To Name :
To Email :
From Name :
From Email :
Comments :

Evidence-Based Reviews

Subsyndromal depression

Help your bipolar patients feel better.

Vol. 7, No. 8 / August 2008

Mr. W, a 53-year-old divorced entrepreneur, presents to you for evaluation of poor concentration, decreased self-esteem, and difficulty making decisions that are interfering with his work. A longtime patient of another psychiatrist, Mr. W has a 26-year history of bipolar I disorder. He has not had a manic episode for 5 years but has had several depressive episodes.

During his last manic episode, Mr. W was hospitalized with expansive and irritable mood, racing thoughts, impulsive sexual behavior, psychomotor agitation, elevated self-esteem, marked distractibility, and paranoid ideas about his business partners. His discharge regimen included lithium titrated to 0.9 mEq/L and divalproex sodium, 1,500 mg/d, with lamotrigine, 200 mg/d, added to reduce depressive relapse risk. After several years of stable treatment, Mr. W complained of cognitive impairment. His psychiatrist discontinued lithium and added a low-dose stimulant—methylphenidate, 20 mg bid—to address Mr. W’s complaints of poor concentration.

Mr. W also is taking zolpidem, 10 mg as needed for onset insomnia, and receives weekly psychodynamic psychotherapy. His work performance problems persist despite these treatments, and his company is failing.

A poor course in bipolar disorder—as in Mr. W’s case—is frequently characterized by persistent or relapsing depression. Bipolar disorder is diagnosed by a manic, mixed, or hypomanic episode, but depression and depressive symptoms are most prominent in clinical practice. Likewise, major observational studies blame depression for most of the time spent ill in bipolar types I and II.1-8

A good deal of bipolar symptom burden is associated with subsyndromal depression—defined as having >2 but <5 DSM-IV-TR symptoms of major depression, with or without depressed mood or anhedonia. Subsyndromal depressive symptoms predict relapse to depression,5 and depressive symptoms are disproportionately responsible—compared with manic symptoms—for the impact of bipolar illness on patients and their families.9

This article offers clinically useful strategies to minimize subsyndromal depression in patients with bipolar disorder (Table 1). These strategies include an evidence-based approach to medication, the use of validated psychotherapies, regular sleep and socialization schedules, and careful monitoring of mood symptoms.

Table 1

How to minimize bipolar subsyndromal depression

Monitor symptoms using validated clinician- and patient-rated tools at all visits

Use evidence-based treatments first

Eliminate ineffective medications

Use adequate doses of medications for different mood states

Monitor and treat adverse effects of successful treatments

Monitor and minimize medications that can worsen symptoms

Watch for the impact of medical conditions on mood

Be attentive to alcohol and substance use (including caffeine, nicotine, and energy drinks)

Monitor psychotherapies for symptom worsening

Address comorbid psychiatric conditions

Regularize social rhythms

Initiate validated psychosocial treatments

Engage the patient as a active participant in treatment

Persistent depression

Randomized, controlled trials designed to obtain FDA approval of bipolar medications inadequately reflect the disabling, confounding nature of bipolar illness. Nearly all of these large studies of acute treatments for mood episodes are placebo-controlled trials with narrow inclusion and broad exclusion criteria. Eliminating subsyndromal symptoms is not their goal, and they are of little help in understanding how to manage residual symptoms.

A more realistic view of bipolar disorder comes from large observational studies that have examined its longitudinal course in outpatients under more or less ideal treatment conditions.10 These studies show that bipolar disorder is almost always recurrent and relapsing, but full recovery and functioning between episodes is not the norm. Most patients never achieve prolonged recovery, complete symptom relief, or return to full functioning.5,8,11

STEP-BD. Most patients in the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) never recovered from a depressed episode during 2 years of prospective follow-up under optimal care. Only 58% of patients who entered the study during an episode of illness achieved 8 consecutive weeks of euthymia.5

Collaborative Depression study. Longitudinal data from the National Institute of Mental Health’s Collaborative Depression Study6,7 showed:

  • patients with bipolar I disorder had depressive symptoms in approximately three-quarters of the weeks in which they reported significant symptoms
  • patients with bipolar II disorder were depressed in nearly all sick weeks.

These findings are consistent with STEP-BD data that showed nearly three-quarters of relapses (72%) occurred with depressed episodes and one-quarter (28%) with manic, mixed, or hypomanic episodes.5

The Stanley Foundation Bipolar Network had similar findings, with bipolar disorder patients reporting 3 times as much time spent with depressed mood as with elevated mood.8 Poor social and occupational functioning predicted poor outcomes, suggesting an interplay between subsyndromal depression, poor functioning, and relapse.

Risk factors

Rapid cycling may be a marker for persistent, subsyndromal symptoms. Rapid cycling is defined clinically as 4 distinct mood episodes—switching to the opposite pole or 2 episodes of the same pole separated by ≥8 weeks of partial or full recovery—in the previous 12 months. Rapid cycling usually is diagnosed retrospectively—introducing patients’ recall bias—but may be more of a marker for symptom persistence than for defined episodes.

In STEP-BD, 32% of study entrants reported ≥4 mood episodes in the previous year, yet only 6% of that subgroup had ≥4 episodes after 1 year of prospective follow-up.12 This suggests:

  • patients who retrospectively report rapid cycling may be chronically and persistently ill, rather than experiencing multiple discrete episodes
  • rapid cycling is a marker for symptom persistence, subsyndromal depression, and lack of sustained remission.

Treatment resistance in bipolar disorder is characterized by symptom persistence, more frequent episodes, and less time spent being healthy. Well-known factors increase the probability of treatment resistance:

  • comorbid anxiety disorders (present in ≤50% of patients with bipolar I and II disorder)
  • active and past substance use disorders (including nicotine dependence)
  • early age of onset of the mood disorder.13-16

Less documented is the likely association between treatment resistance and environmental stress, disrupted social rhythms, and irregular sleep. Clinical experience suggests, however, that patients feel better and stay in remission longer if they sleep regular hours, increase contact with a support network, and adhere to a daily structure. Hypnotics are not well-studied in bipolar disorder, but there is no evidence to suggest that they are not safe. Improved sleep hygiene, nonetheless, is a cornerstone of regularizing sleep, and pharmacologic treatment of sleep difficulties is not likely a replacement for it.

CASE CONTINUED: Restoring the cornerstone

You review Mr. W’s records. Recent lab values were essentially normal, with thyroid stimulating hormone 2.3 mIU/mL and stable renal function. He scores 11 on the Quick Inventory of Depressive Symptoms–self-rated version (QIDS-SR), indicating mild to moderate depressive symptom burden.

His mood chart and interview reveal that he has been depressed and anhedonic most of the day for 4 of the last 10 days. By systematically asking the depression questions in the DSM-IV-TR, you find that he does not meet criteria for depressed mood or anhedonia but has difficulty concentrating most of the day, persistent low self-esteem, and feeling “slowed.”

After you discuss lithium’s pros and cons with Mr. W, he agrees to try this mood stabilizer again. You explain the importance of preventing relapse to mania and of monitoring his cognitive performance at work.

Over time, you titrate lithium to a moderate serum level (0.5 to 0.7 mEq/L) and treat a resulting mild tremor with propranolol, 20 to 40 mg/d. Mr. W is tolerating lamotrigine well, so you continue this medication because of its potential to decrease the probability of relapse to depression. You also continue zolpidem, as needed, but discontinue methylphenidate because you think it may be contributing to sleep difficulties.

Managing medication

Nine drugs are FDA-approved for acute bipolar mania, but treatments for bipolar depression, maintenance treatment, and relapse prevention are far fewer, often partially effective, or effective for a limited number of patients. When depressive symptoms fail to resolve, a reasonable approach is to review patients’ medications and suggest alternatives with proven efficacy for bipolar disorder (Table 2). Patients can then accept or reject various options based on personal preference.

Combination strategies. Antimanic treatment is the cornerstone of treating bipolar I disorder, and preventing manic episodes should be a primary treatment goal. Thus, consider continuing treatments that have prevented mania for your patient—as lithium did in Mr. W’s case—while adding treatments aimed at depression. For example, adding lamotrigine to any antimanic agent is reasonable, especially if doing so does not add substantially to your patient’s side-effect burden.

Minimize antidepressants. Given the predominance and persistence of depressive symptoms in bipolar disorder, one can understand why clinicians and patients might try standard antidepressants without clear evidence supporting this practice. Antidepressants—especially venlafaxine and tricyclic antidepressants—are the most common and likely suspects when patients experience switching to mania, rapid cycling, and symptom persistence.17 Antidepressants’ negative effect has not been clearly defined, however, and may be patient-specific (related to patient factors rather than intrinsic to the compound).

In my clinical experience, minimizing antidepressant use in bipolar depression hastens rather than delays patients’ recovery. A prudent approach would be to use the minimum dose necessary and discontinue the antidepressant if possible. Also minimize medical pharmacotherapies—including corticosteroids and oral contraceptives—that may worsen mood symptoms, especially in patients with this history.

Avoid under-dosing. Inadequate dosing and duration often are overlooked as causes of treatment resistance in bipolar disorder and other illnesses.18 Bipolar disorder medications are hardly benign; every drug approved for any phase of bipolar disorder has a black-box warning. Understandably, clinicians and patients try to choose medications and dosages perceived to be most tolerable. Full-dose treatment trials may be warranted, however, given the high probability of incomplete recovery, impaired functioning, and risk of relapse with ineffective dosing.

Address iatrogenic causes. In addition, identify and eliminate medications and treatments that may be perpetuating patients’ bipolar symptoms. Stimulants such as methylphenidate and amphetamines may contribute to sleep disturbance and manic relapse and might be minimized or eliminated in a patient with continued symptoms and sleep disturbance.19

Antipsychotics. Quetiapine and the combination olanzapine/fluoxetine are FDA-approved for acute bipolar depression episodes, but not all atypical antipsychotics show antidepressant effects in bipolar disorder:

  • Two trials of aripiprazole for bipolar depression failed to show benefit.20
  • A trial that compared risperidone with lamotrigine and inositol for treatment-resistant bipolar depression suggested that risperidone may have hindered recovery.21

Other agents. Lamotrigine’s benefit in acute bipolar depression is controversial, as no trial has shown unequivocally that it is more effective than placebo. Modafinil, 100 to 200 mg/d, was significantly more effective than placebo as an adjunct to mood stabilizer therapy in a 6-week study of bipolar depression.22 This result in a cohort of 85 patients has not been replicated, however, and modafinil’s long-term safety in bipolar disorder is unknown.

Table 2

Subsyndromal bipolar depression: Recommended medications*


Initial and maximum dosages

Clinically important side effects


Start at 50 mg and titrate to 300 mg within 4 to 7 days; maximum 600 mg

Sedation, somnolence, weight gain, gastrointestinal side effects, lipid abnormalities, increased fasting glucose, increased risk of diabetes


Start at 6 mg/25 mg; maximum 12 mg/50 mg

Weight gain, sedation, gastrointestinal side effects, lipid abnormalities, increased fasting glucose, increased risk of diabetes


Must be titrated per package labeling; start at 25 mg and titrate to 200 mg (12.5 mg titrated to 100 mg if patient is on valproate, 50 mg titrated to 400 mg if on carbamazepine or other enzyme inducer); maximum (per label) 500 mg

Rash, headache, balance difficulties, clumsiness; Stevens-Johnson syndrome or toxic epidermal necrolysis are rare but potentially fatal


Start at 300 to 600 mg and use moderate blood levels (0.4 to 0.7 mEq/L); if no improvement in 4 to 8 weeks, titrate to 0.8 to 1.1 mEq/L

Tremor, nausea, diarrhea, increased thirst, increased urination, hair loss, thyroid abnormalities, weight gain, acne, worsening of psoriasis, diabetes insipidus, renal insufficiency


Start at 500 to 750 mg and increase to 15 to 20 mg/kg; usual target blood levels are >50 mg/dL

Nausea, abnormal liver function tests, weight gain, hair loss


Start at 5 mg; maximum 30 mg

Weight gain, sedation, somnolence, lipid abnormalities, increased fasting glucose, increased risk of diabetes


Start at 50 to 100 mg and increase to 200 mg; higher dosages have not been systematically studied in bipolar disorder

Nervousness, insomnia

EPS: extrapyramidal symptoms

* Medications are listed in from most to least evidence supporting their use in treating bipolar depression

CASE CONTINUED: Distressed by psychotherapy

You ask Mr. W about his psychodynamic psychotherapy, and he says that exploring his early life experiences and his work difficulty is increasing his anxiety. You recommend switching to cognitive-behavioral therapy (CBT) to work on delegating tasks that are not his strong areas and focusing on his marketing talents. You also encourage him to maintain regular sleep-wake cycles.

Did you miss this content?
A depressed adolescent who won’t eat and reacts slowly