Prognostic implications of serum ferritin levels in non-anemic women with stage 3 chronic kidney disease

Introduction: Chronic kidney disease (CKD) is often accompanied by iron deficiency and persistent inflammation, both of which complicate the evaluation of iron metabolism and its clinical relevance. Although serum ferritin is commonly used to assess iron status in anemic CKD patients, data on its prognostic value in non-anemic, iron-deficient individuals remain limited and inconclusive.

Methods: This retrospective cohort study utilized the TriNetX database to evaluate 5-year clinical outcomes in adult women with stage 3 chronic kidney disease (CKD), normal hemoglobin levels (≥12 g/dL), normal mean corpuscular volume (MCV, 80–100 fL), and varying serum ferritin concentrations. Patients were stratified into two groups based on ferritin levels: <100 ng/ml (low ferritin) and 100–700 ng/ml (adequate ferritin). Primary outcomes included all-cause mortality, major adverse cardiovascular events (MACE), acute kidney injury (AKI), pneumonia, fractures, and progression to advanced CKD (estimated glomerular filtration rate < 30 ml/min/1.73 m²).

Results: A total of 66,768 eligible non-anemic women with stage 3 CKD, low serum ferritin levels, and normal MCV were identified. Propensity score matching (1,1) based on demographic variables was performed prior to comparing outcomes between low ferritin (n = 52,295) and adequate ferritin (n = 52,295) cohorts. Over 5 years, low ferritin—relative to adequate ferritin—was associated with significantly lower hazards of AKI, CKD progression, and pneumonia (HRs 0.909, 0.953, and 0.956; log-rank p < 0.05, <0.01, and <0.005, respectively). By contrast, low ferritin—relative to adequate ferritin—was associated with a significantly higher fracture hazard (HR 1.125; log-rank p < 0.05). No significant differences were observed in all-cause mortality or MACE. Low ferritin was associated with lower all-cause mortality at years 1–3 (ORs: 0.739, p < 0.001; 0.842, p < 0.05; 0.895, p = 0.038) and to a lower cumulative incidence of CKD progression at years 2–5 (ORs: 0.888 at year 2, 0.898 at year 3, 0.907 at year 4, 0.914 at year 5; all p < 0.05). Subgroup analysis revealed that low ferritin levels were especially protective against AKI and pneumonia in patients with elevated CRP (>10 mg/L), postmenopausal women, and those aged 18–64. The association of low ferritin levels with enhanced renal function preservation was more pronounced in postmenopausal women, individuals with vitamin D ≥ 30 ng/ml, and patients with diabetes. Conversely, adequate ferritin was associated witha lower fracture risk in older adults (>65 years) and those with vitamin D < 20 ng/ml.

Conclusion: This study of non-anemic female patients with stage 3 CKD found that adequate ferritin levels correlated with a heightened risk of AKI, renal disease progression, and pneumonia. In contrast, low ferritin levels were associated with a higher fracture risk but a lower likelihood of renal function deterioration.

Introduction

Iron deficiency is a widespread condition, affecting approximately 25% of the general population and 30–45% of chronic kidney disease (CKD) patients (1). Although bone marrow iron staining is the gold standard for assessing iron stores, its invasiveness limits routine use. Consequently, serum ferritin and transferrin saturation (TSAT) are widely used biomarkers—ferritin reflecting iron storage and TSAT indicating iron availability. Iron is vital for hemoglobin and myoglobin function, DNA synthesis, enzymatic activity, and mitochondrial energy production (2). Ferritin functions not only as an iron storage protein but also as an acute-phase reactant, rising in response to inflammation, malignancy, and systemic disease. Elevated ferritin has been associated with poor outcomes in sepsis (3), acute myocardial infarction (4), cerebrovascular disease (5), and rheumatoid disorders (6), highlighting its dual role as a marker of both iron status and inflammation. To better distinguish these effects, future studies should incorporate inflammatory markers such as CRP and IL-6, along with immunomodulators like vitamin D, to determine whether adverse outcomes are driven by iron imbalance or underlying inflammation (7).

The KDIGO guidelines recommend initiating iron therapy in non-dialysis CKD patients when ferritin is <100 ng/ml with transferrin saturation (TSAT) < 40%, or ferritin is 100–300 ng/ml with TSAT <25%. Iron supplementation is generally avoided if ferritin >700 ng/ml or TSAT >40% (8). A large retrospective study of 18,878 non-dialysis-dependent-CKD patients found no link between ferritin levels and mortality in females, while in males, ferritin <100 ng/ml was associated with a 23% increase in all-cause mortality, and TSAT ≤20% was associated with a 121% increase in all-cause mortality, suggesting that TSAT may be a superior prognostic marker (9). Another study found that low TSAT was a stronger predictor of mortality and MACE in NDD-CKD patients, while ferritin ≥300 ng/ml was associated withhigher mortality, with no consistent associations at lower ferritin levels (10). These findings underscore the limitations of ferritin as a standalone biomarker, as it may reflect inflammation rather than true iron status. Despite growing interest, data on iron deficiency in non-anemic CKD patients remain limited, and it is still unclear whether serum ferritin levels influence clinical outcomes in non-anemic female CKD patients.

We hypothesize that the protective effect of lower ferritin levels against adverse renal and infectious outcomes is not solely attributable to its role as an inflammatory marker but is rooted in fundamental iron biology (11). High iron stores can lead to excess labile iron, which catalyzes the Fenton reaction to generate reactive oxygen species, inducing oxidative stress that directly damages renal tubular cells and promotes fibrosis (12). Therefore, a state of lower iron storage, reflected by lower ferritin, may mitigate this iron-mediated nephrotoxicity and slow the progression of kidney disease (13). Furthermore, by restricting systemic iron availability, lower ferritin levels may enhance nutritional immunity, a host-defense strategy that limits pathogen access to essential iron, thereby reducing susceptibility to infections like pneumonia (14). This study aims to explore whether our observed clinical associations are consistent with these proposed biological mechanisms (15).

We assess the impact of serum ferritin levels below 100 ng/ml on five-year clinical outcomes in non-anemic women with stage 3 CKD, using ferritin levels between 100 and 700 ng/ml as the adequate control group. Using TriNetX data, it examines associations with all-cause mortality, major adverse cardiovascular events (MACE), acute kidney injury (AKI), pneumonia, fractures, and CKD progression (GFR ≤ 30 ml/min). These findings help to clarify the prognostic value of ferritin for clinical outcomes in patients with CKD.

Patients and methods

Study design

This retrospective cohort study was conducted by analysing data from the TriNetX network and adhered to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. TriNetX is a global federated health research network¹ providing access to electronic medical records—covering diagnoses, procedures, medications, laboratory results, and genomic data—from numerous large healthcare organizations (HCOs). This study used data from a subset of 139 HCOs within the Global Collaborative Network. The de-identification of patient data by TriNetX complies with Section §164.514(b)(1) of the HIPAA Privacy Rule (16). The study protocol was approved by the Institutional Review Board of Taipei Tzu Chi Hospital (Approval Number: 14-IRB027; approval date: 06/03/2025). 140 healthcare organizations were queried at the time of analysis using ICD-10 codes aligned with the study’s inclusion and exclusion criteria.

Study cohorts

Because ferritin reference intervals and diagnostic cut-offs for iron deficiency are sex specific (and under active re-evaluation), restricting to women avoids misclassification bias from applying mixed-sex thresholds (17, 18). Patient selection criteria, cohort characteristics, and the analytical methodology for assessing the long-term (5-year) outcomes in females with CKD stage 3 can be seen in Figure 1. This study included women aged ≥18 years with moderate CKD (ICD-10 N18.3), hemoglobin ≥12 g/dL, normal MCV (80–100 fL), and recorded serum ferritin levels between January 1, 2010, and January 1, 2025. Patients were excluded if they had a history of kidney transplantation (ICD-10 Z94.0), genitourinary malignancies (ICD-10 C51–C58, C64–C68), pregnancy (ICD-10 O00–O08, O09, O10–O16, O20–O29, O30–O48, O60–O77), or gastrointestinal bleeding (ICD-10 K92.2). Initially, 66,768 patients had low ferritin and 59,740 had adequate ferritin; after 1:1 propensity score matching, each group included 52,295 patients. Moderate CKD (stage 3) was defined using ICD-10 N18.3 and TriNetX code 8001, with eGFR calculated via the creatinine-based MDRD equation.

Figure 1

Figure 1. Flowchart outlining the patient selection process for the study.

The study compares patients with a low ferritin group (<100 ng/ml) and an adequate ferritin group (100 ~ 700 ng/ml) follow the latest KDIGO guideline (8). To reduce the effect of demographic confounders, 1:1 propensity score matching was performed based on demographic factors. The primary outcomes assessed over 5 years include all-cause mortality (ICD-10: R99), major adverse cardiovascular events (MACE) (ICD-10: I20-I25, I21, I46, I49, I50, I61, I63, R99), acute kidney injury (ICD-10: N17), pneumonia (ICD-10: J18.9), fractures (ICD-10: M81.0, S12, S22, S32, S42, S52, S62, S72, S82, S92), and progression to GFR < 30 ml/min (TNX: 8001).

Data analysis

The data analysis for this study involved evaluating clinical outcomes in propensity-matched cohorts of stage 3 CKD, non-anemic women with low or adequate serum ferritin levels over a follow-up period of 5 years. The presence of additional comorbidities or covariates such as diabetes mellitus, hypertension, menopause, age, 25-vitamin D and C-reactive protein (CRP) levels were also analyzed compared between the two cohorts.

To address the analytical limitations of the TriNetX platform, we implemented a two-stage analytical strategy. The first stage involved conducting an exploratory Cox proportional hazards analysis to screen for potential outcomes. Due to platform constraints, this preliminary model was adjusted for four principal confounders: age, sex, diabetes, and hypertension. Outcomes that demonstrated a statistically significant association in this exploratory model were advanced to a more comprehensive secondary analysis. In this stage, we performed Kaplan–Meier survival analysis and applied a Cox proportional hazards model that was fully adjusted for the complete set of covariates used for propensity score matching. This included detailed demographics (age, sex, race, ethnicity), as well as comorbidities, medications, and laboratory results. This two-step methodology allowed for an initial screening of relevant endpoints, which were then subjected to a more rigorously adjusted survival analysis.

In the Kaplan–Meier analysis, patients whose last recorded clinical event occurred within the observation period were censored by the TriNetX platform. The Kaplan–Meier survival data derived from the original TriNetX platform are presented in Supplementary Table 1. However, when calculating cohort risks for a defined timeframe, these patients are included rather than excluded. To account for this, we calculated the number of patients with the outcome and the corresponding risk for each 90-day interval (excluding repeated measures) and summed these across 21 intervals (90 × 21 = 1,890 days) to determine the total number of patients with the outcome over 5 years. Using these manually aggregated data, cumulative survival rates were computed using the Kaplan–Meier estimator to compare cohorts using the log-rank test. Raw data from the Kaplan–Meier analyses conducted on the TriNetX platform were extracted for each outcome, covering twenty-one 90-day intervals, and compiled into an Excel file (Supplementary Table 2). The survival analysis was then performed using the ‘survival’ package for R software (version 4.4.2, Vienna, Austria).

To assess statistical significance, independent two-sample t-tests were conducted, and the Bonferroni-Holm correction was applied to account for multiple comparisons across three tests, with adjusted α thresholds iteratively defined (e.g., α = 0.0083 for the smallest p-value of six tests). The six pre-specified primary outcomes were mortality, MACE, AKI, pneumonia, fractures, and CKD progression. Both unadjusted and Holm–Bonferroni adjusted p-values (e.g., α = 0.0083 as the smallest threshold across six tests) were calculated to control the family-wise error rate. Given the reduced sample size after propensity score matching, all reported statistically significant results had unadjusted p-values below their respective thresholds. Subsequent statistical analyses and data visualizations were performed using GraphPad Prism (Version 8.0) and SigmaPlot (Version 10.0).

Sensitivity analysis

To assess the robustness of our findings, we conducted two sensitivity analyses. First, we evaluated clinical outcomes across more granular ferritin ranges (100–300, 300–500, and 500–700 ng/ml). Second, to account for the influence of follow-up duration, we compared results from the primary cohort (enrolled 2010–2025) with those from a time-restricted cohort (enrolled 2010–2020), which ensured a potential follow-up period of at least 5 years for all patients.

Competing risk analysis and cumulative incidence curve

To account for the competing risk of mortality, we conducted a competing risk analysis on the TriNetX platform. This analysis provided the unadjusted cumulative incidence for pre-specified outcomes in each cohort, thereby isolating the event of interest from the influence of patient death. The platform utilizes the Aalen-Johansen estimator to generate cumulative incidence curves. By extracting the cumulative incidence values at defined follow-up intervals, we can constructed cumulative incidence curve.

Results

Patient characteristics

A total of 66,768 low ferritin patients and 59,740 adequate ferritin patients met the inclusion and exclusion criteria for the study. Table 1 provides a detailed comparison of patient baseline characteristics before and after propensity score matching. Demographic data, diagnosis comorbidity prevalence, medication usage, and laboratory values, were compared between patients with low or adequate ferritin levels. Before matching, the adequate ferritin group was older (63.4 ± 12.4 years vs. 59.6 ± 13.9 years, p < 0.001, standardized difference = 0.290) and had a lower proportion of White patients (70.9% vs. 62.6%, p < 0.001). Hypertensive diseases and Diabetes mellitus were significantly more prevalent in the adequate ferritin group (28.3% vs. 26.0%, p < 0.001; 10.9% vs. 10.3%, p < 0.001). To better balance biochemical test values, some laboratory variables were subdivided into multiple levels to help fully adjust for residual confounders. After matching, all demographic and laboratory differences were minimized, with standardized differences below 0.1, indicating that propensity score matching effectively eliminated demographic disparities between the two cohorts.

Table 1

Table 1. Baseline patient characteristics before and after propensity score matching were reported.

The association of serum ferritin level with clinical outcomes

Cox proportional-hazards analysis (Supplementary Table 3).

Cox proportional-hazards models compared six outcomes between patients with ferritin <100 ng/ml and those with ferritin 100–700 ng/ml (reference group). Low ferritin was associated with lower risks of all-cause mortality (HR 0.897; 95% CI 0.832–0.968; p = 0.0048), acute kidney injury (HR 0.857; 95% CI 0.812–0.904; p < 0.0001), and pneumonia (HR 0.893; 95% CI 0.845–0.943; p < 0.0001). No significant difference was observed for progression to eGFR < 30 ml/min/1.73 m² (HR 0.973; 95% CI 0.93–1.018; p = 0.2359). In contrast, low ferritin was associated withhigher risks of MACE (HR 1.042; 95% CI 1.018–1.067; p = 0.0007) and fractures (HR 1.236; 95% CI 1.202–1.27; p < 0.0001).

Kaplan–Meier survival analyses (Figure 2) demonstrated differential long-term risks between low ferritin and adequate ferritin patients. Over a 5-year follow-up:

Figure 2

Figure 2. Five-year Kaplan–Meier survival curves showing the association between serum ferritin levels and clinical outcomes: (A) all-cause mortality, (B) major adverse cardiovascular events (MACE), (C) acute kidney injury (AKI), (D) GFR < 30 ml/min, (E) pneumonia, and (F) fractures, in patients stratified by ferritin levels (100–700 ng/ml vs. <100 ng/ml).

All-cause mortality and MACE

Overall risk for all-cause mortality and MACE across the 5-year observation period were not statistically significantly different between the two groups (Figures 2A,B). However, Kaplan–Meier analysis for all-cause mortality (Figure 2A) showed a clear separation starting at year 1, with the adequate ferritin group exhibiting a significantly higher mortality risk during the first 2 years of follow-up. As illustrated in Figure 3A, the cumulative incidence of all-cause mortality was consistently higher in the adequate ferritin group at year 1 (p < 0.001) and year 2 (p < 0.05), but the difference became non-significant beyond year 2 (year 3, p = 0.038; year 4, p = 0.120; year 5, p = 0.156). The difference in mortality risk between the two groups dissipated over time, with yearly odds ratios progressively converging on 1.

Figure 3

Figure 3. Association of serum ferritin levels with five-year cumulative incidence of (A) all-cause mortality and (B) renal function decline (GFR < 30 ml/min). P-values and odds ratios (ORs) reflect comparisons between ferritin levels of 100–700 ng/ml and <100 ng/ml at each annual follow-up.

AKI and renal function decline

In patients with ferritin <100 ng/ml, the 5-year AKI risk was 0.020 versus 0.025 in those with ferritin 100–700 ng/ml, corresponding to a risk difference of −0.005 and an odds ratio of 0.807; the p-value for the risk difference was <0.001. Likewise, patients with ferritin levels below 100 ng/ml had a 5-year risk of 0.029 for their GFR falling below 30 ml/min, compared to a risk of 0.032 in those with ferritin levels between 100 and 700 ng/ml. This corresponds to a statistically significant risk difference of −0.004 (p = 0.001) and an odds ratio of 0.887.

Patients with ferritin levels below 100 ng/ml had a 5-year risk of 0.029 for their Kaplan–Meier survival analysis showed that log-rank p-values for both AKI and renal function decline were <0.05 & < 0.01 respectively, indicating significantly lower incidence rates in patients with lower ferritin levels (Figures 2C,D). CKD progression was defined as a decline in GFR to <30 ml/min/1.73 m². Patients in the adequate ferritin group experienced a significantly faster decline in renal function (log-rank p < 0.01). As shown in Figure 3B, the cumulative incidence of CKD progression over the 5-year follow-up was consistently higher in the adequate ferritin group compared to the low ferritin group. Although the between-group difference was not significant at year 1 (p = 0.134), it became statistically significant from year 2 onward (p < 0.05). The difference in CKD progression risk between the two cohorts dissipated over time, with ORs of 0.888 at year 2, 0.898 at year 3, 0.907 at year 4, and 0.914 at year 5.

Pneumonia

In the ferritin <100 ng/ml group, the 5-year pneumonia risk was 0.018 versus 0.022 in the 100–700 ng/ml group, yielding an absolute risk difference of −0.003 and an odds ratio of 0.846 (p < 0.001 for the risk difference). Kaplan–Meier analysis showed a significantly lower incidence of pneumonia in patients with low ferritin levels (log-rank p < 0.05; Figure 2E).

Fractures

A 5-year fracture risk of 0.071 was observed in patients with ferritin levels below 100 ng/ml, representing a statistically significant increase compared to the 0.066 risk found in patients with ferritin levels of 100–700 ng/ml (risk difference: 0.005, p = 0.007; odds ratio: 1.075). Kaplan–Meier analysis showed a significantly higher fracture risk in the low ferritin group (p < 0.05; Figure 2F). A comparison of the cohort’s overall risk at 5-years follow-up confirmed this association, suggesting that low ferritin may contribute to increased fracture risk, possibly due to impaired bone metabolism related to iron deficiency.

Subgroup analyses of clinical outcomes by ferritin levels (100–700 ng/ml vs. <100 ng/ml)

The forest plots present subgroup analyses of 5-year clinical outcomes—including acute kidney injury (AKI; Figure 4A), renal function decline (GFR < 30 ml/min; Figure 4B), pneumonia (Figure 4C), and fractures (Figure 4D) —in female patients with moderate CKD and normal hemoglobin levels. The analysis examines the influence of key variables such as CRP levels (≤10 vs. >10 mg/L), age (18–64 vs. ≥65 years), menopausal status, vitamin D levels (<20 vs. ≥30 ng/ml), and the presence or absence of hypertension (HTN) and diabetes (DM) on these outcomes.

Figure 4

Figure 4. Forest plots illustrating subgroup analyses of selected covariates associated with specific clinical outcomes among patients stratified by serum ferritin levels (100–700 ng/ml vs. <100 ng/ml): (A) acute kidney injury (AKI), (B) GFR < 30 ml/min/1.73m², (C) pneumonia, and (D) fractures.

Acute kidney injury and renal function decline

The analysis shows that lower ferritin levels were generally associated with reduced risks of AKI and renal function decline, suggesting a potential protective effect. The greatest reduction in AKI risk was observed in patients with elevated CRP (>10 mg/L), postmenopausal women, and individuals aged 18–64 years (Figure 4A). Similarly, the strongest protective effects against renal decline were seen in postmenopausal women, those with vitamin D ≥ 30 ng/ml, patients with diabetes, and older adults (Figure 4B).

Pneumonia

The low ferritin group was associated with a reduced risk of pneumonia across most subgroups (Figure 4C). The strongest protective effects were seen in patients with elevated CRP (>10 mg/L), younger age (18–64 years), and postmenopausal women. Notably, protection was also observed in those with vitamin D deficiency and diabetes, suggesting that systemic inflammation, hormonal status, and metabolic or nutritional factors may influence pneumonia risk in female CKD patients.

Fractures

The adequate ferritin group (100–700 ng/ml) was associated with a lower fracture risk, suggesting a protective role of iron stores in bone health (Figure 4D). The protective effect of adequate ferritin on fracture risk was significant in patients with vitamin D deficiency, older age (≥65), and without diabetes. No significant difference was observed in younger patients, indicating that age-related bone loss may be more influential than ferritin levels in fracture risk among younger CKD women.

Collectively, Kaplan–Meier analyses demonstrated that while the adequate-ferritin group had higher early mortality (first 2 years), this disparity attenuated over time (ORs ≈ 1). In contrast, associations with AKI, pneumonia, and renal decline remained significant throughout the 5-year follow-up. The effects were strongest in high-risk subgroups—elevated CRP (>10 mg/L), postmenopausal women, and younger patients—where low ferritin showed the clearest protective association. Although mortality differences weakened longitudinally, the consistent patterns across outcomes and subgroups support plausible inflammatory and iron-mediated mechanisms, warranting further prospective investigation.

Sensitivity analysis of ferritin-level-dependent clinical outcome heterogeneity

To evaluate ferritin-dependent outcome variability, we performed sensitivity analyses using stratified ranges (100–300, 300–500, and 500–700 ng/ml), which yielded consistent results (Supplementary Table 4). In stage 3 CKD women without anemia, ferritin <100 ng/ml showed no mortality difference versus the 100–700 ng/ml reference. However, comparative analyses revealed significantly higher mortality hazards in the 301–500 and 501–700 ng/ml groups than in the <100 ng/ml cohort, indicating that hyperferitinemia may confer greater mortality risk than iron deficiency.

The second sensitivity analysis comparing the full (2010–2025) and time-restricted (2010–2020) enrollment cohorts was conducted to assess for heterogeneity in 5-year outcomes (Supplementary Table 5). The associations for mortality, AKI, eGFR decline, and pneumonia remained consistent, favoring the low-ferritin group. However, the MACE association shifted from neutral to modestly protective in the time-restricted analysis, while the elevated fracture risk seen in the full cohort was attenuated to null. Although absolute risks varied, the comparative hazard ratios for most outcomes remained stable, indicating minimal heterogeneity introduced by the enrollment period.

Nutritional status and use of iron supplements

Nutritional status may be an unmeasured confounder affecting the association with low ferritin levels. To address this, we adjusted for baseline albumin levels post-propensity score matching, which partially mitigates, but does not fully eliminate, its influence. Regarding the potential protective effect of low ferritin, our hypothesis is supported by our recent publication (19). In that study, we found that iron supplementation in non-anemic women with stage 3 CKD and low ferritin was associated with increased risks of MACE, adverse renal events, and pneumonia, without offering a survival benefit. These findings reinforce the possibility that a low ferritin state may be protective in this population.

Discussion

This study examined the relationship between serum ferritin levels and clinical outcomes in non-anemic, non-dialysis female patients with stage 3 CKD. Low ferritin levels (<100 ng/ml) were associated with reduced risks of AKI, renal decline, and pneumonia, particularly in postmenopausal women. Additionally, low ferritin was associated withlower all-cause mortality before year 2 and continued protection against renal decline beyond year 2. However, it was also associated with a higher risk of fractures, whereas adequate ferritin levels (100–700 ng/ml) were protective, especially in older adults and those with vitamin D deficiency. These findings underscore the complex role of ferritin in inflammation, bone health, and kidney function, demonstrating the importance of a personalized approach to CKD management.

Previous studies suggest that transferrin saturation (TSAT) is a more reliable predictor of mortality risk than serum ferritin. For instance, Guedes et al. found low TSAT, not ferritin, was more strongly associated with MACE, and serum ferritin ≥300 ng/ml showed no significant link to cardiovascular outcomes (3, 4). Rostoker et al. also highlighted inflammation as a key confounder affecting the interpretation of ferritin and TSAT, with significant impacts on mortality and cardiovascular outcomes in non-dialysis CKD patients (20). The study did not identify a statistically significant association between ferritin levels and 5-year all-cause mortality or major adverse cardiovascular events (MACE) (9, 10). Further analysis showed that low ferritin levels were associated with significantly lower mortality within the first 2 years, suggesting elevated ferritin—possibly reflecting inflammation—may contribute to short-term mortality in female CKD patients. These findings highlight the need for monitoring ferritin and investigating inflammatory markers to improve risk stratification and management.

In non-anemic CKD patients, the link between serum ferritin and kidney disease reflects the complex interaction of iron metabolism, inflammation, and kidney damage. A retrospective study in critically ill patients found ferritin levels >680 ng/ml were associated with higher 28-day mortality, regardless of sepsis, suggesting ferritin may indicate disease severity rather than cause injury directly (21). The Korean National Health Survey found high ferritin levels associated withincreased CKD risk in men but not women, suggesting a possible sex-specific susceptibility to renal decline (22). Consistent with prior studies, the findings indicated that adequate ferritin levels (100–700 ng/ml) were associated with a higher incidence of AKI, while low ferritin levels (<100 ng/ml) were associated witha reduced risk of AKI—particularly among postmenopausal women, individuals aged 18–64, and those with CRP > 10 mg/L—supporting the role of inflammation in increasing AKI risk.

Recent studies have clarified the link between ferritin and CKD progression. Tsai et al. found that elevated ferritin and hsCRP were independently associated with faster CKD progression and initiation of renal replacement therapy, with the highest ferritin tertile showing a 1.4-fold increased risk of renal decline (23). Eisenga et al. reported J- or U-shaped associations, with both low and high ferritin levels independently predicting CKD progression (24). Findings align with previous studies, indicating that female stage 3 CKD patients with adequate ferritin levels experienced greater renal decline over time, with the protective effect of low ferritin becoming most evident after year 2. This supports the role of inflammation in influencing ferritin’s impact, as CKD-related inflammation elevates hepcidin, promoting iron retention and ferritin synthesis. Pro-inflammatory cytokines like IL-1β and TNF-α further drive kidney damage by increasing ferritin expression and disrupting iron balance (25–27).

Ferroptosis, an iron-dependent form of regulated cell death driven by lipid peroxidation and mitochondrial dysfunction, has emerged as a key mechanism in AKI. Unlike apoptosis or necrosis, ferroptosis contributes to renal tubular injury in animal models through iron overload and oxidative stress (28). Clinically, this is relevant as nearly 50% of patients with hospital-associated AKI develop new-onset CKD within 3.3 years (29). The transition from AKI to CKD is primarily driven by maladaptive repair, involving proximal tubule injury, mitochondrial dysfunction, immune activation, and chronic inflammation (30). Emerging preclinical data suggest that inhibiting ferroptosis may protect mitochondrial function, reduce inflammation, and prevent tubular injury, offering potential benefits in diabetic nephropathy and AKI (31–34). These mechanisms may partly explain the long-term renal decline observed in this study. However, clinical evidence linking ferroptosis to AKI-to-CKD progression remains limited, warranting further investigation.

The link between serum ferritin and pneumonia risk in non-anemic CKD patients remains unclear due to limited direct evidence (22, 35). In CKD, elevated ferritin often reflects chronic inflammation, complicating the distinction between true iron overload and inflammation-driven hyperferritinemia (36). Emerging evidence suggests elevated ferritin may indicate inflammatory burden and increased infection risk, including infection by influenza and SARS-CoV-2 (37). In CKD, a pro-inflammatory state, high ferritin likely reflects immune dysregulation rather than iron overload (38, 39). Studies in critically ill patients further support this, linking hyperferritinemia to poorer sepsis outcomes, reinforcing its role as a marker of immune dysfunction (40). The study demonstrates a significant association between relatively high ferritin levels and increased pneumonia risk, while lower risk was observed in patients with reduced ferritin—especially among postmenopausal women, individuals with CRP > 10 mg/L, and those aged 18–64. Prior research suggests iron overload may raise infection risk by promoting bacterial growth, especially in CKD patients on iron therapy (41).

Lower ferritin levels (<100 ng/ml) were associated with an increased fracture risk in non-anemic CKD patients, particularly among those with low vitamin D, elevated CRP, and older age—suggesting a potential link between low ferritin and bone fragility. This aligns with prior research showing that iron deficiency impairs collagen synthesis and osteoblast function, reducing bone mineral density and increasing fracture risk (42). Iron deficiency leads to low bone turnover, impaired mineralization, and greater fracture risk, providing a mechanistic explanation for the increased fractures observed in the low-ferritin group (43). Patients with iron-deficiency anemia have also shown a higher risk of fractures, highlighting the essential role of iron in supporting skeletal integrity (44). Iron deficiency can impair bone metabolism, while excess iron may contribute to bone fragility through oxidative stress and reduced osteoblast activity (45). Additionally, CKD-related mineral and bone disorders independently affect bone health (46). These findings highlight the need for longitudinal studies incorporating bone turnover, inflammation, and mineral metabolism markers to clarify ferritin’s role in CKD-related bone fragility.

Our study reveals a critical trade-off associated with low ferritin levels. While a low ferritin state appears protective against adverse renal and infectious outcomes (47)—potentially through reduced oxidative stress and ferroptosis—it may simultaneously reflect an iron-deficient state that compromises bone health (43). This dichotomy underscores the clinical challenge of balancing the systemic benefits of lower iron stores against the potential risk to skeletal integrity.

In this retrospective cohort study, we collected baseline ferritin levels in a carefully selected population of stable, non-anemic women with stage 3 CKD, excluding participants with conditions known to affect ferritin metabolism (including kidney transplantation history, genitourinary malignancies, pregnancy, or active gastrointestinal bleeding). Following this cohort for 5 years ensured proper temporal sequencing between exposure and outcomes. To further address potential confounding, we performed 1:1 propensity score matching on demographic characteristics, clinical comorbidities, medication profiles, and relevant laboratory measures. This comprehensive approach significantly reduces concerns about reverse causation, a conclusion supported by the consistency of our findings across multiple sensitivity analyses.

Limitations

This study has several important limitations. Its retrospective design using a large database network introduces inherent challenges, including the absence of transferrin saturation (TSAT) data, which precluded a full assessment of iron status, the potential for misclassification bias from diagnostic codes, and heterogeneity in laboratory assays across institutions. These issues are compounded by a high proportion of missing data for key laboratory values, particularly serum vitamin D (>60%) and C-reactive protein (≈90%), which reduced statistical power and introduced a significant risk of selection bias (Supplementary Table 6). Finally, the cohort was predominantly White and U.S.-based, which may limit the generalizability of our findings to other populations.

Conclusion

This study underscores the multifaceted role of serum ferritin in non-anemic female patients with stage 3 CKD. Over 5 years, low ferritin levels (<100 ng/ml) were consistently associated withreduced risks of AKI, renal decline, and pneumonia—particularly in patients with inflammation, younger age, or postmenopausal status. Conversely, adequate ferritin levels (100–700 ng/ml) were associated with lower fracture risk, especially in older adults and those with vitamin D deficiency, suggesting a protective role in bone health. While overall mortality and MACE risks were comparable, early mortality was higher in the adequate ferritin group, possibly reflecting inflammatory burden. These findings highlight the need to interpret ferritin not only as an iron marker but also as an indicator of inflammation. Individualized ferritin targets based on age, comorbidities, inflammation, and bone health may improve clinical risk assessment. Further research is warranted to confirm these associations and inform iron management strategies.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

Ethics statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Taipei Tzu Chi Hospital (approval number: 14-IRB027; approval date: 06/03/2025). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements.

Author contributions

M-TL: Writing – review & editing, Conceptualization, Data curation. C-LL: Writing – review & editing, Conceptualization, Formal analysis, Data curation. Y-CH: Conceptualization, Writing – review & editing, Formal analysis, Data curation. JW: Writing – review & editing, Conceptualization. K-WT: Data curation, Investigation, Writing – review & editing. L-JS: Data curation, Validation, Formal analysis, Writing – review & editing. C-CW: Formal analysis, Project administration, Data curation, Methodology, Writing – review & editing. Y-CW: Conceptualization, Writing – original draft, Data curation. K-CL: Writing – review & editing, Conceptualization, Writing – original draft, Formal analysis.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This study was supported by grants from Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation [TCRD-TPE-113-RT-3(1/3)] and Taoyuan Armed Forces General Hospital (TYAFGH-A-114003).

Acknowledgments

The authors would also like to thank the Core Laboratory at the Department of Research, Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, for their technical support and the use of their facilities.

Conflict of interest

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.

Generative AI statement

The author(s) declare that no Gen AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

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.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnut.2025.1682003/full#supplementary-material

Footnotes

^https://live.trinetx.com/

References

1. GBD Chronic Kidney Disease Collaboration. Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the global burden of disease study 2017. Lancet (London, England). (2020) 395:709–33. doi: 10.1016/S0140-6736(20)30045-3

PubMed Abstract | Crossref Full Text | Google Scholar

2. Lopez, A, Cacoub, P, Macdougall, IC, and Peyrin-Biroulet, L. Iron deficiency anaemia. Lancet (London, England). (2016) 387:907–16. doi: 10.1016/S0140-6736(15)60865-0

PubMed Abstract | Crossref Full Text | Google Scholar

3. Brandtner, A, Tymoszuk, P, Nairz, M, Lehner, GF, Fritsche, G, Vales, A, et al. Linkage of alterations in systemic iron homeostasis to patients' outcome in sepsis: a prospective study. J Intensive Care. (2020) 8:76. doi: 10.1186/s40560-020-00495-8

PubMed Abstract | Crossref Full Text | Google Scholar

4. Brinza, C, Floria, M, Popa, IV, and Burlacu, A. The prognostic performance of ferritin in patients with acute myocardial infarction: a systematic review. Diagnostics (Basel, Switzerland). (2022) 12:476. doi: 10.3390/diagnostics12020476

PubMed Abstract | Crossref Full Text | Google Scholar

5. Balla, J, Balla, G, and Zarjou, A. Ferritin in kidney and vascular related diseases: novel roles for an old player. Pharmaceuticals (Basel, Switzerland). (2019) 12:96. doi: 10.3390/ph12020096

PubMed Abstract | Crossref Full Text | Google Scholar

6. Seyhan, S, Pamuk Ö, N, Pamuk, GE, and Çakır, N. The correlation between ferritin level and acute phase parameters in rheumatoid arthritis and systemic lupus erythematosus. Eur J Rheumatol. (2014) 1:92–5. doi: 10.5152/eurjrheumatol.2014.032

PubMed Abstract | Crossref Full Text | Google Scholar

7. Fenercioglu, AK, Gonen, MS, Uzun, H, Sipahioglu, NT, Can, G, Tas, E, et al. The association between serum 25-hydroxyvitamin D3 levels and pro-inflammatory markers in new-onset type 2 diabetes mellitus and prediabetes. Biomolecules. (2023) 13:1778. doi: 10.3390/biom13121778

PubMed Abstract | Crossref Full Text | Google Scholar

8. Garabed Eknoyan, NL, and Winkelmayer, WC. KDIGO 2025 clinical practice guideline for anemia in chronic kidney disease (CKD). Kidney Int Suppl. (2025) 15:1–140. doi: 10.1016/j.kisu.2024.12.001

Crossref Full Text | Google Scholar

9. Yu, H, Shao, X, Guo, Z, Pang, M, Chen, S, She, C, et al. Association of iron deficiency with kidney outcome and all-cause mortality in chronic kidney disease patients without anemia. Nutr J. (2025) 24:7. doi: 10.1186/s12937-025-01072-1

PubMed Abstract | Crossref Full Text | Google Scholar

10. Guedes, M, Muenz, DG, Zee, J, Bieber, B, Stengel, B, Massy, ZA, et al. Serum biomarkers of Iron stores are associated with increased risk of all-cause mortality and cardiovascular events in nondialysis CKD patients, with or without Anemia. J Am Soc Nephrol. (2021) 32:2020–30. doi: 10.1681/ASN.2020101531

PubMed Abstract | Crossref Full Text | Google Scholar

11. Dhondge, RH, Agrawal, S, Kumar, S, Acharya, S, and Karwa, V. A comprehensive review on serum ferritin as a prognostic marker in intensive care units: insights into ischemic heart disease. Cureus. (2024) 16:e57365. doi: 10.7759/cureus.57365

PubMed Abstract | Crossref Full Text | Google Scholar

12. Zhao, Z. Iron and oxidizing species in oxidative stress and Alzheimer's disease. Aging Med (Milton). (2019) 2:82–7. doi: 10.1002/agm2.12074

PubMed Abstract | Crossref Full Text | Google Scholar

13. Xie, T, Yao, L, and Li, X. Advance in iron metabolism, oxidative stress and cellular dysfunction in experimental and human kidney diseases. Antioxidants (Basel). (2024) 13:659. doi: 10.3390/antiox13060659

PubMed Abstract | Crossref Full Text | Google Scholar

14. Shah, AA, Donovan, K, Seeley, C, Dickson, EA, Palmer, AJR, Doree, C, et al. Risk of infection associated with Administration of Intravenous Iron: a systematic review and Meta-analysis. JAMA Netw Open. (2021) 4:e2133935. doi: 10.1001/jamanetworkopen.2021.33935

PubMed Abstract | Crossref Full Text | Google Scholar

15. McCullough, K, and Bolisetty, S. Ferritins in kidney disease. Semin Nephrol. (2020) 40:160–72. doi: 10.1016/j.semnephrol.2020.01.007

PubMed Abstract | Crossref Full Text | Google Scholar

16. Palchuk, MB, London, JW, Perez-Rey, D, Drebert, ZJ, Winer-Jones, JP, Thompson, CN, et al. A global federated real-world data and analytics platform for research. JAMIA Open. (2023) 6:ooad035. doi: 10.1093/jamiaopen/ooad035

PubMed Abstract | Crossref Full Text | Google Scholar

17. Truong, J, Naveed, K, Beriault, D, Lightfoot, D, Fralick, M, and Sholzberg, M. The origin of ferritin reference intervals: a systematic review. Lancet Haematol. (2024) 11:e530–9. doi: 10.1016/S2352-3026(24)00103-0

PubMed Abstract | Crossref Full Text | Google Scholar

18. Jager, L, Rachamin, Y, Senn, O, Burgstaller, JM, Rosemann, T, and Markun, S. Ferritin cutoffs and diagnosis of Iron deficiency in primary care. JAMA Netw Open. (2024) 7:e2425692. doi: 10.1001/jamanetworkopen.2024.25692

PubMed Abstract | Crossref Full Text | Google Scholar

19. Chen, HC, Liao, MT, Wang, J, Tsai, KW, Wu, CC, and Lu, KC. Clinical outcomes of Iron supplement therapy in non-Anemic female CKD stage 3 patients with low serum ferritin level: a multi-institutional TriNetX analysis. J Clin Med. (2025) 14:5575. doi: 10.3390/jcm14155575

PubMed Abstract | Crossref Full Text | Google Scholar

20. Rostoker, G, Lepeytre, F, and Rottembourg, J. Inflammation, serum iron, and risk of mortality and cardiovascular events in nondialysis CKD patients. J Am Soc Nephrol. (2022) 33:654–5. doi: 10.1681/ASN.2021081044

PubMed Abstract | Crossref Full Text | Google Scholar

21. Ren, X, Jiang, Z, Liu, F, Wang, Q, Chen, H, Yu, L, et al. Association of serum ferritin and all-cause mortality in AKI patients: a retrospective cohort study. Front Med. (2024) 11:1368719. doi: 10.3389/fmed.2024.1368719

PubMed Abstract | Crossref Full Text | Google Scholar

22. Kang, HT, Linton, JA, Kwon, SK, Park, BJ, and Lee, JH. Ferritin level is positively associated with chronic kidney disease in Korean men, based on the 2010-2012 Korean national health and nutrition examination survey. Int J Environ Res Public Health. (2016) 13:1058. doi: 10.3390/ijerph13111058

PubMed Abstract | Crossref Full Text | Google Scholar

23. Tsai, YC, Hung, CC, Kuo, MC, Tsai, JC, Yeh, SM, Hwang, SJ, et al. Association of hsCRP, white blood cell count and ferritin with renal outcome in chronic kidney disease patients. PLoS One. (2012) 7:e52775. doi: 10.1371/journal.pone.0052775

PubMed Abstract | Crossref Full Text | Google Scholar

24. Fujisawa, H, Nakayama, M, Haruyama, N, Fukui, A, Yoshitomi, R, Tsuruya, K, et al. Association between iron status markers and kidney outcome in patients with chronic kidney disease. Sci Rep. (2023) 13:18278. doi: 10.1038/s41598-023-45580-8

PubMed Abstract | Crossref Full Text | Google Scholar

25. Torti, FM, and Torti, SV. Regulation of ferritin genes and protein. Blood. (2002) 99:3505–16. doi: 10.1182/blood.V99.10.3505

PubMed Abstract | Crossref Full Text | Google Scholar

26. Lemos, DR, McMurdo, M, Karaca, G, Wilflingseder, J, Leaf, IA, Gupta, N, et al. Interleukin-1β activates a MYC-dependent metabolic switch in kidney stromal cells necessary for progressive tubulointerstitial fibrosis. J Am Soc Nephrol. (2018) 29:1690–705. doi: 10.1681/ASN.2017121283

PubMed Abstract | Crossref Full Text | Google Scholar

27. Wen, Y, Lu, X, Ren, J, Privratsky, JR, Yang, B, Rudemiller, NP, et al. KLF4 in macrophages attenuates TNFα-mediated kidney injury and fibrosis. J Am Soc Nephrol. (2019) 30:1925–38. doi: 10.1681/ASN.2019020111

PubMed Abstract | Crossref Full Text | Google Scholar

28. Ni, L, Yuan, C, and Wu, X. Targeting ferroptosis in acute kidney injury. Cell Death Dis. (2022) 13:182. doi: 10.1038/s41419-022-04628-9

PubMed Abstract | Crossref Full Text | Google Scholar

29. Bucaloiu, ID, Kirchner, HL, Norfolk, ER, Hartle, JE 2nd, and Perkins, RM. Increased risk of death and de novo chronic kidney disease following reversible acute kidney injury. Kidney Int. (2012) 81:477–85. doi: 10.1038/ki.2011.405

PubMed Abstract | Crossref Full Text | Google Scholar

30. Sato, Y, Takahashi, M, and Yanagita, M. Pathophysiology of AKI to CKD progression. Semin Nephrol. (2020) 40:206–15. doi: 10.1016/j.semnephrol.2020.01.011

PubMed Abstract | Crossref Full Text | Google Scholar

31. Sui, M, Xu, D, Zhao, W, Lu, H, Chen, R, Duan, Y, et al. CIRBP promotes ferroptosis by interacting with ELAVL1 and activating ferritinophagy during renal ischaemia-reperfusion injury. J Cell Mol Med. (2021) 25:6203–16. doi: 10.1111/jcmm.16567

PubMed Abstract | Crossref Full Text | Google Scholar

32. Sha, W, Hu, F, Xi, Y, Chu, Y, and Bu, S. Mechanism of ferroptosis and its role in type 2 diabetes mellitus. J Diabetes Res. (2021) 2021:9999612. doi: 10.1155/2021/9999612

Crossref Full Text | Google Scholar

33. Wang, WJ, Jiang, X, Gao, CC, and Chen, ZW. Salusin-β participates in high glucose-induced HK-2 cell ferroptosis in a Nrf-2-dependent manner. Mol Med Rep. (2021) 24:12313. doi: 10.3892/mmr.2021.12313

PubMed Abstract | Crossref Full Text | Google Scholar

34. Wang, J, Liu, Y, Wang, Y, and Sun, L. The cross-link between ferroptosis and kidney diseases. Oxidative Med Cell Longev. (2021) 2021:6654887. doi: 10.1155/2021/6654887

PubMed Abstract | Crossref Full Text | Google Scholar

35. Kernan, KF, and Carcillo, JA. Hyperferritinemia and inflammation. Int Immunol. (2017) 29:401–9. doi: 10.1093/intimm/dxx031

PubMed Abstract | Crossref Full Text | Google Scholar

36. Ueda, N, and Takasawa, K. Impact of inflammation on ferritin, hepcidin and the Management of Iron Deficiency Anemia in chronic kidney disease. Nutrients. (2018) 10:1173. doi: 10.3390/nu10091173

PubMed Abstract | Crossref Full Text | Google Scholar

37. Hegelund, MH, Glenthøj, A, Ryrsø, CK, Ritz, C, Dungu, AM, Sejdic, A, et al. Biomarkers for iron metabolism among patients hospitalized with community-acquired pneumonia caused by infection with SARS-CoV-2, bacteria, and influenza. APMIS. (2022) 130:590–6. doi: 10.1111/apm.13259

PubMed Abstract | Crossref Full Text | Google Scholar

38. Ebert, T, Pawelzik, SC, Witasp, A, Arefin, S, Hobson, S, Kublickiene, K, et al. Inflammation and premature ageing in chronic kidney disease. Toxins. (2020) 12:227. doi: 10.3390/toxins12040227

PubMed Abstract | Crossref Full Text | Google Scholar

39. Rohr, M, Brandenburg, V, and Brunner-La Rocca, HP. How to diagnose iron deficiency in chronic disease: a review of current methods and potential marker for the outcome. Eur J Med Res. (2023) 28:15. doi: 10.1186/s40001-022-00922-6

PubMed Abstract | Crossref Full Text | Google Scholar

40. Schuster, FS, Nyvlt, P, Heeren, P, Spies, C, Adam, MF, Schenk, T, et al. Differential diagnosis of hyperferritinemia in critically ill patients. J Clin Med. (2022) 12:192. doi: 10.3390/jcm12010192

PubMed Abstract | Crossref Full Text | Google Scholar

41. Ganz, T, Aronoff, GR, Gaillard, C, Goodnough, LT, Macdougall, IC, Mayer, G, et al. Iron administration, infection, and anemia management in CKD: untangling the effects of intravenous iron therapy on immunity and infection risk. Kidney Med. (2020) 2:341–53. doi: 10.1016/j.xkme.2020.01.006

PubMed Abstract | Crossref Full Text | Google Scholar

42. Chon, SJ, Choi, YR, Roh, YH, Yun, BH, Cho, S, Choi, YS, et al. Association between levels of serum ferritin and bone mineral density in Korean premenopausal and postmenopausal women: KNHANES 2008-2010. PLoS One. (2014) 9:e114972. doi: 10.1371/journal.pone.0114972

PubMed Abstract | Crossref Full Text | Google Scholar

43. von Brackel, FN, and Oheim, R. Iron and bones: effects of iron overload, deficiency and anemia treatments on bone. JBMR Plus. (2024) 8:ziae064. doi: 10.1093/jbmrpl/ziae064

PubMed Abstract | Crossref Full Text | Google Scholar

44. Liu, X, and An, J. Dietary iron intake and its impact on osteopenia/osteoporosis. BMC Endocr Disord. (2023) 23:154. doi: 10.1186/s12902-023-01389-0

PubMed Abstract | Crossref Full Text | Google Scholar

45. Cai, H, Zhang, H, He, W, and Zhang, H. Iron accumulation and its impact on osteoporotic fractures in postmenopausal women. J Zhejiang Univ Sci B. (2023) 24:301–11. doi: 10.1631/jzus.B2200519

PubMed Abstract | Crossref Full Text | Google Scholar

46. Ketteler, M, Evenepoel, P, Holden, RM, Isakova, T, Jørgensen, HS, Komaba, H, et al. Chronic kidney disease-mineral and bone disorder: conclusions from a kidney disease: improving global outcomes (KDIGO) controversies conference. Kidney Int. (2025) 107:405–23. doi: 10.1016/j.kint.2024.11.013

PubMed Abstract | Crossref Full Text | Google Scholar

47. Lyu, G, Liao, H, and Li, R. Ferroptosis and renal fibrosis: mechanistic insights and emerging therapeutic targets. Ren Fail. (2025) 47:2498629. doi: 10.1080/0886022X.2025.2498629

PubMed Abstract | Crossref Full Text | Google Scholar