Commentary on Symptom Management of the Patient with CKD: The Role of Dialysis

Introduction

This evidence-based review by Cabrera et al. (1) examines the effect of timing of dialysis initiation on control of uremic symptoms in patients with advanced CKD. The authors consider the effects of dialysis initiation and its timing on health-related quality of life (HRQoL), malnutrition, fatigue, cognitive impairment, pain, neuropathy, sleep disturbances, and depression. They note that, overall, there is quite limited evidence quantifying these effects. There are unique challenges to examining the effect of dialysis initiation in interventional clinical trials (see below), and longitudinal observational studies have suffered from high rates of loss to follow-up and missing data. However, acknowledging the limitations of this evidence base, the authors report the following important observations. (1) A single randomized clinical trial (the Initiating Dialysis Early and Late [IDEAL] Trial) examined the effect of early (creatinine clearance =10–14 ml/min) versus delayed (creatinine clearance <7 ml/min or uremic symptoms) hemodialysis initiation on survival in initially asymptomatic patients, finding no beneficial effect of early initiation. (2) HRQoL was likewise not improved in early versus delayed hemodialysis initiation in the IDEAL Trial. Furthermore, in the longitudinal Choices for Healthy Outcomes in Caring for ESRD Observational Study, a majority of patients reported either no improvement or even worsening of HRQoL 1 year after initiation of maintenance dialysis. (3) Markers of nutritional status generally improve after dialysis initiation, but malnutrition remains highly prevalent and is not affected by a greater dose of hemodialysis. (4) Self-reported cognitive difficulties improve in only a minority (17%) of patients 1 year after initiating maintenance dialysis, with roughly one quarter of patients actually reporting worsening of self-perceived cognitive function. (5) Fatigue is reported by a large majority of patients with advanced CKD, but only one quarter report improvements in overall vitality after 1 year of dialysis. The hemodialysis treatment itself frequently worsens fatigue symptoms acutely. (6) Sleep disturbances (both self-reported and objectively measured) are highly prevalent in advanced CKD; only 19% of patients on maintenance dialysis reported improvement in sleep symptoms after 1 year. The authors suggest that—in the setting of this incomplete evidence—decisions regarding the timing of dialysis initiation for individual patients be made using the principles of shared decision making, in which patient preferences are incorporated after full discussion of potential benefits and also, risks and burdens imposed by dialysis treatment.

Delayed Treatment Clinical Trials

Clinical trials are the most definitive method for determining the effects of maintenance dialysis on uremic symptoms and patient-oriented outcomes. Observational studies to address this question are limited by the potential for confounding by indication, in which the reasons to initiate maintenance dialysis distort the true causal effects of dialysis itself. A delayed treatment trial is useful for evaluating treatments that cannot be completely withheld for ethical or practical reasons. Equipoise dictates that a rational and informed person should have no clear preference among the treatments in a clinical trial. A trial of early versus late dialysis initiation is justified on ethical grounds, because existing evidence cannot clearly distinguish whether a patient who has asymptomatic late-stage CKD should start dialysis or remain on medical therapies.

Delayed treatment trials may be limited by the inability to blind participants and investigators to the interventions. Blinding is used to reduce systematic differences in participant reporting of subjective outcomes and other activities outside of a trial. In a trial of early versus late dialysis initiation, study participants will be aware of their dialysis status, potentially influencing their perception and reporting of uremic symptoms. Participant awareness of dialysis may also affect their usual activities outside of the trial, such as physical activity or employment. Likewise, physician awareness of participants’ dialysis status could influence their decisions to prescribe specific medications or provide dietary counseling. Nonetheless, in real world clinical practice, patients and physicians are fully aware of dialysis treatments; blinding is not part of usual clinical care. Therefore, delayed treatment trials can be considered as a type of pragmatic trial design, which tests the practical effect of study treatments in broadly generalizable groups of people under realistic conditions.

To detect the effect of study treatments, the outcomes of clinical trials should be measured with a high degree of certainty. Uremic symptoms are complex constructs that may be difficult to disentangle from similar-appearing manifestations of other disease processes and that may be influenced by cultural expectations. Single-item point-in-time response questionnaires may not fully characterize symptoms that are unique to kidney disease. Newer uremic symptom assessment tools that include prospective assessment and crosscultural validation are critical to serve as outcomes of future clinical trials.

Existing trials of early dialysis initiation represent a necessary step toward understanding the true risks and benefits of maintenance dialysis treatments. The lack of clear benefits of early dialysis should not discourage additional trials in this critically important area. Instead, these results should inspire future trials that include larger populations, validated patient-oriented uremic symptom assessment tools, and greater contrasts in the timing and types of RRTs.

Disclosures

None.