What Special Considerations Are Needed in Treating Symptoms in Patients With End-Stage Renal Disease, and How Should Patients Who Stop Dialysis Be Managed?


Introduction and Scope of the Problem

Some degree of reduced kidney function is common in both the general population and people with have other serious illnesses. Chronic kidney disease (CKD), defined as the presence of either kidney damage or decreased kidney function for at least 3 months, regardless of cause, is present in over 10% of the population worldwide. Loss of kidney function, as manifested through a reduction in the glomerular filtration rate (GFR), can ultimately result in the most severe form of CKD, stage 5, when the estimated GFR is below 15 mL/min/1.73 m 2 and/or the patient starts kidney replacement therapy (KRT) (dialysis or, for eligible patients, kidney transplant). Patients who have received a kidney transplant, undergo maintenance dialysis, or have opted for nondialytic conservative management are said to have end-stage kidney disease ( Fig. 50.1 ). Recent research has revealed that many patients think “end-stage kidney disease” implies that they are near death or do not have treatment options. A movement is therefore underway to call end-stage kidney disease “kidney failure” instead, and the kidney community is trying to make its terminology more widely understandable by avoiding the terms renal and nephro when possible. For the purposes of this chapter, the term advanced CKD encompasses CKD stages 4 and 5 as well as kidney failure, or end-stage kidney disease (see Fig. 50.1 ).

Fig. 50.1, Classification of abnormal kidney function. Based on: National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis . 2002;39 (2 Suppl 1):S1-266.; KDIGO. Summary of recommendation statements. Kidney Int. Suppl . 2013;3(1):5-14.

It is important to note that kidney failure is not the only consequence of CKD; having an earlier stage of CKD is also a risk factor for cardiovascular disease and premature death. Thus clinicians treating patients with any serious illness, not just kidney failure, will often encounter CKD patients. These clinicians will also frequently encounter patients with acute kidney injury (AKI), a sudden decrease in kidney function, especially in the inpatient setting (see Fig. 50.1 ). Regardless of how it is defined and whether or not KRT is performed, AKI is significantly associated with in-hospital mortality. As a result, besides the practical consequences (e.g., need for alteration in medication dosing strategies), AKI also has important prognostic implications.

Data on the symptoms experienced by people with AKI are limited, while those with CKD, regardless of whether they undergo dialysis, suffer from a significant symptom burden. This symptom burden is associated with lower reported health-related quality of life and depression. Though the particular symptoms vary widely depending upon the population surveyed and the instrument used, symptoms are often not specific to CKD and may be the result of the CKD itself, the treatment for the CKD, comorbid conditions, or some combination of these. (See Table 50.1 for a description of some of the symptom assessment instruments intended for use in patients with kidney failure.) The quantity and severity of symptoms experienced are not clearly linked to level of kidney function (in patients not on dialysis), to stage of kidney disease (predialysis vs. on dialysis), to type of dialysis (hemodialysis vs. peritoneal dialysis), to choice of dialysis versus conservative nondialytic management, or to specific parameters of dialysis—for example, Kt/V, the most widely used measure of dialysis dose, which is the clearance of urea during dialysis divided by the volume of distribution of urea in the body. Unfortunately the key role that the kidneys play in drug metabolism and elimination means that many of the medications commonly employed for symptom management can either not be used or must be used with extra caution in patients with kidney dysfunction or failure.

Table 50.1
Selected Symptom Assessment Tools Designed for Use in Kidney Failure
Tool Patients for Whom It Was Developed Number of Items Content Pros Cons
Dialysis Symptom Index (DSI) Patients with kidney failure on maintenance hemodialysis 30
  • Symptom presence in past 7 days (yes/no)

  • How bothersome a symptom was (1, “not at all bothersome,” to 5, “bothers very much”)

  • Comprehensive

  • Too long for some patients to complete

  • Too much information for clinicians to act upon

Modified Edmonton Symptom Assessment System (Modified ESAS) Patients with kidney failure on maintenance hemodialysis or peritoneal dialysis 10
  • Extent to which a particular symptom is present at that moment (marked on a visual analog scale or a scale of 0, “none,” to 10, “severe”)

  • Very brief and easy to complete

  • Has a free-text response for patients to add a symptom

  • Limited number of included symptoms

  • Only elicits symptom presence and severity at the moment of completion

Integrated Palliative care Outcome Scale-Renal (IPOS-Renal) Patients with kidney failure on maintenance hemodialysis, peritoneal dialysis, or conservative nondialytic management 23
  • 15 symptoms: how much each has affected the patient over the past 3 or 7 days (“not at all” to “overwhelmingly”)

  • 6 questions: frequency of mood, anxiety, family’s/friends’ anxiety, sharing with family/friends, getting information (“not at all” to “always”)

  • 2 questions: practical problems, wasting time on health care

  • Goes beyond symptoms to ask about illness impact on psychosocial well-being

  • Has free-text responses for patients to add symptoms

  • Available in a version for completion by clinicians

  • Different response structures for different items may confuse patients

Across studies of patients with advanced CKD, whether or not patients are on maintenance dialysis, the most prevalent symptoms tend to be fatigue (also referred to as tiredness or lack of energy), which is present in somewhere between two-thirds and 90% of patients, difficulty falling or staying asleep, pruritus, and various types of pain, such as bone or joint aches and muscle cramps. The most severe symptoms tend to be fatigue, sleep problems, and pain. In the few studies that elicit information about sexual dysfunction, decreased interest in sex, difficulty becoming sexually aroused, and trouble with various aspects of sex are found in a significant minority or even majority of patients, both male and female, and are among the symptoms patients find most severe and concerning. Symptoms of mood and anxiety disorders are also common in patients with advanced CKD; as with other conditions, kidney dysfunction complicates the management of behavioral health problems.

Of particular concern given the high symptom burden of patients with advanced CKD is the fact that nephrologists and nephrology advanced practitioners are often unaware of the presence and severity of symptoms. In a seminal study of patients on maintenance hemodialysis, provider responses were less than 50% sensitive for the detection of 27 of 30 symptoms. Patient–provider agreement on the presence of symptoms was also exceptionally low for 25 of 30 symptoms, and providers underestimated the severity of 19 of 30 symptoms. Other research has demonstrated that though hemodialysis nurses and technicians believe patients report most of their symptoms, the patients themselves say otherwise. A study of large-scale implementation of a standardized symptom screening tool in hemodialysis units found that while it may have improved patient and provider symptom awareness, particularly for psychosocial symptoms, and facilitated patients raising issues with providers, it did not result in better symptom management, patient–provider communication, or interdisciplinary communication. Notably, providers’ confidence in managing symptoms actually dropped following implementation of this routine symptom screening, suggesting that when providers realized the true extent of patients’ symptoms, they also realized their shortcomings as symptom managers.

In addition to the symptoms that people with CKD experience, they also face a shortened life expectancy compared to people of the same age without CKD. This is especially true in kidney failure patients. For example, in 2018, a female aged 40 to 44 years in the general US population had 40.3 years of remaining life expectancy, while a female of the same age undergoing maintenance dialysis had only 10.2 years of remaining life expectancy. Though the life expectancy gap is less pronounced at older ages, it is still present (e.g., 9.9 remaining years for a male aged 75 to 79 years in the general US population vs. 3.4 remaining years for a male of the same age undergoing maintenance dialysis). Five-year survival for a patient initiating maintenance dialysis in the United States in 2013 was only 41.4% on hemodialysis and 46.9% on peritoneal dialysis, with the difference most likely explained by the conditions of the patients who chose each type of dialysis, such as functional status and comorbidities, rather than the dialysis type itself.

While over half of maintenance dialysis patient deaths with known causes are attributed to cardiovascular diseases, a notable minority are attributed to “withdrawal,” that is, a patient’s or surrogate’s choice to stop dialysis. In 2018, dialysis withdrawal was said to be responsible for 18.7% of deaths in maintenance hemodialysis patients and 17.8% of deaths in peritoneal dialysis patients. Given that most decisions to stop dialysis are made in the context of another serious illness, it might not be correct to say that dialysis withdrawal, rather than the other serious illness, is the true cause of death. Nonetheless clinicians must be aware that the end-of-life trajectory typically observed in a particular illness will be different in a patient with coexisting kidney failure, whether or not that patient withdraws from dialysis.

Symptoms seen after a patient withdraws from dialysis result not only from the lack of dialysis but also from other underlying conditions. Whether symptoms of volume overload, such as dyspnea and swelling, develop depends upon the patient’s volume status at time of withdrawal, how much urine the patient still produces, and how much salt and water the patient continues to consume. Uremic toxins, which are poorly characterized and understood, will accumulate in patients following dialysis withdrawal, causing anorexia, nausea and vomiting, a tendency toward bleeding, and a metabolic encephalopathy that might include somnolence, confusion, hallucinations, myoclonus, asterixis, seizures, and ultimately coma.

Death after dialysis withdrawal tends to occur in a more predictable time frame than death without withdrawal of organ support, though survival duration still significantly varies among patients. The most recent available data from the United States demonstrated a median survival of 7 days after dialysis withdrawal, with 70% of deaths in less than 10 days and 98% of deaths in less than 30 days. The variation is in part due to the patients’ degree of residual kidney function, which can be very roughly approximated by assessing urine output. Residual kidney function at the time of withdrawal depends upon how much patients had when they started dialysis and how long they have been on dialysis. It is important to note that almost all data on survival after dialysis withdrawal come from long-term kidney failure patients who underwent maintenance dialysis. Patients with AKI who recently initiated dialysis may still spontaneously recover some kidney function, meaning these patients’ life expectancy following dialysis withdrawal can be more difficult to predict than that of kidney failure patients who have stopped maintenance dialysis.

Relevant Pathophysiology

The kidney plays a major role in the maintenance of the extracellular environment that the body’s cells require to work properly. Though kidney function tends to be equated with GFR, glomerular filtration is in fact just the first step in an intricate process that demands coordination between several parts of the organ and results in formation of the kidney’s most well-known product, urine. The kidney puts into urine some of the waste products of metabolism, such as urea, creatinine, and uric acid. It also specifically adjusts the amount of water and electrolytes in the urine in response to net intake and endogenous generation. Through changes in reabsorption or secretion in the tubules, the kidney is able to fine-tune the excretion of water and solutes such as sodium, potassium, and hydrogen. In addition to urine, the kidney also produces several hormones. These include renin, prostaglandins, and bradykinin, which participate in the regulation of systemic and renal hemodynamics; erythropoietin, which stimulates red blood cell production; and 1,25-dihydroxyvitamin D3 (calcitriol), which is active in calcium, phosphorus, and bone metabolism.

As kidney function diminishes, patients will often suffer from the classic “AEIOU” that health sciences students learn as the major indications for renal replacement therapy: metabolic Acidosis, hyperkalEmia, drug Intoxication, volume Overload, and Uremia. In addition, they might have hypertension, hypocalcemia, hyperphosphatemia, elevated parathyroid hormone levels, and anemia. It is often not clear, however, how these abnormalities, many of which can only be detected through laboratory testing rather than by speaking with or examining the patient, translate into the symptoms that patients with reduced kidney function actually feel. Consequently, it is also not clear that fixing these abnormalities, as various medications and dialysis can do, results in improvement in symptoms. How the physiological disturbances seen with reduced kidney function might cause specific symptoms is addressed in the next section on treatment recommendations. In addition to the more obvious physiological disturbances associated with reduced kidney function, there are also less obvious ones, including chronic inflammation and malnutrition. These also might play a part in patients’ symptoms.

Since the kidney is responsible for the elimination of many drugs, dose adjustments are often necessary in patients with kidney dysfunction. Some drugs are potentially toxic to the kidneys themselves (e.g., NSAIDs), while others are toxic to another organ system when systemic concentrations rise too high because of decreased clearance by the kidneys (e.g., neurotoxicity from certain opioids). The risk–benefit ratio of a given drug often depends upon the patient’s level of kidney function, and clinicians may have to decide on drug choice and dosing with limited empirical data. For patients undergoing dialysis, clinicians must also consider whether dialysis will remove a substantial amount of the drug or its metabolites and therefore whether a supplemental dose is needed after dialysis. Clearance of a drug or its metabolites depends upon the type and amount of dialysis and how much residual kidney function the patient retains, but empirical data to inform clinical practice are sparse. When prescribing medications for patients with reduced kidney function, whether for symptom management or otherwise, clinicians should consult a reputable, regularly updated reference guide (see Chapter 4 for discussion of opioid pharmacology in the setting of kidney failure).

Summary of Evidence Regarding Treatment Recommendations

General Pain Management

People with kidney dysfunction can have pain from any etiology. In addition, pain can be related to the underlying kidney disease, particularly polycystic kidney disease, or the consequences of kidney dysfunction. For example, advanced CKD patients may develop neuropathic pain from uremic polyneuropathy, painful skin lesions from calcific uremic arteriolopathy (also known as calciphylaxis), and bone and joint pain from secondary or tertiary hyperparathyroidism or dialysis-related amyloidosis. For patients who undergo dialysis, aspects of the dialysis procedure itself can be painful. At every hemodialysis session, patients with arteriovenous (AV) fistulas or grafts as their dialysis access endure cannulation with two large-bore needles that are left in place for hours. Cramping during hemodialysis is also common. Patients on peritoneal dialysis can experience pain when dialysate is infused into or drained out of the peritoneal cavity, even in the absence of peritonitis.

The overall approach to pain in patients with kidney dysfunction is similar to that in patients with normal kidney function. Prior to treatment, the clinician must assess pain intensity, chronicity, possible reversible causes of the pain, type of pain (nociceptive, neuropathic, or combined), and treatment goals. Nonpharmacological treatment, especially for chronic pain, remains important in management. When pharmacological management is necessary, however, the choice of agents will often be different than it is in patients with normal kidney function. The WHO analgesic ladder, introduced in 1986 for patients with chronic cancer pain, with its recommendation to first try nonopioids before trying opioids and to consider use of adjuvant agents at every step, has some applicability for patients with kidney dysfunction. It should be noted, though, that the validity of the WHO analgesic ladder’s distinction between “weak opioid” and “strong opioid” has been questioned; moreover, this distinction is of much less relevance in patients with kidney dysfunction since most of the weak opioids (e.g., codeine) are not recommended for use in this population.

For more information on the safe and effective use of opioids in patients with kidney dysfunction, see Chapter 4 . Unfortunately, the two opioids considered safest for these patients, fentanyl and methadone, are also the two with which clinicians lacking specialized training in pain management are usually least familiar. Because of its pharmacokinetic and pharmacodynamic properties, methadone in particular requires a prescriber specifically educated in its use. For this reason, the threshold to refer a patient with diminished kidney function to a pain management or palliative care specialist, or to call an inpatient consult, should be low. The clinician without specialized training in pain management who needs to treat a patient with kidney dysfunction with an opioid should probably use hydromorphone, which conveniently is available in oral, rectal suppository, and intravenous/subcutaneous formulations.

Beyond its usual risks, acetaminophen carries no additional risks to patients with kidney dysfunction. Though many clinicians, especially nephrologists, have long advocated for completely avoiding the use of NSAIDs in patients with any degree of kidney dysfunction or even patients at risk for kidney dysfunction, this NSAID prohibition may lead to greater use of opioids. However, use of opioids can also be associated with adverse outcomes. Recent research has demonstrated that in the United States, maintenance dialysis patients who filled outpatient opioid prescriptions were more likely to die, discontinue dialysis, or be hospitalized than those without outpatient opioid prescriptions; the risk for altered mental status, fall, and fracture after filling an opioid prescription increased in a dose-dependent manner; and longer duration of opioid prescription was associated with hip fracture. For all these studies, however, one cannot rule out confounding by indication; that is, patients with conditions already placing them at higher risk for these outcomes were perhaps more likely to be prescribed opioids.

Meanwhile, though the adverse effects of NSAIDs, including gastrointestinal bleeding, cardiovascular events, worsening of hypertension, volume overload, hyperkalemia, and AKI in those with residual kidney function, are well described in other populations, there is little evidence, positive or negative, regarding specific adverse effects of NSAIDs in patients with kidney dysfunction, except that NSAID use has been linked to more rapid progression of preexisting CKD. Clinicians considering prescribing NSAIDs to patients with any degree of kidney dysfunction, including kidney failure, should take into account that these patients have baseline higher risk for many adverse events associated with NSAIDs, so the risk with NSAID exposure could be additive. In patients with CKD, the more advanced the CKD, the more caution that must be exercised. Some experts believe that NSAIDs are absolutely contraindicated in patients with stage 5 CKD not yet on dialysis if they are still pursuing life-extending treatments, as the adverse effects could either be fatal or precipitate the need for maintenance dialysis. For patients who are already undergoing maintenance dialysis, the risks to the kidney remain as long as residual kidney function does. These risks include loss of that residual kidney function, which can have negative prognostic implications. However, other adverse effects of NSAIDs, such as volume overload or hyperkalemia, are more easily treatable in patients undergoing regular dialysis. Taken together, the available facts suggest that a short course of NSAIDs may be reasonable in a patient with any degree of kidney dysfunction, but only after a thorough conversation with the patient about the risks and benefits of these drugs versus the alternatives. Of note, the preceding discussion applies only to systemic NSAIDs. Topical NSAIDs (e.g., gels, patches), which have limited systemic absorption, are a safer option, though effectiveness varies widely by formulation and is more pronounced in acute than in chronic musculoskeletal pain.

Three nonopioid, non-NSAID medications often used as pain control adjuvants deserve special mention in the context of patients with kidney dysfunction. The first, the muscle relaxant baclofen, is sometimes prescribed for pain thought to be secondary to muscle spasms. The medical literature is replete with case reports of patients with kidney dysfunction, many of them receiving maintenance hemodialysis, who experience altered mental status, ataxia, abnormal muscle tone, and falls following baclofen use. The majority of baclofen is excreted unchanged in the urine, prolonging the half-life and rendering the therapeutic index especially narrow in patients with reduced kidney function. Retrospective cohort studies have found an increased risk of hospitalization for encephalopathy after initiation of baclofen in patients with both nondialysis CKD and kidney failure. The risk in kidney failure patients—1 in 14 being hospitalized for encephalopathy within a median of 3 days of dispensing —is especially striking and argues for extreme caution with baclofen, including both lower doses and longer dosing intervals, if used for these patients. The other two medications, the anticonvulsants gabapentin and pregabalin, are popular choices for the treatment of neuropathic pain and are also given to advanced CKD patients for restless legs syndrome and pruritus, though neither drug has an FDA-labeled indication for these latter two symptoms. Dizziness and drowsiness are major adverse effects of both drugs. The primary route of elimination for both is the kidneys, meaning the half-life is extender in patients with advanced CKD. A large retrospective cohort study from the United States demonstrated an association between use of gabapentin or pregabalin and altered mental status or falls in maintenance hemodialysis patients, even at widely recommended doses. Careful administration, including starting with a low dose, slowly up-titrating, and closely monitoring for adverse effects, is warranted.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here