Lewis M. Cohen, MD

Suicidal Behaviors and Ideation in Oncology Patients

” by Carey J. Myers, MD, PhD, and Carolina Retamaro, MD

We understand that the authors focused on assessments for psychopathology and suicide risk in patients undergoing disease-directed cancer treatments. In the section “Is Refusal of Treatment a Form of Suicide?” they touch on the concept that (consistent with long-held personal health values and in the absence of contributing psychiatric illness) some patients may choose to refuse—or start and eventually discontinue—disease-modifying therapies with the aim of letting their disease take its natural course, even though that might hasten death. What they do not explore are more active forms of hastening death.

We wish to address this omission. Ten states and the District of Columbia, representing roughly 22% of the United States population, have now legalized Medical Aid in Dying (MAID). While Myers and Retamaro do not directly raise the relationship between cancer suicides and MAID, we believe it important enough to pen a response.

Data from MAID-permitting states shows that patients whose advanced malignancies are likely to lead to their deaths within 6 months comprise the vast majority of those pursuing MAID: 70% to 80% of MAID deaths 

 Broadly, MAID laws allow terminally ill adults with decision making capacity to request and receive from a willing prescriber a lethal prescription to self-administer at the time of their choosing. All current state laws explicitly stipulate that such deaths are not suicide. The District of Columbia statute, for example, asserts that, “actions taken in accordance with this act do not constitute suicide, assisted suicide, mercy killing, or homicide.”

 Maine’s statute adds that state reports may not refer to acts committed under this Act as 

 and Washington State’s statute instructs that “state reports shall refer to practice under this chapter as obtaining and self-administering life-ending medication.”

 Such stipulations guide how to understand the practice, how to complete death certificates, and how to prevent insurers from refusing to pay death benefits. Most importantly, they are intended by lawmakers to help avoid the stigma of suicide.

A large and growing literature outlines the differences between decisions to decline disease-modifying treatments (or directly hasten one’s own death) and the psychopathology associated with suicidal behavior.

 The latter may trigger efforts at suicide prevention and intervention, sometimes even against an individual’s will.

 And while cases with characteristics of both conventional suicides and MAID do occur, clinicians must be prepared to distinguish between them in order to 

This does not challenge the undeniable fact that some cancer patients die by (conventional) suicide. We simply argue that MAID is not the same as those deaths.

We recommend that professionals in jurisdictions with MAID familiarize themselves with relevant statutes, and that terminally ill oncology patients receive accurate information about their rights under the law. Together with accurate predictions about prognosis and expected outcomes of available treatments, this information is essential for physicians who want to work skillfully and humanely with cancer patients, especially those who are expressing doubts about proposed treatment, or articulating a preference for hastened death.

is Maddie Katz Professor of Palliative Care Research and Education at the UCLA David Geffen School of Medicine, where he is Vice Chair for Clinical Affairs in the Department of Psychiatry. He is Board Certified in Psychiatry and Hospice/Palliative Medicine. 

is an emeritus professor of psychiatry at Tufts University School of Medicine and an emeritus professor of psychiatry at the University of Massachusetts-Baystate Medical School. He is the author of 

 and a Guggenheim fellow in Health & Medicine

is a professor of bioethics and psychiatry in the School of Medicine, Case Western Reserve University

is Distinguished Professor of Philosophy and Medical Ethics at the University of Utah. She has authored, co-authored, edited, or co-edited some twenty books, including

The Ethics of Suicide: Historical Sources

 She has been named one of the “Mothers of Bioethics.” 

 has been in the private practice of general and forensic psychiatry for over 35 years. Active in the American Psychiatric Association at both national and local levels, she is past president of Washington Psychiatric Society. 

is professor of psychiatry at the Mayo Clinic.

1. Oregon Health Authority. Oregon death with dignity act. February 26, 2021. Accessed June 10, 2021. 

2. California Department of Public Health. California end of life option act 2019 data report. July 2020. Accessed June 10, 2021. 

3. Washington State Department of Health. 2018 death with dignity act report. July 2019. Accessed June 10, 2021. 

4. District of Columbia Law 21-182(16). Death with Dignity Act of 2016.

5. LD 1313(HP 948). An Act to Enact the Maine Death with Dignity Act.

6. 70.245.180(1). The Washington Death with Dignity Act.

7. Puts MT, Tapscott B, Fitch M, et al. A systematic review of factors influencing older adults' decision to accept or decline cancer treatment. 

8. Strouse TB. Requests for PAD and the assessment of capacity. 

9. Cohen LM, Steinberg MD, Hails KC, et al. Psychiatric evaluation of death-hastening requests. Lessons from dialysis discontinuation. 

10. Yager J, Ganzini L, Nguyen DH, Rapp EK. Working with decisionally capable patients who are determined to end their own lives. 

11. Bostwick JM, Cohen LM. Differentiating suicide from life-ending acts and end-of-life decisions: a model based on chronic kidney disease and dialysis. 

12. Creighton C, Cerel J, Battin M. Statement of the American Association of Suicidology: “Suicide” is not the same as “Physician Aid in Dying.” American Association of Suicidology. October 30, 2017. Accessed March 30, 2021. 

Articles

Patients with terminal illnesses may choose to refuse treatment or even hasten their own deaths, but is that the same thing as suicide?

A New Question in End-of-Life Ethics

Psychiatrists are often asked to medicate patients with chronic kidney disease (CKD). Currently, there are 350,000 Americans who receive maintenance dialysis for renal failure, and this predominantly elderly population with multiple comorbidities is growing. In selecting psychotropic medications, psychiatrists need to be aware of the effects of renal dysfunction on medication absorption (ie, bioavailability), volume of distribution, metabolism, and excretion of the parent drug and the respective metabolites.

A few psychotropic medications, such as lithium, are primarily excreted by the kidneys, whereas most others are fat-soluble and are metabolized by the liver. Although the 

routinely recommends the use of partial dosages of psychotropic medications for renal patients, the data that support these recommendations are limited. There are even fewer empirical data available about psychotropic medication use for children. Thus, the clinician will find it necessary to engage in medication trials that are influenced by individual patient tolerability, response, and changes in clinical status.

In 2005, more than 106,000 patients with newly diagnosed end-stage renal disease (ESRD) began dialysis therapy, joining a dialysis population of 341,000 people and a transplant population of 144,000.

 Kidney disease involves people of all ages, including children, but most individuals fall into the adult and geriatric categories.

, the National Kidney Foundation Kidney Disease Outcomes Quality Initiative has divided chronic kidney disease into 5 stages. This article emphasizes the management of persons who fall into stages 4 and 5, in which renal function is greatly compromised, necessitating renal replacement therapy. In the nephrology literature, they have traditionally been labeled as having ESRD (the stages of CKD are a more recent taxonomy)-the term we will use throughout this article.

Although the psychiatric epidemiological literature has glaring limitations, clinicians need to familiarize themselves with the prevalence and hazards posed by psychiatric disorders in renal patients.

 For example, precisely quantifying the rate of depression in dialysis patients is limited by heterogeneity in study design, assessment procedures, and sample composition. Reported rates of depression have ranged from 0% to 100%. This variation underscores the need for a more refined and systematic approach to evaluating depression in renal patients. Furthermore, medication-related problems are extraordinarily frequent and were identified in 98% of a sample of 133 hemodialysis patients.

 In this study, patients had a mean of 6 ± 2.3 comorbid illnesses; 11 ± 4.2 different drugs were prescribed per patient, and a total of 475 medication-related problems were detected.

Patients with renal failure may evince alterations in any pharmacokinetic parameter-absorption, volume of distribution, metabolism, or excretion (

 Consequently, medication-related problems are frequent occurrences. Two other factors may be of equal or even greater importance, and both influence pharmacokinetics: Many ESRD patients are elderly, and the prevalence of associated severe comorbid disorders (eg, diabetes, hypertension) is high.

Bioavailability is the extent to which a dose of drug enters the systemic circulation. An oral dose is first absorbed from the GI tract; it subsequently passes through the liver, where metabolism and biliary excretion may occur. A decrease in bioavailability for some agents occurs at the level of drug absorption from the GI tract. Alterations in gastric alkalinity (caused by excessive urea generation by the internal urea-ammonia cycle) result in alterations in gastric pH that affect the absorption of psychotropic drugs.

What is already known about psychiatric treatment for patients with renal failure? 
? Clinical research and pharmacokinetic data about the psychiatric management of patients with renal failure are lacking. However, experience suggests that in this patient population, competent care may be delivered when the unique physiological, pharmacokinetic, dynamic, and risk/benefit factors are considered.

What new information does this article provide? 
? This article reviews the indications and dosing considerations for the most commonly used psychotropic medications. Clinical pearls and practical tips are offered to provide guidance and to assist clinicians as they initiate medication interventions for renal patients with comorbid psychiatric disorders.

What are the implications for psychiatric practice? 
? Patients with chronic renal disease may receive timely and effective psychiatric care for their debilitating conditions with psychiatric consultation and the judicious use of psychoactive medications.


Rival mechanisms that affect absorption in patients with renal failure include nausea and vomiting as well as delayed gastric emptying (from diabetic gastroparesis). Diabetes mellitus is a common cause of ESRD and is an important factor to consider when prescribing because of the relationship between atypical neuroleptics and increased serum glucose levels.

Volume of distribution entails the effects of dilution or concentration of medications in the body. The 2 overarching factors that influence distribution are the volume of distribution and protein binding. It is not uncommon in ESRD for elderly patients to become cachectic. These patients have less fluid and less body mass and a decreased volume of distribution. Consequently, a given dose results in a greater concentration of medication in a cachectic patient than in a non-cachectic patient.

The issue of protein binding is especially significant in ESRD. The principle plasma protein responsible for binding to acidic drugs is albumin, while for alkaline drugs, it is α

 Proteinuria and hypoalbuminemia are frequently found in renal failure, and there is an accumulation of endogenous binding inhibitors (eg, organic acids, uremic toxins). Binding inhibitors compete with drugs for the carrier protein binding site, and albumin undergoes conformational changes with hypothesized changes in binding properties. The results are diminished protein binding and an increase in the bioactive free fraction of acidic drugs in plasma. All other factors remaining constant, the greater the protein binding of a medication, the lower the dose of the drug required in ESRD, since there is a greater risk of toxicity when increased amounts of the free, unbound forms of acidic drugs are in the plasma.

Another factor to consider is that the circulating concentration of α

-acid glycoprotein may increase in patients maintained with renal transplant or hemodialysis. The effect may be a decrease in the unbound circulating fraction and diminished biological activity of the drug.

The metabolism or degradation of a medication is probably the area of pharmacokinetics that is least familiar, particularly because the intermediate products of metabolism are difficult to isolate and identify. The rate of renal metabolism by the renal brush border is predicted to decrease as the glomerular filtration rate (GFR) declines. In addition, evidence suggests that metabolism by the liver is variably altered in ESRD. Both the expression and function of cytochrome P-450 (CYP) 2C9 and CYP3A4 are decreased in severe ESRD.

 Although the literature is sometimes contradictory, it appears that in renal failure there is a general slowing of chemical reduction and hydrolysis, but there are normal rates of glucuronidation, microsomal oxidation, and sulfate conjugation.

Excretion takes place through the GI tract and from the kidney to the bladder, and this latter avenue is obviously absent following renal failure. Glomerular filtration, active tubular secretion, and passive tubular reabsorption are the 3 distinct mechanisms of excretion by the kidney, and all of them are affected in ESRD. The artificial substitution by a dialysis machine results in intermittent excretion rather than the continual process performed by healthy kidneys. Drugs are only occasionally metabolized to pharmacologically active compounds that are then normally excreted in urine. Fortunately, most psychotropic medications are metabolized by the liver, eliminated in bile, and excreted in feces.

There is a paucity of data documenting the efficacy and pharmacokinetics of psychotropic agents in patients with renal impairment.

 Most research is based on a handful of subjects, and well-controlled trials in psychopharmacology are rare. There are very few in vivo studies of pharmacokinetics in persons without kidney disease or patients with ESRD. Accordingly, in the absence of empirical data, clinical experience suggests that the majority of psychotropic medications can be safely used in the ESRD population. Some important considerations and exceptions are noted below.

Most psychotropic medications are fat-soluble and have large volumes of distribution. They are metabolized in the liver and metabo-lites are eliminated in urine and bile. Most of these drugs easily pass the blood-brain barrier, and they are not dialyzable. Attention needs to be paid to medications that have active metabolites-those that are highly plasma protein–bound, and those that have altered pharmacokinetics or pharmacodynamics.

Lithium is the only known psychotropic medication in which long-term use is associated with nephrotoxicity, renal insufficiency, and even ESRD. Although renal function will often improve if the drug is stopped, some patients require maintenance dialysis or kidney transplant. Attempts should be made to transition to the classic anticonvulsant medications divalproex and carbamazepine, but they may be inadequate mood stabilizers.

For the minority of bipolar patients with ESRD who continue to require lithium, treatment involves administration of a single dose (usually 600 mg) after each dialysis session. Because lithium is a readily dialyzable, small molecule, it is entirely eliminated by dialysis, and a single post-dialysis dose will result in a steady serum level. Serum lithium levels should be measured immediately before dialysis. If lithium is used to augment the therapeutic effects of antidepressants in treatment-resistant unipolar depression, a smaller dose (eg, 300 mg) may be given after dialysis sessions.

The ultimate goal for children with ESRD is kidney transplant. However, many will live years before transplant occurs, and many may need psychiatric care before and after the transplant. With the exception of recommended dosing of anticonvulsants, 2 treatment reviews do not list any psychotropic medications in their Tables of dosing considerations for children with significant renal impairment.

 predates the dramatic increase in use of psychoactive medications for children with a variety of internalizing and externalizing behavioral conditions.

Pediatric ESRD differs from the adult version in which diabetes mellitus and hypertension are more frequent long-standing comorbid disorders.

 Pediatric psychopharmacology ideally determines dosing on the basis of body surface area instead of body weight, but minimal information is available concerning either application with psychotropic medications.

 For children on dialysis and in need of psychiatric management, rely on the pediatric axiom for dosing: “start low and go slow.”

Many of the psychotropic medications-with the exception of the typical neuroleptics and somnolent tranquilizers-are used in pediatric psychiatry. In this population, some are relied on more often than others (eg, fluoxetine is preferred over paroxetine because of its longer half-life and research data that support its indication and use). Conservative initial dosing for pediatric indications would frequently be a 50% to 75% lower dose than the typical adult dose, with allowances to increase dosing during subsequent care, depending on tolerability, response, and changes in clinical status. This is especially the case for stimulant compounds such as amphetamine and methylphenidate, which have a higher proportion of elimination through the kidneys than many other medications.

Because there are no available data, even the package inserts for the stimulant medications do not specify any dosing adjustments to be made for children with ESRD. It appears that methylphenidate or the clearance of its inactive hepatic metabolite, ritalinic acid, is not as sensitive to the alkalinization of urine as amphetamine products.

With renal dysfunction and ESRD, lower dosing and slower increases during the clinical use of mixed amphetamine salts and dextroamphetamine is prudent. The alkalinization of urine associated with renal disease reduces the clearance of amphetamine and its metabolites to as low as 1%, while acidification increases clearance and can there-fore reduce levels. Consulting psychiatrists, in coordination with the primary care team and nephrolo-gists, can be of assistance with assessing the indications for and judicious use of psychiatric medications in any child with ESRD and psychiatric comorbidities. Families need to be fully informed as to treatment shortcomings.

Psychopharmacological pearls and renal failure

There are no formal guidelines for the treatment of psychiatric disorders in patients with renal failure, so the following points are based on clinical experience.

• ESRD is defined as a GFR of less than 15 mL/min/1.73 m

, in most cases accompanied by signs and symptoms of uremia

• Individuals with ESRD are in need of dialysis/transplant; some patients may require dialysis or transplant at a GFR of 15 mL/min/1.73 m

 or greater because of symptoms of uremia

• Practitioners who prescribe psychotropic agents should consider obtaining postdialysis blood drug levels to inform decisions regarding dose titration when reference levels are known

• Drug dialyzability is determined by the physical and chemical characteristics of the agent (eg, molecular size, protein binding, water solubility, volume of distribution) and the dialysis procedure (eg, dialysis membrane and flow rates)

• Peritoneal dialysis is generally much less efficient at removing agents than hemodialysis

• An agent incapable of being removed by hemodialysis cannot be removed by peritoneal dialysis

• The predominant form of CKD associated with lithium therapy is chronic tubulointerstitial nephropathy, which often presents with insidious development of renal insufficiency with minimal proteinuria

• The prevalence of reduced GFR associated with long-term lithium therapy is approximately 15%; a much smaller number of lithium-treated patients eventually have renal insufficiency that leads to dialysis

• Patients with mild to moderate chronic renal insufficiency associated with lithium can expect gradual improvement in GFR after lithium discontinuation

• Uninterrupted lithium exposure decreases the kidney’s endogenous ability for cellular regeneration; although a consensus does not currently exist as to the discontinuation threshold for lithium treatment because of diminished kidney function, it is suggested that repeated serum cre-atinine concentrations exceeding 140 μmol/L (1.6 mg/dL) indi-cate the need for nephrologist consultation

• Individuals with serum creatinine concentrations of more than 2.5 mg/dL are at very high risk for progression to ESRD compared with individuals who have serum creatinine concentrations of less than 2.5 mg/dL. Moreover, in individuals with a creatinine clearance of 40 mL/min or less, there is a higher probability of continued renal deterioration at lithium discontinuation than in those who have a creatinine clearance of more than 40 mL/min

• In patients with ESRD, a reduction in the lithium dose is required to prevent further toxicity, and a single dose of lithium is recommended after dialysis sessions; serum lithium levels obtained before and after dialysis are used to established appropriate dosing

• In treating depression, SSRIs (eg, fluoxetine) appear to be generally safe and well-tolerated; venlafaxine levels markedly increase in patients with renal failure, which theoretically could intensify hypertension, a common morbidity in ESRD; tricyclic antidepressants are generally avoided because of concern for safety and tolerability; duloxetine levels increase markedly in patients who have ESRD compared with patients who have normal renal function, and this medication is not recommended

Sizable numbers of patients have compromised renal function, and it is highly likely that psychiatrists will need to collaborate with primary care providers and nephrologists when prescribing psychotropic medications. There are unfortunately few formal pharmacological research studies that have investigated this subject, and clinicians must rely on basic knowledge of the factors that influence the effects and metabolism of psychoactive substances. This requires an appreciation of pharmacokinetics, including absorption, volume of distribution, metabolism, and excretion of the parent drug and metabolites.

Most psychotropic medications are not metabolized or excreted by the kidneys. In the absence of systematic data, caution is always necessary, but most of these medications are well tolerated and produce their accustomed beneficial effects. Renal dysfunction should not theoretically be associated with any unusual drug interactions.

Currently, there are 350,000 Americans who receive maintenance dialysis for renal failure, and this predominantly elderly population with multiple comorbidities is growing.

Lewis M. Cohen, MD

Articles

A New Question in End-of-Life Ethics

Psychiatric Issues for Patients With Renal Disease