Acute kidney injury (AKI) is a sudden decline or loss of renal function. The number of hospital stays involving a principal diagnosis of AKI increased by a factor of 1.8 from 2005 to 2014, from 281,500 to 504,600 stays (1). About 50% of critically ill patients in intensive care unit (ICU) develop AKI (2, 3). In a multi-country prospective study, 11.0% of patients with severe AKI died in ICU, and AKI was associated with an increased risk of 28-day death (OR = 1.77, 95%CI = 1.17–2.68) (4). AKI is associated with an increase of ICU length of stay (LOS), use of renal replacement therapy (RRT), duration of mechanical ventilation, and mortality (5–8), which threatens human life and health and imposes a heavy burden on the global economy.

Diabetes, the most common cause of renal failure, is also associated with AKI and a higher risk of death from AKI (9, 10). Diabetes is one of the important risk factors for AKI (11). A study showed that patients with diabetes were eight times more likely to develop AKI than those without diabetes (12). AKI in diabetic patients is also an important cause of prolonged hospital stay, renal failure requiring RRT, development of chronic kidney disease (CKD) or end-stage renal disease, and eventual death (13, 14).

Iron is an essential trace element in the human body and plays an important role in many physiological and biochemical processes. When excess iron accumulates in organs such as the heart, kidney, and pancreas, the catalytic iron can produce toxic reactions leading to organ dysfunction (15). Iron overload has been linked to a higher risk of diabetes and insulin resistance (16, 17). In addition, ferroptosis plays an important role in the pathogenesis of AKI (18). Iron may increase the risk of kidney injury in diabetes by increasing oxidative/nitrification stress and decreasing antioxidant capacity (19, 20). However, the relationship between iron metabolism and AKI in diabetic patients remains unclear. The aims of this present study were to analyze the risk factors of AKI and all-cause death in critically ill patients with diabetes and to explore the association between iron metabolism and AKI in patients with diabetes.

In this study, the clinical data from MIMIC-III database were retrospectively analyzed, and the results showed that the critically ill patients with diabetes complicated with AKI were more likely to have abnormal iron metabolism indicators than those in non-AKI group, and the increase of SF was closely related to AKI and all-cause death in critical patients with diabetes. Previous studies have shown that the predictors for AKI in diabetic patients were advanced age, CHF, CKD, hypertension, smoking, alcohol consumption, diabetic nephropathy, infection, aminoglycosides, diabetic ketoacidosis, non-steroidal anti-inflammatory drugs, long course of diabetes, diuretics, beta blockers, and delay of follow-up (14, 24–26). However, whether abnormal iron metabolism is related to the occurrence of AKI in diabetic patients seems to be unproven in previous studies. This present study, for the first time, confirmed that increase of SF is an important risk factor and predictor of all-cause death and AKI, so as to provide certain reference for the prevention and treatment of AKI in critical patients with diabetes. Early correction of the iron metabolism disorder may improve the renal prognosis and survival rate of critical patients with diabetes.

Our results showed that critical diabetic patients with AKI were prone to increased SF and TRF saturation and decreased TRF and TIBC. Iron has many biological functions, such as regulating cell cycle progression, DNA synthesis and repair, mitochondrial function, and inflammatory responses (27). SF is an acute-phase reactant and can be elevated in a wide range of conditions, including acute liver injury, chronic liver disease, sepsis, systemic inflammatory conditions, inherited haemoglobinopathies, and secondary iron overload syndrome (28). Therefore, iron metabolism disorder and acute inflammatory response can both cause the increase of SF levels. In addition, our study showed that the increase of SF was an independent risk factor for AKI and 6-month all-cause death in critically ill patients with diabetes (the ORs were 1.002 and 1.380, respectively). Studies have also shown that increased SF was associated with decreased residual renal function and increased mortality in patients on peritoneal dialysis (29, 30), which were consistent with our results. The high glucose toxicity in diabetic patients could reduce the release of serum iron and increase the levels of SF. Iron overload can catalyze the production of excess hydroxyl free radicals and promote lipid peroxidation, resulting in oxidative stress and tissue damage and promoting the occurrence of AKI (31). Ferroptosis is also a type of cell death characterized by intracellular iron overload and the accumulation of reactive oxygen species, leading to lipid peroxidation, which is involved in the occurrence and development of AKI (18, 32, 33). In addition, excessive iron deposition in the liver, heart, pancreas, brain, skin, and other organs and tissues will not only lead to extensive fibrosis and impaired function of organs but also seriously affect the quality of life of patients and even lead to death (34). Therefore, diabetic patients with increased SF were more likely to develop AKI and have a high risk of all-cause death. However, there was also a study that did not match our results. In a prospective cohort study, SF levels at admission could be used as a prognostic indicator of renal recovery in patients with AKI, in which SF alleviated oxidative stress by sequestration of iron and limited its involvement in chemical reactions that produce reactive oxygen species, thus helping to improve the renal prognosis (35). The possible reasons were the difference in basic characteristics, basic diseases, and therapeutic measures for the renal protection of patients. The TRF is the main iron-containing protein in plasma, which is responsible for the transport of iron released by erythrocyte degradation and absorbed by the digestive tract. When TRF decreases, the catalytic transport of iron will be inhibited. In addition, the concentration of TRF will also decrease in inflammatory and malignant lesions, which further affects the occurrence of renal inflammation (36). Combined with the results of our study, the increased SF in critically ill diabetic patients with AKI might be caused by inflammation and iron metabolism disorder at the same time. However, which kinds of mechanism play a dominant role in the pathogenesis of AKI in critically ill diabetic patients still need to be further studied.

In addition, our study also showed that CHF was an independent risk factor for diabetic AKI. In a retrospective study with a median follow-up time of 30.75 months, 19.76% of patients with type 2 diabetes had AKI, and CHF (adjusted OR = 2.89, 95%CI = 1.62–5.13) was an important predictor of AKI (37). CHF is often associated with the state of low cardiac output resulting in renal hypoperfusion. Morever, blocked venous return also leads to ischemic renal injury, which is also referred to as acute tubular necrosis (38). Ultimately, decreased renal blood flow leads to epithelial and necrotic fragmentation obstruction and loss of filtration fluid through damaged tubular epithelium, and decreased glomerular ultrafiltration pressure leads to decreased glomerular filtration rate and increased Scr (39, 40). In addition, we also included in the regression model of the OASIS the Scr levels as predictors of AKI in diabetes. A study showed that the OASIS had a good predictive effect on the assessment and prognosis of critically ill patients (41). The baseline Scr values could also predict the occurrence and progression of AKI (42). These results were all consistent with our findings.

However, several limitations to our study should be acknowledged. First, our study was a single-center retrospective analysis, and the subjects were all critically ill patients in ICU. Second, the diagnosis of AKI was determined only by changes in Scr; some AKI patients with only changes in urine volume might be missed. Third, due to data limitations, only the results of iron metabolism laboratory parameters were collected in this study, without collecting the results of iron deposition in organs (such as through magnetic resonance imaging). Fourth, the study did not include levels of inflammation indicators (C-reactive protein, procalcitonin), use of vasopressor, duration of mechanical ventilation, anti-infective drugs, and other confounding risk factors that might be associated with adverse outcomes. Therefore, our results are still being validated in a prospective multicenter cohort enrolling non-critical patients.

Critically ill diabetic patients with AKI and poor prognosis were more likely to have abnormal iron metabolism. Increased SF and decreased TRF are important risk factors for AKI and 6-month all-cause death in critical diabetic patients. The renal prognosis and survival rate of diabetic patients may be predicted by SF levels clinically.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material.

MM: study design, data analysis, and writing the manuscript. YG: study design, data analysis, data collection, and writing the manuscript. LD: study design and revision of the manuscript. YL: data collection and data analysis. NX and YL data collection and revision of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the National Natural Science Foundation of China (no. 8176030057), the Guangxi Natural Science Foundation (no. 2018GXNSFBA050040), Guangxi Clinical Research Center for Urology and Nephrology (no. 2020AC03006), and the Scientific Research and Technological Development Program of Guangxi (no. GuiKeGong 1598011-6).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.