Acute renal failure (ARF) results from an abrupt decline in renal function and is usually caused by an ischemic or toxic insult. Unfortunately in many cases, ARF inadvertently develops in the hospital setting in conjunction with diagnostic or therapeutic procedures. For example, ARF may result from decreased renal perfusion associated with anesthesia and surgery or with the use of vasodilators and nonsteroidal anti-inflammatory drugs (NSAIDs). Similarly, ARF frequently occurs in patients treated with potential nephrotoxicants like gentamicin, amphotericin, and cisplatin. The nephron damage that occurs with ARF is not always reversible; animals that do recover adequate renal function usually require prolonged and expensive intensive care. Two recent retrospective studies have documented the poor prognosis associated with ARF in dogs. In a study of hospital-acquired ARF, the survival rate was 40%, whereas in another study of all types of ARF, the survival rate was 24%. (Toxicant and ischemic-induced ARF in people is associated with a mortality rate of 50 to 60%, despite the widespread availability of hemodialysis.) Obviously, prevention of hospital-acquired ARF is important. Several risk factors have been identified that predispose dogs to gentamicin-induced ARF (Table), however it is likely that many of these risk factors also predispose dogs and cats to other types of toxicant-induced ARF as well as ARF induced by ischemia. A combination of decreased renal perfusion and/or use of nephrotoxic therapeutic agents superimposed on more chronic, pre-existing risk factors is usually responsible for ARF in the clinical setting.

Acute renal failure has three distinct phases, which are categorized: 1) initiation, 2) maintenance, and 3) recovery. During the initiation phase, therapeutic measures that reduce the renal insult can prevent development of established ARF. Tubular lesions and established nephron dysfunction characterize the maintenance phase. Therapeutic intervention during the maintenance phase, although often life saving, usually does little to diminish existing renal lesions or improve dysfunction. The recovery phase is the period when renal lesions resolve and function improves. Tubular damage may be reversible if the tubular basement membrane is intact and viable epithelial cells are present. Although additional nephrons cannot be produced and irreversibly damaged nephrons cannot be repaired, functional and morphologic hypertrophy of surviving nephrons can often adequately compensate for the decrease in nephron numbers. Even if renal functional recovery is incomplete, adequate function may be reestablished.

Risk factors for ARF

Dehydration and volume depletion are perhaps the most common and most important risk factors for ARF (Table). Studies in human beings indicate volume depletion increases a patient's risk of developing ARF by a factor of ten. Hypovolemia not only decreases renal perfusion, but also decreases the volume of distribution of nephrotoxic drugs and results in decreased tubular fluid flow rates and enhanced tubular absorption of toxicants. In addition to hypovolemia, renal hypoperfusion may be caused by decreased cardiac output, decreased plasma oncotic pressure, increased blood viscosity, systemic hypotension, and decreased renal prostaglandin synthesis. Any of these conditions can increase the risk of ARF in the hospital setting.

Pre-existing renal disease and advanced age, which is often associated with some degree of decreased renal function, can increase the potential for nephrotoxicity by several mechanisms. For example, the pharmacokinetics of potentially nephrotoxic drugs can be altered in the face of decreased renal function. Gentamicin clearance is decreased in dogs with subclinical renal dysfunction, and the same is probably true for other nephrotoxicants. Animals with renal insufficiency also have reduced urine concentrating ability and, therefore, decreased ability to compensate for prerenal influences. Renal disease may also compromise production of prostaglandins that help maintain renal vasodilation and blood flow. Additionally, the hyperphosphatemia that can occur in patients with renal insufficiency/failure is thought to increase the risk of ARF.

Decreased serum concentrations of several electrolytes can increase the risk of ARF. Hyponatremia exacerbates gentamicin nephrotoxicity in rats and potentiates contrast media-induced ARF in dogs. Hypocalcemia, hypomagnesemia, and hypokalemia are additional electrolyte abnormalities that may potentiate nephrotoxicity. Calcium and magnesium compete with gentamicin for anionic phospholipid membrane binding sites so that attachment, binding, and uptake of gentamicin in various tissues is inversely proportional to local divalent cation concentrations. These cations also tend to suppress parathyroid hormone production and release leading to decreased production of membrane phospholipid and, therefore, gentamicin binding to tissues. Studies in dogs have demonstrated that dietary potassium restriction exacerbates gentamicin nephrotoxicity, possibly because potassium depleted cells are more susceptible to necrosis. It is important to note that gentamicin administration in dogs is associated with increased urinary excretion of potassium. This increased urinary excretion of potassium could result in potassium depletion and increased nephrotoxicity in clinical patients. Therefore, serum electrolyte concentrations should be closely monitored in patients receiving potentially nephrotoxic drugs, especially if these patients are anorexic, vomiting, or have diarrhea.

Administration of potentially nephrotoxic drugs or a drug that may enhance nephrotoxicity obviously increases the risk of ARF. For example, concurrent use of furosemide and gentamicin in dogs is associated with increased risk and severity of ARF. Furosemide probably potentiates gentamicin-induced nephrotoxicity by causing dehydration, reducing the volume of distribution of gentamicin, and increasing the renal tubular absorption of gentamicin. Fluid repletion minimizes, but does not avoid, the potentiating effect of furosemide on gentamicin nephrotoxicity in the dog, because furosemide facilitates cellular uptake of gentamicin independent of hemodynamic changes. By similar mechanisms, furosemide has been shown to enhance intravenous radiocontrast-induced nephrotoxicity in human beings. Use of NSAIDs can also increase the risk of ARF. Anesthesia, sodium and/or volume depletion, sepsis, congestive heart failure, nephrotic syndrome, and hepatic disease are conditions in which prostaglandin-induced renal vasodilatation becomes important and the susceptibility to NSAIDs is increased.

Recent evidence in dogs suggests that the quantity of protein fed prior to a nephrotoxic insult can significantly affect the subsequent renal damage and dysfunction. Feeding high dietary protein prior to and during gentamicin administration reduces nephrotoxicity, enhances gentamicin clearance, and results in a larger volume of distribution compared with feeding medium or low protein. The beneficial effects of high dietary protein are likely associated with increased glomerular filtration and, therefore, improved toxicant excretion. High dietary protein also results in increased urinary excretion of protein, which may compete for nephrotoxicant reabsorption by tubular epithelial cells. Further research in the area of dietary protein conditioning is needed; if dietary protein conditioning can be shown to have renal protective effects it may have important clinical implications. Once renal damage has occurred however, high dietary protein would likely result in increased serum urea nitrogen and phosphorus concentrations and, therefore, would not be recommended.

Risk factors are additive and any complication occurring in high-risk patients increases the potential for ARF. Patients with shock, acidosis, sepsis, and major organ system failure are at increased risk for ARF, and these are the patients that are likely to require aggressive treatment including prolonged anesthesia, surgery, or chemotherapeutics, which are potentially damaging to the kidneys. For example, ARF is common in dogs with pyometra and E. coli endotoxin-induced urine concentrating defects, especially if fluid therapy is inadequate during anesthesia for ovariohysterectomy or during the recovery period. Trauma, extensive burns, vasculitis, pancreatitis, fever, diabetes mellitus, and multiple myeloma are additional conditions associated with a high incidence of ARF in veterinary medicine.

Early recognition of renal damage/dysfunction

Since therapeutic intervention is most successful when initiated during the induction phase of ARF, early recognition of renal damage/dysfunction is important. Physical examination of the patient at risk for ARF should include evaluation of pulse quality and hydration status. Monitoring body weight, packed cell volume and plasma total solids in comparison to baseline values may indicate subtle changes in hydration status. Blood pressure measurement will identify hypotensive and hypertensive patients, both of whom may be at increased risk for renal injury. In patients with palpable kidneys, renal swelling and/or pain may be associated with an acute ischemic or toxic insult.

Numerous urine parameters can herald the development of ARF. Urine output should be monitored in all high-risk patients that undergo anesthesia. Urine production may be quantitated using a closed indwelling catheter collection system. Normal urine output is approximately 1-2 ml/hr/kg body weight. Oliguria (< 0.25 ml/hr/kg) or anuria requires prompt attention and treatment. Nonoliguric ARF is being recognized with increasing frequency; increases in urine production, therefore, may also signal the onset of renal damage. Examples of nonoliguric ARF include those induced by gentamicin and cisplatin. Increased urine turbidity or changes in urine sediment (increasing numbers of WBCs, RBCs, renal epithelial cells, or cellular or granular casts) are other indications of acute renal damage along with increases in the fractional clearance of sodium and chloride. Finally, the acute onset of glucosuria or proteinuria may also be indicative of early glomerular or tubular damage. The interpretation of all of the above parameters is enhanced by knowledge of baseline values.

Detection of enzymes such as gamma-glutamyl transpeptidase (GGT) and N-acetyl-beta-D-glucosaminidase (NAG) in the urine has proven to be a sensitive indicator of renal tubular damage. These enzymes are too large to be normally filtered by the glomerulus, and, therefore, enzymuria indicates cell leakage, usually caused by tubular epithelial damage or necrosis. Urinary GGT originates from the proximal tubule brush border and NAG is present in proximal tubule lysosomes. In a study of gentamicin-treated dogs, increased urinary GGT activity was the earliest marker of renal damage/dysfunction. Interpretation of enzymuria is aided by baseline values obtained prior to a potential renal insult; 2 to 3-fold increases over baseline suggest significant tubular damage. Recently, urine enzyme/creatinine ratios have been shown to be accurate in dogs prior to the onset of azotemia. False positive results can occur with severe glomerular damage, resulting in increased glomerular filtration of serum enzymes. False negative results can occur after tubular damage depletes tubular enzyme stores.

Knowledge of the predisposing risk factors allows the clinician to assess the risk-benefit ratio in individual cases in which an elective anesthetic procedure is considered or the use of potentially nephrotoxic drugs is indicated. In some cases, predisposing risk factors can be corrected prior to any potential renal insults. In other cases, more intensive monitoring of the patient at risk may allow detection of ARF in the induction phase prior to the onset of established failure.

Table 1. Potential risk factors for ARF in dogs and cats

*Risk factors that are potentially correctable - see text.