Radiation-induced brachial plexus injury: follow-up of two different fractionation schedules.

All 449 breast cancer patients treated with post-operative radiotherapy to the breast and lymph nodes between 1982 and 1984 have been followed for 3-5.5 years. In this group two different fractionation schedules were used, one five times a fortnight and one daily, both over 6 weeks. The calculated dose to the brachial plexus was 45 Gy in 15 fractions or 54 Gy in 30 fractions. These schedules are equivalent doses using the standard NSD formula. The diagnosis of a brachial plexus injury was made clinically and computed tomography was used to distinguish radiation injury from recurrent disease. The actuarial incidence of a radiation-induced brachial plexus injury for the whole group was 4.9% at 5.5 years. No cases were seen in the first 10 months following radiotherapy. The incidence rises between 1 and 4 years and then starts to plateau. When the large fraction size group is compared with the small fraction size group the incidence at 5.5 years is 5.9% and 1.0%, respectively (p = 0.09). Two different treatment techniques were used in this group but were not found to contribute to the probability of developing a brachial plexus injury. It is suggested that radiation using large doses per fraction are less well tolerated by the brachial plexus than small doses per fraction; a commonly used fractionation schedule such as 45 Gy in 15 fractions may give unacceptably high brachial plexus morbidity; and the use of small doses per fraction or avoiding lymphatic irradiation is advocated.