THE MANAGEMENT OF RENAL FAILURE: RENAL REPLACEMENT THERAPY AND DIALYSIS

Published on 10/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1672 times

CHAPTER 88 THE MANAGEMENT OF RENAL FAILURE: RENAL REPLACEMENT THERAPY AND DIALYSIS

Joseph M. Gutmann, Christopher McFarren, Lewis M. Flint, Rodney Durham

Acute renal failure (ARF) is a common and devastating problem that contributes to morbidity and mortality in critically ill patients. ARF prolongs hospital stays and increases mortality. Although effective renal replacement therapy (RRT) is available, it is not ideal and the best therapy is prevention.

The kidneys are the primary regulators of volume and composition of the internal fluid environment and their excretion. Renal failure leads to regulatory function impairment, causing retention of nitrogenous waste products and disturbance in fluid, electrolyte, and acid-base balance. Renal injury in intensive care unit (ICU) patients is a progressive process, usually starting with a prerenal insult—which progresses to severe renal injury. Other systemic issues can worsen the renal injury.

Acute renal failure in critically ill patients is a growing clinical problem. Options for RRT in these patients use convective and diffusive clearance, which may be intermittent (as in classic hemodialysis) or continuous. RRT needs to be tailored to the needs of each patient. Current and future research studies are essential in improving outcomes.

INCIDENCE

Acute renal failure is defined as an abrupt and sustained decline in the glomerular filtration rate (GFR),¹ which leads to accumulation of nitrogenous waste products and uremic toxins. In critically ill patients, more than 90% of the episodes of ARF are due to acute tubular necrosis (ATN) and are the result of ischemic or nephrotoxic etiology (or a combination of both). ARF affects nearly 5% of all hospitalized patients and as many as 15% of critically ill patients.² Like many other medical conditions, there is no gold standard of diagnosis, no specific histopathologic confirmation, and no uniform clinical picture.

The mortality rate of an isolated episode of ARF is approximately 10% to 15%. When it occurs in association with multiple-organ dysfunction, as in the ICU setting, mortality rates are much greater and vary in published series between 40% and 90%.³

In some cases, preexisting conditions may worsen. New major complications, such as sepsis and respiratory failure, may also develop after the onset of renal failure. Although ARF that requires RRT carries a high mortality,⁴ there is emerging evidence to suggest that milder forms of ARF that do not require supportive therapy with RRT have better patient outcomes.⁵

Many aspects of surgical diseases and their care have the potential to impair renal function, either by toxic effects on the renal parenchyma or by reducing renal perfusion (or a combination of the two). The prevention of ARF in critical patients consists of minimizing toxicity and ensuring adequate blood flow. Avoidance of renal failure is preferred to any treatment. Therefore, renal function should be monitored closely so that adverse circumstances can be limited.

Given the impact of ARF on mortality, it is important to prevent or hasten the resolution of even the mildest forms of ARF. The goals of a preventive strategy for the syndrome of ARF are to preserve renal function, to prevent death, to prevent complications of ARF (volume overload, acid–base disturbances, and electrolyte abnormalities), and to prevent the need for chronic dialysis (with minimum adverse effects).

This chapter explores preventive strategies, the major challenges ARF presents, and key issues to be considered. Can the patient be managed conservatively or will RRT be needed? If RRT is required, which form of RRT is most appropriate?

MECHANISM OF INJURY/ETIOLOGY

Diagnosis

Renal failure is measured routinely and easily in the ICU: the excretion of water-soluble waste products of nitrogen metabolism, urea and creatinine, and the production of urine. To understand renal failure, we need to reflect on some important aspects of renal physiology.

Water and Fluid Homeostasis

Because body water is the primary determinant of the osmolality of the extracellular fluid (ECF), disorders of body water homeostasis can be divided into hypo- and hyperosmolar disorders depending on whether there is an excess or deficiency of body water relative to body solute. The end result of any change in circulating blood volume is a change in sodium excretion by the kidneys. This is brought about by the activation of the sympathetic nervous system, the reninangiotensin-aldosterone axis, and release/suppression of natriuretic peptides.

Assessment of Renal Function

Serum concentrations of blood urea nitrogen (BUN) and creatinine are the most commonly used markers of renal function. Urea is the end product of protein and amino acid catabolism. Under normal conditions, 80%–90% of total nitrogen excretion is by the kidneys. Creatinine is formed in muscle by the nonenzymatic degradation of creatine and phosphocreatine, and is excreted primarily by glomerular filtration. A small percentage of creatinine is actively secreted into the glomerular filtrate and tubular reabsorption of creatinine is negligible.⁵

Circulating concentrations of BUN and creatinine are determined not only by how efficiently they are excreted by the kidneys but by their rate of production. Urea formation depends on the amount of protein and amino acids catabolized. It is increased with high-protein diets, reabsorption of hematomas, and digestion of blood in the gastrointestinal (GI) tract. It is reduced in starvation. These factors may change the value of the BUN, even though renal function is adequate. Creatinine production reflects muscle mass. It is constant over the short term, and steadily diminishes if muscle mass is lost. Muscle mass diminishes with age, together with intrinsic renal function. Therefore, serum creatinine stays relatively constant over time.

Creatinine Clearance

Determination of the creatinine clearance (Ccr) provides a measure of renal function. Creatinine secretion and reabsorption in the kidneys is negligible. Clearance is defined as the volume of plasma or serum cleared by the kidneys over a period of time. It is calculated as

where Ucr is urine creatinine, Pcr is serum creatine, and V is volume.

The clearance reflects the net effect of GFR, which is the amount of fluid filtered from the plasma in a given time by the kidneys. The most commonly used method for estimating Ccr is the Cockcroft-Gault formula:

Normal GFR is 125 ± 15 ml/min/1.73 m² body surface area (BSA).

Sodium has the highest serum concentration of all cations in the ECF. Any transport of sodium necessarily involves the transport of water. Renal sodium clearance is an important mechanism for the regulation of ECF volume and tonicity. Aldosterone promotes tubular reabsorption of sodium, and it is elaborated in response to changes in hydrostatic pressure within the glomerular arterioles. If renal blood flow or pressure is reduced, tubular sodium reabsorption is increased—thus preserving ECF volume. The ratio of sodium clearance to Ccr is known as the fractional excretion of sodium (FENa):

Here, Una and Ucr are the urinary concentrations of sodium and creatinine, and Pna and Pcr are the serum levels of sodium and creatinine, respectively. If the FENa is very low (<1%), it may indicate inadequate renal arteriolar pressure—suggesting that factors other than intrinsic renal dysfunction are responsible for clinically inadequate renal function.⁶

Urine Production and Output

The end result of renal function is the production of urine. Quantitative measurements of urine are important for assessing renal function. Urine output is highly sensitive to renal blood flow, making it a key indicator of renal function and total body vascular perfusion ⁷ (Table 1).

Table 1 Adequate Urine Output

Age	Urine Output (ml/kg/min)
Infant (<10 kg)	2.0
Toddler (10–20 kg)	1.5
Child (20–50 kg)	1.0
Adult (>50 kg)	0.5

MANAGEMENT OF PATIENTS

Conservative Management

Despite many advances in medical technology, the mortality and morbidity attributed to ARF in the ICU remains high. Primary strategies to prevent ARF still include adequate hydration, maintenance of mean arterial pressure (preferably MAP >60), and minimization of exposure to potentially nephrotoxic agents. Although hydration was shown to be beneficial, the type of fluid to be used in a hydration regime remains controversial.

Considering its low cost, low toxicity, and consistent benefit, NAC (N-acetylcysteine) administration with IV hydration should be considered to decrease the prevalence of nephropathy in high-risk patients. Note that the routine use of NAC is controversial and is not well studied.

Nonpharmacologic Strategies for Acute Renal Failure Prevention

Nonpharmacologic strategies to prevent ARF include ensuring adequate hydration (limiting dehydration), maintenance of adequate mean arterial pressures, and minimizing exposure to nephrotoxic agents. Four particular strategies are worth reviewing: fluids, aminoglycoside dosing, lipid-soluble preparations of amphotericin, and nonionic contrast agents.

Fluids

Adequate hydration is the cornerstone of renal failure prevention. One randomized controlled trial (n = 1620) compared hydration using 0.9% saline infusion with 0.45% saline in dextrose for prevention of radiocontrast-induced nephropathy in patients who underwent coronary angiography.⁸ Hydration with 0.9% saline infusion significantly reduced contrast nephropathy compared with 0.45% saline in dextrose hydration (0.7% vs. 2%, respectively; p = 0.04). This effect was greater in women, diabetics, and patients who received a large volume (>250 ml) of a contrast agent. A recent single-center randomized controlled trial compared the efficacy of sodium bicarbonate with 0.9% saline hydration in preventing contrast nephropathy.⁹ In this study, 119 patients who had stable serum creatinine of at least 1.1 mg/dl were randomized to 154 mEq/l infusion of sodium chloride (n = 59) or sodium bicarbonate (n = 60) before and after contrast (iopamidol) administration. One of 59 patients (1.7%) in the group that received bicarbonate developed contrast nephropathy (defined as an increase of ≥25% in serum creatinine from baseline within 48 hours) compared with 8 of 60 patients (13.3%) in the group that received saline (p = 0.02).

Nephrotoxin Exposure

Minimizing exposure to potentially nephrotoxic agents is an important strategy to prevent ARF in the ICU setting. Aminoglycosides, other antibiotics, amphotericin, and radiocontrast are the nephrotoxins encountered most commonly in the ICU. A systematic review in patients who had neutropenic fever and received aminoglycosides, however, found no significant differences in efficacy or nephrotoxicity between once daily and three times daily dosing.¹⁰

The use of lipid formulations of amphotericin B seems to cause less nephrotoxicity compared with standard formulations, but direct comparisons of long-term safety are lacking. With regard to contrast media, one systematic review (31 randomized controlled trials, 5146 patients) compared low osmolality contrast media with standard contrast media.¹¹ The study showed that low osmolality contrast media did not influence the development of ARF or the need for dialysis.

Pharmacologic Strategies for Acute Renal Failure Prevention

Loop Diuretics

Multiple small clinical trials studied the efficacy of loop diuretics in preventing ARF and have provided conflicting results. They have been underpowered, nonrandomized, or methodologically flawed. One systematic review that compared fluids with diuretics in people who were at risk for ARF from various causes did not show any benefit from diuretics with regard to prevalence of ARF, need for dialysis, or mortality.¹²

N-Acetylcysteine

Systematic reviews found that NAC plus hydration reduced the incidence of contrast nephropathy more than hydration alone in people who had baseline renal impairment and underwent radiocontrast studies.¹³ A recent study, however, suggested that NAC could decrease serum creatinine independently without any effect on GFR (as evaluated by other surrogate outcomes, such as serum cystatin C levels).¹⁴ Hence, the current implications of reduction in serum creatinine after contrast administration with the use of NAC remain unclear and need to be explored further.

INDICATIONS FOR RENAL REPLACEMENT THERAPY IN ACUTE RENAL FAILURE

As in chronic kidney disease, overt disturbances of ECF volume and body fluid composition remain the objective indications for initiation of RRT in patients with ARF (Table 2). These include volume overload, hyperkalemia, severe metabolic acidosis, uremia, and azotemia.

Table 2 Indications for Renal Replacement Therapy

Volume Overload

Volume overload is generally recognized as an indication for RRT in ARF. All modalities of RRT are effective at diminishing intravascular volume. Subjective criteria for initiation of therapy include impairment of cardiopulmonary function by pulmonary vascular congestion or compromise of cutaneous integrity and wound healing by peripheral edema.

Mehta and colleagues ¹⁵ performed a retrospective analysis of data from 522 critically ill patients who had ARF. Fifty-nine percent of these patients had been treated with diuretics. After adjustment for relevant covariates and the propensity for diuretic use, they observed a significant increase in the risk of death or nonrecovery of renal function (odds ratio 1.77, 95% confidence interval 1.14–2.76). On the basis of this, they concluded that diuretic therapy was potentially deleterious in patients who had ARF. They noted, however, that the increased risk was borne largely by patients who were unresponsive to diuretics. This suggested that this increased risk might reflect selection for a more severe degree of renal injury.

Hyperkalemia

The treatment of hyperkalemia with evidence of myocardial toxicity was one of the early indications for hemodialysis in ARF. Hyperkalemia is a well-recognized complication of ARF, which, if not treated, may be rapidly fatal. Most medical therapies for hyperkalemia (e.g., intravenous calcium to directly antagonize the effects of hyperkalemia on the myocardial cell membrane, and intravenous insulin/dextrose and intravenous or inhaled β-adrenergic agonists to shift potassium into the intracellular compartment) are primarily temporizing measures. Three modalities are available to decrease the total body potassium burden: diuretic therapy, enteric potassium-binding resins, and dialysis.

In patients who have severe renal failure, diuretic therapy is generally ineffective in promoting kaliuresis due to lack of diuretic response. Although sodium polystyrene sulfonate can enhance fecal potassium losses, its use is limited in patients with recent intraabdominal or GI surgery, ileus, or bowel ischemia. Dialysis provides the most rapid means of decreasing the serum potassium concentration. However, because of variability in study design and evolution of dialysis techniques it is difficult to determine the expected potassium removal during a single dialysis treatment.¹⁶

Even greater clearances of potassium may be achieved by using more permeable synthetic hemodialysis membranes and greater blood flow rates. However, the rate of potassium removal is ultimately limited by the rapid decrease in the concentration gradient between plasma and dialysate.¹⁷ As with volume status, a specific threshold level of serum potassium cannot be established as an indication for initiation of RRT. Myocardial toxicity from hyperkalemia is uncommon when the serum potassium concentration is less than 6.5 mmol/l.¹⁶ Therefore, decisions regarding the initiation of treatment for control of hyperkalemia must take into consideration the absolute level and rate of increase of serum potassium, the patient’s overall condition, and the likely efficacy of medical therapy.

Metabolic Acidosis

The role of alkali therapy in the treatment of metabolic acidosis, particularly lactic acidosis, is controversial.¹⁸ The use of RRT as an alternative to alkali replacement in metabolic acidosis can avoid some of the deleterious effects ascribed to aggressive alkali replacement, specifically volume overload and hypernatremia. Although progressive metabolic acidosis is a generally accepted indication for RRT, clinical trials to establish a threshold blood pH or serum bicarbonate concentration or to demonstrate improved patient outcomes have not been performed.

Other Electrolyte Disturbances

RRT may be used for the treatment of a variety of other electrolyte disturbances that can occur in the setting of ARF. These include severe hypo- and hypernatremia, hyperphosphatemia, hypo- and hypercalcemia, and hypermagnesemia. In the treatment of hyponatremia, caution must be used to ensure that rapid correction does not predispose to the development of the osmotic demyelination syndrome. A rapid decrease of serum phosphate and uric acid levels and control of acidemia using RRT are necessary in patients who have the tumor lysis syndrome to support recovery of renal function.

Uremia

The development of overt uremic signs or symptoms represents an obvious indication for initiation of RRT in ARF. Early manifestations of uremia, such as anorexia, nausea and vomiting, and pruritus, are nonspecific and may be difficult to differentiate from other comorbid conditions in patients who have critical illness. Mental status changes, which may represent uremic encephalopathy, also may be difficult to differentiate from other etiologies of delirium in the critically ill patient. Uremic pericarditis is usually a late complication, but requires urgent initiation of renal support given the high risk of intrapericardial hemorrhage and tamponade. As was emphasized more than four decades ago by Teschan et al.,¹⁹ optimally RRT should be initiated before the onset of overt uremic manifestations.

Azotemia

In many patients, the sole indication for initiation of RRT in ARF is the presence of progressive azotemia in the absence of uremia or other indications for renal support. There is no consensus, however, on the degree of azotemia that warrants initiation of therapy. In a multicenter trial that evaluated the dosing strategies for RRT in critically ill patients who had ARF, we observed substantial variation in practice regarding the degree of azotemia deemed appropriate for initiation of treatment between practitioners within individual institutions and between institutions (unpublished data). There are many experts in nephrology who feel that RRT should be initiated in critically ill patients with a BUN >60.

TIMING OF INITIATION OF RENAL REPLACEMENT THERAPY

Beginning with the studies by Paul Teschan and colleagues,¹⁹ in the years following the Korean War numerous studies have attempted to define the criteria for timing of initiation of RRT in ARF. These studies attempted to determine the balance between three major competing risks: the inherent risk that results from delay in therapy; the potential risk of harm as a result of RRT, including complications of therapy and the potential that dialysis may prolong the course of ARF; and the risk that early initiation of therapy will result in patients undergoing treatment who, if managed conservatively, might recover renal function without requiring RRT.

In their landmark report, Teschan et al.¹⁹ described a prospective uncontrolled series of 15 patients who had oliguric ARF who were treated with “prophylactic” hemodialysis defined as the initiation of dialysis before the serum urea nitrogen reached 100 mg/dl.¹⁸ Patients received daily dialysis (average duration 6 hours) using twin-coil cellulosic dialyzers at a blood flow of 75–250 ml/min to maintain a predialysis serum urea nitrogen of less than 75 mg/dl. Caloric and protein intake were unrestricted. All-cause mortality was 33%. Mortality due to hemorrhage or sepsis was 20%. Although no control group was studied, the investigators reported that the results contrasted dramatically with their own past experience in patients in whom dialysis was not initiated until “conventional” indications were present.

Acute Renal Failure

ARF is a common complication in critically ill patients and is associated with a mortality rate greater than 50%.²⁰ As many as 70% of these patients require RRT, making it an important component of the management of ARF in the ICU. Ideally, RRT controls volume, corrects acid-base abnormalities, improves uremia through toxin clearance, promotes renal recovery, and improves survival without causing complications (such as bleeding from anticoagulation and hypotension). The available RRT options include intermittent hemodialysis (IHD), continuous RRT (CRRT), and sustained low-efficiency dialysis (SLED). Currently, there is insufficient evidence to establish which modality of RRT is best for ARF in the critically ill patient. There is a general consensus that patients receiving CRRT using lower blood flow rates and lower fluid removal rates have less cardiovascular instability/morbidity. Clearly, there is no significant difference in mortality rates with any of the available modalities. Understanding the advantages and limitations of the various dialysis modalities is essential for appropriate RRT selection in the ICU setting.

Principles of Renal Replacement Therapy

All forms of RRT rely on the principle of allowing water and solute transport through a semipermeable membrane and then discarding the waste products. Ultrafiltration is the process by which water is transported across a semipermeable membrane. Diffusion and convection are the two processes by which solutes are transported across the membrane. The available RRT modalities use ultrafiltration for fluid removal and diffusion, convection, or a combination of diffusion and convection to achieve solute clearance.

Ultrafiltration achieves volume removal by using a pressure gradient to drive water through a semipermeable membrane. This pressure gradient is known as the transmembrane pressure gradient and is the difference between plasma oncotic pressure and hydrostatic pressure. Determinants of the ultrafiltration rate include the membrane surface area, water permeability of the membrane, and transmembrane pressure gradient.²¹

Diffusion occurs by movement of solutes from an area of higher solute concentration to an area of lower solute concentration across a semipermeable membrane. The concentration gradient is maximized and maintained throughout the length of the membrane by running the dialysate (an electrolyte solution usually containing sodium, bicarbonate, chloride, magnesium, and calcium) countercurrent to the blood flow. Solutes with a higher concentration in the blood, such as potassium and urea, move down their concentration gradient across the membrane to the dialysate compartment. Conversely, solutes with a higher concentration in the dialysate (such as bicarbonate) diffuse into the blood. Solute concentrations that are nearly equivalent in the blood and dialysate, such as sodium and chloride, move very little across the membrane. Because smaller solutes (such as urea and creatinine) diffuse more rapidly than larger solutes, lower-molecular-weight molecules (<500 daltons) are cleared more efficiently than heavier molecules. The rate of solute diffusion depends on blood flow rate, dialysate flow rate, duration of dialysis, concentration gradient across the membrane, and membrane surface area and pore size.²¹

Convection occurs when the transmembrane pressure gradient drives water across a semipermeable membrane (as in ultrafiltration) but then “drags” with the water both small-molecular-weight (BUN, creatinine, potassium) and large-molecular-weight (inulin, β₂-microglobulin, tumor necrosis factor, vitamin B₁₂) solutes. Membrane pore diameter limits the size of the large solutes that can pass through. Increasing the transmembrane pressure difference allows more fluid and solutes to be “pulled” through the membrane. Because the efficiency of solute removal depends mainly on the ultrafiltration rate, typically at least 1 l of water needs to be pulled through the membrane each hour. The process of increasing the ultrafiltration rate to provide convective clearance of solutes is known as hemofiltration. Ultrafiltration rate is determined by the transmembrane pressure, water permeability of the membrane, and membrane surface area and pore size.²¹

CLASSIFICATION OF RENAL REPLACEMENT THERAPIES

RRT for ARF can be classified as intermittent or continuous, based on the duration of the treatment. The duration of each intermittent therapy is less than 24 hours, whereas the duration of continuous therapy is at least 24 hours. The intermittent therapies include IHD and SLED. The continuous therapies include peritoneal dialysis and CRRT.²² Peritoneal dialysis is rarely used in the acute setting because it provides inefficient solute clearance in critically ill catabolic patients, increases the risk of peritonitis, compromises respiratory function by impeding diaphragmatic excursion, and is contraindicated in patients with recent abdominal surgery or abdominal sepsis.²³

Intermittent Hemodialysis

Traditionally, nephrologists have managed ARF with IHD—empirically delivered three to six times a week, 3–4 hours per session, with a blood flow rate of 200–350 ml/min and a dialysate flow rate of 500–800 ml/min. In IHD, solute clearance occurs mainly by diffusion—whereas volume is removed by ultrafiltration. The degree of solute clearance, also known as the “dialysis dose,” is largely dependent on the rate of blood flow. Increasing the blood flow increases solute clearance. Decisions regarding dialysis duration and frequency are based on patient metabolic control, volume status, and presence of any hemodynamic instability.

Advantages of IHD include rapid solute and volume removal. This results in rapid correction of electrolyte disturbances, such as hyperkalemia, and rapid removal of drugs or other substances in fatal intoxications within a matter of hours. IHD also has a decreased need for anticoagulation compared with other types of RRT because of the higher blood flow rates and shorter duration of therapy.

The main disadvantage of IHD is the risk of systemic hypotension caused by rapid electrolyte shifts and fluid removal. Hypotension occurs in approximately 20%–30% of hemodialysis treatments. Sodium modeling, cooling the dialysate, increasing the dialysate calcium concentration, IV albumin, and intermittent ultrafiltration may be used to improve hemodynamic stability during IHD. Despite this, approximately 10% of ARF patients cannot be treated with IHD because of hemodynamic instability. Systemic hypotension can limit the efficacy of IHD and result in poor solute clearance, insufficient acid-base correction, and persistent volume overload, because the rate of ultrafiltration necessary to maintain fluid balance is seldom achieved within the 4-hour dialysis session.

Rapid solute removal from the intravascular space can cause cerebral edema and increased intracranial pressure. ARF patients with head trauma or hepatic encephalopathy are at a significant risk of brain edema and even herniation.²⁴ Finally, there is a lack of consensus as to how to assess solute clearance (dialysis dose) and what constitutes an adequate dose in ARF because the kinetics of urea in the end-stage renal disease patient cannot be extrapolated to patients with ARF.

Although the results of some studies suggested an advantage of daily HD over conventional IHD, it is unclear whether the increased dialysis dose improved outcome by improving uremic control or by reducing the volume of fluid removed during each dialysis session and resulting in less hemodynamic instability.

Continuous Renal Replacement Therapy

Although the worldwide standard for RRT is IHD, CRRT has emerged over the past decade as a viable modality for management of hemodynamically unstable patients with ARF. Continuous therapies have evolved from systems that relied on arterial access and blood pressure to maintain blood flow through the extracorporeal circuit to pump-driven systems that use double-lumen venous catheters. The arteriovenous (CAVH) circuit is now rarely used in CRRT because of poor solute removal and complications from arterial cannulation. Unlike IHD, CRRT is a continuous treatment occurring 24 hours a day—with a blood flow of 100–200 ml/min and a dialysate flow of 17–40 ml/min if a diffusive CRRT modality is used. The different CRRT modalities can use diffusion, convection, or a combination of both for solute clearance.

All types of CRRT use membranes that are highly permeable to water and low-molecular-weight solutes. CRRT modalities are classified by access type and method of solute clearance. Venovenous circuits are now the standard, and the various venovenous modalities of CRRT differ by their mechanism of solute removal. The four main types of CRRT in order of increasing complexity are slow continuous ultrafiltration, continuous venovenous hemofiltration (CVVH), continuous venovenous hemodialysis (CVVHD), and continuous venovenous hemodiafiltration (CVVHDF).²⁵

In slow continuous ultrafiltration, low-volume ultrafiltration at a rate of 100–300 ml/hr is performed to maintain fluid balance only and does not result in significant convective clearance of solutes. No fluids are administered either as dialysate or replacement fluids, and the purpose of treatment is for volume overload with or without renal failure. Indications include volume overload in patients with congestive heart failure refractory to diuretics.

In CVVH, solute clearance occurs by convection. Solutes are carried along with the bulk flow of fluid in a hydraulic-induced ultrafiltrate of blood. No dialysate is used. Clearances are similar for all solutes that have a molecular weight in the range at which the membrane is readily permeable. The rate at which ultrafiltration occurs is the major determinant of convective clearance. The ultrafiltration rate is determined by the transmembrane pressure, water permeability, pore size, surface area, and membrane thickness. Typically, hourly ultrafiltration rates of 1–2 l/hr are used to provide adequate solute removal. These high ultrafiltration rates rapidly cause volume contraction, hypotension, and loss of electrolytes. Intravenous “replacement fluid” is provided to replace the excess volume being removed and to replenish desired solutes. Replacement fluid can be administered prefilter or postfilter.

In CVVHD, a dialysate solution runs countercurrent to the flow of blood at a rate of 1–2.5 l/hr. Solute removal occurs by diffusion. Unlike IHD, the dialysate flow rate is slower than the blood flow rate, allowing small solutes to equilibrate completely between the blood and dialysate. As a result, the dialysate flow rate approximates urea and Ccr. Ultrafiltration is used for volume control but can allow for some convective clearance at high rates. CVVHDF combines the convective solute removal of CVVH and the diffusive solute removal of CVVHD. As in CVVH, the high ultrafiltration rates used to provide convective clearance require the administration of intravenous replacement fluids.

Replacement fluids can be administered prefilter or postfilter. Postfilter replacement fluid results in hemoconcentration of the filter and increased risk of clotting, especially when the filtration fraction is greater than 30%. The filtration fraction is the ratio of ultrafiltration rate to plasma water flow rate and is dependent on the blood flow rate and hematocrit.²⁵ Prefilter replacement fluid dilutes the blood before the filter, resulting in reduced filter clotting. Dilution of solutes before the filter reduces solute clearance by up to 15% by lowering the diffusion driving force and convective concentration.

Advantages and Disadvantages

The advantages of CRRT include hemodynamic tolerance caused by slower ultrafiltration rates.²⁷ The gradual continuous volume removal makes control of volume status easier and allows administration of medications and nutrition with less concern for volume overload. Because it is a continuous modality, there is less fluctuation of solute concentrations over time and better control of azotemia, electrolytes, and acid-base status. The improved hemodynamic stability may be associated with fewer episodes of reduced renal blood flow, less renal ischemia, and more rapid renal recovery. Mehta et al.²⁸ examined this issue in a prospective study in which 166 ICU patients with ARF were randomized to IHD or to CRRT. CRRT patients who survived were significantly more likely to show renal recovery than those treated with IHD. Because CRRT does not cause rapid solute shifts, it does not raise intracranial pressure like IHD.

The cumulative solute removal with CRRT is greater than that achievable with IHD. Ronco et al.²⁹ provided convincing evidence that increasing solute clearance with CRRT can improve outcome in critically ill patients with ARF. In a prospective randomized controlled trial, 425 critically ill patients with ARF were assigned to CVVH using ultrafiltration rates of 20 ml/kg/hr (group 1), 35 ml/kg/hr (group 2), or 45 ml/kg/hr (group 3). The ultrafiltration rate of 20 ml/kg/hr was based on the average rate used in clinical practice as reported in the literature at the time of the study. The blood flow rates ranged from 120 to 240 ml/min and the replacement fluid was administered postfilter. The primary study outcome was survival at 15 days after discontinuation of CVVH. Secondary outcomes were recovery of renal function and CRRT-related complications. Patient survival after discontinuing CVVH was 41, 57, and 58% in groups 1, 2, and 3, respectively. Survival in group 1 was significantly lower than group 2 (p = 0.0007) and group 3 (p = 0.001), demonstrating a survival advantage for patients treated with CVVH at an ultrafiltration rate of at least 35 ml/kg/hr. It is unclear, however, whether the reduction in mortality was solely caused by small-molecule (urea) clearance or by both small-molecule clearance and increased middle-molecule clearance.

Intermittent Hemodialysis versus Continuous Renal Replacement Therapy: Outcomes

There are few prospective studies comparing IHD with CRRT with respect to outcomes, such as mortality or recovery of renal function. Mehta et al.²⁸ randomized 166 patients to CRRT (CVVH or CVVHDF) or IHD. Univariate intention-to-treat analysis revealed a higher mortality among patients receiving CRRT. Patients randomized to CRRT had higher APACHE III scores and had a higher prevalence of liver failure, confounding the results. Multivariate analysis revealed no impact of RRT modality on all-cause mortality or recovery of renal function. Instead, severity of illness scores (such as APACHE III scores and number of failed organs) were more important prognostic factors. The authors concluded that insufficient data existed to draw strong conclusions, mainly because of the lack of randomized controlled trials and the influence of biases and confounding variables.

Sustained Low-Efficiency Dialysis or Extended Daily Dialysis

SLED and extended daily dialysis are slower dialytic modalities run for prolonged periods using conventional hemodialysis machines with modification of blood and dialysate flow rates. Typically, sustained low-efficiency dialysis and extended daily dialysis use low blood-pump speeds of 200 ml/min and low dialysate flow rates of 300 ml/min for 6–12 hours daily. Sustained low-efficiency dialysis and extended daily dialysis combine the advantages of CRRT and IHD. They allow for improved hemodynamic stability through gradual solute and fluid removal, as in CRRT. At the same time, they are able to provide high solute clearances (as seen in IHD) and eliminate the need for expensive CRRT machines, costly customized solutions, and trained staff.

Because sustained low-efficiency dialysis and extended daily dialysis can be done intermittently based on the needs of the patient, they also avoid the interruption of therapy for various diagnostic and therapeutic procedures that may be required in such patients. Kumar et al.³⁰ described their prospective experience of 25 patients treated with extended daily dialysis and 17 patients treated with CVVH at University of California Davis Medical Center. No significant differences in mean arterial pressure or inotrope requirements were observed between the two groups. Mortality was higher in the extended daily dialysis group (84% vs. 65%). The APACHE II scores were higher, however, in the extended daily dialysis group at the onset of treatment. The authors argued that extended daily dialysis was more cost effective by removing the need for constant monitoring of dialysis equipment and reducing nursing workload.