RESUMEN
The purpose of this study is to obtain a better operational knowledge of Stereotactic Body Radiotherapy (SBRT) treatments with CyberKnife(r). An analysis of both In-room Times (IRT) and technical interventions of 5 years of treatments was performed, during which more than 1600 patients were treated for various indications, including liver (21%), lung (29%), intracranial (13%), head and neck (11%) and prostate (7%). Technical interventions were recorded along with the time of the failure, time to the intervention, and the complexity and duration of the repair. Analyses of Time Between Failures (TBF) and Service Disrupting TBF(disr) were performed. Treatment time data and variability per indication and following different system upgrades were evaluated. Large variations of IRTs were found between indications, but also large variations for each indication. The combination of the time reduction Tool (using Iris(r)) and Improved Stop Handling was of major impact to shortening of treatment times. The first implementation of the Iris collimator alone did not lead to significantly shorter IRTs for us except during prostate treatments. This was mostly due to the addition at the same time of larger rotational compensation for prostate treatments (58 instead of 1.58). Significant differences of duration between the first fraction and following fractions of a treatment, representing the necessity of defining imaging parameters and explanation to patients, were found for liver (12 min) and lung treatments using Xsight(r) Spine (5 min). Liver and lung treatments represent the longest IRT's and involve the largest variability's in IRT. The malfunction rate of the system followed a Weibull distribution with the shape and scale parameters of 0.8 and 39.7. Mean TBF(disr) was 68 work hours. 60 to 80% of the service disrupting interventions were resolved within 30-60 min, 5% required external intervention and 30% occurred in the morning. The presented results can be applied in the evaluation of the required machine time in order to implement robotic radiosurgery for different indications. The analytical distributions of IRTs and technical interruptions can be used for simulations.