National cohort study reveals challenges in diagnosing iron deficiency anaemia in cirrhosis, suggesting improvements in testing practices.
• Current testing practices for iron deficiency anaemia (IDA) reveal missed opportunities for diagnosis and management in cirrhosis. • Ferritin and MCV remain useful for IDA diagnosis in cirrhosis but current thresholds are unable to exclude hemoglobin response to oral iron. Diagnosing iron deficiency anaemia (IDA) in patients with co-morbidities is challenging due to inflammatory effects on biomarkers. Using real-world data, we described testing practices, IDA incidence, and oral iron response among cirrhosis patients. We conducted a retrospective cohort study using Clinical Practice Research Datalink (CPRD) Aurum (2015–2021), including 43,083 adults with cirrhosis in primary care and a haemoglobin measurement. We assessed ferritin testing determinants using Cox regression with time-dependent covariates. IDA was defined using ferritin thresholds (<30, <50, <70, <100 μg/L) within 90 days of anaemia. Haemoglobin response (≥10 g/L increase at 3 months) to oral iron was analysed using multivariable logistic regression. Ferritin testing occurred in 54.5% of patients; 59.6% had prevalent anaemia. Only 56.7% of anaemic patients had ferritin measured within 90 days of low haemoglobin. Males aged 18–49 were less likely to be tested (HR 0.75, 95%CI 0.68–0.84). Among tested patients, 28.4% had ferritin <30 μg/L. IDA incidence increased with higher thresholds, from 4.68 per 100 person-years (<30 μg/L) to 8.27 per 100 person-years (<100 μg/L). Mean cell volume (MCV) <80 fL increased testing likelihood (HR 1.40, 95%CI 1.30–1.50), however MCV 80–95 fL did not. MCV sensitivity for detecting IDA declined from 91.1% to 80.2% with higher ferritin thresholds. Oral iron was associated with increased odds of haemoglobin response (adjusted OR 2.51, 95%CI 1.79–3.54), particularly in patients with low ferritin or MCV. IDA is frequently under-evaluated in cirrhosis and consistent testing approaches are needed across ages and sexes. Research examining clinical outcomes and treatment impact is required.
No takes yet. Share an insight, caveat, or question.
Maynard et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: