Safety and Efficacy of Non-Irradiated Granulocyte Transfusions (GTX) in Neutropenic Patients with Severe or Refractory Abdominal Infections: A Single Center Retrospective Analysis of 119 Transfusions in 22 Patients

Background: Neutropenic patients with hematologic malignancies are at high risk of infectious related-deaths. Granulocyte transfusions (GTX) are used in neutropenic patients to treat severe refractory infections. These blood products are usually irradiated to eliminate T-cells that can cause transfusion-associated graft-versus-host disease (TA-GVHD) in severely immunocompromised patients. We adopted a strategy of rapid utilization of non-irradiated GTX in neutropenic patients with severe abdominal infection to accelerate recovery and prevent infectious-related deaths. We report here the safety and efficacy data of 119 transfusions administered in 22 patients with abdominal infections.

Methods: We retrospectively analyzed the safety and efficacy data of patients with abdominal infections who received non-irradiated GTX in our institution between January 1^st 2015 and December 31^st 2017. Patients with hematological malignancies, severe neutropenia (<0.5 x 10⁹/L) and severe abdominal infection would receive GTX until neutrophils increase > 1.0 x 10⁹/L or significant clinical improvement. Granulocyte donors were mobilized using subcutaneous G-CSF 480 mcg and 8 mg of oral dexamethasone regardless of ABO and HLA compatibility. ABO incompatible GTX were drained of red cells prior to transfusion. We collected increments of absolute neutrophil counts (ANC) after each GTX, number of days with ANC > 1.0 x 10⁹/L and transfusion-related complications. Controlled infection was defined as significant clinical improvement or antibiotic de-escalation or end.

Results: A total of 119 non-irradiated GTX were administered in 22 patients for 25 different infectious episodes. Baseline characteristics of patients are summarized in table 1. Median age of patients was 53.5 year-old (range 21 to 73) and 12 patients (54.5%) were male. Most patients had acute myeloid leukemia (14/22 patients, 63.6%) and relapsed disease (13 patients). GTX were administered in 11 patients (44%) for neutropenic enterocolitis, 6 (24%) for small bowel obstruction and/or perforation, 5 (20%) for clostridium difficile colitis, 2 (8%) for abdominal abscesses and 1 (4%) for appendicitis. Patients received a median of 3 GTX during their infectious episode (range 1 to 17). The median ANC before first GTX was 0.1 x 10⁹/L (range 0.0 to 4.5). After GTX, there was a median increase in ANC of 0.3 x 10⁹/L (range -0.8 to 26.1), with the peak of ANC occurring at a median of 24 hours after first GTX and returning to baseline at a median of 2 days after the last GTX. In 15 patients (68%), ANC was above 1 x 10⁹/L for 3 days or more. GTX contributed to control infection in 19 infectious episodes (76%) after a median of 5 days following first transfusion (range 2 to 32 days). There was no significant association between sex, infectious diagnosis, initial ANC, number of GTX, mean ANC increment per patient or ANC > 0.1 x 10⁹/L for 3 days and the rate of controlled infection 7 days after first GTX. Twelve patients (54.5%) were able to pursue their treatments without significant delays (> 7 days) after their infectious episode. Five patients (22.3%) died with the active infection for which they had received a GTX; all these patients also had active refractory disease and severe comorbidities. The median overall survival (OS) from first GTX was 7.7 months (95% CI; 2.7 to 17.4). Patients who had a controlled infection within 7 days of first GTX had significantly better OS (median OS 12.4 months vs. 1.6 months, log-rank p < 0.001) (Figure 1). Pulmonary events leading to death were reported in 2 patients; one had acute respiratory distress syndrome and cardiac arrest the day after the first GTX, but had pneumonia before transfusion, and one had worsening of pre-existing lung infiltrates 3 days after GTX and eventually died of respiratory failure. Importantly, no TA-HSCT were reported in the medical notes, neither suspected by the review of the patients' charts.

Conclusion: Non-irradiated GTX are safe and effective to control severe abdominal infection. Noteworthy, no TA-GVHD occurred in this cohort of severely immunocompromised patients. Pulmonary complications after GTX are rare but can be serious, mostly seen in patients with pre-existing pulmonary infiltrates. GTX in such instances may pose higher risks. Further studies are needed to evaluate the comparative safety and efficacy of non-irradiated GTX with irradiated GTX.

View largeDownload slide

View largeDownload slide

Close modal