Study designs of evaluations included in the review
Randomised controlled trials (RCTs). One Japanese trial was excluded due to translation difficulties, while another trial was excluded because the comparison was between G-CSF and G-CSF plus thymostimulin, with no control group.
Specific interventions included in the review
Myelosuppressive chemotherapy with or without G-CSF for prophylactic or therapeutic indication. The dosages of G-CSF tested by the included studies were 2, 5, 6 (230 microg/m2) and 50 microg/kg. The chemotherapy regimens used included: cyclophosphamide plus doxorubicin; adriamycil, cyclophosphamide, vincristine, bleomycin, etoposide, prednisolone, cotrimaxazole and ketoconazole; cyclophosphamide, doxorubicin and etoposide; carboplatin, ifosfamide (with mesna), etoposide, and vincristine (plus radiotherapy during first cycle); doxorubicin, ifosfamide and dacarbazine; any chemotherapy agent with the antibiotic agents ceftaxidime and amikacin (for the treatment of neutropenia); fluorouracil, epirubicin and cyclophosphamide, with or without radiotherapy; mitomycin, mitoxantrone and methotrexate; mitomycin, vindesine and cisplatin; cisplatin, vincristine, doxorubicin and etoposide. Some studies also included placebo therapy.
Participants included in the review
Patients receiving myelosuppressive chemotherapy for cancer. Studies that included children were excluded. The included studies examined the following types of cancer: small-cell lung cancer (SCLC), non- Hodgkin's lymphoma, sarcoma, inflammatory and advanced breast cancer, non-SCLC, and various other.
Outcomes assessed in the review
Important adverse clinical outcomes due to infection and the maintenance of chemotherapy were assessed. The outcome measures reported for the included studies were: the mean or median number of hospital days; the mean or median days on intravenous antibiotics; percentage with febrile neutropenia (FN) or incidence of infection; the mean or median number of days with FN; tumour response rate; percentage alive or median value for disease-free survival or progression-free survival; percentage alive or median value for overall survival; the median number of days to recovery to an absolute neutrophil count (ANC) greater than or equal to 0.5, or greater than or equal to 1.0; median days with an ANC of less than 0.5; and median dose intensity of chemotherapy.
How were decisions on the relevance of primary studies made?
The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.