Abstract
Background and Objectives
Granulocyte transfusion supports patients with severe neutropenia. Maintaining a pool of eligible donors and optimizing donation frequency are essential for ensuring an adequate supply while safeguarding donor well‐being. This study investigates the impact of donation frequency on erythrogram parameters, focusing on sex‐specific differences.
Study Design and Methods
We conducted a retrospective analysis of 343 successive granulocyte collections from 65 apheresis donors over 11 years (2012–2023). Donors were categorized by sex, and erythrogram parameters were analysed in relation to donation frequency and intervals.
Results
Frequent donations within a short inter‐donation interval (≥3 in 14 days) affected subsequent pre‐donation haemoglobin levels. Each additional donation within 14 days led to a decrease of 0.81 g/dL in haemoglobin (p = 0.017). A significant interaction between sex and donations within 14 days (β = 0.76, p = 0.018) indicated that frequent donations had a more pronounced negative effect on haemoglobin levels in female donors. The proportion of donations meeting the pre‐donation haemoglobin eligibility criteria declined with each successive donation within 14 days (100% at first, 85.8% at second, 25% at third). Female donors showed a significant haemoglobin reduction over three donations within 14 days (13.4–11.6 g/dL, p = 0.005) compared to males (14.4 –14 g/dL, p = 0.95).
Conclusion
Short inter‐donation intervals have a more pronounced negative effect on pre‐donation haemoglobin levels in female donors, underscoring the need for individualized donation guidelines to ensure donor safety.
Keywords: donor eligibility criteria, donor well‐being, granulocyte transfusion, repeat granulocytapheresis
Highlights.
Female donors exhibited a more pronounced decline in pre‐donation haemoglobin levels compared to male donors, particularly when donating frequently within short intervals.
Successive granulocyte donations within a 14‐day period significantly reduced haemoglobin levels in both sexes, with a marked decrease in the eligibility of female donors for subsequent donations.
The findings of this study suggest the need for sex‐specific donation guidelines to optimize donor safety and maintain an adequate supply of granulocyte products, especially in female donors.
INTRODUCTION
Granulocyte transfusion has been used in treating bacterial and/or fungal antibiotic‐refractory infections in patients with severe neutropenia or with impaired neutrophil function [1, 2]. The efficacy of granulocyte transfusions is primarily determined by the transfusion of an adequately high dose of granulocytes [2]. Advanced cell separation technology and the use of granulocyte colony‐stimulating factor (G‐CSF)/dexamethasone stimulation enable the collection of these high‐dose granulocytes [3, 4, 5].
To ensure an effective course of granulocyte transfusion, it is crucial to have a pool of eligible donors to maintain frequent transfusions considering that allogeneic granulocytes have a very short shelf life [6]. However, this must be balanced with the need to safeguard donor well‐being. Frequent blood donation raises health concerns for repeat donors, especially women, as the physiological demands can affect haemoglobin levels and iron stores, potentially leading to anaemia [7]. Existing guidelines from the American Association of Blood Banks (AABB) and European Directorate for the Quality of Medicines & HealthCare (EDQM) provide a framework to minimize health risks for granulocyte donors through regulated donation frequency and intervals [8, 9]. However, the impact of these factors on erythrogram parameters, particularly in the context of potential sex‐specific differences, remains understudied.
Previous studies have investigated the impact of repeated granulocytapheresis on donor health [10, 11, 12, 13, 14, 15, 16]. However, only a few have focused on longitudinal haematopoietic changes associated with repeat G‐CSF‐stimulated granulocyte donation [11, 14, 16]. Notably, the role of donor sex in these longitudinal haematopoietic changes has been examined in just one study to date [11].
The findings of these studies reveal a combination of consistent and variable results. Reported decreases in haemoglobin ranged from modest reductions of about 5% to more pronounced haematocrit declines of 12%–15% following leukapheresis [10, 14]. Under more intensive donation protocols, progressive declines in haemoglobin levels were documented, dropping from 14.2 to 12.5 g/dL over five consecutive days of daily donations [16]. In contrast, long‐term studies showed no significant persistent anaemia, with haemoglobin levels remaining stable over extended periods post‐donation [13].
These variations highlight how donor and donation‐specific factors influence haematological outcomes. Sex is a key determinant, as female donors are more susceptible to significant declines in haemoglobin levels during frequent donations. Worel et al. reported that female donors are more likely to experience haemoglobin declines below eligibility thresholds following frequent donations [16]. In contrast, Quillen et al. employed sex‐matched controls in their analysis but did not explore sex‐specific haematological differences [13]. Studies by Axdorph Nygell et al. and Bensinger et al. have not explicitly addressed sex‐based effects [12, 15].
Donation frequency has also been a focal point in understanding erythrogram parameters post‐donation. Buchholz et al. reported that donors can safely undergo up to eight granulocytapheresis procedures within a 10‐day span without severe haematological effects [14]. However, Worel et al. reported progressive declines in haemoglobin with daily serial donations [16]. Furthermore, lifetime donation history adds further complexity. Szymanski et al. found persistent declines in granulocyte and lymphocyte counts correlated with increased donation numbers, although the interplay with erythrogram parameters was not explored [11]. Additionally, the type of apheresis device used has also been implicated in modulating haematological responses, with older apheresis devices associated with greater losses of red blood cells (RBCs) compared to newer models [17, 18].
Given these complex interactions, a comprehensive longitudinal analysis is needed to evaluate individual erythrogram parameters in response to granulocytapheresis frequency and to examine the interplay between donor and donation variables. This study aims to provide insights into the sex‐specific effects of donation frequency and interval on donor haemoglobin levels, with the goal of informing donation guidelines that prioritize donor safety while maintaining an adequate blood supply for patients in need.
STUDY DESIGN AND METHODS
Study cohort
Donors included in this study were healthy, volunteer, allogeneic granulocyte donors at the National Institutes of Health (NIH) Clinical Center, between 1 April 2012 and 31 December 2023, under a single institutional review board (IRB)‐approved research protocol (NCT01553214), which entailed written informed consent (Table 1). In addition to meeting the eligibility requirements for allogeneic blood donors, individuals with pre‐existing medical conditions such as uncontrolled hypertension, peptic ulcer disease and posterior subcapsular cataracts were not considered as suitable candidates. Donors who did not meet the pre‐donation haemoglobin eligibility criterion were excluded from the study, except in cases of urgent medical need where the treating physician could request an authorized medical exemption subject to review and approval by the Blood Bank medical director. All donors received 480 mcg subcutaneous injection of G‐CSF (filgrastim) 18–24 h pre‐leukapheresis alone or in combination with 8 mg dexamethasone orally 12 h before leukapheresis.
TABLE 1.
Donor and donation demographics.
| Male | Female | |
|---|---|---|
| n (%) or mean (range) | ||
| Donor characteristics | ||
| Mean age at donation (years) | 54.9 (26–73) | 57.3 (33–74) |
| Number of donors | 46 (70.8%) | 19 (29.2%) |
| Total number of donations | 251 (73.2%) | 92 (26.8%) |
| Number of donations per donor | 5.4 (2–17) | 4.8 (2–13) |
| Time between donations (days) | 319 (3–2230) | 351 (3–3203) |
| Previous platelet donors | 46 (100%) | 17 (89.5%) |
| Number of platelet donations per donor | 41 (1–118) | 34 (1–85) |
| Product characteristics | ||
| Haemoglobin level (g/dL) | 2.7 (0.6–8.1) | |
| Haematocrit (%) | 8.9 (2.7–28.1) | |
| RBC volume (106/μL) | 0.9 (0.3–2.8) | |
| Platelet count (103/μL) | 405 (145–803) | |
| Granulocytes count (1010/L) | 6.2 (3.9–20.6) | |
| Volume (mL) | 341 (171–565) | |
| Apheresis device | ||
| COBE Spectra | 226 (65.9%) | |
| Spectra Optia | 117 (34.1%) | |
Abbreviation: RBC, red blood cell.
Collection procedures
Granulocyte collections were performed using continuous flow centrifugation on an apheresis device (Spectra Optia or COBE Spectra, Terumo BCT) with peripheral venous access. Each procedure processed 7–7.5 L of whole blood at a flow rate of 50–70 mL/min. The collected products were analysed for granulocyte content, haemoglobin level, RBC count and haematocrit percentage.
Statistical analysis
The normality of data from independent sample groups was evaluated using the Shapiro–Wilk test, and variances were compared using Levene's test. Two‐sample comparisons were performed using either the unpaired t‐test or the Mann–Whitney U test, depending on the data distribution. When analysing multiple groups, either the one‐way analysis of variance (ANOVA) (parametric) or the Kruskal–Wallis test (nonparametric) was applied. For analyses involving two independent variables, a two‐way ANOVA was used to assess the main effects and interaction effects. The Tukey test or the Dunnett T3 test was used to adjust for multiple comparisons, based on the assumption of homoscedasticity. The probability of events between subgroups was compared using Fisher's exact test. The Pearson correlation coefficient was used to measure the linear correlation between donation frequency and pre‐donation haemoglobin levels. A linear mixed‐effects model was constructed with fixed effects including sex, age, number of donations within 14 days, lifetime number of granulocyte donations, lifetime number of platelet donations, time between donations, apheresis device and interactions between sex and key variables (number of donations within 14 days, lifetime number of granulocyte donations, age and time between donations). Sex was coded as a binary variable, with male coded as 1 and female as 0. Independent variables were selected based on their clinical relevance and potential to confound the relationship between donation frequency and haemoglobin levels. Results were considered statistically significant when p < 0.05 and labelled with asterisks: *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001. All statistical analyses were performed using Prism 9 (GraphPad Software 9.5.0, Carlsbad, CA) and Python (version 3.10.0), leveraging the statsmodels package (version 0.13.5) for mixed‐effects modelling.
RESULTS
In this study, we included all consecutive granulocyte donations between April 2012 and December 2023 at the NIH Clinical Center, comprising 343 granulocyte collections from 65 volunteer apheresis donors (Table 1).
Donor characteristics
The study included 65 donors, of whom 19 were females (29.2%) and 46 were males (70.8%). The mean age at donation was 54.9 years for males (range 26–73) and 57.3 years for females (range 33–74). Male donors contributed a total of 251 donations (73.2%), while female donors accounted for 92 donations (26.8%). The mean number of donations per donor was 5.4 for males (range 2–17) and 4.8 for females (range 2–13). The mean time between donations was 319 days for males (range 3–2230) and 351 days for females (range 3–3203). All subsequent analyses were conducted based on the total number of granulocyte collections.
Product characteristics
The mean haemoglobin concentration, haematocrit and RBC content of the collected granulocyte products were 2.7 g/dL (range 0.6–8.1), 8.9% (range 2.7–28.1) and 0.9 × 106/μL (range 0.3–2.8), respectively. The mean platelet count was 405 × 103/μL (range 145–803), and the mean granulocyte content was 6.2 × 1010 (range 3.9–20.6). The mean product volume was 341 mL (range 171–565) (Table 1). Spectra Optia and COBE Spectra devices were used for 34.1% and 65.9% of the donations, respectively.
Pre‐donation haemoglobin levels determined donor eligibility, with a higher proportion of female donors (19.6%) falling below the 12.5 g/dL cutoff compared to males (10.4%) below the 13.0 g/dL cutoff (Figure 1a,b). As described in the Study Design and Methods section, these donors were included in the study through authorized medical exemptions approved by the Blood Bank's medical director.
FIGURE 1.
(a) Donor haemoglobin levels across donations. Pre‐apheresis haemoglobin concentrations are indicated by sex‐specific markers: Blue for males and red for females. The eligibility threshold for haemoglobin concentration is demarcated by sex‐specific dashed lines, with the blue line representing the male threshold at 13 g/dL and the red line indicating the female threshold at 12.5 g/dL. (b) Percentage of donations below pre‐apheresis haemoglobin threshold by sex. The proportion of donations above (black) and below (red) the haemoglobin cutoff levels for eligibility. For females, 19.6% of donations were below the 12.5 g/dL threshold, while for males, 10.4% were below the 13 g/dL threshold.
Sex‐specific impact of lifetime donation frequency on erythrogram parameters
Analysis of donation frequency categories revealed differences between male and female donors. Among female donors, the majority (73.6%) had donated between 1 and 5 times, 21.1% had donated 6–9 times and only 5.3% had donated ≥10 times. In comparison, male donors showed a slightly lower proportion in the 1–5 donations category (67.4%), with 21.7% donating 6–9 times and a higher proportion (10.9%) donating ≥10 times (Figure 2a).
FIGURE 2.
(a) Sex‐based differences in donation frequency. Female donors account for 73.6% of the ≤5 donation category compared to 67.4% for males. In the 6–9 donation category, females make up 21.1%, while males represent 21.7%. For ≥10 donations, males account for 10.9% compared to 5.3% for females. (b) Impact of repeat donations on haemoglobin levels. Mean pre‐donation haemoglobin concentrations (g/dL) are plotted against the number of cumulative granulocytapheresis donations for female donors (red circles) and male donors (black squares). Haemoglobin levels progressively decline with increasing donation frequency in both sexes, with a more pronounced decrease observed in male donors. (c) Haemoglobin levels in male and female donors across donation frequency categories. The mean haemoglobin concentration is displayed for each donation frequency category: ≤5 donations (blue), 6–9 donations (red) and ≥10 donations (green). Among female donors, no significant (ns) differences in haemoglobin levels were observed between donation categories (p = 0.26). In contrast, male donors showed significant differences in haemoglobin levels across categories (p < 0.0001). Tukey's post hoc test revealed that male donors with ≤5 donations had significantly higher haemoglobin levels (14.76 ± 0.16 g/dL) compared to those with ≥10 donations (13.74 ± 0.38 g/dL, p < 0.0001) and those with 6–9 donations (14.61 ± 0.29 g/dL) compared to those with ≥10 donations (p = 0.0033).
The longitudinal changes in pre‐donation haemoglobin levels were analysed across cumulative donations for female and male donors (Figure 2b). Pre‐donation haemoglobin levels decreased with increasing donation frequency in both sexes. Male donors showed a more pronounced decline, with a maximum decrease of 15%, from an initial mean haemoglobin level of 14.8 g/dL (±0.18) at the first donation to 12.6 g/dL at the 17th donation. Female donors also exhibited a reduction in haemoglobin levels, with the most significant decline of 7.5%, from an initial mean haemoglobin level of 13.7 g/dL (±0.28) at the first donation to 12.7 g/dL at the 12th donation.
To address the limitation of having different ranges of donation numbers between males and females, donation numbers were categorized. Analysis by lifetime donation frequency (≤5 donations, 6–9 donations or ≥10 donations) further revealed significantly lower haemoglobin levels in males with ≥10 donations compared to those with ≤5 donations (p < 0.0001, Figure 2c, Table 2). This difference was not observed in female donors.
TABLE 2.
Erythrogram parameters in male and female donors across donation frequency categories.
| Gender | Donation frequency (n) | Erythrogram parameters (mean ± SD) | |||||
|---|---|---|---|---|---|---|---|
| Haemoglobin (g/dL) | Haematocrit (%) | RBC volume (1012/L) | |||||
| Pre | Post | Pre | Post | Pre | Post | ||
| Female | ≤5 donations (70) | 13.4 (0.9) | 11.6 (0.9) | 40.5 (2.9) | 34.8 (2.6) | 4.4 (0.3) | 3.8 (0.32) |
| 6–9 donations (16) | 13.7 (0.9) | 12.0 (0.9) | 41.7 (3.3) | 36.2 (2.4) | 4.5 (0.31) | 3.9 (0.27) | |
| ≥10 donations (6) | 13.6 (1) | 12.3 (1.6) | 41.5 (2.8) | 36.8 (4.6) | 4.4 (0.32) | 3.9 (0.47) | |
| Male | ≤5 donations (172) | 14.7 (1.1) | 13.4 (1.2) | 43.9 (3.3) | 39.1 (3.5) | 4.7 (0.37) | 4.2 (0.40) |
| 6–9 donations (52) | 14.6 (1.1) | 13.5 (1.1) | 43.7 (3.2) | 39.1 (3) | 4.6 (0.34) | 4.2 (0.33) | |
| ≥10 donations (27) | 13.7 (1) | 12.3 (1.3) | 42.4 (2.8) | 37.6 (3.6) | 4.5 (0.40) | 4 (0.47) | |
Abbreviations: post, post‐donation; pre, pre‐donation; RBC, red blood cell; SD, standard deviation.
Effect of short inter‐donation intervals on erythrogram parameters in male and female donors
All donors who donated ≥2 times within a 14‐day period were identified. Successive donations within this window significantly reduced pre‐donation haemoglobin levels, decreasing the percentage of eligible donations from 100% at the first donation to only 25% by the third donation (Figure 3a, Table 3).
FIGURE 3.
(a) Impact of short inter‐donation intervals on donor eligibility. The percentage of donations meeting pre‐donation haemoglobin eligibility criteria (black bars) significantly decreased with successive donations within 14 days (100% at first, 85.8% at second, 25% at third). (b) Sex‐dependent changes in haemoglobin with short inter‐donation interval. Haemoglobin levels in female donors (left) and male donors (right) across three donations within 14 days. The eligibility thresholds for haemoglobin concentration are demarcated by sex‐specific dotted lines, with the blue line representing the male threshold at 13 g/dL and the red line indicating the female threshold at 12.5 g/dL. Pre‐donation and post‐donation haemoglobin levels are presented as mean ± standard deviation. (c) Correlation between donation frequency and pre‐donation haemoglobin levels. The correlation coefficient between the number of donations within specified periods (≤7 days; >7 and ≤14 days; >30 days) and pre‐donation haemoglobin levels is shown for male (blue) and female (red) donors. For female donors, a significant negative association was observed for donations within ≤7 days (β = −0.24, p = 0.019). In contrast, coefficients for >7 and ≤14 days (β = −0.05) and >30 days (β = 0.03) were not statistically significant. Male donors exhibited non‐significant coefficients across all donation frequency categories (≤7 days: β = −0.04; >7 and ≤14 days: β = −0.03; >30 days: β = −0.02).
TABLE 3.
Erythrogram changes following multiple donations within 14 days.
| Donor number (gender) | Donation | Haemoglobin (g/dL) | Haematocrit (%) | RBC volume (1012/L) | |||
|---|---|---|---|---|---|---|---|
| Pre | Post | Pre | Post | Pre | Post | ||
| Donor 1 (female) | 1 | 13.4 | 11.9 | 41.4 | 36.4 | 4.4 | 3.91 |
| 2 | 13.3 | 11.9 | 41.6 | 36.5 | 4.4 | 3.9 | |
| 3 | 12.2 | 10.9 | 38.1 | 33.2 | 4.0 | 3.5 | |
| Donor 2 (female) | 1 | 13.9 | 12.8 | 41.8 | 36.3 | 4.6 | 4.2 |
| 2 | 13.4 | 11.0 | 38.1 | 30.9 | 4.4 | 3.6 | |
| 3 | 11.1 | 9.7 | 32.5 | 28.1 | 3.7 | 3.2 | |
| Donor 3 (female) | 1 | 13.2 | 11.8 | 41.0 | 35.8 | 4.4 | 3.9 |
| 2 | 13.0 | 11.0 | 40.6 | 33.1 | 4.4 | 3.6 | |
| Donor 4 (female) | 1 | 13.1 | 11.4 | 38.4 | 33.4 | 4.4 | 3.8 |
| 2 | 12.4 | 11.1 | 37.3 | 32.9 | 4.2 | 3.7 | |
| Donor 5 (male) | 1 | 14.5 | 12.8 | 40.7 | 35.5 | 4.5 | 3.9 |
| 2 | 13.6 | 11.7 | 39.5 | 33.0 | 4.3 | 3.6 | |
| 3 | 12.9 | 11.5 | 37.9 | 33.0 | 4.2 | 3.6 | |
| Donor 6 (male) | 1 | 14.3 | 13.3 | 43.2 | 39.1 | 4.8 | 4.4 |
| 2 | 15.3 | 14.4 | 46.2 | 42.4 | 5.2 | 4.7 | |
| 3 | 15.1 | 13.6 | 45.3 | 39.7 | 5.1 | 4.5 | |
| Donor 7 (male) | 1 | 14.5 | 13.5 | 44.4 | 40.0 | 4.9 | 4.4 |
| 2 | 14.5 | 13.6 | 44.6 | 42.2 | 4.9 | 4.7 | |
| Donor 8 (male) | 1 | 15.1 | 13.4 | 44.2 | 39.0 | 4.7 | 4.2 |
| 2 | 13.7 | 12.5 | 42.0 | 37.2 | 4.5 | 4.1 | |
Abbreviations: post, post‐donation; pre, pre‐donation; RBC, red blood cell.
The change in pre‐donation haemoglobin levels with each subsequent donation from the initial donation differed between male and female donors, with female donors experiencing a more pronounced decline compared to male donors (Figure 3b). For female donors, pre‐donation haemoglobin levels decreased significantly from a mean of 13.4 ± 0.36 g/dL at the first donation to 13.03 ± 0.45 g/dL at the second donation, and further to 11.6 ± 0.78 g/dL at the third donation. Statistical analysis revealed significant differences between donations 1 and 3 (p = 0.0055) and between donations 2 and 3 (p = 0.026), but not between donations 1 and 2 (p = 0.69). Post‐donation haemoglobin levels in female donors followed a similar trend, decreasing from 12.2 ± 0.55 g/dL at the first donation to 11.2 ± 0.44 g/dL at the second donation and to 10.3 ± 0.85 g/dL at the third donation, with a significant difference between donations 1 and 3 (p = 0.0050), but not between donations 1 and 2 (p = 0.10) or between donations 2 and 3 (p = 0.15).
In contrast, male donors showed relatively stable pre‐donation haemoglobin levels across the three donations: a mean of 14.4 ± 0.12 g/dL at the first donation, 14.5 ± 0.85 g/dL at the second donation and 14 ± 1.56 g/dL at the third donation. Post‐donation haemoglobin levels also remained stable. Statistical analysis revealed no significant differences in pre‐donation or post‐donation haemoglobin levels between donations for male donors (p > 0.05 for all comparisons).
Correlation analysis underscored this sex difference, revealing a strong negative correlation between pre‐donation haemoglobin and frequency of donations within a 7‐day period for females. This correlation was absent in males (Figure 3c). Multivariate regression analysis confirmed the strong negative impact of frequent donations within the previous 14 days on pre‐donation haemoglobin levels in females (estimate: −0.78, p = 0.039). Factors such as age, prior granulocyte or platelet donations and apheresis device did not significantly influence haemoglobin recovery (Table 4).
TABLE 4.
Multivariate analysis of factors affecting pre‐donation haemoglobin in female donors.
| Variable | Estimate (range) | p‐value |
|---|---|---|
| Age | 0.019 (−0.003, 0.042) | 0.08 |
| Lifetime number of granulocyte donations | −0.023 (−0.102, 0.055) | 0.4 |
| Lifetime number of platelet donations | −0.24 (−0.90, 0.42) | 0.4 |
| Apheresis device (Spectra Optia) | 0.0051 (−0.42, 0.43) | 0.9 |
| Number of donations within 14 days | −0.78 (−1.20, −0.36) | 0.039 |
Multivariate analysis of factors influencing erythrogram parameters using a mixed linear model
To control for the variance associated with random factors without data aggregation, a linear mixed‐effects model was built (Table 5). This approach allows for a more accurate assessment of the impact of donation frequency and inter‐donation intervals on erythrogram parameters while accounting for individual donor characteristics and repeat donations [19].
TABLE 5.
Pre‐donation haemoglobin linear mixed‐effects model.
| Variable | Coefficient | SE | p | 95% CI |
|---|---|---|---|---|
| Intercept | 13.2 | 1.016 | <0.001 | 11.241, 15.225 |
| Sex | 0.54 | 1.218 | 0.65 | −1.850, 2.926 |
| Age | 0.022 | 0.018 | 0.21 | −0.013, 0.058 |
| Number of donations within 14 days | −0.81 | 0.21 | 0.017 | −1.241, −0.389 |
| Lifetime number of granulocyte donations | −0.062 | 0.034 | 0.071 | −0.129, 0.005 |
| Lifetime number of platelet donations | −0.001 | 0.004 | 0.80 | −0.008, 0.006 |
| Time between donations | −0.000 | 0.000 | 0.33 | −0.000, 0.000 |
| Apheresis device (Spectra Optia) | 0.26 | 0.11 | 0.015 | 0.053, 0.482 |
| Sex: Total donations within 14 days | 0.76 | 0.321 | 0.018 | 0.132, 1.391 |
| Sex: Lifetime number of granulocyte donations | −0.008 | 0.033 | 0.82 | −0.073, 0.058 |
| Sex: Age | −0.001 | 0.021 | 0.96 | −0.042, 0.040 |
| Sex: Time between donations | 0.000 | 0.000 | 0.18 | −0.000, 0.001 |
Abbreviations: CI, confidence interval; SE, standard error.
The mixed‐effects model revealed that the number of donations within a 14‐day period had a significant negative impact on pre‐donation haemoglobin levels (p = 0.017). The significant positive interaction between sex and donations within 14 days (β = 0.76, p = 0.018) indicates that the negative effect of frequent donations on haemoglobin levels is less severe in male donors compared to female donors.
The interaction between sex and the number of donations within 14 days was also significant (β = 0.76, p = 0.018), indicating that the negative effect of frequent donations within short intervals on haemoglobin levels was more pronounced in female donors compared to male donors.
The type of apheresis device used showed a significant association with pre‐donation haemoglobin levels. Use of the Spectra Optia device was associated with higher haemoglobin levels compared to the COBE Spectra device (β = 0.26, p = 0.015).
Other variables, including the lifetime number of granulocyte donations, the time interval between donations and donor age, did not show statistical significance in this model. Additionally, interactions between sex and other factors such as lifetime donation count, age and time between donations also failed to reach statistical significance.
DISCUSSION
This study provides a comprehensive analysis of the impact of granulocytapheresis donation frequency and inter‐donation intervals on erythrogram parameters, with a particular focus on sex‐based differences. Our findings reveal that short inter‐donation intervals significantly affect pre‐donation haemoglobin levels in female donors more than in male donors. Specifically, while both sexes experience declines in haemoglobin levels with closely spaced donations, females exhibit a larger decrease per additional donation within a 14‐day period.
Female donors experienced significant declines in haemoglobin levels when making multiple donations within a 14‐day period, as demonstrated by our multivariate analysis identifying the number of donations within 14 days as the only significant predictor of pre‐donation haemoglobin levels in female donors (Table 4). Moreover, the significant positive interaction between sex and the number of donations within 14 days suggests that male donors experience a less pronounced decline compared to female donors under similar conditions. These findings indicate that shorter inter‐donation intervals may not allow sufficient time for haemoglobin and iron stores to replenish in female donors.
In contrast, while male donors showed a statistically significant decline in haemoglobin levels with increasing lifetime donations (Figure 2c), the linear mixed‐effects model revealed that the lifetime number of granulocyte donations was not a significant predictor of pre‐donation haemoglobin levels when controlling for other variables (Table 5). Additionally, the interaction between sex and lifetime number of donations was not significant, indicating that over the long term, the cumulative impact of multiple donations on haemoglobin levels was similar for both sexes. Importantly, despite the observed decline, mean haemoglobin levels in male donors remained above the eligibility threshold, rendering the reduction clinically insignificant.
These sex‐specific differences in haemoglobin recovery align with previous studies reporting a higher prevalence of anaemia among female apheresis donors compared to males [20, 21, 22]. Factors contributing to this increased susceptibility in females include menstrual blood loss, higher iron requirements during reproductive years and physiological differences in iron metabolism [7, 23, 24, 25, 26]. Additionally, haemoglobin levels can decrease during the first few days post‐donation because of physiological compensation for blood volume loss, leading to a dilutional effect [27]. These factors, combined with sex‐specific physiological differences, highlight the importance of implementing sex‐specific guidelines when establishing donation criteria and monitoring donor health.
The mean age of 57.3 years among female donors suggests that many were beyond their reproductive years, potentially mitigating some of the iron depletion effects seen in younger women. Menopausal status can influence haemoglobin levels as a result of hormonal changes affecting iron metabolism, often resulting in higher haemoglobin levels after menopause as menstrual blood loss ceases. Previous studies have shown that menstrual blood loss is an independent determinant of haemoglobin and ferritin levels in pre‐menopausal blood donors [28] and that iron deficiency is more prevalent among younger, pre‐menopausal women due to ongoing menstrual blood loss and higher iron requirements [29, 30, 31]. However, age was not found to be a significant predictor of pre‐donation haemoglobin levels in our mixed‐effects model, indicating that within our donor population, age did not significantly impact haemoglobin recovery rates following granulocyte donation. Without specific data on menopausal status, we cannot definitively assess its impact, but it remains a factor worth considering in future studies.
Despite current recommendations from organizations like the AABB and EDQM that aim to minimize health risks for granulocyte donors through regulated donation frequency and intervals with sex‐based thresholds [8, 9], our findings suggest that these guidelines may not fully account for the sex‐specific differences observed. The pronounced decline in haemoglobin levels among female donors with short inter‐donation intervals underscores the need for further research to better understand the impact of these factors on donor safety and well‐being.
Implementing longer inter‐donation intervals or restricting donation frequency, particularly for female donors, may minimize the risk of iron deficiency and anaemia associated with frequent donations. This approach aligns with recommendations from the AABB Association Bulletin on strategies to monitor and prevent iron deficiency in blood donors [32]. Tailoring donation schedules to consider both sex‐specific haematological responses and appropriate intervals between donations could optimize donor health while ensuring a sustainable granulocyte supply for patients in need.
The strengths of our study include the longitudinal design and the use of a linear mixed‐effects model analysis to control for the variance associated with random factors. However, several limitations should be considered. The retrospective design and relatively small sample size, particularly among female donors in higher donation frequency categories, may limit the robustness of our conclusions. There is also potential selection bias, as donors deferred for low haemoglobin levels who did not return for future donations were excluded, potentially underestimating the effects of donation frequency and intervals on haemoglobin levels and donor eligibility. Additionally, the retrospective nature of the study precluded the evaluation of sensitive iron deficiency markers, such as serum ferritin or transferrin saturation, which could provide deeper insights into the mechanisms underlying haemoglobin declines. Future prospective, multi‐centre studies with larger cohorts will be needed to validate these findings and further investigate the long‐term effects of frequent granulocyte donations on donor health.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
ACKNOWLEDGEMENTS
A.A. designed the study with input from K.A.W.‐M. and L.C., collected and analysed the data, performed the statistical analysis and wrote the manuscript and K.A.W.‐M. and L.C. contributed to the writing and editing of the manuscript.
Alswied A, Chen LN, West‐Mitchell KA. Longitudinal assessment of erythrogram parameters in response to granulocytapheresis frequency: A sex‐based analysis. Vox Sang. 2025;120:268–276.
Funding information The authors received no specific funding for this work.
DATA AVAILABILITY STATEMENT
The datasets generated and/or analysed during the current project are not publicly available due to sensitive information and restrictions related to identifiable private health information. Minimal de‐identified aggregate data in the form of tables are available from the corresponding author on reasonable request and subject to institutional approval.
REFERENCES
- 1. Cesaro S, Chinello P, De Silvestro G, Marson P, Picco G, Varotto S, et al. Granulocyte transfusions from G‐CSF‐stimulated donors for the treatment of severe infections in neutropenic pediatric patients with onco‐hematological diseases. Support Care Cancer. 2003;11:101–106. [DOI] [PubMed] [Google Scholar]
- 2. Estcourt LJ, Stanworth SJ, Hopewell S, Doree C, Trivella M, Massey E. Granulocyte transfusions for treating infections in people with neutropenia or neutrophil dysfunction. Cochrane Database Syst Rev. 2016;4:Cd005339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Stroncek DF, Yau YY, Oblitas J, Leitman SF. Administration of G‐CSF plus dexamethasone produces greater granulocyte concentrate yields while causing no more donor toxicity than G‐CSF alone. Transfusion. 2001;41:1037–1044. [DOI] [PubMed] [Google Scholar]
- 4. Strauss RG, Rohret PA, Randels MJ, Winegarden DC. Granulocyte collection. J Clin Apher. 1991;6:241–243. [DOI] [PubMed] [Google Scholar]
- 5. Brockmann F, Kramer M, Bornhäuser M, Ehninger G, Hölig K. Efficacy and side effects of granulocyte collection in healthy donors. Transfus Med Hemother. 2013;40:258–264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Hubel K, Rodger E, Gaviria JM, Price TH, Dale DC, Liles WC. Effective storage of granulocytes collected by centrifugation leukapheresis from donors stimulated with granulocyte‐colony‐stimulating factor. Transfusion. 2005;45:1876–1889. [DOI] [PubMed] [Google Scholar]
- 7. Newman B. Iron depletion by whole‐blood donation harms menstruating females: the current whole‐blood‐collection paradigm needs to be changed. Transfusion. 2006;46:1667–1681. [DOI] [PubMed] [Google Scholar]
- 8. American Association of Blood Banks , editor. Standards for blood bank and transfusion services. 33rd ed. Bethesda, MD: AABB; 2022. [Google Scholar]
- 9. European Directorate for the Quality of Medicines & HealthCare . Guide to the preparation, use and quality assurance of blood components: Recommendation Number R (95) 15. Strasbourg: European Directorate for the Quality of Medicines & HealthCare (EDQM); 2015. [Google Scholar]
- 10. Strauss RG, Goeken JA, Imig KM, Greazel C. Effects on immunity of multiple leukapheresis using a rapidly excreted analog of hydroxyethyl starch. Transfusion. 1986;26:265–268. [DOI] [PubMed] [Google Scholar]
- 11. Szymanski J, Troendle J, Leitman S, Hong H, Yau YY, Cantilena C. The effect of repeated stimulated granulocyte donations on hematopoietic indexes in donors: a 24‐year donor center experience. Transfusion. 2019;59:259–266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Axdorph Nygell U, Sollén‐Nilsson A, Lundahl J. Eighteen years experience of granulocyte donations—acceptable donor safety? J Clin Apher. 2015;30:265–272. [DOI] [PubMed] [Google Scholar]
- 13. Quillen K, Byrne P, Yau YY, Leitman SF. Ten‐year follow‐up of unrelated volunteer granulocyte donors who have received multiple cycles of granulocyte‐colony‐stimulating factor and dexamethasone. Transfusion. 2009;49:513–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Buchholz DH, Schiffer CA, Wiernik PH, Bwits SW, Reilly JA. Granulocyte harvest for transfusion: donor response to repeated leukapheresis. Transfusion. 1975;15:96–106. [DOI] [PubMed] [Google Scholar]
- 15. Bensinger W, Price T, Dale D, Appelbaum F, Clift R, Lilleby K, et al. The effects of daily recombinant human granulocyte colony‐stimulating factor administration on normal granulocyte donors undergoing leukapheresis. Blood. 1993;81:1883–1888. [PubMed] [Google Scholar]
- 16. Worel N, Kurz M, Peters C, Höcker P. Serial granulocytapheresis under daily administration of rHuG–CSF: effects on peripheral blood counts, collection efficiency, and yield. Transfusion. 2001;41:390–395. [DOI] [PubMed] [Google Scholar]
- 17. Beyan C, Çetin T, Kaptan K, Nevruz O. Effect of plateletpheresis on complete blood count values using three different cell separator systems in healthy donors. Transfus Apher Sci. 2003;29:45–47. [DOI] [PubMed] [Google Scholar]
- 18. Das SS, Chaudhary R, Verma SK, Ojha S, Khetan D. Pre‐ and post‐donation haematological values in healthy donors undergoing plateletpheresis with five different systems. Blood Transfus. 2009;7:188–192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Verbeke G, Molenberghs G, Verbeke G. Linear mixed models for longitudinal data. New York: Springer; 1997. [Google Scholar]
- 20. Schreiber GB, Brinser R, Rosa‐Bray M, Yu ZF, Simon T. Frequent source plasma donors are not at risk of iron depletion: the Ferritin Levels in Plasma Donor (FLIPD) study. Transfusion. 2018;58:951–959. [DOI] [PubMed] [Google Scholar]
- 21. Bier‐Ulrich A‐M, Haubelt H, Anders C, Nagel D, Schneider S, Siegler KE, et al. The impact of intensive serial plasmapheresis and iron supplementation on iron metabolism and Hb concentration in menstruating women: a prospective randomized placebo‐controlled double‐blind study. Transfusion. 2003;43:405–410. [DOI] [PubMed] [Google Scholar]
- 22. Li C, Feng Q, Zhang J, Xie X. A multivariate analysis of the risk of iron deficiency in plateletpheresis donors based on logistic regression. Transfus Apher Sci. 2023;62:103522. [DOI] [PubMed] [Google Scholar]
- 23. Custer B, Bravo M, Bruhn R, Land K, Tomasulo P, Kamel H. Predictors of hemoglobin recovery or deferral in blood donors with an initial successful donation. Transfusion. 2014;54:2267–2275. [DOI] [PubMed] [Google Scholar]
- 24. Nasserinejad K, de Kort W, Baart M, Komárek A, van Rosmalen J, Lesaffre E. Predicting hemoglobin levels in whole blood donors using transition models and mixed effects models. BMC Med Res Methodol. 2013;13:62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Mast AE, Schlumpf KS, Wright DJ, Johnson B, Glynn SA, Busch MP, et al. Hepcidin level predicts hemoglobin concentration in individuals undergoing repeated phlebotomy. Haematologica. 2013;98:1324–1330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Baart AM, de Kort WL, Atsma F, Moons KG, Vergouwe Y. Development and validation of a prediction model for low hemoglobin deferral in a large cohort of whole blood donors. Transfusion. 2012;52:2559–2569. [DOI] [PubMed] [Google Scholar]
- 27. Schotten N, Pasker‐de Jong PCM, Moretti D, Zimmermann MB, Geurts‐Moespot AJ, Swinkels DW, et al. The donation interval of 56 days requires extension to 180 days for whole blood donors to recover from changes in iron metabolism. Blood. 2016;128:2185–2188. [DOI] [PubMed] [Google Scholar]
- 28. Ekroos S, Karregat J, Toffol E, Castrén J, Arvas M, van den Hurk K. Menstrual blood loss is an independent determinant of hemoglobin and ferritin levels in premenopausal blood donors. Acta Obstet Gynecol Scand. 2024;103:1645–1656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Milman N, Kirchhoff M. The influence of blood donation on iron stores assessed by serum ferritin and hemoglobin in a population survey of 1359 Danish women. Ann Hematol. 1991;63:27–32. [DOI] [PubMed] [Google Scholar]
- 30. Cable RG, Glynn SA, Kiss JE, Mast AE, Steele WR, Murphy EL, et al. Iron deficiency in blood donors: analysis of enrollment data from the REDS‐II Donor Iron Status Evaluation (RISE) study. Transfusion. 2011;51:511–522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Simon TL, Garry PJ, Hooper EM. Iron stores in blood donors. JAMA. 1981;245:2038–2043. [PubMed] [Google Scholar]
- 32. AABB Donor Health and Safety Committee . Association Bulletin #17‐02: Strategies to Monitor, Limit, or Prevent Iron Deficiency in Blood Donors. Available from: https://www.aabb.org/docs/default‐source/default‐document‐library/resources/association‐bulletins/ab17‐02.pdf. Last accessed 6 Apr 2024.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and/or analysed during the current project are not publicly available due to sensitive information and restrictions related to identifiable private health information. Minimal de‐identified aggregate data in the form of tables are available from the corresponding author on reasonable request and subject to institutional approval.