Iron Levels Shift in Fatty Liver Disease, but Fibrosis Link Stays Unclear

Patients with metabolic dysfunction-associated fatty liver disease show a distinctive pattern of changes in routine blood iron markers, according to a prospective case-control study conducted at Ain Shams University Hospitals in Egypt. In a cohort of 60 age- and sex-matched individuals, the 30 patients with the condition had significantly higher serum ferritin and total iron-binding capacity than 30 healthy controls, together with significantly lower serum iron and transferrin saturation. The study, published in the Egyptian Liver Journal, also found that these iron profile abnormalities were not significantly associated with liver stiffness measurements in this cohort, tempering hopes that a simple iron panel could flag fibrosis risk. The findings, the authors argue, point to altered systemic iron homeostasis in metabolic fatty liver disease rather than definitive evidence of iron overload.

Metabolic dysfunction-associated fatty liver disease, often abbreviated MAFLD, is a common chronic liver condition closely linked to insulin resistance, obesity, type 2 diabetes, and metabolic syndrome. The term was introduced in 2020 to reframe what had been called non-alcoholic fatty liver disease, shifting the emphasis from the exclusion of alcohol toward the presence of metabolic dysfunction. A diagnosis requires objective evidence of hepatic steatosis together with overweight or obesity, type 2 diabetes, or metabolic dysregulation. The disease spans a broad clinical spectrum, from uncomplicated fat accumulation in the liver to steatohepatitis, fibrosis, cirrhosis, and hepatocellular carcinoma, and its global burden has risen substantially in recent decades in parallel with the spread of metabolic disease. This burden is particularly relevant in the Middle East and North Africa, where fatty liver disease and metabolic risk factors including obesity are highly prevalent.

Iron has long been suspected to play a role in the disease process. The element is essential for cellular respiration and oxygen transport, but excess labile iron can promote lipid peroxidation and the formation of reactive oxygen species, mechanisms that could plausibly feed hepatic inflammation and fibrogenesis. Altered iron handling has been described in metabolic fatty liver disease before, most notably in dysmetabolic iron overload syndrome, which is generally characterized by hyperferritinemia with normal or only mildly increased transferrin saturation. Hepcidin signaling, systemic inflammation, and intracellular iron trafficking may all influence iron homeostasis. Yet, as the study’s authors note, differences in standard iron indices and their relationship with the disease remain incompletely characterized, particularly within Egyptian cohorts, a population that remains underrepresented in the existing literature.

The research team enrolled participants prospectively between December 2025 and March 2026 from the inpatient wards and outpatient clinics of the Department of Internal Medicine, Gastroenterology and Hepatology at Ain Shams University Hospitals. The study included 30 patients with MAFLD and 30 healthy controls matched for age and sex, with controls required to have normal baseline liver biochemistry, no ultrasonographic evidence of hepatic steatosis, and no metabolic syndrome criteria. Adults aged 18 years or older were eligible if they fulfilled the established MAFLD diagnostic criteria, with hepatic steatosis confirmed either by a Hamaguchi ultrasonographic score of at least 2 or by vibration-controlled transient elastography with a controlled attenuation parameter of at least 248 dB/m. Exclusion criteria were designed to keep iron metabolism clean of confounders: significant alcohol consumption, chronic liver disease of another etiology, secondary iron-storage disorders, therapeutic phlebotomy or iron supplementation within the preceding 90 days, active malignancy, acute inflammatory conditions, pregnancy, and decompensated cirrhosis were all grounds for exclusion.

Each participant underwent a detailed medical history and physical examination, with anthropometric assessment covering height, body weight, body mass index, waist circumference, and resting blood pressure. After an overnight fast of at least eight hours, venous blood was drawn for a complete blood count, renal and liver function tests, fasting lipids, fasting glucose and insulin, and a full iron profile comprising serum iron, serum ferritin, total iron-binding capacity, and transferrin saturation, the last calculated as serum iron divided by TIBC multiplied by 100. Analyses were performed in the routine clinical laboratories of Ain Shams University Hospitals on Beckman Coulter DxC 700 AU and/or Roche cobas 6000 chemistry analyzers. Liver fat and fibrosis were assessed non-invasively using vibration-controlled transient elastography, with an XL probe used in participants with obesity, and fibrosis was staged using NAFLD-specific liver stiffness thresholds ranging from under 7.0 kilopascals for absent or mild fibrosis to 10.3 kilopascals or above for cirrhosis.

The metabolic profile of the patient group was consistent with the disease’s known signature. Compared with controls, patients with MAFLD had significantly greater body weight, body mass index, and waist circumference, along with significantly higher glycemic indices, triglyceride levels, high-sensitivity C-reactive protein, and gamma-glutamyl transferase, together with significantly lower HDL cholesterol and serum albumin, although the albumin median remained within the conventional reference range. No significant differences emerged in cell counts, aminotransferases, bilirubin, coagulation measures, renal function, total cholesterol, or LDL cholesterol. On liver imaging, the Hamaguchi ultrasound score, controlled attenuation parameter, liver stiffness measurement, and NAFLD fibrosis score were all significantly higher in the patient cohort, while FIB-4 scores did not differ significantly between groups. Controlled attenuation parameter staging showed S3 steatosis, the most severe grade, in 63.3 percent of the patients with MAFLD.

The central iron findings were unambiguous in direction. Serum ferritin and total iron-binding capacity were significantly higher in patients with MAFLD, whereas serum iron and transferrin saturation were significantly lower than in healthy controls. Notably, the authors point out that this biochemical pattern does not correspond to classical dysmetabolic iron overload syndrome, which typically features elevated ferritin with normal or only mildly increased transferrin saturation; in this cohort, transferrin saturation was reduced. Receiver operating characteristic analysis showed statistically significant discrimination between cases and controls for all four iron parameters, with transferrin saturation achieving the highest area under the curve at 0.878, followed by serum iron at 0.847, serum ferritin at 0.823, and total iron-binding capacity at 0.776. At an internally derived cutoff of at least 339 micrograms per deciliter, total iron-binding capacity yielded 100 percent specificity, though the authors caution that because this cutoff was derived and evaluated in the same cohort, the estimate may be optimistic and requires external validation.

The fibrosis story was more sobering. Vibration-controlled transient elastography indicated significant fibrosis, defined as liver stiffness of at least 7 kilopascals, in 26.7 percent of the patients, or eight individuals. Patients with significant fibrosis had numerically lower serum iron and transferrin saturation and higher ferritin than those with absent or mild fibrosis, but none of the individual iron profile parameters, including total iron-binding capacity, differed significantly between the fibrosis strata. Within the MAFLD group, Spearman correlation analysis showed no statistically significant associations between circulating iron markers and liver stiffness, the controlled attenuation parameter, the Hamaguchi ultrasound score, FIB-4, or the NAFLD fibrosis score. Most anthropometric and metabolic variables also failed to correlate significantly with iron markers, with only scattered weak correlations emerging, including weak positive links between total iron-binding capacity and AST, total cholesterol, and LDL cholesterol, a weak positive correlation between transferrin saturation and HbA1c, and a moderate inverse correlation between ferritin and systolic blood pressure.

Regression analyses added a metabolic dimension to the iron story. Univariate logistic regression identified serum ferritin, serum iron, transferrin saturation, body mass index, HOMA-IR, and triglycerides as significantly associated with MAFLD case status. An exploratory multivariable model, restricted to three predictors to limit overfitting given the small number of events, found that HOMA-IR and serum ferritin remained independently associated with case status, whereas serum iron did not. Transferrin saturation was excluded from the model because it is mathematically derived from serum iron and total iron-binding capacity, and body mass index and triglycerides were excluded by the three-variable restriction. The authors emphasize that the HOMA-IR estimate carried a wide confidence interval, reflecting substantial imprecision and possible model instability, and that the control group’s deliberate freedom from metabolic syndrome criteria accentuated metabolic differences between cases and controls. These findings should therefore be considered associative and hypothesis-generating rather than evidence of independent predictive performance.

Several limitations frame the conclusions. The study was conducted at a single center with a modest sample, particularly within the fibrosis subgroups, which limited the power of subgroup analyses, and the cross-sectional design precludes causal inference. Fibrosis was assessed using elastography-derived liver stiffness rather than histology, so the absence of a significant association with iron parameters should not be read as evidence against a relationship with histologically defined fibrosis. Hepcidin was not measured and hepatic iron stores were not assessed directly, so the study cannot establish hepatic iron overload or confirm dysmetabolic iron overload syndrome, and the case-control design may have introduced spectrum bias that overestimated discriminative performance compared with heterogeneous clinical populations. The authors call for further studies incorporating hepcidin measurement and direct assessment of hepatic iron stores in larger, clinically representative cohorts to clarify whether the observed alterations reflect altered iron trafficking, functional iron restriction, iron accumulation, or another disturbance of iron homeostasis.

Subject of Research: Alterations in serum iron profile parameters in metabolic dysfunction-associated fatty liver disease and their association with liver fibrosis

Article Title: Serum iron profile alterations in metabolic dysfunction-associated fatty liver disease and their association with liver fibrosis: a case-control study

Article References: Elmarashly, B., Wilson, M. S., Elkarmouty, K., & Raafat, K. (2026). Serum iron profile alterations in metabolic dysfunction-associated fatty liver disease and their association with liver fibrosis: a case-control study. Egyptian Liver Journal, 16(1), Article 75. https://doi.org/10.1186/s43066-026-00560-1

Image Credits: AI Generated

DOI: 10.1186/s43066-026-00560-1

Keywords: MAFLD, serum ferritin, transferrin saturation, total iron-binding capacity, liver fibrosis, transient elastography, insulin resistance, HOMA-IR, case-control study, iron homeostasis, hepcidin, Egyptian Liver Journal