[Drs. Shyu, Lin and Li respond:]
Following the guidelines on using G-CSF for stem cell mobilization,1 we carefully adjusted the dose of G-CSF (15 μg/kg) for up to 5 days to avoid the leukocyte numbers rising above 70 × 109/L (in fact, the leukocyte count for all patients was below 61.3 × 109/L). In addition, we also prescribed moderate hydration and antiplatelet medicine, as suggested by LeBlanc and colleagues,2 to minimize hyperosmolarity and hypercoagulability. During the clinical course of G-CSF treatment, there were no abnormal findings for biochemistry, bleeding time, coagulation time or C-reactive protein. Therefore, we conclude that no patient receiving G-CSF therapy experienced a thrombotic complication. We agree that more factors need to be monitored when using G-CSF treatment for thrombophilic, atherosclerotic and immobilized patients. Given that thrombotic complications possibly induced by G-CSF have been reported, strict selection of patients, adequate hydration and appropriate dosage of G-CSF are needed in the treatment of individual stroke patients.
In our trial, the mean baseline score on the National Institutes of Health Stroke Scale was 12.0 (standard deviation [SD] 4.3) for patients treated with G-CSF and 12.0 (SD 2.6) for the control group. However, for the other scales, some of the individual scores for patients who received G-CSF were higher than those of the controls at the time of recruitment; as a result, the means for the EES, ESS Motor Subscale and Barthel Index were higher for patients in the treatment group. If these outliers are removed, the average scores at recruitment were similar between the treatment and control groups. Similarly, at the 12-month follow-up, scores for the G-CSF-treated patients (excluding those with higher scores at the time of recruitment) remained higher than those of controls. Therefore, although there was some bias in the initial selection of patients, we conclude that G-CSF could be an important treatment for acute stroke.
We emphasize that more randomized, double-blind, placebo-controlled trials are needed to confirm the preliminary findings that we have reported.3
REFERENCES
- 1.
- 2.
- 3.