Tolerance of autologous and allogeneic bone grafts to therapeutic radiation in humans

Purpose: To examine the effect of perioperative irradiation on bone graft h ealing and functional integrity. Methods and Materials: Fifty-five bone grafts (10 autologus and 45 allogene ic) performed between 1978 and 1995 were evaluated retrospectively. Sixteen received preoperative radiation, 11 received postoperative, and 13 were tr eated with a combination of pre- and postoperative radiation. Fifteen nonir radiated grafts were randomly selected to serve as controls, Twenty-three o f the grafts were placed in patients who received chemotherapy in the perio perative period. Functional graft survival and radiographic healing quality were evaluated. Results: Overall rates of graft survival at 1 year were 89% for autografts and 79% for allografts. Graft survival rates were 86% and 68% at 1 and 5 ye ars for the irradiated group, and 67% and 58% for the control group. No sig nificant difference was seen in the Kaplan-Meier graft survival curves of t he two groups. There was a nonsignificant trend toward improved radiographi c healing quality in the control group. No significant differences in outco me based on treatment chronology were found with survival rates of 88% for preoperative treatment and 100% for postoperative treatment. No relation be tween outcome and bone dose (preoperative + postoperative dose), graft dose (postoperative dose), or mean dose/day was found. There was a trend (p = 0 .0525) toward worse outcome seen in the Kaplan-Meier curves of patients who received chemotherapy, This difference, however, was not seen in the I-yea r survival rates or healing quality. Tobacco use tended toward predicting f ailure, with 63% graft survival compared to 85% in nonsmokers (p = 0.09), H ealing quality was significantly lower in the smoking group. Conclusion: The low failure rate of grafts in irradiated sites, overall and compared to controls from this study and relevant literature, as well as t he lack of dose and time effects, does not support significant deviation fr om the indicated treatment regimen for patients who have received or are ex pected to receive a graft. The trend toward decreased quality of radiograph ic bone healing, and data published in relevant literature indicating impro ved healing when radiation is withheld until 3-4 weeks postoperatively sugg est this delay should be attempted when not expected to otherwise compromis e patient outcome. A nonsignificant trend only for the effect of chemothera py on bone grafts was seen, thus we do not recommend changes in its use as appropriate for disease management other than a preference against use duri ng the immediate perioperative period. (C) 1999 Elsevier Science Inc.