Abstract
Abstract
Palliative care (PC) providers often prescribe psychotropic medications to address psychological and physical suffering of patients with serious medical illness. Consideration must be given to the significant medical comorbidities of the patient when selecting a medication. This article seeks to provide guidance on how to safely and effectively select a psychotropic agent for depression, anxiety, and other distressing symptoms for patients with serious illness. To do so, we draw upon a team of physicians and a pharmacist with training in psychiatry and PC to highlight the “Top 10” tips for selecting a psychotropic medication to provide relief for patients with serious medical illness.
Introduction
Psychiatric illnesses are common throughout the world. Diseases ranging from anxiety to depression to post-traumatic stress disorder (PTSD) are frequently seen either alone or concomitantly with medical illnesses. Unfortunately, being diagnosed with or suffering progression of a serious or life-limiting illness can lead to the new onset of mental illness or worsen preexisting psychiatric illnesses.
Across the population, primary care providers typically manage mental illnesses, with difficult cases being seen by providers with specialized training in psychiatry. In the setting of serious or life-limiting illness, however, palliative care (PC) clinicians are frequently called on to manage all forms of psychological suffering. This support can include counseling around coping, improving illness understanding, or strategies for managing physical and psychological symptoms. As PC has moved further upstream in medical illness and away from brink-of-death care, PC clinicians have been asked to develop expertise in managing complex psychiatric illnesses. Given that most PC providers receive limited training in psychiatry, managing the wide range of psychiatric illnesses experienced by patients with serious illnesses can be overwhelming.
This article aims to increase PC providers' comfort around managing psychiatric illnesses, and to harness the pharmacologic properties of psychiatric medications to manage mental health needs and physical symptoms. Specifically focused on psychopharmacology, our physician and pharmacist team trained in PC and psychiatry have created 10 tips that aim to help PC providers to overcome commonly encountered difficulties in managing psychiatric symptoms for patients with serious illness. Indeed, comprehensive PC can only be delivered by providers able to manage all forms of patients' suffering.
Tip 1. The QT Interval Is Only One Data Point and Must Be Cautiously Interpreted, Perhaps with the Help of a QT Interval Nomogram. When Concern Remains, Consider Duloxetine, Bupropion, Olanzapine, Aripiprazole and Avoid IV Haloperidol
Psychotropic medications are associated with prolonged QTc (interval from start of Q wave to end T wave, corrected for heart rate). Torsades de pointes (TdP) is associated with a prolonged QT interval, with one study reporting a 1.20 (95% confidence interval [CI] = 1.15–1.26) increase in relative risk for TdP for every 50 ms increase in the QT interval, 1 raising concern for increased cardiac mortality caused by psychotropic medications.2,3 In clinical practice, however, the benefits of the QT prolonging medication may outweigh the risks. For example, one large cohort study with Veterans Health Administration data showed that higher doses of citalopram (40 mg or greater) were associated with fewer hospitalizations with no increase in mortality. In this study, when citalopram was reduced below 40 mg to follow Food and Drug Administration (FDA) guidelines, there was a fivefold increased hazard for hospitalization within 180 days (hazard ratio = 4.5, 95% CI = 4.1–5.0). 4
When weighing risks and benefits for a hospitalized patient, a QTc >500 should prompt an evaluation for reversible risk factors, like hypocalcemia, hypokalemia, hypomagnesemia, hypoglycemia, and effects from other QT-prolonging medications.3,5 In addition, to more accurately determine the risk of TdP, use a QT nomogram that has a high sensitivity (96.9%; 95% CI = 93.9–99.9) and specificity (98.7%; 95% CI = 96.8–100) for TdP.6,7
Unfortunately, it is very difficult to rank the relative risk of QTc prolongation for antidopaminergic agents, with multiple studies showing conflicting results.8–11 In general, the highest risk of inducing TdP likely comes from intravenous haloperidol. The second generation agents, quetiapine, aripiprazole, and olanzapine, all have relatively low risk of TdP.8,9 For antidepressants, consider duloxetine or bupropion, neither of which have been shown to have significant effects on the QTc.9,12
Tip 2. If Side Effects of Dronabinol, Megestrol, and Dexamethasone Are Limiting Their Use, Consider Mirtazapine or Olanzapine to Help with Appetite and Weight Gain
To date, there are no effective treatments that have been approved by the U.S. FDA for cancer cachexia. Dronabinol and megestrol acetate are only FDA approved for AIDS-related cachexia.13,14 Dronabinol may cause delirium and only increases appetite, not weight, in patients with cancer. 15 Similarly, megestrol causes weight gain mainly by increasing edema, with the cost of significant thromboembolic risk.13,14 Off label, clinicians may turn to steroids, which do have a short-term benefit for quality of life, but are also associated with well-known adverse effects.13,16
There are two psychotropic medications to consider as alternatives. Mirtazapine is an excellent antidepressant with ∼13% of patients reporting significant weight gain of >7% body weight over one year in placebo-controlled trials studying depression. 17 In a small phase II trial in patients with cancer, mirtazapine helped 24% of patients gain ≥1 kg at four weeks. 18
Olanzapine seems to have even more significant effects, with patients in one recent meta-analysis gaining an average of 3 kg within 6 weeks and 10 kg at 38 weeks. 19 Weight gain is also more pronounced for patients with initial BMI <25, possibly because of increased lean body mass.20,21 In one randomized controlled trial (RCT) in patients with cancer, after eight weeks, 85% of patients receiving the combination of olanzapine and megestrol had weight gain of >5% compared with 41% receiving megestrol alone. 22 More importantly, 59% had improvement in their overall quality of life, compared with only 14% of patients on megestrol alone. Interestingly, no deep vein thromboses were reported over the eight-week study period.
Tip 3. Benzodiazepines Should Be Used for Acute Treatment of Insomnia Only When Other Alternatives Are Relatively Contraindicated or Have Not Worked
Nonpharmacologic interventions are first line for the treatment of insomnia. 23 Stimulus control, sleep restriction, relaxation, sleep hygiene, and cognitive therapy are more efficacious than medications, with slower onset but more durable effect. 24 The goal is to improve sleep quality/quantity and reduce insomnia-related daytime impairments. 25 There is high level of evidence for triazolam in sleep-onset insomnia, moderate evidence for temazepam in sleep onset or maintenance insomnia, and low to very low quality of evidence for all others. 26 Melatonin, trazodone, and diphenhydramine are not recommended. 26 When selecting an agent, consider compelling indications. With nausea and vomiting, consider olanzapine to also improve sleep, appetite, and mood. Quetiapine is an option for sleep and delirium. Mirtazapine, 7.5–15 mg, may also improve mood or appetite. In uncontrolled pain, manage the pain first with medications like gabapentin or tricyclic antidepressants (TCAs) with lower anticholinergic burden like nortriptyline and desipramine.
Benzodiazepines may be considered for short-term relief of acute, severe insomnia. Benefits wear off within 7–28 days with risk of rebound insomnia, disrupted sleep architecture, and cognitive effects. 23 One meta-analysis revealed that the number needed to treat sleep benefit in the older adult is 13 and the number needed to harm is 6—therefore, many groups recommend against the use of benzodiazepines in this population. 27 Although lorazepam may be one of the only intravenous options, valproic acid can be considered, especially with concomitant delirium or anxiety or in the setting of a prolonged QTc interval. 28
Tip 4. Dopamine Antagonists (Like Haloperidol or Olanzapine) Target Symptoms of Psychosis and Agitation in Hyperactive Delirium, Have No Benefit in Hypoactive Delirium and, Other Than Quetiapine, Should Be Avoided in Parkinson's Dementia
Evidence supporting the use of medication in the management of delirium is mixed, with several trials and meta-analyses establishing that dopamine antagonists, like first- and second-generation antipsychotics, do not alter incidence or duration of delirium, whereas others support their use.29–32 Nevertheless, dopamine antagonists can be used to target symptoms of delirium (e.g., psychosis and agitation). Hypoactive delirium is unresponsive to dopamine antagonists. 33
Dopamine antagonists should be avoided when caring for patients with low-dopamine states (e.g., Parkinson's disease or dementia with Lewy bodies) as they can precipitate neuroleptic malignant syndrome and worsen motor symptoms. If a dopamine antagonist must be used, quetiapine can be considered. Low-potency first-generation antipsychotics (e.g., chlorpromazine) and many second-generation antipsychotics (e.g., olanzapine and clozapine) have anticholinergic effects that may worsen delirium. 34 Effects on other receptors must also be considered. For example, quetiapine blocks alpha receptors possibly resulting in orthostatic hypotension (with rates as high as 27%), but this receptor binding may be helpful when sedation is intended (e.g., antihistamine effects of chlorpromazine and quetiapine). 3
In older adults with dementia, first- and second-generation antipsychotics have been associated with increased mortality and have a black box warning for their use when treating behavioral symptoms of dementia in this population. Those patients with hypoalbuminemia may require lower doses as most dopamine antagonists are extensively protein bound. Parenteral olanzapine should be used with caution with concomitant parenteral benzodiazepines as the combination can lead to hypotension, respiratory depression, and death.
Tip 5. When Picking an Antidepressant, Choose Escitalopram or Sertraline for Tolerability, Duloxetine If There Is Comorbid Pain, Duloxetine or Bupropion If a Melancholic Presentation of Depression, and Methylphenidate If at the End of Life
A recent large meta-analysis of RCTs involving 21 different antidepressants with 116,477 total patients demonstrated no clinically significant differences in terms of efficacy among the most commonly used antidepressants. 36 In PC, a meta-analysis of antidepressants for the treatment of depression showed that antidepressants were more efficacious than placebo after only four to five weeks (odds ratio [OR] = 1.93, 95% CI = 1.15–3.42). 37 This effect increased with time with an OR of 2.71 (1.50–4.91) at 9–18 weeks.
Given equivalent efficacy, we recommend choosing an antidepressant based on the following considerations: (1) activating versus sedating features of the antidepressant, (2) tolerability of the antidepressant, and (3) potentially useful side effects or other indications of the antidepressant. Table 1 provides a summary of useful properties for our top 10 antidepressants. For example, patients with a more melancholic presentation with severe anhedonia, weight loss, and early morning awakening would likely benefit more from activating antidepressants like duloxetine, bupropion, or nortriptyline. One recent meta-analysis supports this recommendation, as TCAs were more likely to lead to depression remission than selective serotonin reuptake inhibitors (SSRIs) for a melancholic presentation (OR = 0.44, 95% CI = 0.24–0.82). 38 Regarding tolerability, escitalopram and sertraline tend to be the best options. 36 Sertraline has the additional benefit of having few drug–drug interactions and demonstrated safety in very ill patients but can cause more gastrointestinal distress. 39 We generally prefer escitalopram to citalopram because it seems to have less QTc prolongation, although it can be more expensive. 10 Finally, in choosing an antidepressant, consider additional patient-related features like neuropathy, which would favor duloxetine, 40 poor appetite, which would favor mirtazapine, 41 or time to onset, which would favor methylphenidate. 42
Authors' Tips for Selecting among Their Top 10 Medications for Depression
For augmentation only.
Short-term benefit only (two to four weeks).
Mild caution for other cardiac side effects, particularly tachycardia and arrhythmia.
5HT, serotonin; DA, dopamine; NE, norepinephrine; QTc, QT interval corrected for heart rate; SNRI, serotonin and norepinephrine reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor; TCA, tricyclic antidepressant.
Tip 6. To Treat Anxiety, Always Consider a Patient's Prognosis, Starting with SSRIs, Serotonin and Norepinephrine Reuptake Inhibitors (SNRIs) or Buspirone, and Using Benzodiazepines at the Lowest Effective Dose for the Shortest Duration Possible
SSRIs are the most commonly used medications in managing anxiety, but they can take four to six weeks to have a full effect. They are generally well tolerated, with typical side effects including gastrointestinal distress, headaches, and sexual side effects. A recent meta-analysis of 89 trials, including 25,441 patients, demonstrated efficacy in the treatment of generalized anxiety disorder (GAD) for SSRIs, SNRIs, buspirone, and mirtazapine with good tolerability. 43 The evidence for efficacy in anxiety was highest for escitalopram, duloxetine, and venlafaxine. Although pregabalin also demonstrated efficacy, it may be costly and have the potential for misuse. 44 Buspirone is best used as monotherapy or to augment SSRIs, but also takes several weeks to have an effect and can cause sedation. 45 Buspirone may be a compelling choice for those patients on high doses of opioids for whom benzodiazepine co-administration may be dangerous.
The use of benzodiazepines in the management of anxiety is controversial. Benzodiazepine use is generally recommended for short-term relief of acute severe panic or anxiety.23,24 Tolerance to the anxiolytic and sedative effects of benzodiazepines develops between 7 and 28 days of treatment.23,46 If a patient has PTSD, benzodiazepines increase the risk for developing substance abuse (adjusted OR of 2.78) and worsening anxiety.47–49 Furthermore, benzodiazepines are associated with harm in older adults. 50 However, benzodiazepines can be used in the short-term while waiting for another medication to take effect or to engage in difficult situations (e.g., radiologic studies; procedures), and can be helpful if the prognosis is shorter.
Tip 7. Benzodiazepines Can Be Effective for Acute Seizures, Terminal Agitation, and Are Second Line for Anticipatory Nausea and Vomiting, Exercising Great Caution with Concomitant Central Nervous System Depressants Like Opioids
The use of benzodiazepines is robust in palliative and end-of-life care but the evidence is not overwhelmingly supportive.24,51 Clonazepam is indicated for treatment of generalized seizures and lorazepam, diazepam, and midazolam can abort acute status epilepticus. 52 The preferred evidence-based treatments for chronic anxiety disorder are cognitive interventions and serotonergic medications. 23 Benzodiazepines are not associated with respiratory depression when used appropriately and without concomitant opioids, even in patients with chronic lung disease, 53 opioids remain the preferred modality for anxiety associated with breathlessness as benzodiazepines are not efficacious.54–56 The National Comprehensive Cancer Network (NCCN) recommends benzodiazepines for anticipatory nausea second-line to behavioral interventions and other first-line treatments. 24 In the treatment of delirium, benzodiazepines have been associated with worsening mental status and should be avoided unless for sedation at the end of life.57,58 Cognitive and functional impairments associated with benzodiazepines are more pronounced in medically frail and older adults. 24 Use caution when co-prescribing opioids and other central nervous system depressants because of dose-related increased risk of overdose, assessing risk versus benefit and considering co-prescribing naloxone for high-risk patients.
Tip 8. Dopamine Antagonists Can Improve Symptoms of Nausea, Hiccups, and Anxiety
Many patients requiring PC specialist interventions for symptom management may not respond to conventional therapies. In particular, nausea and vomiting because of anticancer therapies can cause refractory symptoms that may respond to off-label interventions such as dopamine antagonist medications. Haloperidol, a potent first-generation dopamine antagonist, has been used extensively in PC for nausea management; however, the data are of poor quality and there are no RCTs to evaluate efficacy in PC patients. 59 Olanzapine, a second-generation dopamine antagonist with broad-spectrum antiemetic activity, has shown promise for the management of chemotherapy-induced nausea and vomiting. 60 To date, there is moderate evidence that oral olanzapine increases the likelihood of not being nauseous or vomiting during chemotherapy in adults with solid tumors compared with placebo. 61
Hiccups can become refractory and debilitating in the setting of advanced cancer and other gastrointestinal illnesses. Chlorpromazine is the only drug approved by the FDA for the treatment of hiccups, although other evidence-based agents are often used in PC settings.62,63 A systematic review demonstrated efficacy in prospective trials for baclofen, gabapentin, and metoclopramide, although RCT data only exist for baclofen and metoclopramide. Case series have shown efficacy for haloperidol as well. At present, there are no standard guidelines for treatment other than current FDA-approved chlorpromazine, although metoclopramide and baclofen have better safety profiles and more robust supportive evidence. 64 Consider trialing baclofen first, unless an agent can be selected based on compelling indications, such as gabapentin with co-occurring pain, chlorpromazine or haloperidol with nausea, or metoclopramide with gastrointestinal reflux or peritoneal carcinomatosis.
Treatment approaches for management of GAD include pharmacotherapy with SSRIs, (SNRIs), and in some settings pregabalin. Unfortunately, some patients may not tolerate these agents or may need further options. A systematic review demonstrated a role for dopamine antagonists for anxiety disorders, including monotherapy with quetiapine, which appears to be efficacious in GAD. The data on olanzapine and risperidone were inconclusive for efficacy in GAD.65,66 Furthermore, evidence is limited for the role of dopamine antagonists for social anxiety disorder, or panic disorder, although aripiprazole and risperidone can be used for treatment refractory obsessive–compulsive disorder as an augmentation to SSRI/SNRI medications.67,68
Tip 9. SSRIs Are First-Line Treatment for PTSD, and Quetiapine Monotherapy May Help with Sleep, Nightmares, and Anxiety Symptoms for Combat-Related PTSD
Patients with serious illnesses often carry scars of previous trauma. Pharmacotherapy for PTSD can significantly reduce symptoms, and current evidence suggests primary drug monotherapy with SSRIs as first line for most veterans with PTSD.69,70 Psychotherapy may be provided both with and without pharmacotherapy, although the evidence for combined therapy has not been proven to be more effective in PTSD. 71 Paroxetine is FDA approved for the treatment of PTSD; however, because of the high risks of anticholinergic and sexual side effects, as well as SSRI discontinuation syndrome because of the short half-life of paroxetine, we do not recommend this in PC patients. 72
In addition to SSRI therapy, quetiapine has been used as an effective adjunct for sleep disturbances and anxiety symptoms for patients with PTSD. In addition, a recent RCT of quetiapine (mean dose, 258 mg/day) as monotherapy in military veterans with combat-related PTSD demonstrated greater improvement in re-experiencing, hyperarousal, depression, and anxiety compared with placebo. 73 Aripiprazole has also been studied in a single placebo-controlled trial and a few open-label studies. Early data suggest that aripiprazole may be a reasonable adjunct to SSRI medications for patients with PTSD as well. 74
Alpha-blockers, in particular prazosin, have been recommended for the management of symptoms of PTSD, given that the drug targets subcortical alpha-adrenergic receptors involved in emotional hyperarousal and norepinephrine-mediated dysregulated sleep. However, a placebo-controlled randomized trial of military veterans with combat-related PTSD treated with prazosin demonstrated no significant improvement or difference versus placebo. 75 Thus, the data remain inconclusive as to whether certain patients would benefit from treatment with prazosin as previous studies had shown efficacy. 55 Given the risks for orthostatic hypotension in PC patients, especially near the end of life, we would recommend against the routine use of prazosin given the conflicting data on efficacy.
Tip 10. Ketamine May Be the Future of Managing Depression and Suicidal Ideation
Ketamine is a noncompetitive N-methyl-D-aspartate (NMDA)-receptor antagonist. At low doses, it has analgesic properties, and at higher doses it acts as an amnestic and anesthetic. 76 It can be administered intravenously, intramuscularly, intranasally, orally, and topically. The topical applications are limited to local pain management (e.g., mucositis).77,78 In recent years, there has been growing interest in the use of ketamine for the relief of depression and suicidal ideation as effects can be seen within days compared to two to six weeks for traditional antidepressants.
Intranasal ketamine was just approved by FDA for use in treatment of refractory depression, with several studies supporting the use of ketamine for this indication.79–81 There are also case reports, small open trials, and retrospective studies reporting its effectiveness in hospice and PC populations82–84 In most trials, patients are given ketamine 0.5 mg/kg intravenously once over 40 minutes. In other trials, patients were given ketamine 50 mg once intranasally.79,85 Clinically, access to ketamine may be limited as it is available as an anesthetic and thus only given intravenously. Results are seen within hours to days and effects can last for weeks. Ketamine has been shown to be safe for short-term use, although some patients complain of dissociative symptoms that can be mitigated by premedication with a benzodiazepine or haloperidol. 86
Conclusion
Psychiatric illnesses are common in PC. The skill needed to manage these conditions, including the role of psychopharmacologic drugs, is a crucial component of providing comprehensive PC. Given that the evidence base is increasing in PC, it is important that PC clinicians remain current to offer as many evidence-based interventions as possible to decrease suffering for our patients.
Footnotes
Author Disclosure Statement
No competing financial interests exist.
