link fluoroquinolone use to suicidality; there are limited clinical data, and much of the evidence is derived from case reports and observational studies.

HMG-CoA (3-hydroxy-3-methylglutaryl–coenzyme A) Reductase Inhibitors

The possible link between lipid-lowering therapy and suicidal ideation was first publicized in 1990 following a review published by Muldoon et al⁷¹ that found that although there was a decrease in cardiovascular mortality, there was an increase in noncardiovascular mortality due to unintentional injury, suicide, or violence. Since this publication, there has been much debate surrounding lipid-lowering therapy—specifically the HMG-CoA reductase inhibitors, otherwise known as the statins—and their relation to suicidal ideation.

A nested, case-control study from Callreus et al⁷² evaluated a wide range of cardiovascular medications and their relation to suicidal ideation. The population was only from one county in Denmark (about 9% of the total Danish population) with data collected from The Danish Registry of Cause of Death and The Odense University Pharmacoepidemiological Database. Neither statins nor other lipid-lowering therapies were implicated in an association with an increased risk of suicide. Another nested, case-control analysis by Yang et al⁷³ used population data from the UK General Practice Research Database to evaluate the effect of lipid-lowering medications on depression and suicide. In this study, it was conversely found that patients with current statin use had a somewhat lower risk of depression with preexisting cardiovascular disease compared with patients without preexisting cardiovascular disease. Another finding in this study showed that patients with statin and nonstatin lipid-lowering therapy use of longer than 1 year had a lower risk of depression than those who received less than 1 year of treatment.

A small, cross-sectional study by Davison and Kaplan⁷⁴ evaluated the correlations between statin use and suicidal ideation, specifically in patients with mood disorders. Mood disorders of depression and bipolar disorder were evaluated using the HAM-D and the Young Mania Reporting Scale to measure the severity of symptoms. The only significant factors associated with suicidal ideation in this study were increased Young Mania Reporting Scale scores and statin use. Only lipophilic statins (eg, simvastatin, lovastatin) were used in this study. Multiple confounding variables in this study make a cause and effect relationship hard to establish.

Statins and nonstatin lipid-lowering therapy have been associated with a variety of neuropsychiatric symptoms, such as depression, impulsivity, aggression, and suicidal ideation, after numerous case reports and observational studies. The main causality question has focused around whether it is reducing cholesterol or low serum cholesterol in general that can increase these psychiatric symptoms. An experimental animal study conducted on monkeys demonstrated that monkeys with a low-fat diet and low serum cholesterol concentrations were more aggressive than those monkeys with a high-fat diet (luxury diet) and high serum cholesterol concentrations.^75,76 The monkeys fed a low-fat, low-cholesterol diet also had a significantly lower serum prolactin response to a fenfluramine challenge, which is an indicator of low central serotonergic activity. This led investigators to believe that brain serotonergic activity was altered.

After the Muldoon et al⁷¹ review was published, researchers attempted to describe the psychologic reasoning for the potential link between lipid-lowering therapy and suicidal ideation. Engelberg⁷⁷ hypothesized that a reduction in serum cholesterol decreases brain cell membrane cholesterol, which lowers the lipid microviscosity; this ultimately could decrease the exposure of protein serotonin receptors on the membrane, resulting in reduced serotonin uptake into the blood and brain cells. Vevera et al⁷⁸ conducted a study to evaluate the effect of lipid-lowering therapy—specifically simvastatin, because of its lipophilicity—on membrane microviscosity, serotonin uptake, and behavioral changes. It was found that serotonin transporter activity was transiently affected by serum cholesterol concentrations. Serotonin transporter activity was significantly increased following 1 month of simvastatin therapy, but this increase declined starting at 2 months and especially at more than 12 months of therapy. This result is consistent with the nested case-control by Yang et al⁷³ mentioned earlier. Once again, there are many confounding factors involved in this association of statins and suicidal ideation, making it difficult to determine a direct cause and effect relationship.