- clinical nephrology
- contrast
- dextran
- hydroxyethyl starch
- mannitol
- osmotic nephropathy
- renal tubular epithelial cells
- SGLT2 inhibitors
- sucrose
Medications are a common cause of kidney disease and can induce injury through a variety of mechanisms (1). The tubulointerstitium is the most common compartment involved in drug-induced kidney disease (DIKD). Drug-induced toxic acute tubular injury is most common, whereas acute tubulointerstitial nephritis from idiosyncratic drug reactions and crystalline nephropathy from precipitation of medications within the tubules also cause DIKD (1). A less common but nonetheless notable cause of DIKD is osmotic nephropathy (1,2). A number of medications (Table 1) have been described to cause this lesion, including infusion of “hyperosmotic” medications such as carbohydrate-stabilized intravenous immune globulin (IVIg), hydroxyethyl starch, dextran, mannitol, and contrast medium (1⇓–3). This lesion has also been described in rats exposed to intravenous glucose where significant amounts of glucose are noted in the urine (4). The recent description of five cases of biopsy-proven osmotic nephropathy with SGLT2 inhibitors fits this experimental observation and will be discussed along with other medications (5⇓–7).
Medications associated with osmotic nephropathy
Osmotic nephropathy was first described in 1906 by Lamy et al. (8), and it was subsequently rediscovered in the 1930s when sucrose infusions were used to combat cerebral and generalized edema (9). It is a histopathologic diagnosis characterized as swollen renal tubular epithelial cells filled with cytoplasmic vacuoles (Figure 1A) (2,10). In contrast to vacuolization of cells due to ischemic injury, osmotic nephropathy cells do not have apical blebbing or nuclear dropout. Electron microscopy demonstrates cytoplasmic vacuoles are lysosomes that can be small and appear empty or large with amorphous electron-dense material (2,10). Osmotic nephropathy occurs in the setting of intravenous administration of an offending agent that undergoes glomerular filtration followed by pinocytosis when the substance encounters the apical membrane of proximal tubular epithelial cells (1,2). After pinocytosis, the offending agents found in the pinocytotic vacuoles fuse with each other and the lysosomes where they accumulate and become engorged and distended (Figure 1B). Importantly, the vacuoles and cellular swelling do not occur due to an “osmotic” effect of the intracellular drugs drawing water into the tubular cells, making the term a misnomer that has persisted (1,2). Rather, it is the pinocytosis and lysosomal accumulation of drug that ultimately promotes intracytoplasmic vacuolization and proximal tubular cell swelling (1,2). Not uncommonly, swollen tubules are seen next to normal-appearing tubules. The severity of vacuolization and cell swelling is dose dependent. For example, depending on the severity of drug exposure, vacuolization of the tubular cells may be mild and focal with minimal swelling or produce a diffuse “clear cell” appearance with marked swelling and basal displacement of nuclei (1,2,10). The straight segment (S2 and S3) of the proximal tubule is primarily involved, although the convoluted segment cells may contain vacuoles in severe cases, whereas the distal and collecting tubules are spared (2).
Osmotic nephropathy. (A) Light microscopy focusing on the renal tubules demonstrates vacuolization and swelling of proximal tubular cells in a patient exposed to hydroxyethyl starch. This finding is characteristic of osmotic nephropathy. (B) Apical membrane handling of various agents by proximal tubular cells increases cellular uptake of this potentially nephrotoxic drug. Drugs such as IVIg (stabilizers sucrose>glucose>maltose), hydroxyethyl starch, dextran, mannitol, and contrast are filtered at the glomerulus, undergo apical cell membrane pinocytosis, and enter the cell where they are translocated into lysosomes. These substances accumulate within lysosomes, which promote cellular swelling, occlusion of tubular lumens, and eventual lysosomal rupture, resulting in tubular cell injury. IVIg, intravenous immune globulin.
Risk factors for development of osmotic nephropathy and kidney injury are related to both the drug and health of the kidney. The dose of the offending drug is critical because larger doses and a longer duration of exposure increases the risk for this lesion (1,2,10). Other factors contribute to the severity and duration of tubular cell vacuolization and swelling. The digestibility of the drug importantly influences vacuolar retention; poorly digestible substances will remain longer within lysosomes and accumulate within the cells (1,2,10). Underlying CKD or acute tubular injury (ischemic or toxic) also affects the severity of osmotic nephropathy (and kidney dysfunction) by impairing lysosomal digestion (1,2,10). In fact, persistent lysosomal alterations are an early sign of cell damage and may be associated with irreversible cell damage. Other risk factors, which likely reflect reduced kidney function, include diabetes mellitus and older age.
The histologic presence of osmotic nephropathy does not always equate to proximal tubular dysfunction and may in fact occur in the absence of clinically overt AKI (1,2,10). However, severe and diffuse tubular involvement is often associated with urinalysis abnormalities (tubular proteinuria, swollen tubular cells with numerous vacuoles alone and within casts) and/or AKI (2,10). AKI and oliguria that develops from osmotic nephropathy are primarily due to tubular cell dysfunction with perhaps only a small contribution from tubular obstruction occurring from massively swollen cells, which occlude tubular lumens (2,10). Because both the duration of drug exposure and patient- and drug-related risk factors for osmotic nephropathy are known, preventive measures can be instituted (1,2). These include primarily using alternative agents, reducing the dose and the duration of exposure, minimizing reduced kidney perfusion from volume depletion, and avoiding concurrent nephrotoxin exposure. In fact, the removal of sucrose as a stabilizer from most IVIg preparations has largely eliminated osmotic nephropathy as a complication. When AKI does develop, it usually resolves rather quickly on its own. However, some patients, especially those with the aforementioned risk factors, may require kidney replacement therapy. Most patients will recover kidney function with discontinuation of the offending agent.
Medications Most Commonly Associated with Osmotic Nephropathy
Osmotic nephropathy is generally associated with the infusion of hyperosmotic or hyperoncotic agents. Some of the most common culprits are as follows. First, IVIg preparations are the most frequently described agents to cause osmotic nephropathy (2⇓–4,10). The different IVIg preparations contain different stabilizing substances (sucrose, glucose, maltose, sorbitol, glycine, and albumin) to reduce immunoglobulin aggregation, but they may precipitate osmotic nephropathy. The vast majority of cases are from IVIg containing sucrose, although rare cases have been described in IVIg containing other stabilizers (2). Second, radiocontrast-associated AKI may also cause osmotic nephropathy despite it being mentioned much less frequently as a mechanism of contrast-associated kidney injury compared with ischemia, oxidative stress, and direct toxicity (2,11,12). Yet, kidney biopsy specimens from patients exposed to contrast frequently show classic osmotic nephropathy. The significance of these lesions in causing AKI is unclear (2,12). Third, intravenous mannitol has also been implicated in osmotic nephropathy. Its infusion into rabbits, cats, dogs, and humans has been described a dose-related osmotic nephropathy lesion (13). Finally, two volume expanders that were widely utilized in the past—low-molecular dextran and hydroxyethyl starch (HES)—have also been associated with dose-related osmotic nephropathy in animals and humans (3,14⇓–16). The small dextran polymers found in the dextrans, and the starch fragments found in the large molecular weight/high substitution ratio HES preparations, are freely filtered and undergo pinocytosis by proximal tubular cells, resulting in classic osmotic lesions. As with the other hyperosmolar agents, underlying risk factors increase the likelihood of the development of AKI.
SGLT2 Inhibitors
SGLT2 inhibitors have revolutionized the treatment of CKD in both patients with diabetes and those with other forms of CKD (17). The addition of this class of medications to state-of-the-art care has significantly reduced the progression of CKD. Although these drugs have multiple beneficial effects, there are a few case reports of AKI and numerous reports in the Food and Drug Administration’s Adverse Events Reporting System of AKI associated with these drugs (18). Although most of these cases are likely due to decreased glomerular capillary pressure, some may be due to the development of osmotic nephropathy. Case reports of biopsy-proven osmotic nephropathy with SGLT2 inhibitors have been published (5⇓–7). The five patients in these reports ranged in age from 41 to 66 years, all had diabetes mellitus and hypertension, and all had mild to moderately advanced CKD with eGFRs (ml/min per 1.73 m2) of 79, 59, 56–60, 36, and 27–32. Three patients were on dapagliflozin, one was on canagliflozin, and one was on empagliflozin (overdose of 50 mg). Serum creatinine on presentation ranged from 1.1 to 10.7 mg/dl (mean 4.6 mg/dl), whereas significant urine glucose was noted in all patients. Kidney biopsy in all patients revealed findings consistent with osmotic nephropathy with vacuolization and cell swelling limited to the proximal tubules. With supportive care and discontinuation of SGLT2 inhibitors, all patient recovered kidney function back to baseline except for one patient who was left with higher-stage CKD. Despite the novelty of these findings, it is humbling to note that in 1935, Homer Smith (19) described these lesions in dogs treated concomitantly with phlorizin and either sucrose or an equivalent amount of glucose (100cc of 50%). Not surprisingly, the sucrose caused a more severe lesion.
Why would this lesion develop with these drugs, which are clearly different from the hyperosmotic agents previously described? A possible explanation lies with experimental and clinical settings where the proximal tubules are exposed to significant amounts of filtered glucose. Glucose infusions have been associated with osmotic nephropathy in experimental studies and in humans exposed to 10% glucose solutions (2⇓–4). Autopsy specimens of kidneys from diabetic patients suffering from severe hyperglycemia and DKA revealed proximal tubular cells packed with intracytoplasmic vacuoles: the Armanni–Ebstein lesion (6,20). The genesis of this lesion is not definitively known, but it is thought to result from severe hyperglycemia causing excessive amounts of urinary glucose, which then undergoes tubular cell pinocytosis. The large volume of pinocytosed glucose overwhelms the cell’s ability to metabolize the glucose load. Along the same line, SGLT2 inhibitors may cause this lesion through their effect of blocking proximal tubular glucose transport in the S1 segment, sending a large amount of glucose to the S3 segment. The S3 segment is also most prominently involved with the other drugs associated with osmotic nephropathy. In one case of SGLT2 inhibitor–associated osmotic nephropathy, the lysosomal vacuoles had amorphous debris on electron microscopy as described with the other agents (6).
In conclusion, osmotic nephropathy is an under-recognized cause of drug-induced kidney injury. Although many of the drugs associated with this lesion are no longer used (dextran, sucrose-containing IVIg) or used less commonly (mannitol, HES) in current times, contrast and SGLT2 inhibitors are widely used. The small number of cases of osmotic nephropathy described with SGLT2 inhibitors may reflect that the lesion is either uncommon or is being missed (most AKI cases are not biopsied). A reasonable approach to assess for this lesion would be to undertake a kidney biopsy in patients with SGLT2 inhibitor–associated AKI who do not recover to baseline within 5–7 days.
Disclosures
L. Juncos reports consultancy for AstraZeneca and Fresenius Critical Care; honoraria from AstraZeneca and Fresenius; is on the editorial boards for Frontiers in Pharmacology, JASN, and Kidney360; and participates in a speakers’ bureau for AstraZeneca and Fresenius/NxStage. M.A. Perazella reports honoraria from UpToDate and is a scientific advisor for AJKD, CJASN, Clinical Nephrology, Journal of Onco-Nephrology, Kidney International, Kidney360, and KI Reports.
Funding
None.
Acknowledgments
The content of this article reflects the personal experience and views of the authors and should not be considered medical advice or recommendation. The content does not reflect the views or opinions of the American Society of Nephrology (ASN) or Kidney360. Responsibility for the information and views expressed herein lies entirely with the authors.
Author Contributions
M.A. Perazella wrote the original draft of the manuscript; and M.A. Perazella and L. Juncos conceptualized the article and reviewed and edited the manuscript.
- Received December 16, 2021.
- Accepted December 20, 2021.
- Copyright © 2022 by the American Society of Nephrology
Jump to section
More in this TOC Section
Cited By...
- No citing articles found.