Adherence to granulocyte-colony stimulating factor (G-CSF) guidelines to reduce the incidence of febrile neutropenia after chemotherapy--a representative sample survey in Germany.

PURPOSE: Febrile neutropenia (FN) after chemotherapy increases complications, morbidity, risk of death, reduction of dose delivery and impairs quality of life. Primary granulocyte-colony stimulating factor (G-CSF) prophylaxis after chemotherapy is recommended in the guideline (GL) if the risk of FN is high (≥20%) or intermediate (≥10-20%) with additional risk factors. This study evaluated the implementation of G-CSF GL.

PATIENTS AND METHODS: Sample size of the survey was calculated at 2% of the incidences of malignant lymphoma, breast cancer, and lung cancer in Germany in 2006. Patients were documented retrospectively over three to nine cycles of chemotherapy and FN risk ≥10%. Professional physician profiles were analyzed by classification and regression tree analysis (CART).

RESULTS: One hundred ninety-five hematologists-oncologists and pulmonologists and gynecologists specialized in oncology documented data of 666 lung cancer patients, 286 malignant lymphoma patients, and 976 breast cancer patients, with 7805 chemotherapy cycles; 85.1% of physicians claimed adhering to G-CSF GL. Adherence to GL in all high-FN-risk chemotherapy cycles was 15.4% in lung cancer, 84.5% in malignant lymphoma, and 85.6% in breast cancer, and in all intermediate-FN-risk chemotherapy cycles, lung cancer it was 38.8%, malignant lymphoma it was 59.4%, and breast cancer it was 49.3%. G-CSF was overused without additional patient risk factors in 7.2% lung cancer cycles, 16.8% malignant lymphoma cycles, and 17.6% breast cancer cycles. The CART analysis split pulmonologists and other specialists, with the latter adhering more to GL. Pulmonologists, trained less than 22.5 years, adhered better to GL, as did also gynecologists or hematologists-oncologists with professional experience less than 8.1 years.

CONCLUSIONS: Acceptance of and adherence to G-CSF GL differed between lung cancer, lymphoma, and breast cancer. Physicians overestimate their adherence to the GL. Physicians adhering to the GL can be characterized.