Substance-Induced Mood Disorder

A number of reports of substance-induced mood disorders (SIMDs) have emerged since the 1950s, when psychiatric complications were described in patients treated with the anti-hypertensive reserpine. In addition to prescription medications, several over–the-counter (OTC) medications as well as recreational and illicit drugs have been implicated in the onset of drug-induced depression or mania. Despite the number of cases reported over the years, few controlled studies of the phenomenon have been conducted to date.

The essential feature of SIMDs is the onset of symptoms in the context of subtance use, intoxication, or withdrawal. Full criteria for a depressive or bipolar spectrum disorder need not be met for a diagnosis.

Hypotheses regarding the etiology of SIMDs are based on the known properties of the medications involved and their potential correlation with current neurophysiologic models of affective disorders. These include models of tryptophan depletion, catecholamine depletion, and alterations in the hypothalamic-pituitary-adrenal axis.

Notably, SIMDs are more likely to occur in individuals with risk factors for primary psychiatric mood disorders, such as major depressive disorder, persistent depressive disorder, or bipolar disorder. One of the most common risk factors is a personal or family history of a mood or substance disorder.

Substance-induced mood disorder, a diagnosis in prior editions of the Diagnostic and Statistical Manual of Mental Disorders, is no longer a diagnosis according to DSM-5. The diagnosis is now subsumed under the diagnosis "substance/medication-induced mental disorders" within the chapter "substance-related and addictive disorders." ^[1]  “Substance-induced depressive disorder” and “substance-induced bipolar and related disorder” are also individual diagnoses in DSM-5. This review focuses on Medication-Induced Mood Disorders.

While a number of drugs have been associated with an increased risk for mood disorders, medications with considerable evidence of an association with depression and/or mania include antidepressants, antiepileptic drugs (AEDs), corticosteroids, interferon (IFN)-alpha, and opiates.

Antidepressants

Antidepressant-induced mania has been observed in 20%–40% of patients with a diagnosed bipolar disorder who are exposed to antidepressants, and may “reveal” a bipolar diathesis by inducing mania in patients treated for presumed unipolar depression. ^[11] A link between antidepressants and treatment-emergent suicidality in children and adolescents up to 24 years old resulted in the FDA issuing a black box warning on all antidepressant medications. However, findings from subsequent prospective data and randomized controlled trials have challenged this association. ^[27]

Antiepileptic drugs

A large FDA-sponsored meta-analysis comprising 199 placebo-controlled trials reported an increased risk of suicidal behavior in individuals taking AEDs. ^[10]  The largest effect was noted with levetiracetam, which was also observed in a subsequent case-control study. ^[9]  A 2010 prospective study comprising more than 5 million patients exposed to AEDs confirmed an increased risk of suicide-related events both in depressed and non-depressed individuals. ^[8]

Corticosteroids

An early systematic review comprising 29 clinical studies reported severe psychiatric “reactions” in approximately 5% of steroid-treated patients. These reactions typically occurred early in treatment and were most frequently depressive and manic syndromes. Mild-to-moderate reactions occur in 28% of treated patients. ^[33] Increased corticosteroid dosage has been associated with a higher rate of adverse psychiatric effects, but does not appear to modify symptom timing, severity, or duration. Manic symptoms or syndromes may be more likely in short-course steroid treatments, while depression may be more common with long-term therapy. ^[6] In a more recent prospective study of high-dose corticosteroids for the treatment of relapse in multiple sclerosis, scale-rated manic and depressive symptoms increased in a large percentage of subjects. Symptom scores were measured three days after completing steroid treatment and were compared to baseline scores. When measured one month later, depression and manic symptom scores returned to baseline. ^[32]

Interferon-alpha

Interferon-alpha (IFN-alpha) treatment has been associated with an increased risk of depression in randomized controlled trials (RCTs) and prospective cohort studies. ^{[2, 3]} Although relevant trials were not designed to study this as an endpoint (with the exception of one) and did not use standardized interviews for psychiatric assessments, the relatively high proportion of depression in this population (20%–45%) raises important questions about IFN tolerability/toxicity. Patients with hepatitis C and a previous history of psychiatric illness (and more specifically depression) before IFN treatment are at an even a greater risk for developing IFN-induced depression. ^[2]

Different types of IFN may have different risk profiles. In a prospective study of 96 patients treated with different types of IFN, the highest incidence of depression was reported with IFN-alpha n1, the lowest incidence with IFN-alpha n3, and the most severe depression and the highest rate of suicidal ideation with IFN-alpha 2b. ^[4]

Despite early evidence suggesting that IFN-beta might also be associated with depression, subsequent evidence based on large, randomized, controlled trials did not support an association of IFN-beta (1a and 1b) and depression. ^[2]

Case reports of mania in the setting of IFN-alpha withdrawal have also been reported. ^{[34, 35]}

Opiates

A retrospective study of more than100,000 US adults with no prior diagnosis of depression reported a 9%–12% incidence of new-onset depression after opiate analgesic use. ^[56]  A longer duration of opiate use, but not daily dose, was associated with a higher risk of new-onset depression. A subsequent analysis of these data suggested a two-fold increased risk of depression recurrence in remitted patients. ^[57]  A 2012 case series described nine reports of hypomanic symptoms presenting in the setting of opiate withdrawal. ^[36]

Weak or conflicting drug associations with SIMDs

Substances with limited or conflicting evidence of a link to depression and/or mania are listed below. Unless otherwise specified, these agents have been linked specifically to depression.

• Angiotensin-converting enzyme inhibitors – Evidence is limited to case reports and small epidemiological studies with limited control of confounding variables. ^[37]

• Beta blockers – Some observational epidemiological studies have associated beta blockers with depression, but most studies demonstrate no such association. ^[5]  Fatigue or anergia, known side effects of beta blockers, may be misinterpreted as depressive symptoms. A 2013 meta-analysis described lower rates of depression in individuals treated with beta blockers as compared to placebo. ^[28]

• Calcium channel blockers

• Clonidine ^[5]  – An early study of clonidine in the treatment of hypertension reported treatment-emergent depression in three of 28 patients, two of which had a history of recurrent depression. One instance of hypomania was also reported in a patient with a history of depression and psychosis. ^[38]

• Digoxin – A small prospective study linked digoxin treatment with increased risk of depression following myocardial infarction. ^[39] Digoxin has also been linked to a modest increase in depressive symptoms in elderly women, but not men, with chronic medical conditions. ^[40] A number of psychiatric effects, including depression, have been reported in patients taking digoxin according to a FDA package insert for Lanoxin (digoxin). ^[7]

• Finasteride – A small, open-label study reported that 19 of 23 patients receiving finasteride developed moderate to severe depression during the course of their treatment; ^[15] However a large prospective study only demonstrated a small but statistically significant increase in depression scores with finasteride treatment. ^[16]

• Gonadotropin-releasing-hormone agonists (GnRH-a) – A small, open-label, randomized study found an association of these agents with depressive symptoms. ^[13] However, no association was reported in a subsequent large, open-label, randomized trial ^[14]

• HMG-CoA (3-hydroxy-3-methylglutaryl-coenzyme A) reductase inhibitors

• A number of case reports including a relatively recent case series have described treatment-emergent depression with statins including depression and suicidality. ^{[41, 42]}  However, two placebo-controlled studies, and other observational studies, have demonstrated no such association. ^{[5, 42]}

• Isotretinoin – Since 1982, a number of case reports have linked isotretinoin with depression and suicidality. ^{[5, 29]} However, a recent systematic review and meta-analysis of 17 prospective studies found that isotretinoin treatment in patients with acne vulgaris was associated with significant improvement in depressive symptoms. ^[30]

• Leuprolide – Treatment-emergent depression occurred in 5.3% of patients treated with leuprolide in randomized controlled trials ^[12]

• Methyldopa – A number of case reports since the 1960s have linked methyldopa to depression. ^[43] However, a subsequent literature review reported no clear association with depression, despite its biological plausibility as a potent inhibitor of DOPA decarboxylase that may reduce levels of catecholamines in the central nervous system. ^[43]

• Progestin-releasing implanted contraceptives – A population-based study reported increased odds ratios for depression in patients treated with medroxyprogesterone acetate; ^[17] however, a large, multicenter, prospective, randomized, open-label study did not find an association between levonorgestrel (Norplant) and depression ^[18]

• Sedative-hypnotics – A small prospective study indicated an increased risk of treatment-emergent depression with benzodiazepines in medical inpatients after adjusting for psychosocial and clinical confounders. ^[44]  Hypnotics (zolpidem, eszopiclone, zaleplon, and ramelton) have also been associated with a slightly higher incidence of depression than placebo in randomized trials. ^[45]

• Stimulants (eg, amphetamines) – Especially at doses associated with abuse, stimulants are known to induce manic or hypomanic states indistinguishable from a classical bipolar disorder although such associations have not been systematically evaluated ^[47]  An FDA review of clinical trials described rare instances of manic or psychotic syndromes in patients treated with therapeutic doses of stimulants for ADHD. ^[47]

• Varenicline – Several case reports and individual clinical studies reported treatment-emergent depression in patients taking varenicline for smoking cessation. ^[46] However, a meta-analysis compromising 39 randomized controlled trials found no increased risk of depression or suicidal behavior in patients treated with varenicline. ^[31]

Researchers have noted several etiologic factors in mood disorders. The amine-depleting effect of antihypertensive medications and the amine-restoring effect of the first successful antidepressants led to the catecholamine-deficit hypothesis. Endocrine factors have been correlated with depressive symptoms. Hyperthyroidism, of natural causes or iatrogenic, may result in clinical mania. Hypercortisolism and hyperreactivity of the hypothalamic-pituitary-adrenal axis have been implicated in patients with mood disorders, which may explain the clinically observable depressive, manic, and psychotic complications of steroid usage.

Although substance-induced mood disorders (SIMDs) have not been well studied, substance-induced depression may be more likely to occur in females, given that female gender is a significant risk factor for depression beginning in early adolescence. ^[1] Geriatric patients are more likely to take medication than younger patients and are at a higher risk for drug-associated side effects, ^[19]  but studies examining whether the incidence or prevalence of SIMDs differs based on age are lacking. ^{[20, 21]}

Diagnostic and Statistical Manual of Disorders, Fifth Edition (DSM-5) criteria for "substance/medication-induced mental disorders" is as follows: ^[1]

A. The disorder represents a clinically significant symptomatic presentation of a relevant mental disorder.

B. There is evidence form the history, physical examination or laboratory findings of both of the following:

1. The disorder developed during or within 1 month of a substance intoxication or withdrawal or taking a medication; and
2. The involved substance/medication is capable of producing the mental disorder.

C. The disorder is not better explained by an independent mental disorder (i.e., one that is not substance- or medication-induced). Such evidence of an independent mental disorder could include the following:

1. The disorder preceded the onset of severe intoxication or withdrawal or exposure to the medication; or
2. The full mental disorder persisted for a substantial period of time (eg., at least 1 month) after the cessation of acute withdrawal or severe intoxication or taking the medication. 

D. The disorder does not occur exclusively during the course of a delirium.

E. The disorder causes clinically significant distress or impairment in social, occupational, or other important areas of functioning.

The DSM-5 describes eight substance groups that may induce a mood disorder to further codify the diagnosis based on ICD-9 or ICD-10: (1) alcohol; (2) phencyclidine; (3) other hallucinogen; (4) inhalant; (5) opioid; (6) sedative, hypnotic, or anxiolytic; (7) amphetamine (or other stimulant); (8) cocaine; and (9) other (or unknown) substance. Additional DSM-5 specifiers for substance-induced mood disorders are (1) “with onset during intoxication” if criteria are met for intoxication with the substance and the symptoms develop during intoxication; and (2) “with onset during withdrawal” if criteria are met for withdrawal from the substance and symptoms develop during, or shortly after, withdrawal. ^[1]

As with many illnesses, a complete history helps to confirm the diagnosis of an episode of substance-induced mood disorder (SIMD). The onset of symptoms must coincide with the administration of the medication, intoxication by the medication, or withdrawal of the medication. Quick resolution of symptoms (eg, days or weeks after cessation of the medication) is presumptive evidence that the drug has induced the mood disturbance.

Perform a mental status examination. In addition, as there are many somatic illnesses that may cause depressive/manic symptoms, a comprehensive physical examination is necessary to exclude organic causes of mood changes.

In cases of suspected substance-induced mood disorders (SIMDs), it is important to understand that many common symptoms of depression (eg, fatigue, sleep changes, gastrointestinal [GI] problems) arise as adverse effects of medication, or as a symptom of a general medical condition. The temporal relationship of the medication to the development of the depressive and/or manic symptoms is essential to the diagnosis of an SIMD.

When evaluating a patient for possible substance-induced depression and mania, consider the following conditions as differential diagnoses:

• Adjustment disorders

• Antidepressant discontinuation syndrome

• Anxiety disorders (eg, generalized anxiety disorder)

• Bipolar affective disorder

• Caffeine-related psychiatric disorders (eg, caffeine intoxiation or withdrawal)

• Delirium

• Depressive disorders (eg, major depressive disorder or persistent depressive disorder)

• Endocrine disorders (eg, Addison disease, Cushing syndrome, adrenal insufficiency, adrenal crisis, and thyroid or parathyroid disease)

• Infectious conditions (eg, viral hepatitis, mononucleosis, human immunodeficiency virus (HIV) infection, and syphilis)

• Organic brain disorders (eg, amnesia, Alzheimer's, delirium, dementia, Huntington's, or Parkinson's)

• Multiple sclerosis ^[54]

• Schizophrenia or schizoaffective disorder

• Substance intoxication or withdrawal (eg, alcohol, benzodiazepine, cannabis, cocaine, nicotine, methamphetamines, and illicit opioids ^{[22, 23]} )

• Systemic lupus erythematosus ^[55]

If the patient is believed to be unreliable or if intoxication or overdose is suspected, obtain a urine or serum drug screen. Drug levels can be particularly helpful when evaluating a person who may be experiencing drug withdrawal.

Common screening tests include a complete blood count (CBC), thyroid-stimulating hormone (TSH) levels, electrolyte tests, blood urea nitrogen (BUN) and creatinine levels, and liver function tests (LFTs).

Other laboratory work is indicated for excluding any other illnesses as suggested by the patient’s history and physical examination findings.

Focal neurologic signs and/or cognitive changes, an altered level of consciousness, and a risk and/or history of head trauma should prompt consideration for imaging studies. Obtain a computed tomography (CT) scan of the head if trauma, bleeding, normal-pressure hydrocephalus, or subdural is suspected. (A lumbar puncture can exclude reversible normal-pressure hydrocephalus if this is suggested by findings from the patient’s history and imaging studies.)

Obtain a magnetic resonance imaging (MRI) and/or MR angiography (MRA) scan of the head to exclude a mass if focal neurologic signs are present.

Electroencephalography (EEG) may be used to differentiate a delirium from a mood disorder. An EEG usually does not differentiate between a delirium and a dementia.

When substance-induced mood disorder (SIMD) is suspected, it is generally advised to discontinue the offending agent. When doing so, the clinician should weigh the risk of withdrawing treatment (eg, on the clinical status of the underlying medical condition being treated with the offending agent). If the patient truly has a medical or psychiatric need for the drug, consider similarly efficacious but less toxic alternative medications. If no alternatives are available, lower the dose of the offending agent and/or shorten the duration of treatment—as medically indicated. Gradually tapering the offending agent should also be considered, particularly if this agent is associated with a withdrawal or discontinuation syndrome (eg, benzodiazepines or serotonergic antidepressants). If the mood symptoms do not subside within 4 weeks of fully stopping the offending agent, consider other etiologies for the mood symptoms. A retrial of the medication under close supervision may be considered in certain circumstances, such as if the medication was uniquely effective in treating a severe or debilitating medical illness.

Patient education

Instruct patients with SIMDs regarding the following, especially if substance abuse is involved:

• Cause - The patient and his or her family (when appropriate) should be made aware of the etiology of the mood disorder; education can prevent additional episodes and family may help monitor the patient for symptom recurrence.

• Prognosis

• Avoidance of offending agent

For patient education information, see the Depression Center, as well as Depression, Suicidal Thoughts, Bipolar Disorder, and Substance Abuse.

Consultations

For the majority of patients with an iatrogenic mood disorder, the place of first contact with the medical system is a primary care or specialty clinic. Consequently, primary care physicians should always be prepared to diagnose a mood disorder. Furthermore, one's threshold for detection of mood disturbances should be low for those patients who are prescribed medications known to have depression/mania as a possible side effect (eg, steroids, certain antihypertensive agents).

If the patient is suicidal, if psychosis or mania is suspected, or if depressive symptoms are severe, consult a mental health professional. Patients may need intensive outpatient or inpatient psychiatric care until the severity of the symptoms decline.

At the same time, for patients using illicit substances, a psychiatrist might be the first physician to diagnose an SIMD. Many times, psychiatrists, especially when working in a consultation-liaison service, are also responsible for the initial workup and diagnosis of an iatrogenic SIMD that could be mistaken for a primary mood disorder. In such cases, close cooperation and coordination with the primary team regarding treatment doses, duration, and alternatives are recommended.

Complications

The following are potential complications in patients with an SIMD:

• Interpersonal problems

• Loss of work time

• Homicide – In some rare situations, a person who is depressed or manic may become homicidal; homicidal behavior is another indication for inpatient treatment

• Prolonged hospital stays

• Suicide

Regular assessment for suicide risk is mandatory in any patient with depression or mania. Other risk factors for suicide include agitation, psychosis, past suicide attempts, a family history of suicide, other psychiatric comorbidity, and recent psychiatric admission.

Inpatient care

If the substance-induced mood disorder (SIMD) symptoms are severe or cause significant risk of harm to the patient or others, inpatient psychiatric care may be warranted. Any evidence of impaired reality or psychosis should lower the threshold for considering inpatient care. 

If there is uncertainty about the diagnosis, a prompt evaluation by the local emergency mental health system or a local emergency department is indicated.

Specific indications for inpatient care include the following:

• Serious suicidal ideation, which may include a plan

• Homicidal ideation

• Severe impairments in judgment leading to a moderate or high risk of danger to the patient or others

• Inability of the patient to care for himself/herself safely

Outpatient care

Periodically monitor the patient after stopping the offending agent until the mood symptoms have abated. If an abnormal mood persists, consider standard treatment for depression or mania, and consider etiologies other than substance-induced mood disorders.Standard pharmacotherapy may also be considered to treat depressive or manic symptoms while the patient is still taking the offending agent (eg, if the substance is being gradually tapered, or the clinician has assessed that the offending agent should be continued at the same or lower dose.

Prophylactic treatment of SIMD can be recommended on a case-by-case basis when there is significant risk associated with a recommended drug (eg, interferon [IFN]-alpha or steroid-induced depression, or steroid-induced mania). The following needs to be considered when prophylactic treatment is recommended: ^[24]

• The patient's past history, in which a history of previous affective episodes is an indicator of increased risk

• The degree of risk associated with the recommended treatment

With regard to the second item above, based on a few randomized, controlled trials, pretreatment with a selective serotonin-reuptake inhibitor (SSRI) can be considered for prevention of IFN-alpha–induced depression. Several case reports and open-label trials reported successful prophylactic use of lithium, valproic acid, chlorpromazine, olanzapine, and lamotrigine for steroid-induced mania. ^[6]

Idiosyncratic drug-induced depression or mania can be severe and life threatening. Cases of suicide have been reported. If a patient reports a history of mood symptoms upon exposure to a medication, avoid that medication in the future if at all possible.

Practice guidelines recommend an SSRI antidepressant for patients treated with IFN-alpha who develop any depressive symptoms. Many clinicians also recommend prophylactic SSRI treatment for patients with a history of depression who need IFN-alpha. For patients with a history of mood symptoms who require steroid treatment, prophylactic treatment with an antipsychotic agent (concomitantly with the steroid treatment) can prevent or decrease steroid-induced manic symptoms.

If a substance or drug is the true cause of a mood disorder, removal of the offending agent usually results in total recovery to predrug functioning. The resolution of symptoms can take time, but it is usually less than 4 weeks. Lack of symptom resolution after some time (eg, 1 month after discontinuation of the offending agent) may suggest a primary mood disorder. Careful monitoring is recommended until the mood symptoms resolve. No data have systematically examined whether morbidity and mortality from drug-induced mood disorders are different from those from any other depressive or bipolar illness. ^{[25, 26]}