Search

Purpose: The accuracy of intensity-modulated radiotherapy (IMRT) delivery may be compromised by random spatial error and systematic anatomic changes during the treatment course. We present quantitative measurements of the spatial variability of head-and-neck organs-at-risk and demonstrate the resultant dosimetric effects., Methods and Materials: Fifteen consecutive patients were imaged weekly using computed tomography during the treatment course. Three-dimensional displacements were calculated for the superior and inferior brainstem; C1, C6, and T2 spinal cord; as well as the lateral and medial aspects of the parotid glands. The data were analyzed to show distributions of spatial error and to track temporal changes. The treatment plan was recalculated on all computed tomography sets, and the dosimetric error was quantified in terms of the maximal dose difference (brainstem and spinal cord) or the mean dose difference and the volume receiving 26 Gy (parotid glands)., Results: The mean three-dimensional displacement was 2.9 mm for the superior brainstem, 3.4 mm for the inferior brainstem, 3.5 mm for the C1 spine, 5.6 mm for the C6 spine and 6.0 mm for the T2 spine. The lateral aspects of both parotid glands showed a medial translation of 0.85 mm/wk, and glands shrank by 4.9%/wk. The variability of the maximal dose difference was described by standard deviations ranging from 5.6% (upper cord) to 8.0% (lower cord.) The translation of the left parotid resulted in an increase of the mean dose and the volume receiving 26 Gy., Conclusion: Random spatial and dosimetric variability is predominant for the brainstem and spinal cord and increases at more inferior locations. In contrast, the parotid glands demonstrated a systematic medial translation during the treatment course and thus sparing may be compromised.

Background: Patients with head and neck cancer are extraordinarily susceptible to depressive traits. Thus, a general screening of these patients at their first admission to the ital is desirable., Methods: From 1997-2001, 133 patients with head and neck tumors filled in the Self-Rating-Depression-Scale (SDS) at the beginning and end of radiotherapy (ti1/ti2), 6 weeks, and 6 months after radiotherapy (ti3/ti4). RESULTS.: The SDS index increased significantly from 46.44 (ti1) to 48.91(ti2) (p =.025) and then remained stable. The subdomain "somatic-eating-related symptoms" at ti1 was significantly lower than ti2 (p <.001). In contrast to inpatients, outpatients and those with conventional instead of hyperfractionated-accelerated radiotherapy were less impaired by eating-related symptoms. Patients with higher education showed a lower SDS index and cognitive scale. Marital status, tumor stage, histologic grading, and substance abuse had no influence., Conclusions: Patients with a higher risk of depression should receive long-term monitoring during and after the end of radiotherapy, and prompt intervention strategies should be applied., (Copyright 2003 Wiley Periodicals, Inc. Head Neck 25: 000-000, 2003)

Purpose: Psychosocial factors influence patient compliance and have an effect on survival. Identifying patients at risk of decreased quality of life (QOL) with no extra expenditure in terms of hospital staff time or resources is mandatory to plan psychosocial support., Methods and Materials: Between 1997 and 2000, 242 patients with head-and-neck cancer (30% pharyngeal, 29% oropharyngeal, and 13% laryngeal cancer) were screened. Of these, 28.5% refused to participate and 19.0% were excluded (Karnofsky performance score <50, language and cognitive deficits, death, or noncompliance). A total of 124 patients were assessed with the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire at ti1 (beginning of radiotherapy [RT]). Eighty-three patients from this group were examined, with complete data from ti1 to ti3 (6 weeks after RT)., Results: The QOL did not change during RT. In logistic regression analysis, medical information, in contrast to sociodemographic variables, turned out to have no influence on the ability to predict low QOL (sensitivity 80% vs. 32%). Four sociodemographic variables were entered in the regression model (children, currently employment, ethanol abuse, level of secondary education) and accounted for 26% of variance in QOL at ti3., Conclusion: By routinely obtaining clinical information from the patient's history, patients at risk of low QOL after RT can be identified and could benefit from early psychosocial support.