Abstract
Iron deficiency (ID) is a common and clinically significant comorbidity in patients with heart failure (HF), contributing to reduced exercise capacity, poor quality of life, and increased hospitalization risk. Although intravenous (IV) iron therapy has demonstrated efficacy in improving functional outcomes, the comparative effectiveness of IV vs. oral (per os, or PO) iron supplementation remains uncertain. We conducted a systematic review and network meta-analysis (NMA) of 13 randomized controlled trials (RCTs) evaluating IV iron (ferric carboxymaltose, ferric derisomaltose, iron sucrose), PO iron (ferrous sulfate, ferrous fumarate, polysaccharide, sucrosomial, ferric polymaltose), and placebo in HF patients with ID, analyzed as route-specific class effects. Outcomes analyzed included six-minute walk distance (6MWD), ferritin, transferrin saturation (TSAT), HF hospitalization, all-cause mortality, and cardiovascular (CV) mortality. We used a Hartung-Knapp random-effects framework with Sidik-Jonkman variance, assessed heterogeneity and inconsistency using I(2), τ(2), design-by-treatment interaction, and node-splitting. Risk of bias was assessed independently by two reviewers using Risk of Bias 2 (RoB 2), and certainty of evidence for all outcomes was graded using GRADE adapted for NMA. Because most contrasts included fewer than 10 RCTs, formal tests for publication bias were not feasible, and potential small-study effects were considered qualitatively in the GRADE assessments. Trials that reported outcomes only as medians and interquartile ranges (IQRs), or baseline values without follow-up data, were excluded from quantitative pooling and described narratively. IV iron significantly improved 6MWD compared to placebo (mean difference (MD) +26.0 m; 95% confidence interval (CI): 18.1 to 33.9), increased ferritin (MD +237.2 μg/L), and reduced the risk of HF hospitalization (risk ratio (RR) 0.79; 95% CI: 0.66 to 0.93), with moderate to high certainty. PO iron showed a comparable, but not statistically significant, mean improvement in 6MWD (MD +35.1 m; 95% CI: -5.2 to +75.4), with wider CIs and inconsistent ferritin and TSAT gains. Neither IV nor PO iron was associated with a significant reduction in all-cause or CV mortality, although a trend toward benefit was observed with IV therapy. Numerical SUCRA values favored IV iron for HF hospitalization (77.9 vs. 57.6 for PO, 14.5 for placebo), ferritin (100.0 vs. 50.0 vs. 0.0), and TSAT (74.0 vs. 75.8 vs. 0.2), while PO iron ranked slightly higher for 6MWD (76.3 vs. 73.7 vs. 0.0). Included PO formulations encompassed both traditional preparations (ferrous sulfate/fumarate, polysaccharide) and newer agents such as sucrosomial iron and ferric polymaltose. Adverse events were comparable across groups: IV iron was not associated with excess mortality or serious adverse events, and PO iron was primarily limited by gastrointestinal intolerance. Sensitivity analyses restricting outcomes to trials with 3-12 months of follow-up showed consistent results, while longer studies mainly influenced event counts rather than the direction of effect. Our findings support the use of IV iron as the preferred strategy to improve symptoms and reduce hospitalizations in HF patients with ID, whereas PO iron may be considered when IV therapy is inaccessible. Further large-scale trials are needed to clarify long-term mortality impact and the role of newer PO formulations.