Data were obtained independently by four authors who carried out a non-systematic search in the PubMed database. Search strategies included Medical Subject Heading terms for “diethylene glycol” and “kidney injury’’. The selection criteria were: i) time frame from 2015 to 2021; ii) observational studies (case series, case-control, cohort, and cross-sectional), clinical trials, recent consensus statements, and guidelines; iii) full text available in electronic format, written in English or Portuguese; iv) studies with adequate methodological rigor; and v) studies about any aspect concerning kidney injury associated with DEG intoxication. The exclusion criteria were: i) studies that did refer to kidney injury associated with DEG; ii) articles other than the specified inclusion criteria. Adopting the aforementioned search descriptors, 31 articles were obtained and had the title and abstract read by four authors. After this step, 14 articles were selected and completely read. In addition, the bibliographic references of the identified articles were also evaluated. To provide an overview of outcomes from selected case reports and case series, we calculated percentages of total cases reviewed that presented with certain clinical features.
The details of the selection process are displayed in Flowchart 1
PATHOPHYSIOLOGY
Pharmacokinetic data from DEG were not obtained directly from humans, since, in addition to the high toxicity of DEG metabolites, the diagnosis of this intoxication is usually delayed, which can make diagnosis and data collection difficult. In this sense, pre-clinical trials present an important tool for the study of the mechanisms of DEG toxicity.9
Although DEG consists of two linked ethylene glycol molecules, its metabolism generates different products, which are not ethylene glycol (EG) and its by-products. Calcium oxalate crystals, one of the metabolites of EG responsible for renal injury, are not observed in tissues of DEG.2,9
After oral ingestion, DEG is rapidly absorbed by the gastrointestinal tract and is subsequently widely distributed to most tissues through blood circulation. According to an experimental study in primates, the peak of serum concentration occurs between 30 to 60 minutes.10 Dermal absorption can also occur, although it is less common.11
Experimental models with rats showed that about 50-70% of DEG is oxidized by alcohol dehydrogenase (ADH) in 2-hydroxyethoxy acetaldehyde and then, by aldehyde dehydrogenase (ALDH), in acetic 2-hydroxyethoxy acid (HEAA), whose excretion is mostly renal. HEAA is responsible for metabolic acidosis and organ dysfunction observed in DEG poisoning.12,13 In addition, HEAA promotes membrane destabilization and intracellular accumulation of osmotically active particles, with a consequent fluid diversion through the plasma membrane.12,13
Another product of DEG metabolism that has been related to renal injury is diglycolic acid (DGA), also known as oxydiacetic acid, formed from HEAA. DGA undergoes glomerular filtration and is transported to proximal tubular cells by atypical sodium dicarboxylate transporters-1 or carriers of organic anions.9,11 In the proximal tubule, the compound inhibits the citric acid cycle enzyme, succinate dehydrogenase, causing cell death by blocking the production of adenosine triphosphate.14 Robinson et al.15 demonstrated that direct administration of DGA to rats produced renal injury at a dose of 300 mg/kg. However, no toxicity was observed at a dose of 100 mg/kg. Renal histopathology of rats receiving the high dose of DGA showed marked degeneration and necrosis of the proximal tubules. These pathologic findings are also reported in cases of human toxicity.2,9,11,15 Furthermore, Landry et al.16 showed that the DGA, at the concentration of 50 mmol/L, promoted a reduction in ATP levels of the proximal tubule cells, besides reducing oxygen consumption by these cells. Therefore, the substance acts by interfering both in energy production and in the use of mitochondrial oxygen.16,17 The increased urinary levels of kidney injury molecule 1 indicated proximal tubule lesions.15
Another mechanism of injury that has been frequently identified in the cells of the proximal tubule due to the use of toxic compounds is the increased production of reactive oxygen species (ROS) in mitochondria.18,19 In addition to NADPH in the cytoplasm, and ketoglutarate dehydrogenase in the mitochondrial matrix, other important sources of ROS include electron carrier chain complexes I and II. In the presence of hypoxia, succinate dehydrogenase (complex II) has been shown to generate superoxide by undergoing the conversion of dehydrogenase to fumarate reductase. Thus, fumarate receives its electrons from complex I in a reverse electron transport mechanism, functioning as a final electron acceptor and generating ROS.20 DGA, by inhibiting the cell's ability to consume oxygen, places the cell in a condition of "pseudohypoxia". This condition, then, may trigger the process described above, culminating in the generation of ROS.17
In many cases, antioxidant compounds such as N-Acetyl-L-Cysteine and L-ascorbic acid were able to inhibit the production of ROS induced by toxic agents, in addition to stimulating the growth and regeneration of mitochondrial functions.21-23 However, Landry et al.16 showed the use of antioxidants in cells exposed to DGA was not able to reduce cell death at the highest concentration of 50 mmol/L. This result indicated that DGA-induced cytotoxicity occurred independently of ROS production, probably by an alternative mechanism. One possible alternative mechanism would be the inhibition of succinate dehydrogenase by DGA, considering that antioxidants are not able to antagonize this process. Inhibition of succinate dehydrogenase results in decreased ATP and cell death. In this case, the production of ROS would be a by-product of cell death. In the case of lower concentrations of DGA (25 mmol/L), the antioxidant can reduce DGA-induced cell death, suggesting that, at smaller concentrations, the production of ROS has a role in cell death. Inhibition of succinate dehydrogenase seems to represent the main mechanism by which DGA produces its toxic effects. This inhibition promotes ROS production, decreased oxygen consumption, and ATP levels, which result in cell death.17
Thus, kidney injury in DEG poisoning is secondary to proximal tubular necrosis caused by DGA. In addition, marked vacuolization and edema of epithelial cells obstruct the lumen, reducing urine flow and, consequently, resulting in anuria and uremia.17
Figure 1 shows metabolic pathways of DEG metabolism.
MORPHOLOGICAL ASPECTS OF DIETHYLENE GLYCOL KIDNEY TOXICITY
Diethylene glycol (DEG) is a similar molecule to ethylene glycol. Both molecules can be produced in analogous processes and can cause acute renal failure. Initially, DEG was thought to be metabolized by endogenous cleavage of any bond to form ethylene glycol, which would be responsible for the adverse effects. Some experimental models with rats showed oxalate crystals in the urine of animals receiving DEG suggesting that toxic effects could be caused by the formation and subsequent metabolism of the ethylene glycol.23 However, other studies in dogs and rabbits did not show any increase in urinary oxalate concentrations after oral administration of DEG.22,26 Successive studies using radiolabeled DEG in rats and dogs confirmed this last observation.12 In addition, DEG-poisoned patients did not show urinary oxalate formation. This finding supports the argument that endogenous conversion of DEG to ethylene glycol does not occur.27-29 Based on these results, it appears that DEG is not metabolized to two ethylene glycol molecules, likely due to its metabolically stable structure. It was hypothesized that the experiments suggesting oxalate formation may have involved products contaminated with ethylene glycol. The overwhelming evidence is that DEG metabolism does not lead to the formation of oxalate crystals within the kidney.27
Renal acute injuries appear to arise mainly from tubular cytoplasm degeneration, markedly presenting as diffuse/severe vacuolation. Cytoplasmic vacuolization (also called cytoplasmic vacuolation) is a well-known morphological phenomenon observed in mammalian cells after exposure to bacterial or viral pathogens as well as to various toxic natural and artificial low-molecular-weight compounds. The vacuolization can be transient, but it is more likely to cause irreversible damage. The vacuolation due to DEG poisoning is not fully understood. This process may be related to the hydropic degeneration (swelling) secondary to increased osmotic pressure, steatosis associated with basic amine-containing lipophilic compounds, complex lipids, and ion pump dysfunction as a consequence of alterations on Na, K-ATPase or Calcium-Activated Potassium Channels.30 Vacuolization often accompanies cell death; however, its role in cell death processes remains unclear. There is an accentuated clear (negative image) dilatation of the cytoplasm like ballooned or fatty degeneration in tubular cells.27,31 Tubular lesions mainly occur in the proximal convoluted tubules and are restricted to the cortical regions of the kidney.24 Toxic lesions of proximal tubules also manifest as cortical infarctions and/or necrosis, with vascular congestion, diffuse interstitial hemorrhage, and edema. In addition, a profound swelling of the tubular epithelium can cause complete obliteration of the lumen.2,32 Depending on the amount or intensity of the exposure to the poison, vascular lesions, such as vascular necrosis and thrombotic microangiopathy, could be found (see images of renal biopsies from our personal archive in Figures 2 and 3). Vacuolization may occur in podocytes and the parietal epithelium of Bowman's capsule cells. Swelling of the organelles, including the mitochondria, may also be seen (see images of renal biopsies from our personal archive in Figures 2 and 3).
Chronic changes are nonspecific, sharing the same findings as other end-stage renal diseases, including globally sclerosed glomeruli, atrophic tubules, diffuse interstitial fibrosis and discrete infiltrate of inflammatory cells.33