Hemodialysis Prescription

Abstract

Chronic hemodialysis has evolved over more than half a century from a fairly crude, but life-saving treatment for the privileged few to a way-of-life treatment for approximately 4.35 million patients globally. Historically, the dialysis prescription was largely generic (typically 4 hours three times a week), but since the measurement of dialysis clearance/dose in the 1980s, it has become increasingly apparent that one size does not fit all and that chronic underdialysis has a poor outcome. It has also become increasingly clear that residual renal function has huge benefits and hemodialysis prescription can be individually tailored on the basis of residual renal function to the patient’s potential advantage.

This chapter covers the calculation of dialysis clearance, the techniques used, and potential flaws. It also explores the importance of individualizing assessment and prescription of hemodialysis.

FormalPara Learning Objectives

• To understand the principles of and factors affecting hemodialysis clearance.

• To cover the ways of measuring hemodialysis clearance and potential limitations.

• To appreciate the need for tailored and responsive dialysis prescription.

1 Introduction

Chronic hemodialysis has evolved over half a century from a fairly crude life-saving treatment for the privileged few to a way-of-life treatment for approximately 4.35 million patients globally. Historically, the dialysis prescription was largely generic (typically 4 hours three times a week), but since the measurement of dialysis clearance/dose in the 1980s [1], it has become increasingly apparent that one size does not fit all and that chronic underdialysis has a poor outcome (◘ Figs. 81.1 and 81.2 which illustrate the acute and chronic ends of the underdialysis spectrum). It has also become clear that residual renal function has huge benefits [2] and hemodialysis prescription can be tailored on the basis of residual renal function to the patient’s potential advantage.

Fig. 81.1

a, b Show a CXR and CT chest, respectively of a patient who found dialysis difficult and repeatedly curtailed dialysis sessions or missed them altogether. The CXR is highly suggestive of a gross pericardial effusion, in this case due to uremia pericarditis secondary to chronic underdialysis. Uremic pericarditis is, perhaps, the most overt clinical presentation of chronic underdialysis and is associated with a high mortality. b CT chest of a patient with uremic pericarditis. The black arrow shows a large, circumferential pericardial effusion secondary to chronic underdialysis. This condition can resolve with intensive, careful dialysis, but prevention is considerably better than cure. Therefore, monitoring and appreciation of underdialysis should provoke an early assessment of the causes and potential solutions

Fig. 81.2

Signs of dialysis amyloid (flexion contractures and carpal tunnel deposition) in an anuric patient who had been on dialysis for 17 years. The patient had no residual renal function, chronically high β₂M, and evidence of neuropathy. Dialysis amyloid illustrates the other, much more chronic end of the spectrum of ‘underdialysis’. Although, this patient received regular dialysis with reasonable urea clearance on standard dialysis, it was not sufficient to clear middle molecules

Assessment of dialysis clearance is based on measuring the removal of waste products of cell metabolism which accumulate in patients with chronic kidney disease. Urea is produced from the breakdown of proteins and recycling of amino acids. Although there is debate as to the toxicity of urea, urea is used as a surrogate measure of retained azotemic molecules [3]. Even though most laboratory assays report a composite of urea and some other small nitrogenous compounds, including ammonia, measurement of urea is still more reliable than that of creatinine. Increasingly, sophisticated and routine assessment of dialysis clearance, alongside measurement of residual renal function permit a more nuanced and bespoke prescription for patients requiring chronic dialysis.

2 Diffusive Dialyzer Clearance

Urea is a small uncharged molecule that readily diffuses. Diffusion is dependent on the concentration gradient which is increased by faster counter current blood and dialysate flow, dialyzer surface area, narrow-diameter capillary fibers reducing the distance molecules have to move, and resistance to passage through the wall of the capillary fiber. Each dialyzer comes with a manufacturer’s measurement of ex vivo urea clearance at different blood pump and dialysate speeds for different sized dialyzers (KoA), but this clearance is greater than that in clinical practice, as it is not derived from experiments using human plasma.

3 Convective Dialyzer Clearance

Convective clearance is generated by applying an external hydrostatic pressure that drives water across the dialyzer membrane, and membrane permeability to water is termed hydraulic permeability, and differs with dialyzer membrane’s composition and structure. Whereas molecular size is important for diffusion, with urea (MWt 64 d) moving 18 times faster than β2-microglobulin (MWt 11, 800 d), provided that the molecule can pass through the membrane, then convective clearance will be the same, so convective clearance will be similar for both urea and β2-microglobulin using a high-flux dialyzer. Dialyzer flux, which depends on membrane pore size, is now defined according to the clearance of β2-microglobulin.

During a standard hemodialysis treatment, there will be some convective exchange if there has been ultrafiltration. In addition, depending upon dialyzer hydraulic permeability, there may be internal diafiltration, which can be as much as 6–9 L during a four-hour dialysis session. Convective movement will potentially reduce the concentration gradient and may have a small effect in reducing diffusive clearance.

4 Sessional Dialyzer Clearance

The National Cooperative Dialysis Study (NCDS) was the first prospective dialysis study to examine the effect of urea clearance by randomizing patients to a lower and higher time averaged blood urea nitrogen target. During a 12-month follow-up, more patients dropped out of the study or had hospital admissions in the higher time averaged urea group. Gotch and Sargent subsequently reviewed the data, and developed the concept of dialyzer urea clearance, which only required pre- and post-dialysis blood sample, compared to three samples need to calculate time-average urea [1]. To be able to compare urea removal during dialysis (Kt) between patients, they argued that urea body distribution was equivalent to the volume of body water (V). This was estimated using equations derived from anthropometric measurements in healthy subjects.

5 Kt/Vurea Model

The simplest model of dialyzer urea removal is to calculate the urea reduction ratio (URR); however, this does not take into account any additional clearance from ultrafiltration losses. This led to the development of a first-generation equation to estimate Kt/V taking into account ultrafiltration losses. Single pool Kt/V (spKt/V) assumes that urea is distributed equally throughout body water and during the dialysis session urea is removed equally from body water. As this is not the case some advocate the use of a dual pool Kt/V model (dpKt/V) (◘ Fig. 81.3) [4, 5].

Fig. 81.3

A schematic of the difference between the single and dual pool model of urea clearance, illustrating the potential for rapid removal of urea (and other small molecules) from the extracellular space, but more limited, and slower access to the larger intracellular space. Single-pool modeling assuming a singular fluid pool, dual-pool modeling endeavors to account for the difference between these two pools

However, during dialysis blood supply to the skin and internal organs (muscle, heart, liver, kidney, and gastrointestinal tract) falls, and blood pools in the larger capacitance veins, such that shortly after the end of the dialysis session, plasma urea concentrations rebound upward (see ◘ Fig. 81.4).

Fig. 81.4

Urea clearance curves demonstrating extracellular and intracellular clearance of urea (using a single pool model) with rebound of urea on cessation of dialysis. Sampling a later urea yields the equilibrated Kt/V which will be less of an underestimate that the spKt/V derived from an immediate post-dialysis urea [6]

To compensate for this effect, firstly, it is important to minimize this effect by taking precautions when taking the “post-dialysis” blood sample by using the slow-flow method, whereby blood flow is slowed down to allow for plasma urea concentration to re-equilibrate before blood sampling. Thus, a further revised equation was developed to take into account urea equilibration (eKt/V) to compensate for the differential removal of urea from body tissue compartments [7].

All these equations are based on the premise that patients dialyze three times a week, and that blood samples are taken pre- and post- midweek dialysis session. To allow comparison between patients dialyzing more or less frequently the concept of standardized Kt/V (stdKt/V) was introduced. To calculate stdKt/V, the derived stdK is multiplied by time (10,080 minutes per week) and adjusted by V.

As many patients now start dialysis with residual renal function, this has led to the practice of incremental dialysis, with patients bringing in 24-hour urine collections which are used to calculate urinary urea clearance (Kr). This is then adjusted according to the volume of urea distribution (body water) to estimate the equivalent urinary renal clearance (EKRc) [8].

6 Alternate Methods of Estimating Urea Clearance

Advances in dialysis machine technology can be used to estimate urea clearance, termed online clearance. One method is to use changes in dialysate conductivity. Although different in molecular weight, sodium and urea have almost identical diffusion coefficients at 37 °C (Na⁺: 1.94 × 10 ⁻⁵ cm²/s, Urea: 2.20 × 10⁻⁵ cm²/s). By comparing the changes in dialysate conductivity of fresh and spent dialysate to short-term pulses of a change in dialysate sodium, designed to achieve a detectable diffusion of sodium ions across the dialyzer membrane, it is then possible to estimate urea clearance. To prevent a positive sodium balance, pulses alternate between increased and decreased sodium concentrations. The frequency of pulses can be altered from every 12 to 30 minutes. Another method is based on ultraviolet light (UV) optical sensors, using C band: 280 nm −100 nm light-emitting diodes, measuring urea by absorbance in the spent dialysate.

7 Errors in Calculating Kt/V

One of the major technical errors in calculating dialyzer Kt/V is failure to allow sufficient time for urea to re-equilibrate when taking the post-dialysis blood sample, so obtaining a lower post-dialysis urea concentration, and thus, an inflated dialyzer Kt/V.

V is traditionally estimated using anthropometric equations which are based on healthy subjectsmainly Europeans from the 1950s. Body composition changes with age, gender, and ethnicity, and as adipose tissue contains less water than muscle this leads to an overestimation of V in large obese patients, and unsderestimation of V in small male patients, and also women. Thus, if all patients were to achieve the same Kt/V, then men would receive relatively more dialyzer clearance than women for any given weight, and heavier men more dialyzer clearance than lighter men. This inequality of V can be overcome using either bioimpedance to actually measure total body water, or scale Kt to total energy expenditure, or body surface area.

Online clearance methods tend to overestimate dialyzer urea clearance, as the intermittent method based on changes in sodium diffusion will potentially miss falls in blood pump speed or dialysate flow in between measurements, and the UV absorbance method will also measure other nitrogen containing molecules in addition to urea.

Urine output and urinary urea excretion fall in 24 hours post a hemodialysis session, and then start to increase. As such urine collections made after a dialysis session will underestimate residual renal function.

8 Incremental Dialysis for Solute Clearances

The majority of patients starting hemodialysis have some residual renal function. Currently, no dialysis treatment effectively removes protein-bound solutes. However, as the rate of production of these solutes is relatively slow, even a small amount of residual renal function is sufficient to clear these protein-bound toxins by renal tubule organic acid transporters. As such, there is an advantage to preserving residual renal function for as long as possible [10,6,7,8,9,14]. It is now recognized that during a standard hemodialysis session, the blood supply to the kidney falls and as there is no renal autoregulation, episodes of intra-dialytic hypotension risk acute ischemic damage to the kidney and premature loss of residual renal function [15].

This has led to a return to initiating hemodialysis in an incremental fashion, so starting dialysis once or twice weekly and for less than the standard 4 hours according to residual renal function [12,13,13]. The key is to sum residual renal function (KRU) to dialyzer urea clearance [8].

The 2006 KDOQI adequacy guideline advises against hemodialysis dose reduction for patients with a KRU of <2 ml/min/1.73 m²,and recommends a sessional spKt/Vurea target 1.93 for a 3.5-hour dialysis session and 1.68 for a twice-weekly schedule for those with a KRU >2 ml/min/1.73 m²), and a KRU > 3 ml/min/1.73 m² for twice-weekly 3.5-hour HD sessions to achieve an equivalent stdKt/Vurea of 1.2 [6]. In addition to advising against an incremental approach for patients with a residual renal function of a KRU of <2 ml/min, equally one has to consider patient volume status, and as such less frequent dialysis is not advised for patients with compromised cardiac function or have a residual urine output of <600 ml/day, or inter-dialytic weight gains of ≥3.5% [11].

As dialyzer urea clearance is not biologically equivalent to the residual renal clearance, one cannot simply add the time-averaged residual urea clearance (KRU) to the time-averaged dialysis urea clearance (Kdurea) as this would underestimate the contribution of KRU to overall clearance. There are proposed methods to add Kdurea to KRU by converting intermittent Kdurea to an equivalent continuous KRU or vice versa, so that they then may be added together [8].

8.1 Equivalent Renal Urea Clearance (EKRurea, ml/min)

EKRurea is the ratio of the net urea generation rate (G, mg/min) to time-averaged urea concentration (TACurea, mg/ml). To compare averaged renal urea clearance and dialysis urea clearance between patients, EKRurea should be normalized to a standardized urea volume (V, L) of 40 L to become corrected EKR (EKRc, ml/min/40 L). The same correction can be used to normalize KRU to a standardized urea volume of 40 L deriving corrected Kr (Krc, ml/min/40 L).

To prescribe Kt/Vurea corresponding to Krc and adjusting for dialysis sessions per week, European clinical guidelines advise using the Casino and Lopez nomogram (Figure) or an estimate using the following equations [8]:

$$ \mathrm{Weekly}\ \mathrm{thrice}\ \mathrm{EKRc}=1+\left(10\times eKt/\left(\mathrm{Vurea}\right)\right)+ Krc $$

$$ \mathrm{Weekly}\ \mathrm{twice}\ \mathrm{EKRc}=1+\left(6\times eKt/\left(\mathrm{Vurea}\right)+ Krc\right) $$

This target EKRc was derived from achieving a weekly thrice eKt/Vurea of 1.2 (regarding eKt/V urea 1.16 as the standard dose from the HEMO study), which is approximately 13 ml/min/40 L. This target is slightly higher than the target EKRc of 11 ml/min/40 L derived from earlier studies. Notably, these values remain similar for different dialysis schedules.

9 Augmented Dialysis

Although incremental dialysis is often interpreted as less dialysis than the traditional thrice weekly 4-hour sessions, incremental dialysis also includes increasing the amount of dialysis to more frequent or longer dialysis sessions to account for loss of residual renal function [16]. This increase in dialysis is also termed augmented dialysis, to include introducing hemodiafiltration, higher permeability dialyzers, adsorption or displacement techniques designed to increase clearance of middle-sized solutes and protein-bound solutes to compensate for the loss of residual renal function.

10 Interpreting Adequacy in Hemodialysis Patients

It must be remembered that Kt/V is a measure of urea clearance, and as such is only an aspect of the dialysis prescription. In the original National Cooperative Dialysis Study, patients became unwell with a sessional target of 0.9 or less. Over time the target has increased by consensus from 1.2 to 1.4. Achieving a sessional target of 1.4 does not necessarily mean that dialysis clearance of the waste products of metabolism is adequate. Urea is a small molecule with a rapid diffusion coefficient, other molecules diffuse more slowly, and may be predominantly intracellular or protein bound, and as such will have a much lower clearances than urea. So, although a Kt/V below target is likely to provide inadequate clearance, achieving a target Kt/V does not imply adequate clearance. In addition, one has to consider the patient as a whole, as patients with a low protein intake, low muscle mass who are physically inactive will generate fewer waste products of metabolism compared to a patient with a high-protein diet who is physically active with preserved muscle mass, and as such will need greater dialytic clearances.

11 Alternative Methods of Assessing Adequacy of Dialyzer Clearance

Traditionally, dialyzer clearance (Kt) has been adjusted for body water (V). However, this adjustment leads to a lower effective dose of dialysis delivered to women compared to men, and lighter men compared to heavier men. The generation of waste products of metabolism depend upon both basal metabolic rate (BMR) and active energy expenditure (AEE). The majority of dialysis patients are elderly and with additional comorbidities, and as such although total energy expenditure (TEE) is greater than BMR, BMR has accounted for 80% or more of TEE. As such, it has been suggested that as BMR is related to body surface area (BSA), scaling Kt for BSA would provide some groups, for example, women with an equivalent amount of dialysis. AEE is associated with appendicular muscle mass, and as such for men, who typically have more muscle mass, scaling Kt for BSA still leads to a lower effective dialysis dose for smallmen compared to larger men.

12 Adjusting Kt/V Targets for Individual Patients

Although the concept of a target Kt/V was to provide an equivalent dose of dialysis for all patients, there are several major confounders, and as such some patient groups receive a greater dialysis dose than others. As such it is important to recognize that some patient groups require a greater dose of dialysis above the standard target sessional Kt/V to guard against delivering a lower effective dialysis dose. The groups most at risk of receiving less dialysis, and as such require a delivered Kt/V above target include women, small men, patients who have less comorbidity, and those who are physically active and at work.

A key element of this whole process is patient engagement; if the only way to achieve adequate dialysis is to increase the dialysis time, then we have a moral obligation to explain to the patient (and possibly their family), why this is important and what, if any, the alternatives are. Dialysis patients are in a vulnerable position and most of the time have pretty limited control over their situation. Engaging with patients over the options for best treatment is much more likely to succeed than a purely prescriptive approach. Similarly, while it is relatively clear how much dialysis is required to keep a patient well for a significant proportion of elderly dialysis patients’ life expectancy is very poor and there is an important balance to be made between what might appear to be optimum treatment to increase quantity of life and quality of life. There is no formula for this, just patient–doctor dialogue.

12.1 Equations

$$ URR=\left( pre\ \mathrm{urea}\ \mathrm{concentration}\hbox{--} \mathrm{post}\ \mathrm{urea}\ \mathrm{concentration}\right)\ast 100/ pre\ \mathrm{urea}\ \mathrm{concentration} $$

$$ \mathrm{Single}\ \mathrm{pool}\ Kt/\mathrm{V}\ \left(\mathrm{spKt}/\mathrm{V}\right) $$

$$ \mathrm{spKt}/\mathrm{V}=-\mathrm{natural}\ \mathrm{logarithm}\ \left(1- URR\right) $$

First-generation equation

$$ \mathrm{spKt}/\mathrm{V}=-\mathrm{In}\ \left(\left(\mathrm{Urea}\ \mathrm{concentration}\ \mathrm{post}/\mathrm{Urea}\ \mathrm{concentration}\ \mathrm{post}\right)-0.008\ast t-1\ast UF/\mathrm{W}\right), $$

t – dialysis session time hours.

UF – ultrafiltration volume L.

W – post-dialysis weight kg.

Second-generation Kt/V equation

$$ \mathrm{spKt}/\mathrm{V}=-\mathrm{In}\ \left(\left(\mathrm{Urea}\ \mathrm{concentration}\ \mathrm{post}/\mathrm{Urea}\ \mathrm{concentration}\ \mathrm{post}\right)-0.008\ast t\right)+\left(4-3.5\ast \mathrm{R}\right)\ast UF/\mathrm{W}. $$

R = URR.

Equilibrated Kt/V.

Rate adjustment method

$$ eKt/\mathrm{V}=\mathrm{spKt}/\mathrm{V}\hbox{--} 0.6\ \left(\mathrm{spKt}/\mathrm{V}\right)/\mathrm{time}\ \mathrm{hours}+0.03 $$

Standardized Kt/V

$$ \mathrm{stdKt}/V=\frac{10,080\times \frac{1-{e}^{- eKt/V}}{t}}{\frac{1-{e}^{- eKt/V}}{eKt/V}+\frac{10,080}{Ft}-1} $$

t = treatment time, min; V = modeled two-pool post-dialysis volume, mL; F = frequency of treatment per week; Uf = weekly fluid gain between dialyses, mL.

Urinary urea clearance (Kr)

$$ \mathrm{Urinary}\ \mathrm{urea}\ \mathrm{clearance}\ (Kr)=\frac{U\times V}{P} $$

U = Urinary urea nitrogen concentration of interdialytic urine sample (mg/dL).

V = Urine volume excretion rate during interdialytic collection (ml/min).

P = Time-averaged serum urea nitrogen concentration during interdialytic collection (mg/dL).

Variable volume model, with Kr

$$ \mathrm{stdKt}/V=\frac{Eq.6}{1-\frac{0.74}{F}\left[\frac{Uf}{V}\right]}+ Kr\frac{10,080}{V} $$

13 Summary

There is clearly a need to tailor dialysis on the basis of residual renal function, habitus, volume status, and many other aspects including patient preference [17]. Modern dialysis machines are able to document the delivery of dialysis in terms of urea clearance (and fluid removal). The challenge for a nephrologist looking after a heterogeneous group of patients is how to identify, recommend, and deliver the best bespoke dialysis for them.

Tips and Tricks

Supporting and training dialysis staff in measurement and recording of dialysis clearance is a critical first step and includes, ensuring appropriately delayed post-dialysis urea sampling, identification of aspects of dialysis that interfere with delivery (such as intermittent attendance), identification of recirculation, skilled needling etc. In short, having accurate data on which to make management decisions is essential.

Monitoring residual urine output and factoring this into dialysis clearance are both important and a reminder that despite being on dialysis, attempts should be made to avoid nephrotoxins and to preserve residual renal function [18].

Engaging staff in the goals and opportunities for achieving good clearance is also important and assists greatly in delivering patient education and engagement in their own treatment. Multi-disciplinary team meetings offer a good opportunity for discussing the options and teaching.

Questions

• 1. What happens to urine output after dialysis and why does it matter?

• 2. What is the relevance of residual renal function?

• 3. What is the KoA of a dialyzer?

• 4. What are the sizes of urea and β₂-microglobulin (β₂M) molecules?

Answers

• 1. Urine output drops postdialysis, a combination of reduced renal perfusion during dialysis and reduction in intravascular volume. This matters because urine collections post dialysis we underestimate residual renal function, and perhaps, because it reminds us that dialysis itself can be damaging to renal function.

• 2. Residual renal function has huge value and loss of residual renal function is associated with worse mortality. Both blood pressure and volume control are important, permitting greater oral intake and potentially fewer dialysis sessions for longer. Residual renal function is associated with improved anemia control, better nutrition, and critically, the removal of middle molecules.

• 3. It is the manufacturer’s estimate of urea clearance characteristics for a dialyzer at specified pump speeds and dialysate flow rates. It is derived from performance ex-vivo.

• 4. Urea is 64 daltons and β₂M 11,800 daltons. β₂M is derived from the MHC class 1 present on all nucleated cells. Because of its size it is not well cleared by conventional dialysis and therefore, accumulates, potentially forming amyloid fibrils which can cause dialysis amyloid. There are of course, many middle molecules that are well cleared by the kidney, but less well so by conventional dialysis. The fact that β₂M is ubiquitous and relatively easy to measure makes it a useful marker.

Case Study

A 67-year-old male patient was reviewed in dialysis clinic with a Kt/V of 1.0 despite 4 hours of dialysis three times a week. He was dialyzing via a line having been very reluctant to have an AV fistula, with a medium kidney (dialyzer), dialysate flows (Qd) of 800 ml/min and an arterial flow (Qb) of 250 ml/min. He was extremely reluctant to increase his dialysis time and the question was whether he would benefit from using a bigger dialyzer. Unfortunately, dialyzer urea clearance is limited by the lowest factor Qb/KoA/Qd and as ◘ Fig. 81.5 illustrates, increasing the kidney size when the Qb is 250mls is a very inefficient way of increasing clearance.

Fig. 81.5

The relationship between blood flow (Qb) and clearance at a fixed dialysate flow (Qd) for different sized kidneys (dialyzers). Larger dialyzers add very little at low blood flows, but offer a significant benefit when blood flows are higher (e.g., 350/ml/min)

It was possible to explain to the patient and his partner that the factors that influence, providing enough dialysis to keep him well are his weight (which was relatively fixed), the time on dialysis, and a combination of the dialyzer characteristics and blood flow. In practical terms the only two modifiable factors were increasing time or improving his access, which would likely have other benefits. After a discussion between him and his wife he elected for a fistular formation which eventually delivered a Qb of 380 ml/min, permitted the use of a larger dialyzer, and delivered much better clearance.