Manifestations of and risk factors for aminoglycoside nephrotoxicity

INTRODUCTION — Acute kidney injury is a relatively common complication of therapy with the aminoglycoside antibiotics, with a rise in the plasma creatinine concentration of more than 0.5 to 1 mg/dL (44 to 88 micromol/L) or 50 percent increase in plasma creatinine concentration from baseline occurring in 10 to 20 percent of hospitalized patients [1-4]. In children, aminoglycoside-induced acute kidney injury can be seen in as many as 20 to 33 percent of treated children [5].

Aminoglycosides are freely filtered across the glomerulus and then partially taken up by, and concentrated in, cortical proximal tubule cells, which is where most of the kidney damage occurs. After administration, up to 5 to 10 percent of the parenteral dose is retained in the kidney cortex, where it can achieve concentrations greatly exceeding the concurrent serum concentration [6]. This preferential sequestration of aminoglycosides in proximal tubule cells accounts for the observation that kidney failure may become clinically apparent as late as several days after the drug has been discontinued. (See "Pathogenesis and prevention of aminoglycoside nephrotoxicity and ototoxicity".)

The manifestations of and risk factors for aminoglycoside nephrotoxicity are reviewed here. Acute tubular necrosis due to ischemia or other insults and the pathogenesis and potential therapy of aminoglycoside nephrotoxicity are discussed separately:

●(See "Etiology and diagnosis of prerenal disease and acute tubular necrosis in acute kidney injury in adults".)

●(See "Pathogenesis and etiology of ischemic acute tubular necrosis".)

●(See "Pathogenesis and prevention of aminoglycoside nephrotoxicity and ototoxicity".)

MANIFESTATIONS

Nonoliguric acute kidney injury — Acute kidney injury from aminoglycoside exposure typically manifests with a rise in serum creatinine after five to seven days of therapy. The nonoliguric aspect of the kidney failure is secondary to a loss in renal concentrating ability believed to be the result of distal tubular damage.

The acute tubular necrosis that occurs from aminoglycoside exposure is unlikely to be severe if detected early, with incremental increases in the plasma creatinine that are usually mild (0.5 to 2 mg/dL [44 to 177 micromol/liter]) [7]. There is also evidence that patients without chronic kidney disease (CKD) are unlikely to develop CKD following a single episode of aminoglycoside-induced acute kidney injury [8]. However, aminoglycoside toxicity can result in progression of preexisting CKD, even precipitating kidney replacement therapy. The urine sediment most commonly shows mild proteinuria, hyaline, and granular casts. The fractional excretion of sodium is generally above 1 percent (calculator 1) or, for standard units, (calculator 2). (See "Fractional excretion of sodium, urea, and other molecules in acute kidney injury".)

Distal tubular dysfunction — Distal nephron segments also can be affected in aminoglycoside nephrotoxicity. The two major manifestations of distal dysfunction are polyuria due to decreased concentrating ability and hypomagnesemia due to enhanced urinary losses (see next section) [1,9]. The polyuria is postulated to be secondary to the attenuation in the sensitivity of the collecting-duct epithelium to endogenous antidiuretic hormone in aminoglycoside nephrotoxicity [7].

Electrolyte abnormalities — Hypomagnesemia, hypokalemia, hypocalcemia, and hypophosphatemia are infrequently observed. This is likely secondary to the decrement in proximal tubule transport that occurs during aminoglycoside-induced nephrotoxicity. A Fanconi-like syndrome can occur with glycosuria, aminoaciduria, phosphaturia, and uricosuria. The loss of phosphorus in the urine results in hypophosphatemia [1]. With aminoglycoside nephrotoxicity, magnesium depletion resulting in hypomagnesemia can lead to secondary hypokalemia and hypocalcemia.

Treatment of aminoglycoside-induced magnesium-wasting consists of the administration of oral or intravenous magnesium supplements. However, the efficacy of this oral regimen is limited by the urinary excretion of most of the extra magnesium. (See "Hypomagnesemia: Clinical manifestations of magnesium depletion" and "Hypomagnesemia: Evaluation and treatment".)

Course — Given that proximal tubules can regenerate, the usual course of aminoglycoside nephrotoxicity is one of recovery of kidney function. The plasma creatinine concentration usually returns to the prior baseline level within 21 days after cessation of therapy. However, resolution of the acute episode may be delayed if the patient remains hypovolemic, septic, or catabolic; in these settings, tubular regeneration cannot occur.

Irreversible kidney damage is uncommon with acute aminoglycoside nephrotoxicity. However, it may occur with prolonged therapy, even in low doses [10].

Studies in other forms of reversible acute kidney injury suggest that normalization of the plasma creatinine concentration and even the glomerular filtration rate may not reflect complete recovery. Irreversible nephron loss in this setting may be masked by compensatory hyperfiltration in the remaining normal glomeruli. (See "Kidney and patient outcomes after acute kidney injury in adults".)

Although a pattern of a nonoliguric acute kidney injury five to seven days after aminoglycoside exposure is suggestive, differentiating aminoglycoside-induced acute kidney injury from other etiologies of acute tubular necrosis can be difficult given that it often occurs in the setting of significant comorbid conditions, sepsis, and concomitant nephrotoxin exposure. In addition, acute kidney injury is frequently multifactorial, and aminoglycoside nephrotoxicity is only one of its components. An essential initial diagnostic step in the diagnosis of aminoglycoside nephrotoxicity is a careful history with attention to potentially reversible etiologies, such as reduced effective arterial volume or urinary tract obstruction. In addition, a review of all imaging procedures for intravenous contrast exposure and medications for concomitant nephrotoxin use is imperative.

RISK FACTORS — Several risk factors have been identified that increase the risk of aminoglycoside nephrotoxicity [1-3]. The major risk factors are:

●Prolonged duration of therapy

●Advanced age

●Comorbid disease

●Reduced effective arterial volume

●Sepsis

●Concomitant medications

●Elevated plasma aminoglycoside concentrations

●Type of aminoglycoside

●Frequency of dosing

●Serum albumin <3 g/dL

●Anemia

Prolonged duration of therapy — The development of acute kidney injury, as determined by serum creatinine, generally requires at least five to seven days of exposure to aminoglycosides in patients with normal hemodynamic status. Prolonged therapy increases the risk that toxic aminoglycoside concentrations will develop and persist in the kidney cortex.

For any given aminoglycoside, a direct relationship exists between serum concentrations of aminoglycosides, renal cortical proximal tubule cell concentrations, and nephrotoxicity [1,11,12]. Owing to the renal cortical sequestration of aminoglycosides, the risk of aminoglycoside-induced nephrotoxicity is also increased in patients given the drug in repeated courses separated by a few days or weeks [1]. In contrast, renal cortical accumulation is less when aminoglycosides are given as one large, daily dose, as in once-daily dosing programs, because of the saturable nature of the renal cortical uptake [13]. (See "Dosing and administration of parenteral aminoglycosides", section on 'Nephrotoxicity'.)

Advanced age — Advanced age has long been held to be an important risk factor in the development of aminoglycoside nephrotoxicity [14-18]. This increased risk is postulated to be secondary to the impaired capacity for cellular repair and regeneration found in this patient population. In addition, advanced age may result in excessive dosing of aminoglycosides from an overestimation of actual glomerular filtration rate (GFR) by estimated GFR equations.

Comorbid disease — A number of comorbid conditions are associated with aminoglycoside-induced kidney disease. Diabetes mellitus and leukemia, for example, increase the risk for aminoglycoside nephrotoxicity [18].

Early studies of aminoglycoside nephrotoxicity reported increased risk of aminoglycoside nephrotoxicity in patients with chronic kidney disease (CKD). In these studies, however, aminoglycoside concentrations were not pharmacokinetically adjusted, and patients were likely exposed to high serum levels of aminoglycosides with associated nephrotoxicity. Interestingly, subsequent studies showed that patients with higher creatinine clearances were associated with an increased risk of nephrotoxicity [2,14].

Overall, patients with CKD are at increased risk for aminoglycoside nephrotoxicity because the risk of inadvertent overdosing is greater than in patients with normal kidney function. This increased risk is due to the imprecision of our bedside tools in estimating kidney function and aminoglycoside dosing. Furthermore, patients with CKD have less kidney functional reserve and an attenuated ability to recover from aminoglycoside nephrotoxicity.

Among patients with advanced liver disease, activation of the renin-angiotensin-aldosterone system may play a pathogenetic role in increasing the risk of aminoglycoside nephrotoxicity [2]. In addition, obstructive jaundice with a plasma bilirubin concentration above 5 mg/dL (85 micromol/L) is associated with an increased risk of aminoglycoside nephrotoxicity [2,19,20]. How obstructive jaundice increases the sensitivity to aminoglycoside toxicity is not known, but either bilirubin or endotoxin (which is absorbed because of decreased bile salt entry into the intestinal lumen) may be involved [20]. The hypoalbuminemia frequently observed in advanced liver disease is also an independent risk factor for aminoglycoside nephrotoxicity [8].

Reduced effective arterial volume — Reduced effective arterial volume results in decreased kidney perfusion, kidney ischemia, and greatly enhances the risk of aminoglycoside nephrotoxicity [21,22]. This can occur in the clinical setting of intravascular volume depletion, heart failure, sepsis, and advanced liver disease. Moreover, kidney ischemia appears to accelerate the time course and severity for aminoglycoside nephrotoxicity.

Two factors have been postulated to mediate this synergism between the aminoglycoside and kidney ischemia:

●The aminoglycoside interferes with cell energetics, thereby increasing the susceptibility to kidney injury during an ischemic episode [23]. This synergism is most prominent in the S3 segment of the proximal tubule; this segment is particularly sensitive to ischemic injury secondary to minimal reperfusion resulting in continuing hypoxia to the area.

●Ischemia alters membrane lipids. In particular, there is enhanced phosphatidylinositol (PI) content in the luminal membrane [24]. This results in increased aminoglycoside binding, uptake, and accumulation in the proximal tubular cells of S1, S2, and S3 segments since filtered cationic gentamicin binds to PI and other acidic phospholipids and is then internalized by endocytosis. This charge interaction has important implications for the development and possible prevention of aminoglycoside nephrotoxicity and is discussed in detail elsewhere. (See "Pathogenesis and prevention of aminoglycoside nephrotoxicity and ototoxicity".)

Sepsis — In addition to volume depletion, as described above, endotoxin also enhances the nephrotoxicity of the aminoglycosides [25]. Although decreased kidney perfusion due to renal vasoconstriction may play a contributory role, endotoxin also appears to be associated with enhanced drug accumulation in the tubules. How this occurs is not known. (See "Pathogenesis and etiology of ischemic acute tubular necrosis".)

Concomitant medications — Administration of potentially nephrotoxic medications with aminoglycosides places a patient at increased risk for nephrotoxicity. Implicated medications include furosemide, nonsteroidal antiinflammatory drugs (NSAIDs), angiotensin-converting enzyme inhibitors, cisplatin, cyclosporine, clindamycin, and vancomycin [1,2,17,18].

Initial studies suggested that cephalothin increased the nephrotoxicity of the aminoglycosides [26]. However, patients with sepsis or hypotension were excluded. In this limited population, cephalothin did have a potentiating effect, but the kidney injury was mild, with the plasma creatinine concentration remaining below 2 mg/dL (176 micromol/L) in almost all patients [26].

In comparison, there seemed to be no increase in toxic effects when all treated patients are included, which more closely simulates the real world [2]. Furthermore, it is unclear if the initial results with cephalothin can be extrapolated to any of the other cephalosporin-type drugs. Evidence suggests that this probably does not occur and that the cephalosporins can be safely given in conjunction with an aminoglycoside [3].

Although it seems likely that cephalosporins do not have a major effect on aminoglycoside nephrotoxicity, there is suggestive evidence that concurrent administration of a penicillin may actually be protective [27,28]. How this occurs is not known, but decreased tubular uptake of the aminoglycoside may be involved. It must be emphasized, however, that penicillins, particularly carbenicillin, piperacillin, and ticarcillin, can physically complex with the aminoglycosides; as a result, these agents cannot be mixed in the same intravenous infusion.

Elevated plasma drug concentrations — Careful monitoring of plasma levels is an important component of aminoglycoside therapy. Impaired kidney function can occur even with perfect control of the plasma level since renal cortical aminoglycoside accumulation increases with time [29,30]. However, the risk is greater in patients receiving divided-dose aminoglycoside therapy who develop high peak plasma drug levels (>10 mcg/mL for gentamicin and tobramycin) [1,30]. In addition, elevated trough levels (>2.5 mcg/mL) have been associated with nephrotoxicity in a once-daily aminoglycoside dosing program of an older cohort of patients [15].

Remarkably, whether an elevated trough or peak level contributes to the increased risk of nephrotoxicity with aminoglycosides is not fully elucidated. In one analysis, no causal relationship could be found between a serum peak or trough aminoglycoside level in several reviewed studies [31].

The therapeutic success observed with once-daily aminoglycoside programs with their associated supratherapeutic levels certainly casts doubt on whether a peak aminoglycoside level increases nephrotoxic risk in this clinical setting. However, an elevated trough level has been shown to increase the nephrotoxic risk in once-daily aminoglycoside programs in older adults [15]. Similarly, in another older patient cohort, an elevated peak level increased aminoglycoside nephrotoxicity with conventional dosing [15].

Although there remains uncertainty about whether peak or trough levels are primarily responsible for nephrotoxicity, serum monitoring of aminoglycosides is beneficial and recommended due to the following reasons:

●The limitations of our bedside estimates of glomerular filtration rates on which we base our aminoglycoside dosing are often imprecise.

●Monitoring programs ensure that less toxic antibiotics will be selected in patients with contraindications to aminoglycosides.

●These programs also ensure that appropriate therapeutic doses of antibiotics are given earlier in a patient's therapy.

Several researchers have demonstrated the benefits of pharmacokinetically monitoring aminoglycoside therapy. This is especially important since few patients have a known glomerular filtration rate allowing for appropriate modeling of the pharmacokinetics. Among the benefits cited were fewer instances of nephrotoxicity and treatment failure and shorter hospital stays [32-35].

Serum peak and trough levels should be measured periodically when conventional aminoglycoside dosing is used. Moreover, peak and trough monitoring in conventional dosing is imperative in older patients with obesity. By contrast, once-daily aminoglycoside programs only require a periodic trough level. This trough level is used to determine the timing of the next dose of aminoglycoside. (See "Dosing and administration of parenteral aminoglycosides".)

Type of aminoglycoside — Not all of the aminoglycosides appear to have the same potential for nephrotoxicity. However, comparing the relative risk of each of the aminoglycosides for nephrotoxicity is problematic due to the variability in the clinical studies regarding the incidence of nephrotoxicity, the use of controls, and the criteria used to define nephrotoxicity.

Overall, when prospective comparative trials are compared, gentamicin has the greatest nephrotoxic potential, followed in decreasing order of nephrotoxicity by tobramycin, amikacin, and netilmicin [1,29,36]. Although amikacin and netilmicin are considered less nephrotoxic than gentamicin and tobramycin, the difference is not considered profound [1]. Few data are available to determine the nephrotoxicity of plazomicin [37,38]. The relative nephrotoxic risk of a specific aminoglycoside will always depend on patient factors, comorbid disease, and the clinical setting in which the aminoglycoside is used.

Frequency of dosing — The frequency of dosing may also be a risk factor for the development of aminoglycoside nephrotoxicity. This is supported by the favorable clinical data observed with the once-daily aminoglycoside dosing programs. In once-daily aminoglycoside therapy, a much larger loading dose is given every 24 to 48 hours. The rationale for this approach is to exploit the concentration-dependent bactericidal activity and postantibiotic effect of aminoglycosides. This approach to aminoglycoside dosing has demonstrated improved clinical efficacy and reduced nephrotoxicity relative to conventional aminoglycoside dosing [21,22]. This is discussed in detail separately. (See "Dosing and administration of parenteral aminoglycosides", section on 'Comparing extended-interval and traditional intermittent dosing'.)

SUMMARY

●Acute kidney injury (AKI) due to aminoglycosides – AKI from aminoglycoside exposure typically manifests after five to seven days of therapy. The nonoliguric aspect of the kidney failure is secondary to a loss in renal concentrating ability believed to be the result of distal tubular damage. The acute tubular necrosis that occurs from aminoglycoside exposure is unlikely to be severe if detected early. (See 'Nonoliguric acute kidney injury' above.)

●Electrolyte abnormalities – Hypomagnesemia, hypokalemia, hypocalcemia, and hypophosphatemia are infrequently observed. (See 'Electrolyte abnormalities' above.)

●Course of AKI – Given that proximal tubules can regenerate, the usual course of aminoglycoside nephrotoxicity is one of recovery of kidney function. The plasma creatinine concentration usually returns to the prior baseline level within 21 days after cessation of therapy. However, resolution of the acute episode may be delayed and incomplete. (See 'Course' above.)

●Risk factors – Several risk factors have been identified that increase the risk of aminoglycoside nephrotoxicity. The major risk factors are prolonged duration of therapy, advanced age, comorbid disease, reduced effective arterial volume, sepsis, hypoalbuminemia, concomitant medications, elevated plasma drug concentrations, type of aminoglycoside, and frequency of dosing. (See 'Risk factors' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Brian S Decker, MD, PharmD (deceased), who contributed to an earlier version of this topic review.