Chapter 30: Acute Kidney Injury and Failure

Acute kidney injury (AKI) is the sudden decline of renal function resulting in the retention of nitrogenous waste products and the inability to regulate electrolytes and extracellular volume. The development of AKI is associated with increased morbidity and mortality. These patients have an increased hospital length of stay and hospital readmissions, all of which translate into increasing resource utilization. The number of deaths resulting from AKI continues to increase even though the treatment of AKI and care of critically ill patients has improved. The reason for this increase in absolute mortality from AKI is twofold: the incidence of AKI is increasing and patients with AKI are older, with a greater number and severity of comorbid conditions.¹ Furthermore, we are pushing the boundaries in medicine; sicker patients who once were untreatable are now undergoing more complex and invasive diagnostic and surgical procedures.

Although the kidneys receive approximately 20% of the total cardiac output, they are still at very high risk for ischemic injury. This dichotomy is explained by the fact that most renal blood flow (80%-90%) is directed toward the renal cortex, where it passes through the glomerulus and is filtered.² The most metabolically active portion of the kidney, however, is not the cortex but the medulla, which is responsible for creating the osmotic gradient that serves to reabsorb water and concentrate urine. In order to maintain this osmotic gradient, blood flow to the medulla is kept low. In fact, the oxygen tension in the medulla is only approximately 10 mm Hg.² This lack of luxury perfusion to a metabolically active portion of the kidney predisposes it to ischemic injury. Furthermore, the kidney concentrates toxins that the body is exposed to, thereby amplifying the exposure of renal cells to toxic substances.

The etiology of AKI is traditionally divided into prerenal, intrinsic renal, and postrenal. Prerenal azotemia occurs as a complication of decreased renal perfusion with preservation of the cellular architecture of the renal parenchyma. Direct injury to the kidneys results in renal azotemia, or intrinsic renal injury. Postrenal azotemia occurs as a result of obstruction of the urinary outflow tract. The overwhelming majority of AKI in critically ill patients is caused by sepsis, whose mechanism does not quite fit within any of the traditional categories of AKI. It was traditionally thought that sepsis-induced AKI was caused by renal hypoperfusion due to systemic hypotension and renal vasoconstriction, along with reperfusion injury to the kidneys. However, this theory has recently come into question. It is now thought that while ischemia may play a role, there are many other pathophysiologic mechanisms at work. Recent human and animal studies have shown that renal blood flow is preserved in sepsis. Current concepts of AKI in sepsis indicate that systemic inflammation, microvascular dysregulation, and mitochondrial alteration, leading to cell death may be more important than global renal hypoperfusion.³ This theory claims that sepsis induces an “inflammatory danger signal” that is initially adaptive and later becomes maladaptive, leading ultimately to, metabolic reprioritization of renal cells which favors cell adaptive processes (like maintenance of cell membrane potential and cell cycle arrest) at the expense of actual filtration of blood.³

Renal hypoperfusion is the hallmark of prerenal AKI. Renal hypoperfusion (resulting from prolonged hypotension, hemorrhage, abdominal compartment syndrome, low cardiac output states, or cirrhosis) is a common cause of prerenal AKI in the ICU. Medications that interfere with normal renal autoregulation (nonsteroidal anti-inflammatory drugs, angiotensin converting enzyme inhibitors, angiotensin receptor blockers, cyclosporine, and tacrolimus) can precipitate AKI in patients with marginal renal perfusion. Prerenal azotemia presents with decreased glomerular filtration rate (GFR) and little evidence of cellular damage in most cases. Therefore, the condition is usually completely reversible. Prolonged prerenal azotemia, however, can lead to evidence of cellular damage, usually in the form of acute tubular necrosis.

Intrarenal AKI has many causes and is best understood when classified by the location of the lesion: glomerulus, tubule, vasculature, or interstitium. Most intrarenal AKI in the ICU setting is due to acute tubular necrosis from prolonged hypotension, nephrotoxic medications, recent exposure to radiographic contrast agents; or interstitial nephritis from nephrotoxic medications, such as antibiotics. Glomerular and vascular causes of intrarenal AKI are more commonly causes of renal failure outside of the ICU.

Urinary indices can be helpful in distinguishing prerenal and renal azotemia in non-ICU settings. A fractional excretion of sodium (FE_Na) less than 1% is commonly used in the non-ICU setting to identify patients with prerenal AKI. However, in the ICU, where many patients receive diuretics, the FE_Na becomes unreliable and should not be routinely used as a diagnostic test. An elevated blood urea nitrogen (BUN)/creatinine ratio is also not necessarily indicative of prerenal AKI in ICU patients due to the lengthy list of comorbidities that may elevate the BUN/creatinine ratio in a euvolemic ICU patient (gastrointestinal hemorrhage, total parenteral nutrition, steroids, catabolic stress). Analysis of urinary sediment can be used to distinguish prerenal and intrarenal AKI, but these tests have limited use in an ICU setting. Generally, the microscopic examination of urine in patients with prerenal or postrenal AKI is normal. Red blood cell casts, protein, white blood cells or leukocyte casts, or hematuria suggest intrarenal causes of AKI.

Postrenal AKI occurs when there is bilateral or unilateral obstruction in urinary flow. This leads to an increase in intratubular pressure, eventually leading to a decrease in glomerular filtration pressure. Postrenal AKI is a very rare cause of AKI in the ICU, since most patients in the ICU have a urinary catheter. Obstruction of the urinary tract can be extrarenal or renal. Extrarenal causes include prostatic hypertrophy, and abdominal/retroperitoneal masses compressing the ureters. Intrarenal obstruction can be caused by deposition of stones, crystals, clots, or tumors.

The RIFLE (acronym stands for: risk, injury, failure, loss, and end-stage renal disease) criteria established a universal definition of AKI, facilitating communication among clinicians and researchers. The criteria include three severity grades (risk, injury, and failure), which are based on changes in serum creatinine and urine output, using the worst of each measure.⁴ The criteria also include two outcome criteria (loss and end-stage renal disease), which are based on the duration of the loss of kidney function.⁴ The RIFLE criteria have been validated with multiple trials showing that the severity of AKI, as defined by the RIFLE criteria, is associated with increased mortality.

The Acute Kidney Injury Network (AKIN) made modifications to the RIFLE criteria in order to take into account the importance of timing in kidney injury. Their diagnostic criteria for AKI include an abrupt (within 48 hours) reduction in kidney function defined by an absolute increase in SCr of more than or equal to 0.3 mg/dL, a percentage increase in SCr of more than or equal to 50% (1.5-fold from baseline) or a reduction in urine output (less than 0.5 mL/kg/h for more than 6 hours).⁴ Their classification system involves a numbered system where stage 1 corresponds with RIFLE R, stage 2 with RIFLE I, and stage 3 with RIFLE F. Loss and end stage kidney disease were removed from the staging system (Table 30–1).

The AKIN and RIFLE criteria are useful measures of AKI, but are also fraught with limitations. Urine output and SCr are at best surrogate markers of renal function, but can be “normal” in the presence of renal dysfunction. The use of SCr can delay the diagnosis of renal injury and increases in SCr are only detected after a substantial reduction in GFR. In this sense, creatinine reflects renal injury that has already occurred. Creatinine is also affected by many factors other than renal dysfunction including muscle mass, catabolic state, rhabdomyolysis, dilutional effects, and drugs. Recent studies have identified new biomarkers of renal injury that can diagnose AKI earlier than SCr, but these are still in the primary stages of clinical adoption.

The risk of developing AKI is dependent on the interplay between a patient's preexisting susceptibility to kidney injury combined with the duration of exposure to insults that result in AKI.⁴ Table 30–2 summarizes the most common risk factors and exposures leading to AKI.

There are some general underlying management strategies that are common to all types of AKI. The mainstay of AKI management relies on early diagnosis and prompt intervention to minimize injury to the kidneys. In addition, adjusting to renal doses of renally metabolized medications is crucial as is avoiding any additional exposure to nephrotoxic drugs.⁵

In general, hemodynamic management, assessment of volume status, and management of fluid administration become even more important once renal function has deteriorated. Hypotension and decreased circulating volume can lead to hypoperfusion and ischemia of the kidneys; especially since injured kidneys lose the ability to autoregulate blood flow and become dependent on the mean arterial pressure to maintain perfusion. Conversely, overaggressive fluid resuscitation can lead to tissue edema, hypoxia, and heart failure, exacerbating poor perfusion to the kidneys. Positive fluid balance in AKI has been correlated with increased 60 days mortality in large multicenter studies.⁶ Strict hemodynamic monitoring and control of volume status in patients with AKI is critical and sometimes requires invasive cardiac monitoring or frequent bedside echocardiographic monitoring.

The type of fluid used for resuscitation in AKI and critically ill patients remains controversial. The Saline versus albumin fluid evaluation (SAFE) trial showed that albumin and crystalloids were equivalent in terms of renal outcomes.⁷ It is probably prudent to avoid volume resuscitation with hetastarch in patients with AKI as recent evidence points to increased renal injury and increased incidence of renal replacement therapy with the use of hetastarches.⁸ It is generally accepted that buffered salt solutions (such as lactated ringers, or Plasmalyte) may be the best agents for intravascular fluid resuscitation.⁴ Some studies indicate that normal saline administered in large quantities can cause hyperchloremia, leading to decreased renal perfusion and increased incidence of AKI and need for renal replacement therapy. There is also a trend towards a hyperchloremic metabolic acidosis, leading to hyperkalemia.⁹

Frequently, patients in the ICU have shock of various etiologies, and after adequate fluid resuscitation, the use of vasopressors is often necessary to maintain an adequate arterial blood pressure. A large randomized study comparing the use of dopamine to norepinephrine infusions in patients with septic shock found that there was no difference in mortality or renal function, but dopamine was associated with more adverse events.¹⁰ Both norepinephrine and vasopressin are suitable agents for the treatment of septic shock in patients that have been adequately volume resuscitated. Vasopressin has been gaining popularity because of some evidence that it may be superior to norepinephrine at reducing the progression of AKI, leading to less patients requiring RRT.¹¹ Finally, inotropes in conjunction with vasopressors also have a place in the treatment of hypotension due to cardiogenic shock as they help improve overall perfusion to all organs.

Glycemic control among the critically ill is a controversial topic. Hyperglycemia and insulin resistance seem to be natural physiologic responses to critical illness, which occur in response to inflammatory mediators that are released during critical illness. Multiple studies have not shown any benefits of intensive insulin therapy and instead have shown an increased risk of hypoglycemia. The largest randomized trial of intensive glucose control in the ICU (NICE-SUGAR trial) failed to show a benefit in prevention of organ dysfunction, mortality, or bacteremia in patients with tight glucose control (81-108 mg/dL) compared to standard glucose control (140-180 mg/dL).¹² The KDINGO guidelines for the management of AKI suggest that critically ill patients should be treated with insulin for the prevention of severe hyperglycemia. Currently the blood glucose target supported by the majority of medical associations is 110 to 180 mg/dL.⁴

Dopamine has been used in the past as prophylaxis against and treatment of renal failure in critically ill patients. Low-dose dopamine (1-3 mcg/kg/min) administered to healthy individuals results in renal afferent arteriole vasodilation, natriuresis, and increased GFR.¹³ However, in patients with AKI, low-dose dopamine has in fact been shown to increase renal vascular resistance.¹³ A large, randomized, placebo controlled, double-blinded study showed that dopamine had no renal protective effects.¹³ There is some evidence that shows that the use of dopamine in AKI can even cause harm. It has been linked with tachy-dysrhythmias, myocardial ischemia, decreased intestinal blood flow, and can suppress immune function. Therefore, the use of dopamine to prevent or treat AKI has been abandoned.⁴

A promising agent for the prevention and treatment of AKI involves the use of fenoldopam. Fenoldopam is a pure dopamine 1-type receptor agonist with no alpha- or beta-receptor properties that is currently only indicated for treatment of hypertension. It reduces systemic vascular resistance and increases renal blood flow. Fenoldopam has been shown in small trials to have a renal protective effect, preventing the development of AKI.¹⁴ While the data is promising, there are still no multicenter, highly powered, randomized controlled trials showing a clinically significant benefit of using fenoldopam in AKI. Furthermore, the hypotension that it causes is a significant side effect particularly in critically ill patients.

Atrial natriuretic peptide (ANP) is another potential therapy for the prevention and treatment of AKI. ANP is an amino acid that is released from the atrium in response to atrial stretch. It has diuretic, natriuretic, and renal vasodilatory activity, and can potentially increase GFR. There have been several negative studies concerning the use of ANP as prophylaxis against AKI.¹⁵ Therefore, until further research proves a benefit, ANP should not be used for treatment of prevention of AKI.

Nesiritide is a recombinant form of brain natriuretic peptide that has been approved by the Food and Drug Administration for use in decompensated heart failure.⁴ It may improve GFR, decrease rise in SCr, and improve urine output in patients, especially those with baseline renal insufficiency. In a pilot trial, nesiritide infusion started during and after cardiopulmonary bypass was shown to improve SCr values in patients with left ventricular dysfunction undergoing cardiac surgery.⁴ Nesiritide is not FDA approved for this indication.

Patients with AKI are frequently administered diuretics, either before or after being diagnosed with AKI. According to one observational study, approximately 59% to 70% of patients with AKI were shown to have been administered a diuretic before the onset of RRT.¹⁶ There are multiple reasons for administering diuretics to a patient with AKI. Nonoliguric renal failure has a better prognosis than oliguric renal failure; therefore, many clinicians administer diuretics to convert oliguric renal failure into nonoliguric renal failure. Diuretics can sometimes ameliorate fluid overload, commonly seen with AKI. Furthermore, some diuretics are thought to have renoprotective effects. However, improper diuretic use can lead to a decrease in intravascular volume, exacerbating prerenal insult to the kidneys.

Loop diuretics, such as furosemide, have a potentially renoprotective profile that may serve to protect the kidneys from AKI. They inhibit the Na²⁺-K-2Cl^– transporter in the ascending limb of the Loop of Henle. Inhibition of this active sodium transporter may decrease the oxygen demand of the renal tubule and protect the kidney against ischemia.¹⁷ Furosemide also increases the flow of urine in the renal tubules, helping to wash out ischemia-related necrotic debris blocking the tubules.¹⁷

A meta-analysis by Ho et al examined the effect of loop diuretics in AKI. They found that furosemide had no impact on hospital mortality, the requirement for RRT, number of dialysis sessions, or the incidence of oliguria. Furosemide was not effective for the prevention of AKI; it did not improve mortality in AKI and did not reduce the severity of AKI. Furosemide also did not decrease the duration of CRRT therapy and did not facilitate the return of renal function.¹⁸ However, furosemide does have a role for the management of hypervolemia, hyperkalemia, and hypercalcemia in AKI.⁴

Mannitol has been used to prevent AKI, however, most of the studies showing a beneficial effect of mannitol are underpowered, retrospective, and of poor quality.⁴ Mannitol may be beneficial when administered just prior to the arterial clamp removal in renal transplantation. In this setting, mannitol has been shown to decrease posttransplant AKI, but has not been shown to have a positive impact on renal graft function after 3 months.¹⁹ There is some underwhelming evidence that mannitol may be beneficial in treating rhabdomyolysis by stimulating osmotic diuresis. But again, the studies were not randomized and are underpowered.²⁰

Protein-calorie malnutrition is an independent risk factor for in-hospital mortality for patients with AKI. The nutritional considerations in patients with AKI must take into account a multitude of patient factors, including metabolic and inflammatory derangements, renal replacement therapy–induced nutrient imbalance, the underlying etiology of AKI, and patient comorbidities.

AKI is associated with hyperglycemia from peripheral insulin resistance, and increased hepatic gluconeogenesis. Hypertriglyceridemia is another consequence of AKI and occurs due to decreased lipolysis and impaired ability to metabolize exogenous lipids.²¹ When considering nutritional supplementation, it should be noted that patients with AKI do not manifest increased energy consumption. The optimal energy to nitrogen ratio is not known. One study of AKI patients on CVVH showed that there was a less negative/weakly positive nitrogen balance with energy intake of 25 kcal/kg/d.²² Another randomized study of AKI patients provided 30 or 40 kcal/kg/d of energy. The higher caloric intake did not improve the nitrogen balance of these patients and instead was associated with worsening hyperglycemia, and hypertriglyceridemia.²³ Therefore, it is reasonable to suggest that patients with AKI be provided with at least 20 kcal/kg/d, but not more than 30 kcal/kg/d.⁴

Protein catabolism is a common finding in critically ill patients, including those with AKI. Consequently, protein intake should not be withheld in order to attenuate the rise in BUN and SCr in order to stave off RRT.⁴ However, administration of high doses of protein have not been proven to improve the nitrogen balance of patients with AKI, and may in fact be associated with increased acidosis, azotemia, and increasing dialysis dose requirements.²⁴ Patients on RRT lose extra protein during therapy in addition to the protein catabolism occurring as a result of their critical illness. Patients on CRRT lose 0.2 g of amino acids per liter of filtrate (5-10 g/d)⁴; patients on intermittent HD may lose less protein per day. The KDINGO guidelines suggest administering 0.8 to 1.0 g/kg/d of protein in AKI patients not on RRT; 1.0 to 1.5 g/kg/d in AKI patients on RRT; and a maximum of 1.7 g/kg/d in patients on CRRT.⁴

Enteral nutrition should be used whenever possible in all critically ill patients, including those with AKI. Critically ill patients may not tolerate enteral nutrition due to delayed gastric emptying and decreased nutritional uptake from bowel wall edema. The use of enteral nutrition has been shown to improve outcome and survival in ICU patients. The provision of nutrients via the gut helps maintain gut integrity, decreases gut atrophy, and decreases bacterial translocation.

Aminoglycoside

Aminoglycosides are very potent bactericidal agents that are being used with increasing frequency because of growing antibiotic resistance. The advantages of aminoglycosides include: predictable pharmacokinetics, and lack of hematologic and hepatic toxicity. However, they are also fraught with considerable side effects, including: nephrotoxicity and ototoxicity.⁴ The risk of aminoglycoside induced AKI can be as high as 25%.²⁶ When used appropriately, with proper dosing and monitoring of levels, the risk of AKI with aminoglycosides can be minimized.

Repeated dosing of aminoglycosides results in the accumulation of aminoglycosides in the renal tubular epithelial cells, leading to a higher incidence of nephrotoxicity with repeated exposure.²⁷ Single daily dosing or extended interval dosing can potentially decrease the amount of aminoglycoside that is taken up in the renal tubules and may limit the potential for nephrotoxicity without sacrificing the therapeutic effect.²⁷ Monitoring of aminoglycoside levels during therapy is essential in critically ill patients with frequent fluctuations in GFR.

Amphotericin B

Amphotericin B is a widely used antifungal drug with many systemic side effects including chills, hypotension, cytokine release, thrombophlebitis, electrolyte disturbances, and hypoplastic anemia. Nephrotoxicity is commonly seen with amphotericin B and is the primary dose-limiting side effect.²⁸

Prevention of amphotericin toxicity relies on judicious use of amphotericin. Amphotericin should be avoided, and azole antifungal agents should be used whenever feasible. One major advance in the avoidance of amphotericin nephrotoxicity has been the advent of lipid formulations of amphotericin. The original formulation of amphotericin used deoxycholate as the solvent.²⁸ Clinical trials have shown that the lipid formulations of amphotericin are less nephrotoxic, while preserving the potent antifungal effect of amphotericin.²⁸

Many critically ill patients receive diagnostic or therapeutic procedures that require exposure to intravenous (IV) contrast medium. The nephrotoxic potential of contrast media is well known and not inconsequential. Diabetes mellitus, Chronic kidney disease (CKD), and advanced age have been identified as the three risk factors that are common amongst patients that develop contrast induced (CI)-AKI. Preexisting renal disease is the most important risk factor, with the incidence of CI-AKI approaching 25% in this setting.³⁰ Precautions to reduce the risk of CI-AKI should be taken in patients whose baseline creatinine is greater than 1.8 mg/dL.²⁹

Prevention of CI-AKI focuses on the use of noniodinated contrast media²⁹, minimizing contrast-media volume, avoiding repeat exposure to contrast media, and expanding plasma volume before administration of contrast media. Using iso-osmolar or low-osmolar iodinated contrast media should be used whenever feasible. In general, repeated exposure to contrast should be delayed for 48 hours in patients without risk factors, and for 72 hours in those with diabetes mellitus or previous AKD/AKI. The risk of AKI seems to be highest after inter arterial injection of iodinated contrast media.²⁹ The published literature suggests that periprocedural dialysis has no protective effect against CI-AKI.

Extracellular volume expansion may serve to prevent CI-AKI by a variety of mechanisms. Exposure of the kidneys to radiocontrast agents causes renal damage via a direct nephrotoxic effect, secretion of vasopressin and stimulation of the renin-angiotensin axis (stimulates renal afferent artery vasoconstriction resulting in renal hypoperfusion).²⁹ Extracellular volume expansion causes decreased secretion of vasopressin and also attenuates the renin-angiotensin axis stimulation. Furthermore, by directly diluting the radiocontrast agent, volume expansion may attenuate the direct nephrotoxicity from the contrast. The type of IV fluid used for extracellular volume expansion plays a role in averting CI-AKI. The two solutions that have shown to help prevent CI-AKI are isotonic normal saline and sodium bicarbonate solutions. The REINFORCE trial, however, found no difference in the incidence of CI-AKI in patients receiving sodium bicarbonate when compared to normal saline.³⁰

There have been many trials examining the use of N-acetyl cysteine (NAC) before contrast exposure in order to decrease the incidence of CI-AKI. Although the evidence supporting the use of NAC in preventing CI-AKI is not particularly strong, considering its negligible side effects, some clinicians include oral NAC along with IV fluid resuscitation as a strategy to prevent CI-AKI.³¹