MRI-detected hepatic and cardiac iron overload despite ferritin-guided chelation in thalassemia: A predictive modelling study
4 affiliations
- 1Department of Pediatric Hematology Oncology, Post Graduate Institute of Child Health (PGICH), Noida, Uttar Pradesh, India
- 2Department of Cardiovascular Radiology and Endovascular Interventions, All India Institute of Medical Sciences, New Delhi, India
- 3Department of Pediatric Hematology Oncology, Post Graduate Institute of Child Health (PGICH), Noida, Uttar Pradesh, India; Corresponding author. Department of Paediatric Hematology Oncology, Post Graduate Institute of Child Health, Noida, India.
- 4Department of Hematology, All India Institute of Medical Sciences, New Delhi, India
Abstract
Background: Serum ferritin while used widely, is often unable reflect organ-specific iron burden in thalassemia. We evaluated the adequacy of ferritin-guided chelation in predicting iron overload and examined determinants of ferritin–MRI discordance. Methods: 65 children and young adults with thalassemia undergoing MRI T2∗ assessment were included. Serial ferritin measurements over prior 6–12 months were used to derive mean ferritin and variability. Primary outcomes were myocardial iron overload (cardiac T2∗ <20 ms) and moderate-to-severe hepatic iron overload. Predictors spanning iron exposure, disease biology, and ferritin dynamics were analysed using Firth penalized logistic regression, with model performance assessed by bootstrap-corrected AUC. Results: Hepatic iron overload was present in 77% and myocardial iron overload in 22% of patients despite ferritin-guided chelation. Ferritin showed moderate correlation with cardiac T2∗ (r = −0.51) and weaker correlation with hepatic T2∗ (r = −0.40), with substantial inter-individual variability. In multivariable analysis, age, transfusion-dependent status, and ferritin variability were independently associated with myocardial iron overload, whereas mean ferritin was not. Model discrimination was good (AUC 0.83). Ferritin–MRI discordance was observed in 38% of patients. Conclusion: Ferritin-guided chelation does not reliably predict organ-specific iron burden. Incorporating MRI is helpful for accurate detection of clinically significant iron overload.
Related
- Iron Chelation Therapy in Iron Overload Disorders: A Systematic Review with Narrative Synthesis of Bioenergetic Dysfunction, Oxidative Stress, and Biomarker-Guided Management.
- Quality by Design (QbD) Empowered by IVIVR Modelling: A Predictive Approach for Drug Formulation Development
- Predictive modelling to differentiate possible bacterial and viral cases of childhood pneumonia in Kilifi, Kenya using protein markers and clinical data
- Predictive modelling approaches for childhood diarrhoeal disease: a systematic review (2000-2025).
- Non-pharmacological interventions as adjuncts to iron chelation therapy in sickle cell disease on chronic transfusion: a scoping review with evidence gap mapping.
BibTeX
@article{verma2026mridetected,
title = {{MRI-detected hepatic and cardiac iron overload despite ferritin-guided chelation in thalassemia: A predictive modelling study}},
author = {Shruti Verma and Aprateem Mukherjee and Nita Radhakrishnan and Vineeta Ojha and Aditi Tulsiyan and Sudipto Bhattacharya and Anuj Singh and Usha Bindal and Tulika Seth and Priya Jagia},
journal = {Pediatric Hematology Oncology Journal},
year = {2026},
month = oct,
doi = {10.1016/j.phoj.2026.101178},
url = {https://doi.org/10.1016/j.phoj.2026.101178},
note = {CC BY},
}
RIS
TY - JOUR
TI - MRI-detected hepatic and cardiac iron overload despite ferritin-guided chelation in thalassemia: A predictive modelling study
AU - Shruti Verma
AU - Aprateem Mukherjee
AU - Nita Radhakrishnan
AU - Vineeta Ojha
AU - Aditi Tulsiyan
AU - Sudipto Bhattacharya
AU - Anuj Singh
AU - Usha Bindal
AU - Tulika Seth
AU - Priya Jagia
PY - 2026
DA - 2026/10/30
JO - Pediatric Hematology Oncology Journal
DO - 10.1016/j.phoj.2026.101178
UR - https://doi.org/10.1016/j.phoj.2026.101178
AB - Background: Serum ferritin while used widely, is often unable reflect organ-specific iron burden in thalassemia. We evaluated the adequacy of ferritin-guided chelation in predicting iron overload and examined determinants of ferritin–MRI discordance. Methods: 65 children and young adults with thalassemia undergoing MRI T2∗ assessment were included. Serial ferritin measurements over prior 6–12 months were used to derive mean ferritin and variability. Primary outcomes were myocardial iron overload (cardiac T2∗ <20 ms) and moderate-to-severe hepatic iron overload. Predictors spanning iron exposure, disease biology, and ferritin dynamics were analysed using Firth penalized logistic regression, with model performance assessed by bootstrap-corrected AUC. Results: Hepatic iron overload was present in 77% and myocardial iron overload in 22% of patients despite ferritin-guided chelation. Ferritin showed moderate correlation with cardiac T2∗ (r = −0.51) and weaker correlation with hepatic T2∗ (r = −0.40), with substantial inter-individual variability. In multivariable analysis, age, transfusion-dependent status, and ferritin variability were independently associated with myocardial iron overload, whereas mean ferritin was not. Model discrimination was good (AUC 0.83). Ferritin–MRI discordance was observed in 38% of patients. Conclusion: Ferritin-guided chelation does not reliably predict organ-specific iron burden. Incorporating MRI is helpful for accurate detection of clinically significant iron overload.
SN - 2468-1245
ER -