Background: Febrile neutropenia (FN) is common in patients with chemotherapy. It requires conventional treatment, however, many studies have reported that G-CSF reduces the incidence of FN; the results were not clear and the physicians use it at their discretion. In this guideline we evaluated the efficacy and safety of the prophylactic use of G-CSF. Methods: We analyzed controlled trials (G-CSF, pegylated form or placebo) given to adult or pediatric patients with chemotherapy for leukemia (LEU), lymphoma and solid tumors (L&ST) or stem cell transplant (SCT), without infections and large radiation ports. Two independent reviewers applied CONSORT to determine the methodological quality; for ranking the evidence we used GRADE and the recommendations were developed by Delphi method. We developed subgroups according to age and type of intervention to analyze the outcomes (risk, duration, severity of FN and adverse events). We performed random-effects or fixed-effects meta-analysis methods according to their heterogeneity. Results: Of 1,776 studies,112 were included. For the risk of FN between C-GSF or pegylated form vs placebo found that G-CSF reduces the risk in adults with LEU (RR 0.89, 95% CI 0.81-0.98; p=0.024), L&ST (RR 0.758, 95% CI 0.68-0.84; p=0.000) and SCT (RR 0.85, 95% CI 0.74–0.97, p=0.017). The risk of developing severe neutropenia reduces in the adults with L&ST with the factor (RR 0.79, 95% CI 0.71-0.88; p=0.000) and pediatric patients with LEU (RR 0.789, 95% CI 0.71-0.88; p=0.000). While the duration of neutropenia in children with L&ST the time reduces with the factor (SMD -0.559; 95% CI -0.841 to -0.28; p=0.000). The G-CSF vs pegylated form, the evidence was inconclusive. Conclusions: When the risk and duration of neutropenia is present we suggest the use of G-CSF in adult and pediatric patients. For adults, we suggest the use of pegylated form, but for pediatric patients we do not have a specific suggestion because the evidence is nonexistent, so it is necessary to carry out clinical trials to obtain evidence.