Abstract
Medications are crucial to the appropriate care of patients. However, acute and chronic kidney injury remains an unfortunate and relatively frequent adverse consequence of drug therapy. Although most medications are generally well tolerated, a subgroup of individuals may develop nephrotoxicity due to innate risk factors that predispose to drug-induced renal toxicity and intrinsic drug nephrotoxicity. Nephrotoxic drugs may be diagnostic agents, therapeutic medications, alternative or complementary substances, or drugs of abuse. The various renal syndromes caused by therapeutic agents include acute and chronic kidney disease, tubular dysfunction, proteinuria/nephrotic syndrome, crystalluria/nephrolithiasis, and hypertension.
Keywords
medications, nephrotoxicity, acute kidney injury, chronic kidney disease, proteinuria, tubulopathy, nephrotic syndrome, hypertension, crystalline nephropathy, osmotic nephropathy
Medications are a mainstay of appropriate patient care, and new agents are being introduced into clinical practice at a rapid pace. Although most drugs are well tolerated, and therapeutic agents are often essential for medical care, kidney injury remains an unfortunate and relatively frequent adverse consequence. This bespeaks the fact that some individuals possess risk factors that predispose to drug-induced kidney toxicity. Not unexpectedly, the general population is regularly exposed to various diagnostic and therapeutic agents with nephrotoxic potential. Although most are prescribed, many other preparations are purchased over the counter. Drugs fall into the categories of diagnostic agents, therapeutic medications, alternative or complementary substances, and drugs of abuse, resulting in a variety of kidney syndromes ( Table 35.1 ).
| Kidney Syndrome | Causative Agents |
|---|---|
| Acute Kidney Injury | |
| Prerenal | Cyclosporine, tacrolimus, radiocontrast, Am B, ACE inhibitors, ARBs, NSAIDs, interleukin-2, exenatide |
| Intrarenal | |
| Vascular disease | Gemcitabine, anti-VEGF drugs, propylthiouracil, interferon |
| ATN | AGs, AmB, cisplatin, tenofovir, ifosfamide, pemetrexed, polymyxins, vancomycin, pentostat, zoledronate, warfarin |
| AIN | Immune checkpoint inhibitors, penicillins, cephalosporins, sulfonamides, rifampin, NSAIDs, interferon, ciprofloxacin, others |
| Crystal nephropathy | Methotrexate, acyclovir, sulfonamides, indinavir, atazanavir, ciprofloxacin, sodium phosphate |
| Osmotic nephropathy | IVIG, HES, dextran, mannitol |
| Postrenal | Methysergide, drug-induced stones, alpha-agonists |
| Proteinuria | Gold, NSAIDs, anti-VEGF drugs, penicillamine, interferon, pamidronate |
| Tubulopathies | AGs, tenofovir, cisplatin, ifosfamide, AmB, pemetrexed, cetuximab |
| Nephrolithiasis | Sulfadiazine, atazanavir, indinavir, topiramate, zonisamide |
| CKD | Li + , analgesic abuse, cyclosporine, tacrolimus, cisplatin, nitrosourea |
Kidney Susceptibility to Nephrotoxic Agents
In addition to clearance of endogenous waste products, excretion of sodium and water, electrolyte and acid-base balance, and endocrine activity, the kidney is responsible for the metabolism and excretion of exogenously administered drugs, making it susceptible to various types of injury. There are several factors that increase the kidney’s susceptibility to these potential toxins, which can be classified into three simple categories (drug-related factors, kidney-related factors, and host-related factors) and often occur in combination to promote nephrotoxicity. As we learn more about drug-induced kidney disease, it appears that these factors explain much of the variability and heterogeneity noted among patients.
Drug-related factors are the critical first step to the development of nephrotoxicity. Innate drug toxicity is important because the drug or its toxic metabolite may cause kidney injury by impairing renal hemodynamics, direct cellular injury, osmotic injury, or intratubular crystal deposition, to name a few conditions. Large doses, extended drug exposure, and nephrotoxic drug combinations further enhance nephrotoxicity.
The kidney’s handling of drugs also determines why certain agents cause nephrotoxicity. As renal blood flow approximates 25% of cardiac output, the kidney is significantly exposed to nephrotoxic drugs. Kidney injury is increased in the loop of Henle where high metabolic rates coexist with a relatively hypoxic environment. Increased drug/metabolite concentrations in the kidney medulla also contribute to direct toxicity. Kidney drug metabolism from cytochrome P450 (CYP450) and other enzymes increases local toxic metabolite and reactive oxygen species (ROS) formation, which promote injury via nucleic acid oxidation/alkylation, DNA-strand breaks, lipid peroxidation, and protein damage.
The kidney pathway of excretion for many therapeutic agents involves proximal tubular cells. Extensive drug trafficking through the cell via luminal and basolateral transporters can lead to cellular injury. Some drugs are endocytosed at the luminal membrane of cells, whereas other drugs are transported into the cell via basolateral ion transporters. Such drug transport can be associated with increased cellular concentrations that injure mitochondria, phospholipid membranes, lysosomes, and other organelles.
Nonmodifiable factors such as older age and female sex increase nephrotoxic risk through reduced total body water leading to drug overdose. Unrecognized reduced glomerular filtration rate (GFR) and hypoalbuminemia, which result in increased toxic drug concentration, also enhance risk. Pharmacogenetic differences likely explain much of the variable response of patients to drugs. Liver and kidney CYP450 enzyme gene polymorphisms are associated with reduced metabolism and end-organ toxicity. Polymorphisms of genes encoding proteins involved in the metabolism and kidney elimination of drugs are correlated with nephrotoxic risk. Another important aspect of genetic makeup is a highly variable host immune response to drugs; one patient reacts with a heightened allergic response, whereas another has a limited reaction with no kidney lesion. Thus, innate host response genes tend to determine the drug reaction.
Kidney susceptibility to drug injury is also enhanced by true and effective volume depletion, including nausea/vomiting, diarrhea, and diuretic therapy, as well as heart failure, liver disease with ascites, and sepsis. This physiology enhances the nephrotoxicity of drugs that are excreted primarily by the kidney, drugs reabsorbed/secreted by the proximal tubule, and drugs that are insoluble in the urine. Nephrotoxic risk is also increased in patients with acute kidney injury (AKI) or chronic kidney disease (CKD) because of a lower number of functioning nephrons, reductions in drug clearance, and a robust kidney oxidative response to drugs and metabolites. Finally, electrolyte and acid-base disturbances present in some patients also contribute to host susceptibility to drug injury.
Kidney Injury Associated With Medications
Therapeutic agents associated with kidney injury can be classified based on the category of the agent or the clinical kidney syndrome. Recognizing that all drugs cannot be covered in this chapter, we describe drug-induced nephrotoxicity by drug category and highlight the clinical kidney syndrome and the segment of nephron injury by the drug within each category. Drug-induced acute interstitial nephritis (AIN) and CKD are discussed elsewhere in the Primer .
Diagnostic Agents
Radiocontrast Agents
Contrast-induced nephropathy (CIN) is the third most common cause of hospital-acquired AKI and is associated with increased risk of hospital morbidity/mortality and long-term adverse outcomes. It is defined by an absolute or percentage rise in serum creatinine from the baseline within 48 to 72 hours. In general, serum creatinine begins to rise within the first 24 hours after exposure, peaks between 2 and 5 days, and returns to baseline by 7 to 14 days. The course of CIN varies depending on the overall patient risk profile.
The incidence of CIN depends on the definition used and the population studied, ranging from 5% to 40%. Two important factors drive this incidence: (1) the increased number of imaging studies and percutaneous procedures with radiocontrast throughout the past decade and (2) the ever-enlarging population of patients with underlying CKD. In the presence of reduced kidney function, the elimination (T 1/2 ) of radiocontrast agents is increased. Thus, the kidney undergoes prolonged contrast exposure that increases the likelihood of kidney injury. In CKD stage 3 or greater, the risk of CIN is twofold to fivefold higher compared with normal kidney function, and the risk escalates as the GFR falls.
Radiocontrast media injure the kidney via multiple mechanisms. First, vasoactive substances, such as adenosine and endothelin, mediate vasoconstriction of the afferent arterioles, thereby reducing renal blood flow and promoting kidney medullary ischemia. Second, renal epithelial cell necrosis also occurs with isoosmolar radiocontrast agents because their high viscosity causes sluggish blood flow through the peritubular capillaries and promotes hypoxic kidney injury. Lastly, radiocontrast causes direct renal tubular toxicity through hyperosmolar injury, which results in vacuolization of proximal tubular cells, and oxidative stress from free oxygen radicals with associated tubular cell apoptosis and necrosis.
The level of kidney function at the time of exposure is one of the most important determinants of the risk for CIN. In addition, patient-specific risk factors include older age, volume depletion, congestive heart failure, diabetes mellitus, both hypertension and hypotension, and anemia. The intraaortic balloon pump is associated with increased AKI risk, primarily because it is a surrogate for severe cardiac disease, tenuous cardiac output, and kidney hypoperfusion. Emergent procedures increase risk because of reduced use of contrast prophylaxis and increased severity of patient illness. The type, volume, and route of contrast administration also affects CIN risk. With regard to radiocontrast type, osmolality and viscosity are the two most important characteristics. The osmolality of a solution varies significantly from high-osmolar contrast media (HOCM) to low-osmolar media (LOCM) to isoosmolar media (IOCM). Viscosity, another contrast property, varies from one product to the next, does not correlate with osmolality, and may be associated with CIN. For example, IOCM solutions are about twice as viscous as LOCM products despite having a lower osmolality.
The incidence of CIN is higher with HOCM than with LOCM, and in CKD patients, the relative risk is doubled. As a result, LOCM and IOCM agents have replaced HOCM. A meta-analysis of 16 randomized controlled trials suggested a benefit of using IOCM instead of LOCM, with the relative risk reduction of CIN greatest in CKD patients. The maximum increase in serum creatinine was less in CKD patients given IOCM compared with LOCM. However, a randomized trial comparing IOCM with LOCM noted no significant difference in CIN incidence. Thus, the benefits of low osmolality may be counterbalanced by the detrimental properties of high viscosity, making these agents equal in their risk for CIN. A larger volume of contrast increases CIN, with a recommended upper limit of 150 mL for patients with a serum creatinine 1.5 to 3.4 mg/dL and maximum dose of 100 mL recommended for patients with a creatinine greater than 3.4 mg/dL. The smallest contrast volume required to perform the procedure should be used. Risk of CIN is highest with intraarterial injection, with the intravenous (IV) route presenting a smaller risk. Coronary angiography has an even higher CIN risk than other arterial studies. CKD outpatients have a low CIN risk with nonemergent computed tomography (CT) scans. In fact, an eGFR greater than 30 mL/min per 1.73 m 2 is not considered a substantial risk for CIN in patients receiving radiocontrast by the IV route.
As radiocontrast exposure is often predictable, measures to reduce kidney injury should be undertaken in patients at risk. In addition to limiting the contrast load and using either IOCM or nonionic LOCM, the most important intervention is IV fluid (IVF) administration. Studies have uniformly demonstrated the benefit of prophylactic isotonic IVF administered both before and after radiocontrast administration. Because urinary alkalinization is hypothesized to reduce kidney oxidative stress, IV sodium bicarbonate has been studied. In an early report, CIN developed in only 2% of CKD patients treated with bicarbonate solution as compared with 14% with IV saline. A later meta-analysis of 23 studies concluded that bicarbonate-containing solutions reduced the risk of RCIN by 38%, but the benefits were noted in small and poor-quality studies. In contrast, larger, higher-quality studies did not demonstrate a reduction in CIN. Furthermore, no benefit was demonstrated for AKI-requiring dialysis, for heart failure, or for total mortality. Thus, sodium bicarbonate is not superior to isotonic saline, and either solution is acceptable for radiocontrast prophylaxis. For outpatient studies, oral fluids with salt tablets before exposure may provide adequate volume expansion to prevent CIN in CKD, but this approach has not been extensively examined.
N-acetylcysteine (NAC) is an antioxidant that is commonly used for CIN prevention. Approximately half of the published randomized controlled trials demonstrate benefit, whereas several meta-analyses suggest either large benefit or no benefit. Beneficial studies are notable for early publication dates, small size, and low quality. Despite the enrollment of nearly 3000 patients, including CKD patients, no beneficial effect of NAC on hard clinical outcomes is noted. Given its favorable safety profile, low cost, easy administration, and wide availability, clinicians could argue for continued use of the drug as prophylaxis. However, the Acetylcysteine for Contrast-Induced Nephropathy Trial casts doubt on this conclusion. This randomized study included 2308 patients and documented no benefit with NAC therapy (the proportion developing CIN was 12.7 among both NAC and placebo recipients). Thus, NAC appears to offer no protection against CIN. Despite a lack of data, it is reasonable to avoid nonsteroidal antiinflammatory drugs (NSAIDs), calcineurin inhibitors, aminoglycosides (AGs), and osmotic agents before radiocontrast exposure. Regarding renin-angiotensin-aldosterone system (RAAS) blockers, some studies note increased CIN risk, whereas others show nephroprotection.
Based on its size, lack of protein binding, and small volume of distribution, radiocontrast is efficiently removed with hemodialysis (HD). In fact, approximately 80% is removed over 4 hours with a high-flux dialyzer. HD after radiocontrast exposure to prevent CIN, especially in patients with advanced CKD, has been examined in several studies. Although all HD studies have been negative, one small study demonstrated that prophylactic HD in stage 5 CKD patients reduced the need for an acute and chronic dialysis requirement after discharge. Hemofiltration performed 4 to 6 hours before and 18 to 24 hours after contrast reduced the incidence of CIN, in-hospital events, need for acute dialysis, and both in-hospital and 1-year mortality. In contrast, the hemofiltration post-procedure alone offered no benefit beyond standard prophylaxis. A systematic review of 11 studies with 1010 patients concluded that one or more sessions of HD, hemofiltration, or hemodiafiltration performed after contrast administration did not reduce the incidence of CIN nor the need for acute or chronic dialysis. Examination of HD and hemofiltration/hemodiafiltration separately shows that HD is associated with increased CIN risk, whereas hemofiltration/hemodiafiltration did not affect the occurrence of CIN but did reduce the receipt of acute dialysis. Therefore HD and hemodiafiltration are not recommended as a prophylactic measure for CIN.
Gadolinium-Based Contrast Agents
Gadolinium-based contrast agents (GBCAs) were considered a safe and effective diagnostic agent, revolutionizing the world of imaging. However, over time, it became clear that GBCAs were not risk free. Rare reports of AKI surfaced, primarily in patients with underlying kidney disease who received large doses via direct arterial injection. Nephrotoxicity may be related to a direct effect on tubules, mediated by osmolarity or some other mechanism. In general, GBCA-induced AKI is rare and typically of minor severity, likely caused by the small volume of contrast required for imaging.
GBCAs began to be used widely for imaging patients with kidney disease in the early to mid-1990s because they offered an outstanding image without the nephrotoxicity of radiocontrast. However, nephrogenic systemic fibrosis (NSF), a severe and largely irreversible sclerosing condition of skin, joints, eyes, and internal organs, was first noted as a complication of GBCAs in 2006. Two factors were required for NSF to develop: GBCA exposure and underlying kidney disease. Other factors that likely further increased the risk for NSF included infection, inflammation, vascular disease, hypercoagulability, hypercalcemia, hyperphosphatemia, erythropoiesis-stimulating agent (ESA), and iron therapy.
The best approach to NSF is prevention because therapeutic interventions are at best suboptimal. The high-risk patient should be identified before GBCA exposure, allowing other imaging options to be explored. Such options include non-GBCA MR imaging, CT scan, ultrasonography, and other techniques that can often provide diagnostic results. When a GBCA is necessary to make the diagnosis, the following approach seems reasonable: (1) use a macrocyclic GBCA; (2) use the lowest dose required to obtain a diagnostic image; (3) optimize metabolic parameters and restrict ESA and iron use immediately before and after GBCA exposure; (4) wait for kidney recovery or stabilization in AKI; and (5) consider performing HD within hours of GBCA exposure in patients already receiving dialysis. The incidence of NSF has essentially disappeared with the implementation of prudent GBCA use in high-risk patients. However, when this disease develops, its consequences are often devastating, and therapeutic options are limited. Although a number of agents have been used, it appears that pain control and physical therapy are most important. Therapies such as extracorporeal photopheresis, sodium thiosulfate, and imatinib show promise; however, only early kidney transplant may offer stabilization or reversal of the fibrosing process.
Oral Sodium Phosphate Preparation
Sodium phosphate preparations are used as purgatives for bowel cleansing before diagnostic colonoscopy and CT virtual colonoscopy. They are administered as a solution or tablets before the procedure and contain approximately 38 g of monobasic sodium phosphate and 9 g of dibasic sodium phosphate.
The adverse events associated with phosphate-containing bowel preparations occur with excessive dosing or use in patients with underlying kidney disease. Hypocalcemia and hyperphosphatemia may complicate therapy, but AKI is of greater concern. The pathogenic mechanism was described in 21 patients with AKI after a phosphate-containing bowel cleansing agent was used for colonoscopy. Patients were predominantly women, had hypertension, and were on RAAS blockade. AKI was recognized at a median of 3 months after colonoscopy. Minimal proteinuria and bland urine sediment were noted. Tubular injury and atrophy with abundant calcium phosphate deposits in distal tubules and collecting ducts were features on kidney biopsy. This entity is termed acute phosphate nephropathy .
Two patterns of kidney injury occur with sodium phosphate administration. First, AKI develops within days of exposure and is associated with hyperphosphatemia and hypocalcemia. A second pattern is seen when AKI is discovered incidentally in patients evaluated weeks or months after exposure. Unfortunately, acute phosphate nephropathy is frequently complicated by CKD. Thus, oral sodium phosphate-based products should not be used in patients with underlying kidney disease, volume depletion, or electrolyte abnormalities. In 2008, following an alert regarding acute phosphate nephropathy issued by the US Food and Drug Administration (FDA), over-the-counter oral phosphate-containing preparations were voluntarily withdrawn from the US market.
Recently, the possibility of CKD developing in patients receiving phosphate-containing enemas was raised. However, because there was no definitive diagnosis (i.e., kidney biopsy) to verify a relationship between the enemas and phosphate-related kidney injury, the association needs further study. Clinicians should remain vigilant for this possibility.
Therapeutic Agents
Analgesics
NSAIDs, including selective cyclooxygenase-2 (COX-2) inhibitors, are widely used to treat pain, fever, and inflammation. More than 20 NSAIDs from seven major classes are approved in the United States, and many are available over the counter. Annually, more than 50 million patients ingest these drugs on an intermittent basis, whereas 15 to 25 million people in the United States use NSAIDs daily.
NSAIDs and selective COX-2 inhibitors are associated with various clinical kidney syndromes ( Box 35.1 ). It has been estimated that 1% to 5% of patients who ingest NSAIDs develop some form of nephrotoxicity, perhaps representing as many as 500,000 persons in the United States alone. These adverse effects are caused primarily by prostaglandin (PG) inhibition; however, other effects are idiosyncratic. PGs are produced by COX enzyme metabolism and are secreted locally in the kidney to modulate the effects of various systemic and local substances. For example, PGs enhance afferent arteriolar vasodilatation in the presence of vasoconstrictors such as angiotensin-II, norepinephrine, vasopressin, and endothelin, thereby providing critical counterbalance to the vasoconstriction that predominates in hypovolemic states. Patients with decreased true or effective circulating volume are at highest risk to develop renal vasoconstriction and reduced GFR. Because CKD is a PG-dependent state, these patients are also at higher risk for NSAID-induced kidney injury. In fact, exposure to NSAIDs doubles the risk of hospitalization for AKI in patients with CKD. Similar rates of AKI with NSAID exposure are noted in the elderly, those with cardiac disease, and patients receiving angiotensin-converting enzyme (ACE) inhibitors. As noted in a nested case-control study, the adjusted relative risk of AKI was 4.1 and 3.2 in current NSAID users versus nonusers in the general population, respectively. Patients with hypertension, heart failure, and diuretic therapy had an adjusted relative risk of 11.6 with NSAIDs.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
