G-CSF Mobilized Granulocyte Transfusions in 32 Pediatric Patients with Severe Infections after Chemotherapy Induced Neutropenia.

Background and Objectives. Infections remain a major obstacle during chemotherapy induced neutropenic episodes. Despite improvements achieved in antibiotic and antimycotic treatment, mortality is still high in patients with severe infections. In this retrospective analysis we studied the effect of granulocyte-transfusions (GTX) in critically ill pediatric patients with life-threatening infections.

Design and Methods. We performed a retrospective review of clinical data, laboratory and microbiological results during chemotherapy induced neutropenia between 2000 and 2003. GTX was considered during neutropenic sepsis not responding to broad-spectrum antibiotic and antimycotic therapy and increasing markers of infection. Granulocytes were collected from unrelated, blood-group matched, healthy and cytomegaly antibody negative donors. All donors received glycosylated G-CSF (Lenograstim™, 1.5 to 6μg/kg BW) plus oral dexamethasone (8mg) and underwent a neutrophil apheresis using the Spectra™ PMN program.

Results. In total, 153 (range 1–19) GTX were transfused in 32 children aged 0.25 to 16 years. The underlying diseases included haematological malignancies (n=28), inborn immunological diseases (n=3) and a solid tumor (n=1). In 15/32 patients an allogenic stem cell transplantation (HCT) was performed. Infections consisted of fever of unknown origin (FUO, n=13), bacterial sepsis (n=8), invasive fungal infections (n=6), and viral infections (n=4). In one patient a double infection with virus and fungi was seen. The mean duration of neutropenia prior to the first GTX lasted 22.2 (range 3–85) days; the mean total duration of neutropenia was 36 (range 5–99) days. In three children, no regeneration of the leukocytes was observed. 18/32 children survived the neutropenic sepsis. With a thrice per week delivery of GTX (mean leukocyte count per transfusion 6.3 (range 1.9–13.9) x 10¹⁰) the mean increase of the leukocyte count 1 hour after GTX was 3480/μl (1580 to 18300). 16 hours after GTX a mean increase of the leukocyte number to 2560/μl (300–13200) was noted. One child developed a respiratory distress syndrome shortly after her 2^nd GTX. Apart from this, no serious adverse events were seen. In 18/32 children the fever / infection was cleared during GTX. Out of 8 children with bacterial infection, 7 children survived. In total, 9 children died in the context of the underlying infection (6 viral, 2 fungal, 1 bacterial, 1 viral+fungal). 8/15 children died after SCT, in 3/8 treatment related toxicity (TRT) was responsible. In the surviving patients we observed a significant decrease in the C-reactive protein (CRP) (mean CRP prior to 1^st GTX 183 mg/l, mean CRP 48 hours after the last GTX 116 mg/l) levels during and after GTX, whereas the CRP levels increased in those children who died (mean CRP prior to 1^st GTX 186 mg/l, mean CRP 48 hours after the last GTX 231 mg/l). HCT or duration of neutropenia were no significant risk factors in this analysis.

Conclusions. GTX were a safe therapeutic measure for critically ill neutropenic patients with progression of sepsis. After initiation of GTX a sustained increase of the leukocyte count could be achieved with beneficial effects on bacterial infections. No effects on viral infections were observed. The CRP-level might serve as a prognostic factor to identify good responders to GTX during sepsis.