Mind the Gap: Underdosing of Intravenous Iron in a Hospitalized Patient Population With Iron‐Deficiency Anemia

It is important for inpatient physicians to recognize and treat iron-deficiency anemia (IDA) given its prevalence in hospitalized patients [1-6]. Inpatient IV iron repletion can improve outcomes in selected populations and reduce the need for subsequent IV iron treatment and outpatient care [7, 8]. However, despite guideline recommendations [1, 3, 9], existing studies on the adequacy of inpatient iron repletion are limited by small sample sizes, narrow clinical focus, or insufficient assessment of iron dosing [10, 11]. We conducted a single-center retrospective observational study to better understand the practice patterns of inpatient physicians in recognizing and adequately treating IDA. The primary target population was adults aged 18 and older hospitalized between September 1, 2018, and September 30, 2023, at Beth Israel Deaconess Medical Center (BIDMC), an urban, academic medical center in Boston, Massachusetts. The primary outcome was the prevalence of IDA among hospitalized adults, defined as a hemoglobin < 10 g/dL and either (1) ferritin < 45 or (2) ferritin 45–100 and total iron saturation (TSAT) < 20% or (3) ferritin < 300 and TSAT < 20% among patients with chronic kidney disease (CKD) not on hemodialysis (HD), congestive heart failure (CHF), or inflammatory bowel disease (IBD) [12]. Secondary outcomes were (1) proportion of patients with IDA who received IV and/or PO iron during their hospitalization and (2) comparison of the total IV iron amount administered (across the entire hospitalization) with the iron deficit as calculated by the Ganzoni equation. There were 219 566 hospitalizations at BIDMC during the study period. After excluding 2714 psychiatric and 10 537 hospitalizations without hemoglobin values, 40 930 hospitalizations (19.8%) among 15 931 unique anemic patients comprised our analytic cohort. Cohort demographics are detailed in Table 1. Four thousand five hundred and sixty-seven individuals (11.2%) met the criteria for IDA, though only 31.5% had iron studies sent during their hospitalization (Figure 1). A higher proportion of admissions to hospital medicine (43.0%) and cardiology (42.6%) had iron studies sent relative to other services. Among patients with IDA, the median ferritin was 48 (IQR, 18–73) and the median MCV was 86 (78.5–98.5). In the setting of either CKD, CHF, or IBD, 47.3% (2159) had a ferritin < 45; 26.7% (1219) had a ferritin 45–100 and a TSAT < 20%; and 26.0% (1189) had a ferritin < 300 and a TSAT < 20%. Among those with a ferritin < 45 and MCV < 80, the median ferritin was 17.0 (IQR, 9.7–27.0) and the median MCV was 73.0 (IQR, 68.0–76.0) (Table S1). Patients with IDA admitted to the Ob-Gyn service had the lowest median ferritin at 16.0 (IQR, 9.9–27.0), followed by those admitted to neurology (32 [IQR, 14.5–67.5]), oncology (43 [IQR, 24–86]), and hospital medicine (44.0 [IQR, 21.9–95.0]) (Table S1). Only 2475 individuals with IDA (54.1%) received any type of iron supplementation during their hospitalization (Figure 1). Of those, 72.5% (1794) received IV iron either as monotherapy or combined with PO iron (Figure 1). Patients admitted to the cardiology and Ob-Gyn services received IV iron at the highest rates (47.4% and 46.1%, respectively) (Table S2). Among the 991 patients with a ferritin < 45 and MCV < 80, the proportion of patients who received any type of iron supplementation during their hospitalization was higher at 66.3% (Figure 1). However, of those who received any iron, the proportion who received IV iron (either as monotherapy or combined with PO iron) was similar to the overall cohort with IDA (76.0%). From 2018 to 2023, a growing proportion received IV iron, from 32.6% to 47.3% for all patients with IDA and from 25.0% to 59.3% for patients with a ferritin < 45 (Table S2). Throughout all study years, ferric gluconate was the most commonly administered IV iron medication, followed by iron dextran and ferumoxytol. The proportion of iron dextran administered rose from 0% in 2018 to 12.7% in 2023 (Table S3). Of 1794 individuals who received IV iron, 71.4% (1280) received a total IV iron amount of < 1000 mg (mg), and the rest received a total amount of ≥ 1000 mg. Among those who received IV iron during their hospitalization, 67.6% (1212) had a weight recorded, allowing calculation of the iron deficit using the Ganzoni equation. The median calculated iron deficit was 1847 mg, and the median total IV iron amount administered was 500 mg, so the median total difference between the calculated iron deficit and the total administered IV iron amount was 1364 mg (Figure 2a). The median total ferric gluconate administered was 375 mg (IQR, 125–750), the median total iron dextran administered was 1025 mg (IQR, 1000–1025), and the median total ferumoxytol administered was 510 mg (IQR, 510–1020). For those with a ferritin < 45 who received IV iron during their hospitalization, 76.0% (379) received a total IV iron amount of < 1000 mg, and 64.9% (324) had a weight recorded, allowing calculation of the iron deficit using the Ganzoni equation. The median calculated iron deficit was 1911 mg, and the median total IV iron amount administered was 375 mg, so the median total dose difference between the calculated iron deficit and the total administered IV iron amount was 1552 mg (Figure 2b). To our knowledge, this is the first study evaluating IV iron use and dosing in a large multi-year and multi-service hospitalized cohort. We found that IDA was both under-recognized and undertreated, as 70% of patients with anemia did not have iron studies sent, and only approximately half of those with IDA received any iron during their hospitalization. Iron studies may not be routinely sent because anemia is rarely the primary problem of hospitalized patients, although in one retrospective study of adults hospitalized with severe gastrointestinal bleeding, iron studies were also only ordered 50% of the time, despite severe anemia [11]. Further, providers relying on microcytosis to prompt testing would underdiagnose IDA, as this is a late finding and most iron-deficient patients are normocytic; indeed, in our cohort of patients with IDA, the median MCV was 86, even though the median ferritin was as low as 48. Finally, providers may believe that ferritin is uninterpretable in the presence of inflammation or use overly stringent cutoffs, which risks underdiagnosing IDA in inpatients. Ferritin functions as an acute-phase reactant and may be normal or elevated despite depleted iron stores. In hospitalized people, iron deficiency is therefore best assessed using ferritin thresholds in combination with TSAT. Ferritin < 100 ng/mL with TSAT < 20% suggests iron deficiency, and higher ferritin thresholds (up to 300 ng/mL) with TSAT < 20% are commonly used in the setting of CKD, CHF, and IBD [12]. Nonetheless, in our cohort, nearly three quarters of patients had a ferritin < 100 despite being hospitalized. Only approximately half of hospitalized patients with IDA received iron supplementation, which might reflect under-recognition of IDA, as approximately two third of those with a ferritin < 45 and MCV < 80 received iron. Most individuals who were treated received IV iron, though the doses administered were substantially lower than the calculated iron deficit (median total dose difference 1364 mg). Even without formally calculating the deficit, almost three fourth received a total IV iron amount of < 1000 mg. The median total dose difference was even more pronounced in patients with a ferritin < 45 (1552 mg), suggesting knowledge gaps in appropriate IV iron repletion. In addition, the majority of IV iron administered was ferric gluconate, which is suboptimal for those not on dialysis as it often requires multiple doses to replete the iron deficit, compared with single-dose infusions with iron dextran, ferumoxytol, or ferric derisomaltose [2]. Given that the median total ferric gluconate administered was 375 mg versus 1025 mg for iron dextran, providers using ferric gluconate often undertreated the iron deficiency despite a similar time investment. This highlights the benefit of single-dose IV iron administration for the treatment of iron deficiency [2]. This pattern of consistent underdosing may lead to anemia symptoms being misattributed to other causes in the future. Furthermore, prior studies have shown that adequate iron repletion is associated with hemoglobin normalization within 8 weeks with fewer outpatient visits for all causes, including for IV iron therapy [7, 8]. Beyond knowledge gaps, underdosing of IV iron may also stem from the current model of inpatient care delivery that favors acute stabilization over chronic disease management. For example, in Medicare-aged patients, diagnosis-related group-based reimbursement does not account for the intensity of therapies administered, potentially disincentivizing full inpatient iron repletion. Taken together, these findings suggest that both provider education and systems-level interventions are needed to improve inpatient diagnosis and management of IDA. Encouragingly, IV iron use, especially IV iron dextran, increased between 2018 and 2023, likely reflecting ongoing educational efforts. Despite evaluating over 200 000 hospitalizations, this was a single-center retrospective study, and results should be validated at other centers with more diverse demographics. Further, we defined anemia with a hemoglobin < 10 g/dL in both men and women to avoid over-diagnosis of anemia and IDA, which has excluded some cases of mild IDA. To conclude, IDA in hospitalized adults is frequently under-recognized and undertreated, with substantial underdosing of IV iron, highlighting the need for standardized approaches to diagnosis and iron repletion. The authors have nothing to report. No previously published material was reproduced in this manuscript. This study was approved by the IRB at Beth Israel Deaconess Medical Center (IRB Protocol # 2023P000254). Patient consent was waived due to the retrospective design and anonymization of all data. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Due to institutional privacy regulations, individual-level patient data cannot be publicly shared. Table S1: Laboratory values by admission service. Table S2: IV iron administration by hospitalization characteristics. Table S3: IV iron formulation administered in patients with IDA, by year. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.