Patterns and Timing of Local Recurrence in Benign Bone Tumors.

BACKGROUND

Benign bone tumors represent a heterogeneous group of nonmalignant lesions often treated surgically when symptomatic or at risk of fracture. Although local recurrence is a recognized risk, comparative data on recurrence timing and presentation across histologies are limited, contributing to the lack of standardized surveillance. We aimed to examine locally recurrent benign bone tumors after surgery to define recurrence patterns by histology.

METHODS

We conducted a single-center retrospective cohort study of 329 curative-intent procedures for benign bone tumors at the University of Iowa Hospitals and Clinics (2010-2024). Included histologies were giant cell tumor (GCT), aneurysmal bone cyst (ABC), unicameral bone cyst (UBC), osteoid osteoma (OO), chondroblastoma (CB), osteoblastoma (OB), non-ossifying fibroma, and chondromyxoid fibroma. Recurrence was defined by histology or characteristic imaging. The primary outcome was mode of recurrence presentation (symptomatic vs. radiographic). Secondary outcomes included time to recurrence and recurrence by bone filler type. Analyses used Wilcoxon rank-sum, Fisher's exact, and Spearman correlation tests.

RESULTS

Forty-two recurrences (12.8%) were identified. Recurrence presentation varied by histology (p = 0.001): OO and OB recurred exclusively symptomatically (100%), CB mostly symptomatically (66.7%), while ABC (20.0%), UbC (21.1%), and GCT (37.5%) recurred predominantly asymptomatically. No recurrences occurred for non-ossifying fibroma or chondromyxoid fibroma. Recurrence timing was not associated with age, sex, tumor site, or surgical method. In GCT treated with curettage, recurrence was more common with allograft than cement (33.3% vs. 3.4%, p = 0.007). OB and OO recurred earliest, with 100% and 83.3% in the first year, respectively. By year three, all OO and most GCT, ABC, and UBC recurrences had occurred; by year five, all tumors except a single CB had fully recurred.

CONCLUSION

Recurrence patterns in benign bone tumors are strongly influenced by histology. OO and OB recur symptomatically within the first year, suggesting limited need for routine imaging. UBC, GCT, and ABC often recur asymptomatically, justifying structured imaging surveillance for 3-5 years. Cement use in GCT may reduce recurrence risk compared with allograft.

LEVEL OF EVIDENCE: III.