Does Local Zoledronate Applied to Pasteurized Bone Autografts Improve the Likelihood of Union of Graft-Host Junctions after Limb-sparing Surgery?

BACKGROUND: Pasteurized bone autograft is a recycling biological reconstruction method for limb-sparing surgery when an allograft or other reconstruction technique is unavailable. Since the application of a local bisphosphonate to morselized allografts can reduce graft resorption and enhance bone formation without systemic complications, adding the local bisphosphonate to pasteurized bone autografts should reduce the graft resorption and improve the graft incorporation to host bone. However, no study that we know of has described the outcomes of local bisphosphonate application to massive allografts or pasteurized bone autografts. Thus, this study compared the outcomes of pasteurized bone autografts with and without local zoledronate. QUESTIONS/PURPOSES: (1) What is the survival of pasteurized bone autografts and what complications lead to graft removal? (2) Does treatment of pasteurized bone autografts with zoledronate alter the survival of pasteurized bone autografts compared with grafts without treatment? (3) Does the local application of zoledronate reduce the proportion of patients with fractures because of metaphyseal graft resorption? (4) Does local application of zoledronate improve union at the graft-host bone junction compared with untreated grafts?
METHODS: Between July 2011 and December 2019, we performed 538 musculoskeletal bone tumor resections. Of these, 101 patients underwent reconstruction with pasteurized bone autografts. Other reconstructions included tumor prostheses (150 patients), allografts (70 patients), reconstruction using a bone cement-plate construct (62 patients), and resection only (155 patients). We generally used pasteurized bone autograft when tumors showed an osteoblastic pattern, had less than one-third cortical destruction, and less than half of metaphyseal bone destruction. Six percent (6 of 101) were lost to follow-up, 6% (6 of 101) had incomplete clinical data, and 16% (16 of 101) had a follow-up period less than 2 years without an event, leaving 73 patients for evaluation. The median (interquartile range) age of the patients was 18 years (15 to 26). Ninety-seven percent (71 of 73) had a diagnosis of bone sarcoma. The median follow-up time was 46 months (33 to 75). From 2011 to 2014, 21 pasteurized bone autografts were prepared without local zoledronate, and from 2014 to 2019, 52 pasteurized bone autografts were prepared with local zoledronate because we thought it might improve union and reduce resorption of the graft. From our tumor registry database, we obtained age, sex, use of chemotherapy, graft length and location, pasteurized bone graft type, fixation methods, the use of local zoledronate, osteotomy gap, complications, proportion of grafts that united by 2 years, and local recurrences. Curves for graft survival were determined using the Kaplan-Meier method with the endpoint of autograft removal and metaphyseal fracture from graft resorption. The probabilities of graft removal were estimated by cumulative incidences using the competing risk analysis, where death was considered as the competing event. Intergroup differences in survival and multivariable analyses were performed using the log-rank test and a Cox regression analysis. A logistic regression model was used to evaluate the association between graft-host osseous union by 2 years and other baseline factors. Union was defined when a callus was seen to bridge the osteotomy line for at least three cortices in both the AP and mediolateral planes.
RESULTS: The 5-year survival rate of all 73 pasteurized grafts was 85% (95% confidence interval 74% to 92%). With the numbers available, we found no difference in the 5-year survival rates between grafts with and without local zoledronate (90% [95% CI 78% to 96%] versus 74% [95% CI 48% to 89%]; p = 0.30). Eleven percent (8 of 73) of patients had metaphyseal fractures because of graft resorption, primarily associated with osteoarticular grafts (5-year fracture-free survival 56% [95% CI 20 to 80]) rather than pasteurized graft-prosthesis composites (94% [95% CI 78% to 98%]) and intercalary grafts (91% [95% CI 50 to 99]; p = 0.001); there was no association with the use of local zoledronate (13%; 7 of 52) compared with those without local zoledronate (5%; 1 of 21) (odds ratio 3.1 [95% CI 0.4 to 27]; p = 0.43). Of the 84 graft-host bone junctions, 85% (71) of the grafts unified within 2 years, 7% (6) unified after 2 years, and 8% (7) of grafts showed nonunion. Union within 2 years was associated with fixation using plate compared with those with stem and with both stem and plate (odds ratio 6.6 [95% CI 1.4 to 31]; p = 0.02) and grafts treated with local zoledronate compared with those without treatment (OR 5.9 [95% CI 1.3 to 28]; p = 0.02).
CONCLUSION: The application of local zoledronate to pasteurized bone autografts for limb-sparing surgery improved the likelihood of graft union compared with untreated grafts, especially when the osteotomy junctions were fixed using plate osteosynthesis, but it did not appear to alter the proportion of patients who experience metaphyseal fracture of the grafts because of graft resorption. Although this is a small study, it suggests that the treatment of pasteurized bone autografts and perhaps bone allografts should be studied further to determine whether bisphosphonates or other adjuncts can improve the union time and return to function in patients undergoing bone tumor resections using these reconstruction types. LEVEL OF EVIDENCE: Level III, therapeutic study.