Initiation of Dialysis Therapy


Acknowledgments

Some data reported in this chapter have been supplied by the USRDS. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy or interpretation of the U.S. government. Dr. Rivara is supported in part by funding from the Extramural Grant Program (EGP) by Satellite Healthcare, a not-for-profit renal care provider.

Introduction

In the United States, the current prevalent population of adults with chronic kidney disease (CKD) exceeds 30 million individuals and continues to grow annually. Among this group are more than 740,000 individuals with kidney failure requiring treatment with renal replacement therapy, defined as the need for maintenance dialysis or kidney transplantation. Although the incidence rate for new kidney failure cases remained relatively stable in the first decade of the 21st century, the period from 2012 through 2017 once again saw a rise in the rate of new end-stage kidney disease (ESKD) cases. In 2017, the last year for which national data from the U.S. Renal Data System (USRDS) are available, over 120,000 individuals with kidney failure-initiated dialysis therapy with either in-center hemodialysis (HD), peritoneal dialysis (PD), or home HD. Due to this continued growth in incident cases, the overall population of patients treated with maintenance dialysis has more than doubled over the past 20 years. Additionally, given the continued growth in the prevalence of obesity and diabetes mellitus, the burden of kidney disease on health outcomes and health care expenditures is likely to continue to increase in the coming decades. Estimates from around the world indicate that growth in the United States is mirrored by similar growth in many other regions in the world, in particular Australia, Asia, and Western Europe. The true incidence of ESKD is less clear in developing countries, where the availability of dialysis technology and financial resources to pay for dialysis and related kidney care may be severely limited.

The period of time immediately preceding and following initiation of dialysis is one characterized by exceptionally high risk for adverse patient outcomes. In particular, the first 3 to 4 months following the start of dialysis has been shown to be characterized by an annualized mortality rate exceeding 40%. The transition period to dialysis therapy is also associated with high rates of other adverse outcomes, including heart failure exacerbations, all-cause hospitalizations, and new strokes. Patient-reported outcomes such as depression, anxiety, and fatigue may also worsen around the time of initiation of dialysis due to the large number of logistical hurdles that must be overcome and the psychosocial adjustment that must occur in the transition from medical management of advanced CKD to dialysis therapy. The appropriate timing of initiation of maintenance dialysis also has profound implications for the financial burden to health care systems—for Medicare beneficiaries, annual per-person per-year costs exceed $75,000 for PD and $90,000 for in-center HD. Thus, for multiple reasons, optimizing the transition to maintenance dialysis is a critical challenge facing the nephrologist and one that encompasses appropriate preparation of patients with CKD for dialysis, timely nephrology referral, appropriate dialysis modality selection, and determination of the best timing for dialysis initiation accounting for clinical and other patient-important factors. This chapter focuses on the indications, timing, modality selection, and process for initiation of dialysis in patients with advancing CKD.

Indications for Initiation of Dialysis in Chronic Kidney Disease Patients

There has long been informal consensus among practicing nephrologists regarding true absolute indications for initiation of dialysis in patients with advanced CKD, largely severe manifestations of the uremic syndrome. Such absolute indications include uremic pericarditis, severe uremic encephalopathy, severe fluid overload leading to respiratory compromise and refractory to stepped diuretic therapy, and severe hyperkalemia refractory to medical management ( Box 3.1 ). Uremic pericarditis, characterized by pleuritic chest or back pain, the finding of a pericardial friction rub, and often a pericardial effusion, has a poorly understood pathophysiology and no definitive diagnostic laboratory biomarker or imaging finding. Prompt initiation of frequent or intensified dialysis is the only treatment shown to be effective for the management of uremic pericarditis, with resolution noted in up to 50%–87% of cases. Early signs of uremic encephalopathy include nausea, restlessness, drowsiness, and slowing of concentration, memory, and executive function. Eventually, accumulation of uremic toxicity will result in myoclonus, asterixis, nystagmus, and obtundation. Regular assessment and physical examination to identify these advanced features are thus critical for patients with advancing CKD, and recognition of any advanced features of uremic encephalopathy should prompt hospital admission and timely initiation of dialysis. Severe fluid overload leading to pulmonary edema and life-threatening hyperkalemia likewise require hospitalization for prompt dialysis therapy.

Box 3.1
Absolute Indications for Dialysis Initiation Among Patients with Advanced Chronic Kidney Disease

  • 1.

    Uremic pericarditis

  • 2.

    Uremic encephalopathy with altered mental status

  • 3.

    Recurrent or persistent hyperkalemia refractory to medical management

  • 4.

    Severe fluid overload leading with pulmonary edema or impaired wound healing

  • 5.

    Refractory nausea with vomiting

  • 6.

    Severe uremic bleeding

  • 7.

    Severe metabolic acidosis

However, the majority of patients who reach kidney failure requiring initiation of renal replacement therapy do not require emergent dialysis for one of these life-threatening complications. Instead, decision making of timing and preparation for dialysis initiation for most patients with advanced CKD involves a complex integration by the treating nephrologist of clinical and laboratory parameters, symptoms attributable to the uremic syndrome, logistic issues centered on access to care, psychosocial factors, assessments of patients’ coexisting illnesses and risk factors for decompensation without dialysis, and patient preferences ( Fig. 3.1 ). Biochemical markers of filtration function include serum creatinine and blood urea nitrogen which are useful in assessing patients’ progression toward the need for dialysis initiation, particularly when coupled with creatinine-based estimates of glomerular filtration rate (GFR); however, as discussed subsequently in greater detail, there remains no absolute threshold of GFR at which patients should be initiated on dialysis in the absence of other indications. Other important biochemical markers include markers of bone and mineral metabolism such as calcium, phosphorus, and parathyroid hormone; blood hemoglobin, ferritin, and iron saturation values, which can indicate worsening anemia and progressive loss of erythropoietin levels as a marker of loss of tissue mass; and increased serum potassium levels reflecting decreased filtration and distal tubular potassium secretary function. With the exception of refractory hyperkalemia, each of these other metabolic abnormalities is not an absolute indication for dialysis initiation but can be informative regarding trajectories of kidney function decline, provide challenges to nondialytic management, and together may help providers and patients work together to decide when to start dialysis. For example, severe hyperphosphatemia accompanied by hypocalcemia and secondary hyperparathyroidism may prompt consideration of activated vitamin D therapy or even calcimimetics, and if such therapy is unsuccessful in controlling these markers for a prolonged period of time, this may provide evidence that initiation of dialysis would benefit the patient from a renal osteodystrophy perspective. Such benefit, of course, would need to be weighed against the increase in treatment burden, cost, and complications associated with dialysis initiation.

Fig. 3.1, Example of a comprehensive patient-centered approach to decision making regarding preparation for and the timing of dialysis initiation. CKD , Chronic kidney disease.

Recently published data suggests that patients initiating dialysis due to refractory fluid overload has a significantly higher 12-month mortality and a greater risk for all-cause hospitalization and hospital readmission compared to patients initiating dialysis for other indications. While there is a paucity of data showing that earlier initiation of dialysis or other dialysis-related interventions may improve outcomes for this high-risk group, such data argue that close attention to volume-related complications is needed for these patients, and initiation of dialysis is a critical tool to be considered in managing refractory fluid overload in advanced CKD. At our center, we consider initiation of dialysis for advanced CKD when high doses of loop diuretics, such as torsemide 100 mg twice daily, are insufficient to control edema such that risk for skin breakdown or decreased patient mobility exists.

Identification of uremic symptomatology is a critical part of regular assessments by the nephrologist in the period of advancing CKD prior to dialysis initiation. Uremic pericarditis and true uremic encephalopathy have already been mentioned and are certainly absolute indications for dialysis initiation. Uremic pericarditis, manifested by a pericardial friction rub on examination and symptoms such as pleuritic or positional chest pain, arm pain, and back pain and sometimes accompanied by a pericardial effusion, should prompt hospital admission and rapid initiation of dialysis in the inpatient setting. High intensity of HD, typically daily treatment sessions until resolution of symptoms, should be considered. Similarly, true uremic encephalopathy, characterized by somnolence or obtundation in the presence of severe azotemia, should likewise prompt hospital admission for dialysis initiation. Such patients may be at high risk for further changes in mental status with dialysis initiation and should be monitored for the occurrence of the dialysis disequilibrium syndrome, discussed in greater detail later.

In addition to these clinical manifestations of the uremic syndrome, other signs and symptoms critical to assess in any patient approaching the need for dialysis include fatigue, lethargy, anorexia, loss of muscle or fat mass, sleep disturbances, bleeding tendency related to platelet dysfunction, memory or other neurocognitive dysfunction, pruritus, musculoskeletal pain, and sexual dysfunction. The accumulation of such signs and symptoms should prompt discussion among the patient, the nephrologist, and patient family members and caregivers to carefully weigh the pros and cons of dialysis initiation. Critically, there is no absolute number or severity of symptoms that necessarily should prompt dialysis initiation of every patient; instead, this assessment and decision making should be tailored to the individual patient and the individual clinical and social context. In some cases, patients may feel that a certain level of uremic symptom burden is acceptable, especially if nondialytic medical therapy is available and preferable over the burden of treatment associated with dialysis. Examples of nondialytic therapies include antiemetics such as ondansetron for the treatment of uremia-associated nausea, appetite stimulants for uremia-associated anorexia, and phosphate binders, ondansetron, naltrexone, or gabapentin for uremic pruritis. In other cases, shared decision making may lead to earlier initiation of dialysis due to valid concerns about medication burden, patient suffering, and family or caregiver support for close monitoring and symptom assessment in the home environment.

Adding to the complexity regarding the optimal timing of dialysis initiation based on uremic symptoms assessment is the uncertainty regarding which signs and symptoms of the uremic syndrome may improve or be ameliorated with the start of dialysis for each individual patient. Certainly, it is true that patients with advanced CKD prior to initiating dialysis have a wide variety and high burden of symptoms, and that these symptoms do prompt initiation of dialysis in many cases. However, a number of recently published studies have indicated that not only does dialysis treatment not eliminate all symptoms of the uremic syndrome but also that the number and severity of physical, mental, and emotional symptoms in patients with ESKD undergoing dialysis are not significantly different from those for patients with predialysis advanced CKD. Studies of patients with ESKD treated with maintenance dialysis have demonstrated a high prevalence of patient-reported treatment-related symptoms, including cramping, pruritus, insomnia, and fatigue. Certain complex symptoms, such as fatigue, are multifactorial in etiology and may be influenced by comorbid or associated illnesses such as anemia, depression, anxiety, sleep disturbances, and chronic pain. Thus, although fatigue is highly prevalent in predialysis patients with advanced CKD, it is similarly prevalent in ESKD patients undergoing maintenance dialysis and does not seem to be impacted by dialysis frequency or intensity. Similarly, although progressive CKD is associated with the appearance and worsening of cognitive impairment, there remains no good evidence that earlier initiation of dialysis impacts the severity or progression of cognitive difficulties and dysfunction. For example, in the Choices for Health Outcomes in Caring for ESKD (CHOICE) study, 26% of incident dialysis patients had worsening in self-reported cognitive difficulties 1 year after dialysis initiation compared to only 17% of patients reporting an improvement in cognitive function. Additional challenges are that many patients may downplay symptoms or may adapt to chronic symptoms or a lower level of physical functioning over time. That is not to say that dialysis does not improve certain classic constituents of the uremic syndrome, in particular anorexia, nausea, and vomiting. However, physicians should incorporate realistic expectations regarding dialysis-related symptoms into conversations with patients regarding preparing for dialysis initiation. Patients should be informed that dialysis may or may not alleviate all of their uremic symptoms and may or may not result in improvement in energy and functional status.

As a further challenge for the nephrologist in monitoring patients progressing toward the need for renal replacement therapy, there is currently no widely adopted and validated symptom assessment patient-reported outcome measure (PROM) or tool for assessment of uremic symptoms. A number of survey instruments assessing symptoms among prevalent dialysis patients have been developed, including the Dialysis Symptom Index (DSI), the modified Edmonton Symptom Assessment System (ESAS), and most recently, the Symptom Monitoring in renal Replacement Therapy–Hemodialysis (SMaRRT-HD) questionnaire. Additionally, the established standard for assessment of health-related quality of life (HRQOL) among dialysis patients, the Kidney Disease Quality of Life 36-item short form survey (KDQOL-36), does contain a symptom/problem subscale that can be used to track common symptom severity among dialysis patients. None of these instruments, however, has been widely validated among patients with advanced-stage 4 or 5 CKD who are approaching but have not yet required initiation of renal replacement therapy. Thus, assessment of uremic symptom burden, frequency, and severity for patients with advanced CKD in the current era is typically accomplished by the individual nephrologist in the course of routine office or telemedicine visits. However, further research is needed to establish best practices for uremic symptom assessment and incorporation into clinical decision making.

The central task of the nephrologist in assessing the optimal timing of renal replacement therapy for patients with advanced CKD is thus to balance the adverse consequences and impact of untreated kidney failure with the potential risk and treatment burden associated with initiation of maintenance dialysis. Excessive delay in the initiation of dialysis may lead to recurrent hospital admission for metabolic and volume complications of kidney failure, including hyperkalemia, pulmonary edema, and uncontrolled hypertension. Delay in dialysis may also expose patients to prolonged periods of protein-energy wasting, malnutrition, and uremia-related weight loss. Conversely, excessively early initiation of dialysis may expose patients unnecessarily to complications of the dialysis procedure itself, including infection-related risk from arteriovenous (AV) dialysis access, hemodynamic fluctuations related to ultrafiltration with subsequent myocardial stunning, and regional alterations in cerebral perfusion, and excess oxidative stress. Additionally, the requirement for daily or thrice weekly in-center or home dialysis treatments will have substantial effects for patients’ ability to work, travel, and function independently and may adversely impact HRQOL.

Clinical Practice Guidelines for Initiation of Dialysis

Given the inherent challenges in the assessment of the optimal time for dialysis initiation in a patient with progressive and advanced CKD, numerous clinical practice guidelines have been published over the past two and half decades to provide individual practicing nephrologists with guidance in the management of patients during this high-risk transition.

Past Recommendations for Timing of Initiation of Dialysis

Prior to 1997, when the National Kidney Foundation Dialysis Outcomes Quality Initiative (DOQI, now KDOQI) released its first comprehensive clinical practice guidelines for dialysis adequacy, there was no clear consensus regarding the optimal timing of dialysis initiation in patients with advanced CKD. In the 1990s, a number of observational research studies had shown a strong correlation between markers of protein-energy wasting and adverse outcomes such as increased mortality risk. It was also observed that objective measures of protein intake, such as the normalized protein catabolic ratio (nPCR), increased over the first 6 months after dialysis initiation, with the presumption that these changes reflected amelioration of uremic toxicity. Thus, when the KDOQI work group on dialysis adequacy released its recommendations in 1997, the specific guidelines commenting on the timing of dialysis initiation focused primarily on rigorous assessment of malnutrition and protein intake. Specifically, the group recommended that patients with advanced CKD should be advised to initiate some form of dialysis when the weekly renal Kt/V urea fell below 2.0, unless either edema-free body weight is stable, the nPCR is greater than or equal to 0.8 g per kilogram of body weight per day, or if there is an absence of clinical signs or symptoms attributable to uremia ( Table 3.1 ). The threshold of renal Kt/V urea of 2.0, which corresponds to a threshold of estimated GFR of approximately 10.5 mL/min/1.73 m 2 , was initially selected based on expert opinion, with supportive evidence over the subsequent decade extrapolated from the findings of two large studies of patients treated with PD: the Canada-USA (CANUSA) study and the Adequacy of PD in Mexico (ADEMEX) trial. Although, the findings from CANUSA, a prospective multicenter cohort study of 680 incident PD patients, indicated that each 0.1-unit/week increase in weekly total Kt/V was associated with a 6% decrease in overall risk for death, the ADEMEX trial subsequently showed no difference in outcomes when patients were randomized to PD prescriptions resulting in weekly Kt/V values of approximately 1.8 versus 2.3.

Table 3.1
Past and Current National Kidney Foundation/KDOQI Clinical Practice Guidelines on the Timing of Initiation of Maintenance Dialysis for Patients with End-Stage Renal Disease
Year of Release Statement
1997 Patients should be advised to initiate some form of dialysis when the weekly renal Kt/V urea falls below 2.0, unless there is (1) stable or increased edema free body weight; (2) nPCR ≥ 0.8 g/kg/day; or (3) absence of clinical signs or symptoms attributable to uremia.
2006 Patients who reach CKD stage 4 (estimated GFR < 30 mL/min/1.73 m2) should receive timely education about kidney failure and options for its treatment, including kidney transplantation, PD, HD in the home or in-center, and conservative treatment. Patients' family members and caregivers also should be educated about treatment choices for kidney failure.
Estimation of GFR should guide decision making regarding dialysis therapy initiation. GFR should be estimated by using a validated estimating equation (Table 1) or by measurement of creatinine and urea clearances, not simply by measurement of serum creatinine and urea nitrogen.
When patients reach stage 5 CKD (estimated GFR < 15 mL/min/1.73 m2), nephrologists should evaluate the benefits, risks, and disadvantages of beginning kidney replacement therapy. Particular clinical considerations and certain characteristic complications of kidney failure may prompt initiation of therapy before stage 5. Such complications include otherwise unexplained decline in functioning or well-being; gastrointestinal dysfunction; or weight loss or other evidence of malnutrition.
2015 Patients who reach CKD stage 4 (GFR < 30 mL/min/1.73 m 2 ) should receive education about kidney failure and options for its treatment, including kidney transplantation, PD, HD in the home or in-center, and conservative treatment. Patients' family members and caregivers also should be educated about treatment choices for kidney failure.
The decision to initiate maintenance dialysis in patients who choose to do so should be based primarily upon an assessment of the signs and/or symptoms associated with uremia, evidence of protein-energy wasting, and the ability to safely manage metabolic abnormalities and/or volume overload with medical therapy rather than on a specific level of kidney function in the absence of such signs and symptoms.
CKD , Chronic kidney disease; GFR , glomerular filtration rate; HD , hemodialysis; KDOQI , Kidney Disease Outcomes Quality Initiative; nPCR , normalized protein catabolic ratio; PD , peritoneal dialysis.

In 2006, KDOQI updated its clinical practice guidelines for dialysis adequacy to reflect widespread adoption of serum creatinine–based equations for estimating GFR and removed the discussion of calculation of the weekly renal Kt/V urea from the guidelines. Instead, the guideline stated that nephrologists should incorporate evaluation of the benefits, risk, and disadvantages of initiation of renal replacement therapy when patients reach stage 5 CKD, corresponding to an estimated GFR of 15 mL/min/1.73 m 2 (see Table 3.1 ). An substantial additional change to the guideline was a recommendation that certain clinical considerations or complications may prompt initiation of renal replacement therapy before stage 5, including a decline in patient functioning or well-being, gastrointestinal dysfunction, or weight loss. The 2006 update, therefore, represented a liberalization of recommendations vis-a-vis dialysis initiation in patients with advanced CKD, placing this decision within the realm of the subjective assessment and clinical judgment of the practicing nephrologist, with little objective data to guide such decision making.

The Initiating Dialysis Early and Late (IDEAL) Trial

Clinical practice guidelines for dialysis initiation practices published through the end of the 20th century were largely based on consensus and expert opinion, with little contribution from research data. In the last years of the 20th century and first decade of the 21st century, a number of observational studies were conducted to investigate the association of estimated GFR at the start of maintenance dialysis with subsequent clinical outcomes. The results of these studies, in general, supported the hypothesis that a higher estimated GFR at the time of dialysis initiation, when calculated using serum creatinine-based estimating equations, was associated with a higher risk for adverse outcomes. Interestingly, a smaller number of studies estimating GFR via timed urine collections and urinary creatinine clearances found no association between GFR at dialysis initiation and subsequent outcomes. These seemingly conflicting results raised concerns among many investigators regarding the influence of residual confounding on observed relationships between estimated GFR and dialysis initiation outcomes. In particular, there emerged concern that since serum creatinine–based estimating equations are subject to misclassification error in patients with substantial sarcopenia, the observed relationships between higher estimated GFR at dialysis start with higher mortality may, in actuality, represent an association of sarcopenia with adverse outcomes.

In light of persistent uncertainty regarding the benefit of using estimated GFR in patients with advanced CKD to time dialysis initiation, and in the absence of other objective measures of kidney function on which to base this critical clinical decision, the publication of the Initiating Dialysis Early and Late (IDEAL) trial in 2010 was a landmark event in nephrology. The IDEAL trial was the first, and to date remains the only, large randomized clinical trial to test the effect of estimated GFR at the time of dialysis initiation on subsequent clinical outcomes. The IDEAL trial, conducted between July 2000 and November 2008, enrolled 828 adults with advanced CKD, defined as an estimated GFR of less than 15 mL/min/1.73 m 2 body surface area, from 32 centers in Australia and New Zealand. Enrolled patients were randomized to initiating dialysis at an estimated GFR of 10–15 mL/min/1.73 m 2 (the “early start” group) or an estimated GFR of 5–7 mL/min/1.73 m 2 (the “late start” group). Patients were followed for a median of 3.6 years for occurrence of death (the primary outcome), cardiovascular events, and hospitalization and were also assessed on metrics of HRQOL. Disappointingly, for many nephrologists who had eagerly awaited the results of IDEAL, the primary finding from the study was that there were no significant differences between the randomized groups in the incidence of any of the primary or secondary outcomes.

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