Prognostic factors for local recurrence in extremity-located giant cell tumours of bone with pathological fracture

Research output: Contribution to journalArticle

View graph of relations Save citation


  • M. R. Medellin
  • T. Fujiwara
  • R. M. Tillman
  • L. M. Jeys
  • J. Gregory
  • J. D. Stevenson
  • M. Parry
  • A. Abudu

Research units


The aim of this paper was to investigate the prognostic factors for local recurrence in patients with pathological fracture through giant cell tumours of bone (GCTB).

Patients and Methods
A total of 107 patients presenting with fractures through GCTB treated at our institution (Royal Orthopaedic Hospital, Birmingham, United Kingdom) between 1995 and 2016 were retrospectively studied. Of these patients, 57 were female (53%) and 50 were male (47%).The mean age at diagnosis was 33 years (14 to 86). A univariate analysis was performed, followed by multivariate analysis to identify risk factors based on the treatment and clinical characteristics.

The initial surgical treatment was curettage with or without adjuvants in 55 patients (51%), en bloc resection with or without reconstruction in 45 patients (42%), and neoadjuvant denosumab, followed by resection (n = 3, 3%) or curettage (n = 4, 4%). The choice of treatment depended on tumour location, Campanacci tumour staging, intra-articular involvement, and fracture displacement. Neoadjuvant denosumab was used only in fractures through Campanacci stage 3 tumours. Local recurrence occurred in 28 patients (25%). Surgery more than six weeks after the fracture did not affect the risk of recurrence in any of the groups. In Campanacci stage 3 tumours not treated with denosumab, en bloc resection had lower local recurrences (13%), compared with curettage (39%). In tumours classified as Campanacci 2, intralesional curettage and en bloc resections had similar recurrence rates (21% and 24%, respectively). After univariate analysis, the type of surgical intervention, location, and the use of denosumab were independent factors predicting local recurrence. Further surgery was required 33% more often after intralesional curettage in comparison with resections (mean 1.59, 0 to 5 vs 1.06, 0 to 3 operations). All patients treated with denosumab followed by intralesional curettage developed local recurrence.

In patients with pathological fractures through GCTB not treated with denosumab, en bloc resection offers lower risks of local recurrence in tumours classified as Campanacci stage 3. Curettage or resections are both similar options in terms of the risk of local recurrence for tumours classified as Campanacci stage 2. The benefits of denosumab followed by intralesional curettage in these patients still remains unclear.


Original languageEnglish
Pages (from-to)1626-1632
JournalBone and Joint Journal
Issue number12
Early online date30 Nov 2018
Publication statusPublished - 1 Dec 2018

Employable Graduates; Exploitable Research

Copy the text from this field...