Prevalence of iron deficiency among youth with sickle cell disease

Abstract Sickle cell disease (SCD) is an inherited hemoglobinopathy affecting up to 100,000 people in the United States. While pain, acute chest syndrome, and anemia are common, silent cerebral infarcts and overt strokes that may result in long-term cognitive deficits can also occur in many affected youth with SCD (Y-SCD). Like SCD, iron deficiency (ID) can also cause anemia and cognitive deficits because of its role in hemoglobin production, myelination, neurogenesis, and connectivity. Despite the high prevalence of ID in the general population and the potential for ID to contribute to anemia and cognitive deficits, the prevalence of ID among Y-SCD remains unclear. Thus, the aims of this study were to determine the prevalence of ID (i.e., ferritin <50 ng/mL) and possible ID (i.e., ferritin 50-99 ng/mL) among Y-SCD at Nationwide Children's Hospital (NCH) who had a ferritin level obtained, and to describe the characteristics of those diagnosed with ID. We also aimed to determine if those with ID had significantly lower hemoglobin (Hb) and mean corpuscular volume (MCV) than those with possible ID. The NCH SCD clinical database was used to identify Y-SCD, ages 6 months-21 years, who had ≥1 outpatient visit from 2015-2023. Those with a ferritin level checked were considered to have been tested for ID. Demographic (e.g., race, age, sex), laboratory, and clinical (e.g. reason for ferritin assessment, hydroxyurea (HU) use) data were abstracted from the electronic medical records of those with ID and possible ID. Those who had a bone marrow transplant prior to having a ferritin level checked were excluded. While ferritin testing for all Y-SCD at NCH is not standard practice, it is ordered for iron overload monitoring among chronically transfused individuals or if youth have symptoms (e.g., PICA) or laboratory findings (e.g., microcytosis) consistent with ID. Descriptive statistics summarized the data. Linear regression models were used to analyze the relationship between ID status and hemoglobin and ID status and MCV. Models were adjusted for age, HU use, and SCD genotype. P-values <0.05 were considered significant. Statistical analyses were carried out using R version 4.5.1. We identified 550 youth followed at NCH during the study period. Of these, 221 (40%) had a ferritin level obtained, and 114 (52%) had a ferritin <100ng/mL (54% male (n=61), 92% Black/African American (n=105), mean age of 11.4 (SD=6.1)). Sixty-five (57%) participants had HbSS disease, and 35 (31%) were taking HU at the time of having their ferritin level checked. Of these, 47 (41%) were ID and 67 (59%) had possible ID. Mean Hb (g/dL) in the ID group was 10.2 (SD: 2.0) and 9.6 (SD: 1.6) in the possible ID group. The estimated mean difference between these groups was <0.01 (95% CI: -0.62, 0.63, p>0.9). Mean MCV (fL) was 69 (SD: 14) in the ID group and 81 (SD:12) in the possible ID group. The estimated mean difference was 6.4 fL higher in the possible ID group (95% CI: 2.3, 10, p=0.003). The most common reason for obtaining ferritin was iron overload monitoring for chronic transfusions (n=37, 32%), followed by undocumented reasons (n=32, 28%), and microcytosis (n=17, 15%). We identified that at least 9% of Y-SCD at our center had ID and more than 50% of Y-SCD who were tested had ID or possible ID. While this prevalence aligns with that of ID among youth in the general population, these findings are likely an underestimate because ferritin can be elevated in those with hyperinflammatory conditions like SCD, screening for ID is not standard for Y-SCD, and it was not performed in 60% of our youth. Surprisingly, chronic transfusions were the most common documented reason why youth with SCD with ID or possible ID were identified, suggesting that the methods used for chronic transfusion (i.e., erythrocytapheresis) might be contributing and/or that there may be underlying features of this subset of Y-SCD that predispose them to ID. Finally, our finding that those with ID had lower MCV than those with possible ID could suggest that improving iron stores has the potential to improve hematologic outcomes. Given our small sample size and single-center retrospective design, future studies are warranted to confirm and characterize the prevalence of ID among Y-SCD, determine the cause(s) of ID in Y-SCD, and determine if normalizing iron stores improves anemia, cognition, and patient-reported outcomes.