Acute Kidney Injury

Acute kidney injury (AKI) can be defined as an acute decrease in the glomerular filtration rate. Oligoanuria is a frequent, but not essential, component of AKI. Common causes of AKI in the critically ill patient include renal hypoperfusion (leading to so-called prerenal azotemia), hypotension or shock causing renal ischemic injury or acute tubular necrosis, nonspecific renal involvement as part of a sepsis syndrome or multiorgan dysfunction, and renal injury from nephrotoxins such as antibiotics and radiocontrast media (see Chapter 71).

Patients with even mild AKI can have volume overload, electrolyte imbalance, and metabolic derangement with associated poor outcomes.^3–5 Such complications can be particularly detrimental to the care of the critically ill patient. While all modalities of renal replacement therapy can correct these abnormalities, certain modalities may be better suited for specific pediatric clinical situations.

Acute Intoxication and Metabolic Disorders

Hemodialysis can rapidly remove many exogenous toxins (e.g., as the result of poisonings and drug overdoses) and endogenous toxins (such as those seen in persons with inborn errors of metabolism), and it is often the therapy of choice for severe life-threatening intoxication. Hemodialysis can be followed by continuous renal replacement therapy (CRRT) for intoxications where a rebound phenomenon may occur (see Chapters 76 and 105).

Renal Support

Diminished renal function is a common component of multiorgan dysfunction syndrome. AKI is highly correlated with poor outcome in hospitalized adults and children.^4,6 In pediatric patients, studies demonstrate increased mortality in critically ill children with excessive levels of fluid overload in the setting of concomitant AKI.^7–10 These observations, coupled with advanced capabilities for renal replacement therapy, lead many clinicians to consider early initiation of renal support. In this model, traditional indicators for dialysis, such as profound azotemia, oligoanuria, massive fluid overload, or significant electrolyte imbalance, may occur too late in the clinical course to alter outcome. Earlier intervention, either with renal replacement therapy or perhaps through careful conservative management, may prevent complications associated with serious metabolic disarray and volume overload and permit vigorous nutritional and medical support. The concept of early “renal support” for patients in the ICU to limit metabolic derangement and prevent volume overload has gained wider acceptance in both the adult and pediatric critical care arenas.

Physiology

It has been known for many years that the peritoneum can be used as a dialyzing membrane.¹⁶ Instillation of a dialysis fluid into the peritoneal space permits diffusion of particles out of the blood and into the dialysis fluid across the peritoneum, which acts as a semipermeable membrane. Through the use of a hypertonic solution, water also passes across the membrane, generating an ultrafiltrate. Water movement will also tend to drag particles across the peritoneum by convection. After the dwell is complete, the spent dialysate is drained from the abdomen and fresh dialysis fluid is introduced.

Indications

PD can remove excess fluid and provide volume control in patients with oligoanuria. Compared with fluid removal with intermittent hemodialysis, fluid removal with PD is much slower. Manipulation of dialysis fluid osmolality and dwell time can adjust the quantity of volume removed. The slow, steady ultrafiltration achieved with PD may be preferable to the rapid fluid removal that occurs in intermittent hemodialysis, particularly in unstable, critically ill patients.

Similar to volume control, PD provides slow and relatively continuous metabolic control. It is an effective method for correction of uremia. Manipulation of the PD prescription can improve clearances.

Technique

The basic technique for PD involves instilling a sterile dialysis fluid into the peritoneal cavity and allowing it to dwell for a specified period, during which particles diffuse across the peritoneal membrane and water moves across by ultrafiltration. At the end of the dwell time, the fluid is removed from the peritoneal space and the process is repeated.

Flexible catheters are most often used for chronic PD in pediatric patients. For acute PD, either this form of a surgically placed catheter or percutaneously inserted temporary PD catheters may be used; some data suggest that fewer complications occur with surgically placed catheters.¹⁷ Catheters come in a variety of sizes, depending on the size of the patient. Local practice often determines who will insert the catheters when they are needed; the procedure requires expertise to ensure proper functioning of the catheter.

PD fluid comes in standardized, sterile bags with premixed formulations, and thus pharmacy preparation is unnecessary. Because most patients with renal failure have metabolic acidosis, the dialysis fluid contains base, usually in the form of lactate. Thus the PD system will remove unwanted particles and also can act as a source of electrolytes. Lactate absorption can lead to confusion regarding acid/base interpretation, especially in critically ill infants. In such settings bicarbonate-based dialysis fluid may be preferable,¹⁸ although this fluid may require extemporaneous preparation by the local pharmacy.

Ultrafiltration in PD is accomplished by osmotic pressure, usually through the presence of dextrose in the dialysis fluid. Dialysis fluids contain standardized concentrations of dextrose; the choices vary somewhat between the United States and other countries. Dialysis fluid with higher concentrations of dextrose will cause greater ultrafiltration for each exchange of fluid.

Dialysis fluid should be warmed to body temperature before instillation. This step is particularly important in small patients, in whom hypotension associated with cold dialysis fluid infusion can develop.¹⁹

Initial exchanges with a new PD catheter should use relatively lower volumes of dialysis fluid (10 to 20 mL/kg; <500 mL/m²) to limit the chance of leakage from the catheter entrance operative site. Volumes may increase gradually during the next few days or weeks to 1100 mL/m².¹⁹

Longer dwell periods between exchanges provide more time for equilibration of dialyzable particles and for ultrafiltration. Although shorter dwell times may not maximize mass transport for given dwell periods, they may permit more dialysis and ultrafiltration in a 24-hour period by allowing more exchanges per day. Initial dwell periods of 30 to 60 minutes can be adjusted later on the basis of clinical status.

PD fluid can be instilled by hand or with the use of a cycler, a device that will automatically fill and empty the patient’s abdomen with dialysis fluid on a preprogrammed schedule. The cycler also contains a warmer for the dialysis fluid and monitoring systems to record effluent volumes. Several brands of cyclers are currently available. Programming limitations may prevent the use of a cycler for some patients who require very small fill volumes or very short dwell times.

Disadvantages and Complications

The PD technique requires placement of a peritoneal catheter and a sufficiently maintained intraabdominal status to permit infusion of dialysis fluid with successful diffusion and ultrafiltration. Patients who have undergone an abdominal operation or have had abdominal complications may be poor candidates for PD.

Invasion of the peritoneal space puts the patient at risk for peritonitis, a potentially serious complication. The importance of sterile technique when performing PD cannot be overemphasized. Appropriate technique limits the risk of infectious complications, which could be fatal in a critically ill patient.

Peritonitis must always be considered in a patient undergoing PD who has a fever or cloudy effluent. Dialysate should be analyzed for cell count, Gram stain, and bacterial culture if infection is suspected. Empirical or specific antibiotic therapy can be placed in the dialysate to treat peritonitis via the intraperitoneal route.²⁰

Dialysate can fail to fill or drain through the PD catheter because of a number of potential problems, including kinking of the catheter, fibrin plugs, omental obstruction, or catheter malposition. Percutaneously inserted temporary dialysis catheters are more prone to malfunction than are surgically placed catheters.²¹ Abdominal x-ray images can confirm appropriate positioning and permit checks for kinks in the catheter. Some success has been reported with thrombolytic agents to treat fibrin plugging.^22–24 Revision or replacement of the catheter may need to be considered if simple maneuvers do not correct the malfunction.

Perforation of abdominal or pelvic structures can occur, either at the time of initial catheter placement or later.²¹ Although this event is relatively uncommon, significant morbidity can result.

PD is a suboptimal choice for patients who require rapid correction of metabolic abnormalities, immediate removal of circulating toxins, or rapid ultrafiltration for acute complications of fluid overload. For the latter indications, hemodialysis would be the preferred modality. Effectiveness of PD may be suboptimal in settings of low cardiac output where splanchnic circulation is compromised. However, this does not preclude the use of PD in selected cases, such as for infants following surgery for congenital heart disease.

Fluid leakage is seen most often with dwell volumes that are too large, especially in the period immediately after catheter placement.²⁵ Lower fill volumes should be used. Fluid leakage into the thorax can compromise respiration. External fluid leakage around the catheter increases the risk of infection.

Intensive Care Unit Issues

Patients undergoing PD can lose protein into the dialysate. Nutritional support must provide sufficient protein to compensate for this loss. High dextrose concentrations in the dialysis fluid can cause hyperglycemia; administration of insulin may be necessary. Indwelling dialysis fluid causes increased intraabdominal pressure that can complicate care of the critically ill patient. Diaphragmatic excursion may be limited, and venous return can be reduced. Stomach compression can lead to gastroesophageal reflux. Although patients undergoing long-term PD who receive fewer daily exchanges usually require maximal fill volumes to achieve adequate dialysis, patients undergoing short-term PD may do better using submaximal fill volumes with more frequent exchanges provided around the clock.

Physiology

The dialyzer used in IHD is an artificial semipermeable membrane. The most common design used today is the hollow-fiber dialyzer. It consists of a plastic cartridge traversed by several thousand thin hollow fibers, each with microscopic fenestrations that permit the passage of water and other small molecules. Dialyzers vary in their surface area, permeability, priming volume, and membrane composition; numerous dialyzers are available commercially. Choosing different dialyzer characteristics permits adjustment of the dialysis prescription to the clinical situation.

Successful hemodialysis requires sufficiently high blood flow to allow adequate dialysis and ultrafiltration with a minimal risk of clotting. Dialysis efficiency falls off dramatically with lower blood pump speeds; this may require extended time on dialysis to compensate. Consequently, high-quality vascular access is essential for successful hemodialysis.

While blood flows through the hollow fibers of the dialyzer, dialysis fluid flows through the cartridge in the space surrounding the hollow fibers. Particles move by diffusion from the blood across the semipermeable membrane into the dialysate. Use of high-flow dialysis fluid with maximal blood flow through the dialyzer permits intermittent hemodialysis to remove particles more efficiently than any other renal replacement therapy.

Increasing blood flow, dialysis fluid flow, or dialyzer size will increase the rate of diffusion. Because diffusion favors the movement of small particles over large particles, large molecules or small molecules bound to larger molecules (such as albumin) will not dialyze well. In addition, intracellular particles will move into the vascular compartment on the basis of individual cell membrane transport characteristics, which may limit the rate at which dialysis can remove particles that do not reside within the vascular compartment.

Ultrafiltration on hemodialysis occurs because of hydrostatic pressure across the membrane that forces water out of the blood. Dissolved particles will travel with the water, leaving by convection. With use of IHD, dialysis staff can control the rate of ultrafiltration with precision and can achieve high rates of ultrafiltration.

Analogous to the process of diffusion, ultrafiltration during IHD removes fluid only from the vascular compartment. Extravascular or “third space” fluid must move into the vascular compartment for removal by ultrafiltration. When the rate of movement from the extravascular space is slower than the rate of ultrafiltration, intravascular depletion can occur even though total body water remains elevated (Figure 72-3). This “two-compartment” model represents a potential limitation of rapid ultrafiltration during IHD, especially in the critically ill patient.

Figure 72–3 Two-compartment model of ultrafiltration. The rate of water removal from the vascular compartment by ultrafiltration (large arrows) exceeds the rate of refilling from the extravascular compartment (smaller arrows), and this process leads to relative vascular volume depletion and hypotension.