Review

Nature Clinical Practice Nephrology (2006) 2, 80-91
doi:10.1038/ncpneph0076  
Received 3 August 2005 | Accepted 26 October 2005

Drug-associated renal dysfunction and injury

Devasmita Choudhury* and Ziauddin Ahmed  About the authors

Correspondence *Dialysis, Veterans Affairs North Texas Health Care System, 4500 Lancaster Road, Dallas, TX 75216, USA

Email devasmita.dev@med.va.gov

Summary

Renal dysfunction and injury secondary to medications are common, and can present as subtle injury and/or overt renal failure. Some drugs perturb renal perfusion and induce loss of filtration capacity. Others directly injure vascular, tubular, glomerular and interstitial cells, such that specific loss of renal function leads to clinical findings, including microangiopathy, Fanconi syndrome, acute tubular necrosis, acute interstitial nephritis, nephrotic syndrome, obstruction, nephrogenic diabetes insipidus, electrolyte abnormalities and chronic renal failure. Understanding the mechanisms involved, and recognizing the clinical presentations of renal dysfunction arising from use of commonly prescribed medications, are important if injury is to be detected early and prevented. This article reviews the clinical features and basic processes underlying renal injury related to the use of common drugs.

Review criteria

We searched PubMed using the following terms: "drug-induced renal failure", "renal toxicity", "medications and renal failure", "acute renal failure and medications", "chronic renal failure and medications" and "renal injury and drugs/medications". We also consulted renal texts.

Keywords:

drugs, electrolyte abnormalities, interstitial inflammation, nephrotoxicity, renal failure

Top

Introduction

Toxic effects on the kidney related to medications are both common and expected, given the kidney's roles in plasma filtration and maintenance of metabolic homeostasis. The renal vascular bed is exposed to a quarter of resting cardiac output. As such, glomerular, tubular and renal interstitial cells frequently encounter significant concentrations of medications and their metabolites, which can induce changes in kidney function and structure. Renal toxicity can be a result of hemodynamic changes, direct injury to cells and tissue, inflammatory tissue injury, and/or obstruction of renal excretion. Markers of early injury are being investigated.1 In the meantime, however, subtle renal damage (e.g. tubulopathy, acid–base abnormalities, electrolyte imbalances and disorders of water balance) and mild urinary sediment abnormalities associated with commonly used medications are frequently unrecognized. Detection is often delayed until an overt change in renal functional capacity is measured as an increase in serum blood urea nitrogen or creatinine.

The true incidence of drug-induced nephrotoxicity is therefore difficult to determine. Studies that evaluated episodes of acute tubular necrosis (ATN) or acute interstitial nephritis (AIN) attributed to medication determined the incidence to be as high as 18.3%. The incidence of nephrotoxic injury due to antibiotics (e.g. aminoglycosides) has been reported to be up to 36%.2, 3 Most episodes of drug-induced renal dysfunction are reversible, with function returning to baseline when the medication is discontinued. Chronic renal injury can, however, be induced by some medications, leading to chronic tubulointerstitial inflammation, papillary necrosis or prolonged proteinuria.4, 5 Heightened physician awareness is necessary if renal injury and associated morbidity from renal failure are to be prevented. As it is impossible to list all drugs associated with nephrotoxicity, this article will summarize the mechanisms of injury associated with particularly common medications, discuss clinical presentations and markers associated with drug-induced renal injury, and evaluate strategies that prevent or minimize renal injury.

Top

Acute renal injury

Prerenal azotemia

Diuretics, alone or in combination with other antihypertensives, are frequently associated with prerenal azotemia.6 Natriuresis is the desired outcome of diuretic use, but the kidney is extremely sensitive to maintaining adequate renal perfusion. Perturbed conservation of NaCl at any one tubular site is usually compensated to some extent by the activity of other renal tubular sites, unless counter-regulatory processes maintaining blood pressure homeostasis are impaired. Use of a single class of diuretics (thiazides, loop, potassium-sparing or osmotic) usually leads to steady-state volume balance as long as the patient's dietary intake of NaCl remains unchanged. Significant volume depletion can occur when oral or parenteral intake changes suddenly (e.g. during episodes of diarrhea, vomiting, massive sweating or bleeding), or when patients suffer compromised cardiac output, cirrhosis or nephrosis.

Concomitant administration of drugs from more than one diuretic class is also associated with massive volume losses. The combination of loop and thiazide diuretics, often used to treat volume overload of heart failure, nephrotic syndrome and/or chronic renal failure, is one example.7, 8, 9, 10 Massive volume losses can result from large doses of osmotic diuretics (e.g. mannitol >300 g) used in neurosurgical patients to decrease brain edema.11 Large volume losses stimulate renal vasoconstriction with marked tubular avidity for NaCl uptake, and decreased urine output. Prolonged vasoconstriction can then lead to tubular dysfunction and tubular necrosis.

Antihypertensive agents—particularly vasodilators—that cause blood pressure to drop suddenly or rapidly can also induce prerenal azotemia. Patients with compromised renal autoregulation are most susceptible. When the rate of renal perfusion decreases, the renal bed autoregulates to vasodilate the glomerular afferent arteriole and vasoconstrict the glomerular efferent arteriole. This process maintains adequate renal filtration pressure despite changes in systemic pressure. This capacity can be impaired in the elderly, in those with significant arterial disease or renal artery stenosis, in the presence of volume depletion, or in those taking vasodilatory prostaglandin inhibitors (e.g. nonsteroidal anti-inflammatory drugs [NSAIDs]), cyclooxygenase (COX) inhibitors, angiotensin-converting enzyme (ACE) inhibitors or angiotensin-receptor blockers (ARBs).

Nonspecific blocking of COX, which inhibits synthesis of prostaglandin from arachidonic acid, can lead to vasoconstriction and reversible renal impairment, particularly in volume-contracted states.12 If these effects are prolonged, intrinsic acute renal failure (ARF) with ATN can ensue. Patients receiving ACE inhibitors might present with a reversible rise in serum blood urea nitrogen and creatinine, because these drugs block the efferent vasoconstriction that is necessary to maintain glomerular filtration in volume-contraction or low-perfusion states.

Vasoconstriction of both afferent and efferent arterioles induced by the immunosuppressive agent ciclosporin can decrease glomerular filtration rate in a dose-dependent and reversible manner.13, 14 Infusion of osmotic contrast dye can also cause acute vasoconstriction and present as prerenal failure with fractional excretion of sodium <1%.15 Arterial vasodilators (e.g. hydralazine, calcium-channel blockers, minoxidil, diazoxide, ACE inhibitors and ARBs), often administered in combination with diuretics, are particularly prone to induce acute changes in renal function with associated tubular sodium avidity. Hypovolemia resulting from obligatory sodium losses in patients with chronic tubulointerstitial disease can exacerbate prerenal failure. Reversal of renal failure and injury is possible if the problem is promptly recognized, the drug discontinued so that renal compensatory processes are restored, and volume infused to restore blood pressure. Changes in urine sediment are relatively benign, although unrecognized prerenal failure can lead to serious tubular injury and tubular necrosis.

Intrarenal toxicity

Medications can have direct toxic effects on cells of the renal vasculature, tubules and glomeruli, and/or induce inflammation of the renal interstitium, leading to acute intrinsic renal failure.

Vascular injury

Primary endothelial damage inducing platelet aggregation and consumption causes thrombotic microangiopathy and renal vascular injury. This effect is associated with the immunosuppressive agents ciclosporin,16, 17 tacrolimus18, 19 and muromonab-CD3,20 the antiplatelet agents ticlopidine, clopidogrel21, 22 and quinine,21 the antivirals interferon and valaciclovir,23 the chemotherapeutic agent mitomycin C (alone or in combination with cisplatin and bleomycin for gastrointestinal malignancies),24, 25 and gemcitabine. Dose-related toxicity is thought to have an important role in development of thrombotic thrombocytopenic purpura–hemolytic uremic syndrome (TTP–HUS) associated with ciclosporin, tacrolimus, muromonab-CD3 and mitomycin C, whereas an immune-mediated reaction might be more prominent in TTP–HUS induced by quinine, clopidogrel or ticlopidine.21, 26 Both immune-mediated and dose-mediated mechanisms are implicated in interferon-induced thrombotic microangiopathy.23 The ability of interferon to stimulate the expression of human leukocyte antigen (HLA)-DR on glomerular and tubular cells, possibly facilitating attack by activated lymphocytes, and to induce production of autoantibodies (e.g. antiphospholipids) indicates an immunologic role.26

Plasmapheresis has been useful in treatment of ciclosporin-induced and tacrolimus-induced HUS, with thrombotic microangiopathy and afferent arteriolar thrombosis.27, 28 Use of this treatment for renal recovery of mitomycin-C-associated HUS has had equivocal results. Nevertheless, plasmapheresis should be attempted.29, 30 HUS has also been reported in women taking estrogen-containing oral contraceptives,31 although a genetic predisposition might underlie this observation.32, 33, 34 Although still rare, HUS associated with the chemotherapeutic agent gemcitabine, a pyrimidine analog, is being reported more frequently as use of this drug becomes more prevalent in solid tumors (e.g. pancreatic, bladder and non-small-cell lung cancer).35, 36

Anticoagulants (such as warfarin and heparin) and thrombolytic agents (such as streptokinase and tissue-plasminogen activator) can cause 'showers' of arterial cholesterol plaques,37, 38 which occlude small-diameter arteries of the kidney (i.e. arcuate and interlobular arteries, terminal arterioles and glomerular capillaries).39 This causes renal tissue ischemia, necrosis and infarction, as well as inflammation of the surrounding interstitium. Anticoagulants might inhibit fibrin-clot remodeling in vessel walls or induce spontaneous hemorrhage within the clot, thereby causing it to weaken and dislodge.40 Cholesterol embolization with anticoagulation therapy is sometimes observed weeks or months after initiation of therapy. Thrombolytic agents often disrupt or dissolve protective thrombi covering ulcerated plaques, thereby releasing cholesterol plaques into the circulation.41 Embolization becomes evident within hours, days or weeks.37 Treatment is supportive, as renal failure is generally irreversible.

Tubular injury

Direct injury of tubules is commonly associated with use of antibiotics, chemotherapeutics, bisphosphonate, immunosuppressives and radiocontrast agents (Table 1). Damage can be toxic, ischemic, inflammatory or obstructive. Urinary sediment abnormalities range from essentially no cells through numerous red cells, white cells and/or brown granular casts, to proteinuria and crystalluria, depending on the site and mechanism of injury. The crucial process of tubular reabsorption depends upon highly regulated apical and basolateral transporters, and intact tight junctions for maintenance of cell polarity. Damage can perturb polarity, leading to insertion of basolateral transporters, such as Na/K-ATPase, into apical sites.42 This is associated with a 'leaky epithelium' and increased concentrations of intracellular calcium. Subsequent signaling via intracellular calcium-dependent cysteine proteases (calpains) disrupts ion homeostasis, leading to cell death.43

Table 1 Common medications associated with acute renal injury.
Table 1 - Common medications associated with acute renal injury.
Full tableFigures & Tables indexDownload PowerPoint slide (328K)

So, damage to the proximal tubule can cause a Fanconi-type abnormality of reabsorption encompassing enhanced saliuresis and kaliuresis, decreased ammonium excretion, and glucosuria, proteinuria, bicarbonaturia and phosphaturia. This abnormality is associated with increased cellular uptake of antiretroviral agents, such as cidofovir and adefovir, by the human organic anion transporter (hOAT).23 Located on the basolateral membrane, hOAT transports anionic, charged metabolites and drugs, including salicylates, urate, methotrexate and nucleoside analogs, into the proximal tubule.44 Probenecid blocks hOAT, thereby minimizing intracellular accumulation of these drugs.45 Prophylaxis with probenecid can be considered in patients receiving cidofovir and adefovir whose baseline serum creatinine is >1.5 mg/dl (132.6 mumol/l); appropriate hydration and dosage adjustment should be performed for those with clearances <55 ml/min (0.9 ml/s).23 Concurrent administration of other nephrotoxic medications with cidofovir, or use within 1 week of cessation of cidofovir therapy, is not recommended.23

Aminoglycosides also commonly target the S1/S2 segments of the proximal tubule and collecting duct. These organic bases with cationic groups are filtered freely, pinocytosed (probably via the multiligand endocytic receptor megalin located on the apical brush border of the proximal tubule),46 and translocated into the lysosomal compartment to inhibit lysosomal enzymes. They accumulate with a prolonged half-life of 100 h; by contrast, their plasma half-life is 3 h.47, 48 Exceeding a threshold level of accumulation induces an injury cascade and progression to cell necrosis. The number of cationic amino moieties seems to correlate with the degree of nephrotoxicity, which is most severe with neomycin, followed by gentamicin, tobramycin, amikacin and streptomycin.49

The urinary magnesium wasting and antidiuretic hormone (ADH) resistance that result from the action of aminoglycosides on the collecting duct lead to hypomagnesemia and nonoliguric ATN.50 Perturbation of glomerular filtration is a late manifestation of aminoglycoside nephrotoxicity. Animal studies indicate that the density and number of glomerular fenestrae are decreased, as is tubular backleak, whereas levels of mesangial platelet-activating factor increase and tubular obstruction develops. Concurrent administration of other nephrotoxins, age, obesity, female gender, hypoperfusion, underlying renal failure or liver disease, hypomagnesemia, hypokalemia and metabolic acidosis, increase the risk of toxicity. Decreasing the frequency of aminoglycoside dosing to at least daily (as dictated by renal clearance) can reduce the risk of toxicity.3 Unfortunately, nephrotoxicity can occur despite appropriate therapeutic monitoring, often within a week of therapy initiation. It is not unusual for nephrotoxicity to manifest after cessation of treatment.

Increased intracellular concentration of the acyclic nucleoside phosphonate foscarnet in the distal tubule causes distal tubular acidosis and nephrogenic diabetes insipidus.23 Similarly, cisplatin-associated tubular necrosis causes enzymuria, kaliuresis and magnesium wasting. Platinum primarily targets and concentrates in the proximal tubular S3 segment, but can also damage the distal tubule and collecting duct in a dose-dependent manner. Oxidative stress and heat-shock proteins are associated with tubular-cell deletion following cisplatin therapy.51 Animal studies of a newer platinum agent, nedaplatin, indicate that lysosomal hyperplasia, necrosis and hyperplasia of the collecting duct and renal papilla can also occur.52 Appropriate volume infusion is important to minimize nephrotoxicity.

Toxic tubular necrosis characterized by deranged Na/K-ATPase, loss of brush border and apoptosis has also been reported following treatment of patients with multiple myeloma and Padget's disease with the bisphosphonate zoledronate at recommended maximal doses of 4 mg given intravenously over 15 min.53 Renal clearance of bisphosphonates requires filtration, active transport and secretion. Extrapolating from its known effects on osteoclasts, it is possible that proximal tubular internalization of bisphosphonate facilitates its incorporation into ATP analogs, thereby inhibiting ATP-dependent pathways and cell energetics.54 Cytoskeletal architecture can also be affected, via inhibition of actin ring assembly. This process could contribute to loss of brush border.55 Vigilance and careful follow-up of renal function are necessary when bisphosphonates are used.

Osmotic nephrosis has been associated with use of intravenous immunoglobulin (IVIG) for several neurologic, rheumatologic, dermatologic and immune-associated disorders, including various glomerulonephrities.56, 57 Osmotic agents, such as mannitol,58 dextran59, 60 and hydroxyethyl starch,61 can induce the isometric vacuolization and swelling of the proximal tubule, which is characteristic of IVIG use. These processes are particularly prevalent when the stabilizing agent of IVIG is sucrose.62 Cellular uptake of nonmetabolizable compounds, such as sucrose, promotes swelling and subsequent tubular cell injury. Underlying renal disease, volume depletion, age and paraproteinemias increase the likelihood of developing these problems. Renal function needs to be monitored closely when significant doses of IVIG and osmotic agents are used. Slow infusion of IVIG over 12 h and/or lower doses (e.g. 0.4 mg/kg) should decrease the risk of renal toxic injury.

Ischemic tubular injury can result from acute vasoconstriction induced by immunosuppressives, radiocontrast agents and the antifungal amphotericin B. The calcineurin inhibitors ciclosporin and tacrolimus can stimulate dose-dependent vasoconstriction of both afferent and efferent arterioles, leading to a drop in glomerular filtration rate that can be reversed during the early stages via dose adjustment or drug discontinuation.14 A decrease in intrarenal prostaglandins, increased vascular renin activity and endothelin are thought to contribute to injury.63 Rates of urinary nitric oxide excretion are decreased. Atrophy, vacuolization, microcalcification and focal mononuclear interstitial infiltrate adversely affect the proximal tubules. All calcineurin inhibitors, including rapamycin, are associated with distal magnesium wasting and tubular collapse, vacuolization and nephrocalcinosis.3, 64

The murine monoclonal antibody muromonab-CD3, used to treat allograft rejection, is thought to induce a cytokine-mediated nephropathy characterized by capillary leak, decreased circulating volume, activated interstitial neutrophils and oxygen radical-mediated ischemic renal injury.3 The antifungal amphotericin B also decreases renal blood flow via dose-dependent acute renal vasoconstriction. At cumulative doses exceeding 2–3 g, amphotericin B causes direct distal tubular injury resulting in nonoliguric renal failure with distal tubular acidosis, concentrating defects and potassium wasting. It can cause ATN at higher doses.3 There is evidence that lipid-based formulations of amphotericin B are less likely to have renal toxic effects.65

The osmotic load of radiocontrast agents induces acute vasoconstriction, increases medullary oxygen consumption and eventually leads to tubular ischemia. For this reason, the fractional excretion of sodium (FENa) is <1% with ARF during radiocontrast-associated nephrotoxicity (RCAN).15 A persistent nephrogram 24–48 h after dye infusion might be noted on abdominal plain film. Underlying risk factors are chronic kidney disease, diabetes, prerenal state, multiple myeloma and age. Intratubular obstruction might also have a role.66 Contrast agents with decreased osmolarity could decrease the osmotic load, but will not necessarily prevent RCAN.

It has been suggested that infusion of saline alone, 0.45% normal saline plus bicarbonate or N-acetylcysteine pre-procedure decreases the likelihood of RCAN developing, particularly in high-risk patients.67, 68, 69, 70 It seems that prophylactic hemodialysis cannot prevent contrast nephropathy.71 A recent trial evaluating the role of hemofiltration before and after injection of contrast media in high-risk patients indicated that deterioration of renal function was ameliorated;72 however, these conclusions remain questionable given the numerous factors that confound the outcome measures of this study.73, 74

Fenoldopam, a selective dopaminergic-1 agonist, was initially thought to improve the renal vasodilation associated with administration of contrast material; however, a randomized trial showed no decrease in the incidence of RCAN with fenoldopam. Indeed, the drug was associated with a higher incidence of hypotension.70 As such, use of fenoldopam for prophylaxis of contrast nephropathy is not recommended. Magnetic resonance studies, when feasible, are preferable for high-risk patients, as gadolinium-associated ARF is rare.75

Other medications, including high-dose aciclovir, ganciclovir, methotrexate and indinavir, can also cause intratubular obstruction, tubular injury and ARF. Any medication associated with significant tubular injury can lead to cell apoptosis and sloughing, active urinary sediment abnormalities, intratubular obstruction and decreased urine output.

Interstitial injury

The common manifestation of medication-associated renal interstitial inflammation is predominance of lymphocytes, monocytes, eosinophils and plasma cells within the interstitium, active urinary sediment with pyuria and/or white blood cell casts, eosinophiluria, hematuria, and mild to moderate proteinuria associated with a clinical presentation of fever, rash, arthralgias and eosinophilia. Only one-third of patients present with these classic symptoms of hypersensitivity; however, renal failure occurs in most patients at relatively early stages. Medications are thought to bind to, or mimic, renal tubular antigens, or to induce an immune reaction following deposition in the interstitium. This immune reaction is probably mediated by T cells, given that these cells are present in the interstitium and that granulomas occasionally form.76 Some drugs (e.g. methicillin) induce deposition of antitubular basement membrane antibodies.

Tubulitis is common. It involves activation of inflammatory cells and release of soluble molecules that promote proinflammatory and profibrotic cytokines, which affect the renal tubules and lead to ARF. Numerous medications have been associated with AIN, most commonly penicillins, cephalosporins, phenytoin, thiazide, furosemide, cimetidine, ranitidine, rifampin, allopurinol, interferon and NSAIDs (Table 1).76 Clarithromycin77 (the newer ketolide semi-synthetic erythromycin-A derivative), telithromycin78 and the COX-2 inhibitor rofecoxib have been implicated in renal biopsy-proven AIN79 in case reports. Proton-pump inhibitors, including omeprazole and pantoprazole, are also reported to cause AIN.80 Drug discontinuation and supportive therapy, including careful volume monitoring and avoidance of hypotension and other nephrotoxins, should be the mainstays of therapy. Some retrospective and observational studies have indicated that the total number of days of renal failure is lower when aggressive disease is treated with steroids, but no differences in patient outcome were detected.76, 81 No controlled trials have justified routine use of steroids for drug-induced AIN. Given that several medications are associated with acute interstitial inflammation, it is prudent to be vigilant. Careful evaluation of renal function should be performed when initiating new medications, particularly if drug hypersensitivity is evident.

Glomerular injury

Medications that alter glomerular histology and permeability often cause proteinuria in the nephrotic range. Toxic lymphokines of interstitial inflammation might be implicated. Some medications (e.g. NSAIDs) stimulate proinflammatory leukotrienes and inhibit the COX pathway, increasing arachadonate catabolism via lipoxygenase.82 Humoral factors might also be involved, given the presence of eosinophils and lymphocytes in the interstitial infiltrate.4 Red cells and white cells might be observed in the urine, even though hypersensitivity is not clinically evident. Minimal change lesions on renal biopsy have been described secondary to NSAID agents, particularly mefenamate and fenoprofen. Membranous lesions are associated with gold, penicillamine, ACE inhibitors and foscarnet. Interestingly, interferon-alpha has been associated with several glomerular lesions, including minimal change, focal segmental hyalinosis with visceral epithelial hyperplasia, crescentic glomerulonephritis and membranous nephropathy.23 Discontinuation of medication usually reverses the clinical findings. Resolution can, however, be delayed for months or years, especially that of gold-induced nephropathy.

Renal biopsies from cancer patients treated with the bisphosphonate pamidronate for hypercalcemia reveal a collapsing focal segmental sclerosis presenting with nephrotic syndrome and widespread tubular injury.55 Given the cytostructural similarities between podocytes and osteoclasts, and the homology between bisphosphonate and certain T-cell ligands, production of interferon-gamma and podocyte injury by other cytokines are mechanisms by which the drug might damage glomeruli.55 Discontinuation of pamidronate does not resolve nephrosis; maintenance renal replacement therapy is frequently necessary.

Postrenal toxicity

Obstruction

Intratubular obstruction leading to both acute renal compromise and chronic renal failure with small atrophic kidneys is associated with high doses of aciclovir (500 mg/m2; frequently used for systemic and genital herpes infections) and ganciclovir (cytomegalovirus retinitis). Indinavir-induced crystalluria and nephrolithiasis also contribute to obstruction (Table 1).23, 83 Indinavir crystals in the cortical and medullary collecting duct can be observed in biopsy tissue from patients with ARF.84 Renal colic with flank or loin pain, or dysuria, are typical presentations of nephrolithiasis associated with indinavir or other drugs. Symptom resolution occurs after spontaneous passage or extraction of the calculi. Increasing daily fluid intake to at least 1.5 l during indinavir therapy is recommended for prevention and/or recurrence.85 Factors that predispose a patient to drug-induced urinary calculi are similar to those for metabolic etiologies of nephrolithiasis, and include history or presence of renal lithiasis, low urine volume, abnormally low or high urine pH, hypercalciuria and hypocitraturia. Medication-specific factors (e.g. high or prolonged dosing regimens, increased urinary drug excretion and poor aqueous drug solubility) also have a role.85

Extrinsic renal blockage due to retroperitoneal fibrosis and ureteral obstruction is commonly associated with methysergide (used to treat vascular headache), and also reported with hydralazine, methyldopa, pindolol, atenolol, ergotamine and dihydroergotamine.86, 87, 88, 89 Renal failure secondary to these medications should prompt evaluation with noncontrast renal imaging to rule out hydronephrosis.

Top

Chronic renal injury

Prolonged exposure of the kidney to analgesics, calcineurin inhibitors or lithium can cause chronic renal damage (Table 2). A case–control study that assessed the risk of kidney failure associated with use of NSAIDs and analgesics (e.g. acetaminophen and aspirin) found that people taking large cumulative doses of acetaminophen and NSAIDs were at increased risk of developing end-stage renal disease.90 Chronic high doses (2–3 g/day) of phenacetin, acetaminophen, aspirin and NSAIDs can cause patchy necrosis and fibrosis of the renal medullary interstitium with occasional mononuclear cell infiltration, and atrophy of Henle's loop.91, 92 Inhibition of vasodilatory prostaglandins by NSAIDs and salicylates can induce medullary ischemia. Drug metabolites tend to become concentrated within the medullary gradient; high levels at the papillary tip and generated via lipid peroxidation induce tissue damage.12, 93, 94

Table 2 Common medications associated with chronic renal injury.
Table 2 - Common medications associated with chronic renal injury.
Full tableFigures & Tables indexDownload PowerPoint slide (138K)

Chronic fibrosis leads to small kidneys in most patients who present with concentrating defects, sterile pyuria or mild proteinuria; 25–50% of patients present with papillary sloughing, which can be accompanied by hemturia and flank pain. A case of papillary necrosis caused by the COX-2 inhibitor celecoxib was recently reported. This patient had, however, a long history of using NSAID agents before switching to COX-2 inhibitors, which were used for several months before papillary necrosis became evident.95 Damage is generally irreversible and progression of renal failure occurs. Urogenital transitional carcinomas and renal cell cancers have also been linked with prolonged analgesic use.96, 97

Chronic fibrosis with an obliterative arteriolopathy and tubular collapse gives rise to a classic biopsy pattern of striped interstitial fibrosis in patients receiving long-term treatment with calcineurin inhibitors and suffering renal failure despite appropriately monitored dosing.14, 98 The distinctive biopsy pattern distinguishes this condition from transplant rejection. The exact details of its pathogenesis remain unclear, but roles for enhanced release of endothelin-1, decreased nitric oxide and increased transforming growth factor-beta have been suggested.14 Animal studies indicate that aldosterone-mediated renal injury is associated with ciclosporin. Fewer pathological changes (e.g. arteriolopathy and tubulointerstitial fibrosis) and decreased levels of transforming growth factor-beta, collagen-1 and fibronectin messenger RNA have been noted when the mineralocorticoid receptor antagonist spironolactone is used. The effect of spironolactone on ciclosporin-associated renal pathology in humans has not been determined.

Chronic use of lithium for bipolar disorder and/or depression can cause interstitial fibrosis and nephrogenic diabetes insipidus. Examination of renal biopsies reveals prevalent cortical and medullary tubular cysts or dilatations (primarily of distal and collecting tubules), both focal segmental glomerulosclerosis and global sclerosis, and chronic tubulointerstitial nephropathy. Evidence for effacement of podocyte foot processes indicates direct glomerular toxicity. In one study, a serum creatinine frequently in excess of 2.5 mg/dl (221.0 mumol/l) predicted progression to end-stage renal disease.99 Lithium also decreases sensitivity to vasopressin by dysregulating the collecting duct water channels aquaporin-2 and aquaporin-3, and epithelial sodium channel subunits. Lithium also perturbs expression of the inner medullary urea transporters UT-A1 and UT-B, resulting in decreased urinary concentration, increased sodium excretion and polyuria.100

Top

Electrolyte and acid–base abnormalities

Changes in serum potassium can be critical. Hypokalemia is associated with most diuretic agents and medications causing distal tubular toxicity, such as gentamicin, cisplatin and carboplatin. Hypomagnesemia is another common effect. Potassium-sparing diuretics can, however, cause significant hyperkalemia in volume-contracted states or when used in combination with other medications that affect renin–angiotensin–aldosterone activity (ACE inhibitors, ARBs, heparin, ciclosporin and NSAIDs, including COX-2 inhibitors [Table 3]). Hyperkalemia is exacerbated by substances that limit distal sodium delivery or block distal sodium channel activity (e.g. NSAIDs, trimethoprim and pentamidine), or decrease adrenergic effects (e.g. beta-blockers).

Table 3 Common medications associated with electrolyte/acid–base abnormalities.
Table 3 - Common medications associated with electrolyte/acid|[ndash]|base abnormalities.
Full tableFigures & Tables indexDownload PowerPoint slide (232K)

Hyponatremia can also be severe during use of thiazide diuretic agents which, while stimulating the release of both ADH and aldosterone secondary to volume loss, increase free water reabsorption with a maintained medullary gradient. Cyclophosphamide and vincristine directly affect the antidiuretic activity of distal tubular free water excretion, leading to hyponatremia.101, 102 NSAIDs potentiate the ADH effect by inhibiting the ADH-blocking action of prostaglandins. Hypernatremia can be a result of the nephrogenic diabetes insipidus induced by lithium and demeclocycline.

Proximal tubular acidosis occurs when medications affect carbonic anhydrase activity at this location. Drugs that have this effect include acetazolamide, dorzolamide, sulfanilamide, mafenide acetate and 6-mercaptopurine. Toxicity associated with aminoglycosides and cisplatin can also induce proximal tubular acidosis. Distal tubular acidosis is associated with agents that affect distal sodium/hydrogen exchange, such as ciclosporin, amphotericin B, lithium and high-dose vitamin D. Demeclocycline, like lithium, perturbs the response to ADH in the collecting tubule. This drug is therefore a useful treatment for hyponatremia related to syndrome of inappropriate ADH.

Top

Pseudo-nephrotoxicity

Competition with creatinine for tubular secretion by medications including trimethoprim and cimetidine can mimic renal failure by increasing serum creatinine. This increase is not, however, associated with any other abnormal clinical findings or urine sediment abnormalities. Steroids and tetracycline are associated with a hypercatabolic response,103 and increase blood urea nitrogen in a manner that is disproportionate to their effect on serum creatinine. Medications that interfere with laboratory determination of serum creatinine, such as ascorbic acid, cefoxitin, cephalothin, flucytosine, levodopa and methyldopa, must be considered as causes when serum creatinine increases in the absence of other clinical signs or findings (Box 1). Medication-related serum creatinine increases are generally limited and tend to stabilize. Return to baseline values usually occurs immediately following drug discontinuation.

Box 1 Drugs that cause pseudo-elevation of blood urea nitrogen and creatinine.


Competitive tubular secretion of creatinine

  • Trimethoprim
  • Cimetidine
  • Probenecid
  • Triamterene
  • Amiloride
  • Spironolactone

Interference with laboratory determination of creatinine

  • Ascorbic acid
  • Cephalosporins (cefoxitin and cephalothin)
  • Flucytosine
  • Levodopa
  • Methyldopa

Hypercatabolic effect

  • Steroids
  • Tetracycline

Top

Conclusions

Renal injury caused by medication is to be expected given the degree to which the kidneys are exposed to total blood volume. Awareness of the varied presentations of drug-associated renal toxicity is important if morbidity is to be minimized. While many medications and classes of drugs inducing nephrotoxicity are discussed, it is impossible to be all-inclusive given the number of drugs in use. The mechanisms of injury presented here, supported by specific examples of the nephrotoxicity of commonly used medications, should allow extrapolation of relevant information to newer pharmacotherapies and similar drug classes that space limitations have prevented us from covering here.

Key points

  • Renal injury caused by medication can usually be reversed if detected early

  • Drug-induced renal damage can be acute or chronic, prerenal, intrarenal (vascular, tubular, glomerular or interstitial) or postrenal

  • Different drug classes share common mechanisms of inducing renal injury (e.g. toxic, ischemic, inflammatory, obstructive or volume depletion)

  • Electrolyte/acid–base abnormalities are common effects of some medications

  • Medications that can cause kidney damage include diuretics, antihypertensives, immunosuppressants, antiplatelet agents, antivirals, chemotherapeutics, antibiotics and radiocontrast agents

Top
Acknowledgments

The authors would like to sincerely thank the Dallas Veterans Affairs Hospital Library and Medical Media Section for their prompt responses in obtaining pertinent papers and creating the tables presented in this article.

Top

References

  1. Han WK and Bonventre JV (2004) Biologic markers for the early detection of acute kidney injury. Curr Opin Crit Care 10: 476–482 | Article | PubMed |
  2. Kleinknecht D et al. (1987) Drug-associated acute renal failure: a prospective collaborative study of 81 biopsied patients. Adv Exp Med Biol 212: 125–128
  3. Kaloyandes GJ et al. (2001) Antibiotic and Immunosupression-Related Renal Failure. Philadelphia: Lippincott, Williams & Wilkins
  4. Kleinknecht D (1995) Interstitial nephritis, the nephrotic syndrome, and chronic renal failure secondary to nonsteroidal anti-inflammatory drugs. Semin Nephrol 15: 228–235
  5. Klinkhoff AV and Teufel A (1997) Reinstitution of gold after gold induced proteinuria. J Rheumatol 24: 1277–1279
  6. Wilcox CS (1999) Metabolic and adverse effects of diuretics. Semin Nephrol 19: 557–568
  7. Ellison DH (2001) Diuretic therapy and resistance in congestive heart failure. Cardiology 96: 132–143 | Article |
  8. Fliser D (1996) Loop diuretics and thiazides—the case for their combination in chronic renal failure. Nephrol Dial Transplant 11: 408–410
  9. Sandhofer A et al. (2002) Severe electrolyte disturbances and renal failure in elderly patients with combined diuretic therapy including xipamid. Wien Klin Wochenschr 114: 938–942
  10. Suki WN (1997) Use of diuretics in chronic renal failure. Kidney Int 59 (Suppl): S33–S35
  11. Cruz J et al. (2004) Successful use of the new high-dose mannitol treatment in patients with Glasgow Coma Scale scores of 3 and bilateral abnormal pupillary widening: a randomized trial. J Neurosurg 100: 376–383
  12. Ejaz P et al. (2004) NSAIDs and kidney. J Assoc Physicians India 52: 632–640
  13. Nankivell BJ et al. (2004) Oral cyclosporine but not tacrolimus reduces renal transplant blood flow. Transplantation 77: 1457–1459 | Article |
  14. Olyaei AJ et al. (2001) Nephrotoxicity of immunosuppressive drugs: new insight and preventive strategies. Curr Opin Crit Care 7: 384–389 | Article | PubMed | ChemPort |
  15. Fang LS et al. (1980) Low fractional excretion of sodium with contrast media-induced acute renal failure. Arch Intern Med 140: 531–533 | Article | PubMed | ISI | ChemPort |
  16. Van Buren D et al. (1985) De novo hemolytic uremic syndrome in renal transplant recipients immunosuppressed with cyclosporine. Surgery 98: 54–62 | PubMed | ChemPort |
  17. Bren A et al. (2005) Follow-up of kidney graft recipients with cyclosporine-associated hemolytic-uremic syndrome and thrombotic microangiopathy. Transplant Proc 37: 1889–1891 | Article |
  18. Abraham KA et al. (2000) Hemolytic–uremic syndrome in association with both cyclosporine and tacrolimus. Transpl Int 13: 443–447 | Article |
  19. Lin CC et al. (2003) Tacrolimus-associated hemolytic uremic syndrome: a case analysis. J Nephrol 16: 580–585
  20. Abramowicz D et al. (1992) Induction of thromboses within renal grafts by high-dose prophylactic OKT3. Lancet 339: 777–778 | Article | PubMed | ISI | ChemPort |
  21. Medina PJ et al. (2001) Drug-associated thrombotic thrombocytopenic purpura-hemolytic uremic syndrome. Curr Opin Hematol 8: 286–293 | Article | PubMed | ISI |
  22. Chinnakotla S et al. (2000) Clopidogrel-associated thrombotic thrombocytopenic purpura/hemolytic uremic syndrome in a kidney/pancreas transplant recipient. Transplantation 70: 550–552 | Article |
  23. Izzedine H et al. (2005) Antiviral drug-induced nephrotoxicity. Am J Kidney Dis 45: 804–817 | Article | PubMed | ChemPort |
  24. Jackson AM et al. (1984) Thrombotic microangiopathy and renal failure associated with antineoplastic chemotherapy. Ann Intern Med 101: 41–44
  25. Giroux L et al. (1985) Mitomycin-C nephrotoxicity: a clinico-pathologic study of 17 cases. Am J Kidney Dis 6: 28–39
  26. Pisoni R et al. (2001) Drug-induced thrombotic microangiopathy: incidence, prevention and management. Drug Saf 24: 491–501 | Article |
  27. Venkat KK et al. (1991) Reversal of cyclosporine-associated hemolytic–uremic syndrome by plasma exchange with fresh-frozen plasma replacement in renal transplant recipients. Transplant Proc 23: 1256–1257
  28. Trimarchi HM et al. (1999) FK506-associated thrombotic microangiopathy: report of two cases and review of the literature. Transplantation 67: 539–544 | Article | PubMed | ISI | ChemPort |
  29. Price TM et al. (1985) Renal failure and hemolytic anemia associated with mitomycin C. A case report. Cancer 55: 51–56
  30. Garibotto G et al. (1989) Successful treatment of mitomycin C-associated hemolytic uremic syndrome by plasmapheresis. Nephron 51: 409–412
  31. Ponticelli C et al. (1977) Late recovery of renal function in a woman with the hemolytic uremic syndrome. Clin Nephrol 8: 367–370
  32. Hauglustaine D et al. (1981) Oestrogen containing oral contraceptives, decreased prostacyclin production, and haemolytic uraemic syndrome. Lancet 1: 328–329 | Article |
  33. Ruggenenti P and Remuzzi G (1990) Thrombotic thrombocytopenic purpura and related disorders. Hematol Oncol Clin North Am 4: 219–241
  34. Berns JS et al. (1992) Inherited hemolytic uremic syndrome in adults. Am J Kidney Dis 19: 331–334
  35. Kintzel PE (2001) Anticancer drug-induced kidney disorders. Drug Saf 24: 19–38 | Article | PubMed | ISI | ChemPort |
  36. Walter RB et al. (2002) Gemcitabine-associated hemolytic–uremic syndrome. Am J Kidney Dis 40: E16 | Article |
  37. Hitti WA and Anderson J (2005) Cholesterol emboli-induced renal failure and gastric ulcer after thrombolytic therapy. South Med J 98: 235–237 | Article |
  38. Ben-Chitrit S et al. (1994) Cholesterol embolization syndrome following thrombolytic therapy with streptokinase and tissue plasminogen activator. Nephrol Dial Transplant 9: 428–430
  39. Eliot RS et al. (1964) Atheromatous Embolism. Circulation 30: 611–618
  40. Pennington M et al. (2002) Cholesterol embolization syndrome: cutaneous histopathological features and the variable onset of symptoms in patients with different risk factors. Br J Dermatol 146: 511–517 | Article |
  41. Fine MJ et al. (1987) Cholesterol crystal embolization: a review of 221 cases in the English literature. Angiology 38: 769–784 | Article | PubMed | ISI | ChemPort |
  42. Lameire N et al. (2005) Acute renal failure. Lancet 365: 417–430 | PubMed | ISI | ChemPort |
  43. Edelstein CL and Schrier R (2001) Pathophysiology of Ischemic Acute Renal Injury. Philadelphia: Lippincott, Williams & Wilkins
  44. Henrich WL (2005) Nephrotoxicity of several newer agents. Kidney Int 67 (Suppl): S107–S109 | Article |
  45. Ho ES et al. (2000) Cytotoxicity of antiviral nucleotides adefovir and cidofovir is induced by the expression of human renal organic anion transporter 1. J Am Soc Nephrol 11: 383–393 | PubMed | ISI | ChemPort |
  46. Nagai J and Takano M (2004) Molecular aspects of renal handling of aminoglycosides and strategies for preventing the nephrotoxicity. Drug Metab Pharmacokinet 19: 159–170 | Article | PubMed | ChemPort |
  47. Kaloyanides GJ and Pastoriza-Munoz E (1980) Aminoglycoside nephrotoxicity. Kidney Int 18: 571–582 | Article |
  48. Cronin RE and Henrich WL (1996) Toxic Nephropathy. Philadelphia: WB Saunders
  49. Maher J (1991) Toxic Nephropathies. Boston: Kluwer Academic Publishers
  50. Patel R and Savage A (1979) Symptomatic hypomagnesemia associated with gentamicin therapy. Nephron 23: 50–52 | PubMed | ChemPort |
  51. Taguchi T et al. (2005) Cisplatin-associated nephrotoxicity and pathological events. Contrib Nephrol 148: 107–121 | PubMed | ISI | ChemPort |
  52. Uehara T et al. (2005) Nephrotoxicity of a novel antineoplastic platinum complex, nedaplatin: a comparative study with cisplatin in rats. Arch Toxicol 79: 451–460 | Article |
  53. Markowitz GS et al. (2003) Toxic acute tubular necrosis following treatment with zoledronate (Zometa). Kidney Int 64: 281–289 | Article | PubMed | ISI | ChemPort |
  54. Hiroi-Furuya E et al. (1999) Etidronate (EHDP) inhibits osteoclastic-bone resorption, promotes apoptosis and disrupts actin rings in isolate-mature osteoclasts. Calcif Tissue Int 64: 219–223 | Article | PubMed |
  55. Markowitz G et al. (2001) Collapsing focal segmental glomerulosclerosis following treatment with high dose pamidronate. J Am Soc Nephrol 12: 1164–1172 | PubMed | ISI | ChemPort |
  56. Orbach H et al. (2004) Intravenous immunoglobulin and the kidney—a two-edged sword. Semin Arthritis Rheum 34: 593–601 | Article |
  57. Khalil M et al. (2000) Macrophagelike vacuolated renal tubular cells in the urine of a male with osmotic nephrosis associated with intravenous immunoglobulin therapy: a case report. Acta Cytol 44: 86–90
  58. Visweswaran P et al. (1997) Mannitol-induced acute renal failure. J Am Soc Nephrol 8: 1028–1033 | PubMed | ChemPort |
  59. Ferraboli R et al. (1997) Anuric acute renal failure caused by dextran 40 administration. Ren Fail 19: 303–306
  60. Erdbruegger U et al. (2004) Quiz page: tubular epithelial injury consistent with osmotic nephrosis. Am J Kidney Dis 44: e1 | Article |
  61. De Labarthe A et al. (2001) Acute renal failure secondary to hydroxyethylstarch administration in a surgical patient. Am J Med 111: 417–418 | Article | PubMed | ISI | ChemPort |
  62. Hansen-Schmidt S et al. (1996) Osmotic nephrosis due to high-dose immunoglobulin therapy containing sucrose (but not with glycine) in a patient with immunoglobulin A nephritis. Am J Kidney Dis 28: 451–453
  63. Gossmann J et al. (2001) Pathophysiology of cyclosporine-induced nephrotoxicity in humans: a role for nitric oxide? Kidney Blood Press Res 24: 111–115 | Article |
  64. Andoh TF et al. (1996) Comparison of acute rapamycin nephrotoxicity with cyclosporine and FK506. Kidney Int 50: 1110–1117 | Article | PubMed | ISI | ChemPort |
  65. Gibbs WJ et al. (2005) Liposomal amphotericin B: clinical experience and perspectives. Expert Rev Anti Infect Ther 3: 167–181 | Article | PubMed | ChemPort |
  66. Porter GA (1989) Contrast-associated nephropathy. Am J Cardiol 64: 22E–26E | Article |
  67. Mueller C et al. (2002) Prevention of contrast media-associated nephropathy: randomized comparison of 2 hydration regimens in 1620 patients undergoing coronary angioplasty. Arch Intern Med 162: 329–336 | Article | PubMed | ISI | ChemPort |
  68. Briguori C et al. (2005) New pharmacological protocols to prevent or reduce contrast media nephropathy. Minerva Cardioangiol 53: 49–58
  69. Tepel M et al. (2000) Prevention of radiographic-contrast-agent-induced reductions in renal function by acetylcysteine. N Engl J Med 343: 180–184 | Article | PubMed | ISI | ChemPort |
  70. Stone GW et al. (2003) Fenoldopam mesylate for the prevention of contrast-induced nephropathy: a randomized controlled trial. JAMA 290: 2284–2291 | Article | PubMed | ISI | ChemPort |
  71. Vogt B et al. (2001) Prophylactic hemodialysis after radiocontrast media in patients with renal insufficiency is potentially harmful. Am J Med 111: 692–698 | Article | PubMed | ISI | ChemPort |
  72. Marenzi G et al. (2003) The prevention of radiocontrast-agent-induced nephropathy by hemofiltration. N Engl J Med 349: 1333–1340 | Article | PubMed | ISI | ChemPort |
  73. Kancha K et al. (2004) Hemofiltration and the prevention of radiocontrast-agent-induced nephropathy. N Engl J Med 350: 836–838; author reply 836–838 | Article |
  74. Forman JP (2004) Hemofiltration and the prevention of radiocontrast-agent-induced nephropathy. N Engl J Med 350: 836–838; author reply 836–838 | Article |
  75. Sam AD II et al. (2003) Safety of gadolinium contrast angiography in patients with chronic renal insufficiency. J Vasc Surg 38: 313–318 | Article | PubMed | ISI |
  76. Rossert J (2001) Drug-induced acute interstitial nephritis. Kidney Int 60: 804–817 | Article | PubMed | ISI | ChemPort |
  77. Baylor P and Williams K (1999) Interstitial nephritis, thrombocytopenia, hepatitis, and elevated serum amylase levels in a patient receiving clarithromycin therapy. Clin Infect Dis 29: 1350–1351 | Article |
  78. Tintillier M et al. (2004) Telithromycin-induced acute interstitial nephritis: a first case report. Am J Kidney Dis 44: e25–e27 | Article |
  79. Szalat A et al. (2004) Acute renal failure and interstitial nephritis in a patient treated with rofecoxib: case report and review of the literature. Arthritis Rheum 51: 670–673 | Article |
  80. Ra A and Tobe SW (2004) Acute interstitial nephritis due to pantoprazole. Ann Pharmacother 38: 41–45
  81. Clarkson MR et al. (2004) Acute interstitial nephritis: clinical features and response to corticosteroid therapy. Nephrol Dial Transplant 19: 2778–2783 | Article | PubMed | ISI | ChemPort |
  82. Clive DM and Stoff JS (1984) Renal syndromes associated with nonsteroidal antiinflammatory drugs. N Engl J Med 310: 563–572 | PubMed | ISI | ChemPort |
  83. Hanabusa H et al. (1999) Renal atrophy associated with long-term treatment with indinavir. N Engl J Med 340: 392–393 | Article |
  84. Martinez F et al. (1998) Indinavir crystal deposits associated with tubulointerstitial nephropathy. Nephrol Dial Transplant 13: 750–753 | PubMed | ISI | ChemPort |
  85. Daudon M and Jungers P (2004) Drug-induced renal calculi: epidemiology, prevention and management. Drugs 64: 245–275 | PubMed |
  86. Demko TM et al. (1997) Obstructive nephropathy as a result of retroperitoneal fibrosis: a review of its pathogenesis and associations. J Am Soc Nephrol 8: 684–688 | PubMed | ChemPort |
  87. Waters VV (1989) Hydralazine, hydrochlorothiazide and ampicillin associated with retroperitoneal fibrosis: case report. J Urol 141: 936–937
  88. Iversen BM et al. (1975) Retroperitoneal fibrosis during treatment with methydopa. Lancet 2: 302–304 | Article |
  89. Damstrup L and Jensen TT (1986) Retroperitoneal fibrosis after long-term daily use of ergotamine. Int Urol Nephrol 18: 299–301 | Article |
  90. Perneger TV et al. (1994) Risk of kidney failure associated with the use of acetaminophen, aspirin, and nonsteroidal antiinflammatory drugs. N Engl J Med 331: 1675–1679 | Article | PubMed | ISI | ChemPort |
  91. Nanra RS et al. (1978) Analgesic nephropathy: etiology, clinical syndrome, and clinicopathologic correlations in Australia. Kidney Int 13: 79–92 | Article | PubMed | ISI | ChemPort |
  92. Gloor FJ (1978) Changing concepts in pathogenesis and morphology of analgesic nephropathy as seen in Europe. Kidney Int 13: 27–33 | Article |
  93. Kincaid-Smith P (1980) Analgesic abuse and the kidney. Kidney Int 17: 250–260 | Article |
  94. Bach PH and Hardy TL (1985) Relevance of animal models to analgesic-associated renal papillary necrosis in humans. Kidney Int 28: 605–613 | Article | PubMed | ISI | ChemPort |
  95. Akhund L et al. (2003) Celecoxib-related renal papillary necrosis. Arch Intern Med 163: 114–115 | Article |
  96. Blohme I and Johansson S (1981) Renal pelvic neoplasms and atypical urothelium in patients with end-stage analgesic nephropathy. Kidney Int 20: 671–675 | Article |
  97. McCredie M et al. (19