Blood donor obesity is associated with changes in red blood cell metabolism and susceptibility to hemolysis in cold storage and in response to osmotic and oxidative stress

Abstract

Background: Obesity is a global pandemic characterized by multiple comorbidities, including cardiovascular and metabolic diseases. The aim of this study was to define the associations between blood donor body mass index (BMI) and RBC measurements of metabolic stress and hemolysis.

Study design and methods: The associations between donor BMI (<25 kg/m² , normal weight; 25-29.9 kg/m² , overweight; and ≥30 kg/m² , obese) and hemolysis (storage, osmotic, and oxidative; n = 18 donors) or posttransfusion recovery (n = 14 donors) in immunodeficient mice were determined in stored leukocyte-reduced RBC units. Further evaluations were conducted using the National Heart, Lung, and Blood Institute RBC-Omics blood donor databases of hemolysis (n = 13 317) and metabolomics (n = 203).

Results: Evaluations in 18 donors revealed that BMI was significantly (P < 0.05) and positively associated with storage and osmotic hemolysis. A BMI of 30 kg/m² or greater was also associated with lower posttransfusion recovery in mice 10 minutes after transfusion (P = 0.026). Multivariable linear regression analyses in RBC-Omics revealed that BMI was a significant modifier for all hemolysis measurements, explaining 4.5%, 4.2%, and 0.2% of the variance in osmotic, oxidative, and storage hemolysis, respectively. In this cohort, obesity was positively associated (P < 0.001) with plasma ferritin (inflammation marker). Metabolomic analyses on RBCs from obese donors (44.1 ± 5.1 kg/m² ) had altered membrane lipid composition, dysregulation of antioxidant pathways (eg, increased oxidized lipids, methionine sulfoxide, and xanthine), and dysregulation of nitric oxide metabolism, as compared to RBCs from nonobese (20.5 ± 1.0 kg/m² ) donors.

Conclusions: Obesity is associated with significant changes in RBC metabolism and increased susceptibility to hemolysis under routine storage of RBC units. The impact on transfusion efficacy warrants further evaluation.

Keywords: BMI; blood donors; hemolysis; obesity; red blood cells.

FIGURE 1

Blood donor BMI is associated with increased hemolysis during cold storage of RBCs and with reduced posttransfusion recovery in NSG mice. Leukoreduced RBC units from 18 donors were stored for 6 weeks and tested weekly as described under methods. Hemolysis measurements were evaluated for each BMI group (<25 kg/m², normal weight, n = 4; 25-29.9 kg/m², overweight, n = 7; ≥30 kg/m², obese, n = 7). A, Percent storage hemolysis. B, Percent osmotic hemolysis. C, Percent AAPH-induced oxidative hemolysis. D, Posttransfusion recovery of 6-week-old leukoreduced RBC transfused into NSG mice. N = 14 of which n = 3 for BMI <25 kg/m², n = 6 for BMI 25-29.9 kg/m², and n = 5 for BMI ≥30 kg/m². Data are represented as mean ± SEM. Asterisks denote significance (P < 0.05, repeated measures two-way ANOVA with Bonferroni’s multicomparison test) of differences between nonobese (<25 or 25-29.9 kg/m²) and obese samples (A); <25 vs ≥30 kg/m² (B); and between 25-29.9 and BMI ≥30 kg/m² (D)

FIGURE 2

Distribution of storage hemolysis, stress-induced hemolysis, or plasma ferritin by blood donor BMI and sex. A-D, Leukoreduced RBCs from male and female donors who participated in NHLBI’s RBC-Omics study (n = 13 197) were stored for 39 to 42 days and tested for storage- or stress-induced hemolysis. A, Percent storage hemolysis by BMI. B, Percent osmotic hemolysis by BMI. C and D, percent oxidative hemolysis by BMI in all donors (C) and first-time donors (D). E and F, Plasma ferritin levels (ng/mL) by BMI in all donors (E) and first-time donors (F). The shaded regions in all plots represent the 95% confidence intervals for the SEM

FIGURE 3

Metabolomics analyses of REDS III RBC-Omics recalled donors as a function of BMI. Blood donors were classified on the basis of BMI (20.5 ± 1.0 kg/m² defined as low BMI vs 44.1 ± 5.1 kg/m² defined as high BMI) before analysis of previously acquired metabolomics data from RBC units stored for 10, 23, and 42 days (A). The heat map in B shows the metabolites significant by ANOVA. A subgroup of metabolites emerged as distinctive between low- and high-BMI subjects, specifically metabolites involved in glutathione homeostasis (and glutathione adducts), lipid metabolism (especially fish oil and free fatty acids), and amino acids (B)

FIGURE 4

Metabolic signatures of low and high BMI in stored RBCs. Representative metabolites are grouped by classes, including (A) lysophospholipids, (B) short-chain fatty acids and short-chain acyl-carnitines, (C) glutathione adducts, (D) oxidant stress markers, (E) arginine metabolism, (F) tryptophan metabolism, and (G) ketogenic and gluconeogenic amino acids. Metabolites are represented as dot plots with superimposed box and whisker plots, on storage Days 10, 23, and 42. Low- and high-BMI subjects are color-coded according to the legend in the bottom right panel of the figure. Nonparametric Wilcoxon test significance for each time point is indicated (* P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001; **** P ≤ 0.0001)