Tumors

69. Physical function is reduced and correlated to quality of life after limb-sparing surgery due to bone sarcoma
Linda Fernandes, Kreutzfeldt Zebis Mette, Michala Skovlund Sørensen , Christina Enciso Holm , Allan Villadsen , Michael Mørk Peteresen
Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet and University College Copenhagen ; Department of Midwifery, Physiotherapy, Occupational Therapy and Psychomotor Therapy, University College Copenhagen; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet


Background: Patients with bone sarcoma report that quality of life (QoL) is affected years after treatment. In rehabilitation, both QoL and different measures of physical function are assessed and targeted. However, few studies have investigated physical function and its relationship to QoL in this patient group.
Purpose / Aim of Study: The purpose was to investigate if physical function in patients with bone sarcoma receiving limb-sparing surgery (LSS) and reconstruction with a tumour- prosthesis was affected and associated with QoL.
Materials and Methods: Thirty patients operated (proximal femur (n=12), distal femur (n=14), proximal tibia (n=4)) between 2006 and 2016 were compared to 30 healthy controls in a cross-sectional design. Differences between groups were tested for objectively measured physical function: isometric muscle strength in gluteus medius and quadriceps, six-minute walk test (6MWT) and 30s sit-to-stand test (STS). In the patient group, measures of physical function were correlated with EORTC QLQ-C30 Global Health (GH), Physical functioning (PF) and Role functioning (RF) subscales. Paired samples t-test was used for analyses of between group differences and Spearman’s rank correlation for associations within the patient group.
Findings / Results: Mean age of participants was 51 years and mean BMI 26. Significant (statistical and clinical) differences were found for all muscle strength tests between the two groups, both for comparisons of the leg that underwent surgery and the contralateral leg. Significant differences were also seen for the 6MWT (499 vs. 607 m, p<0.001) and the STS (12 vs. 18, p<0.001). Physical function was correlated to QoL when comparing muscle strength of gluteus medius with GH (rho 0.43, p=0.019) and with PF (rho 0.51, p=0.006), and quadriceps strength with PF (rho 0.40, p=0.034) and with RF (rho 0.40, p=0.038).
Conclusions: Patients with bone sarcoma receiving LSS and reconstruction with a tumour-prosthesis showed deficits in muscle strength, 6MWT and STS when compared to controls. Decreased strength of gluteus medius and quadriceps muscles correlated to QoL subscales, indicating that strength training could be a potential factor for improving physical components of QoL.

70. Surgical Treatment of Proximal Femoral Metastases: A Systematic Review and Meta-Analysis of Reoperations and Implant Failures
Mads Sten Andersen, Michala Skovlund Sørensen, Michael Mørk Petersen
Orthopedic surgery, Rigshospitalet; Orthopedic surgery, Rigshospitalet; Orthopedic surgery, Rigshospitalet


Background: The continuous improvements in survival of cancer patients increases the prevalence of metastatic bone disease, while also altering the demands for the durability of the surgical implants used. In the appendicular skeleton the metastatic lesions requiring surgery are most often located in the proximal femur, and are currently treated with either open reduction and internal fixation (ORIF), intramedullary nailing (IMN) or endoprosthetic reconstruction (EPR).
Purpose / Aim of Study: To investigate whether there is a difference in reoperation or implant failure rates between the currently used methods of surgical treatment of metastatic bone disease of the proximal femur.
Materials and Methods: The literature was systematically reviewed through a PubMed search in order to identify studies concerning surgical treatment of proximal femur metastases in the period 01.01.2009-04.01.2020. The primary outcome measure was reoperation rate, while implant failure rate was registered as the secondary outcome measure. Eighteen studies, with a total of 2489 patients, were included for analysis, all retrospective and non-randomized. In a meta-analysis pooled estimates for the odds-ratios (OR) between treatments regarding both reoperation and implant failure were calculated with 95 % confidence intervals (CI).
Findings / Results: Meta-analysis showed more patients treated with ORIF needed further surgery, with the pooled OR estimate being 3.07 (95% CI: 1.98-4.76, p<0.001) for reoperation and 8.47 (95% CI: 5.05-14.21, p<0.001) for implant failure compared to EPR. No difference was found when comparing reoperation after IMN to EPR (OR = 1.04 (95% CI: 0.77-1.42, p=0.79)), but the analysis showed a higher risk of implant failure after IMN with an OR of 2.65 (95% CI: 1.74-4.02, p<0.001).
Conclusions: The higher reoperation rate found in patients treated with ORIF, suggests that this method of treating proximal femur metastases should be considered obsolete. Analysis of implant failure further indicates, that EPR provides a more mechanically stable solution compared to IMN. The mediocre evidence that the included studies provide calls for further research reporting time-adjusted reoperation rates, allowing for more optimal comparison of methods between studies.

71. Development and comparison of one-year survival models in patients with primary bone sarcomas - External validation of a Bayesian belief network model and creation and external validation of a new Gradient Boosting Machine model.
Christina Holm, Clare f. Grazal, Mathias Raedkjaer, Thomas Baad-Hansen, Rajpal Nandra, Robert Grimer, Jonathan A. Forsberg, Michael Mørk Petersen, Michala Skovlund Sørensen
Orthopedic surgery, Rigshospitalet; Orthopaedics , USU-Walter Reed Department of Surgery, Bethesda, MD, USA; Department of Orthopaedic Surgery, Aarhus University Hospital, Denmark; Department of Orthopaedic Surgery, Aarhus University Hospital, Denmark; Orthopedic, The Royal Orthopaedic Hospital, Birmingham, UK; Orthopedic, The Royal Orthopaedic Hospital, Birmingham, UK; Orthopaedics , USU-Walter Reed Department of Surgery, Bethesda, MD, USA; Orthopedic surgery, Rigshospitalet; Orthopedic surgery, Rigshospitalet


Background: Bone sarcomas often presents late with advanced stage at diagnosis, resulting in varying short-term survival. In 2016 Nandra et al., generated a Bayesian belief network model (BBN) for 1-year survival of patients with bone sarcomas.
Purpose / Aim of Study: The purpose of present study is to: 1) External validate the prior 1-year BBN prediction model for survival of patients with bone sarcomas, 2) To develop a Gradient Boosting machine (GBM) model using Nandra et al.’s cohort and evaluate if the GBM model outperform the BBN model suggested by Nandra et al. when externally validated on an independent Danish population cohort.
Materials and Methods: The training cohort comprised 3493 patients newly diagnosed with bone sarcoma from the institutional prospectively maintained database at The Royal Orthopaedic Hospital, Birmingham UK. The validation cohort comprised 771 patients with newly diagnosed bone sarcoma included from The Danish Sarcoma Registry between January 1st, 2000 and June 22sd, 2016. Predictive performance of models was evaluated by area under receiver operator characteristic curve (AUC ROC) analysis, Brier score and decision curve analysis (DCA).
Findings / Results: External validation of the BBN 1-year prediction model demonstrated AUC ROC of 68% (95%CI, 62%-73%). AUC ROC of the GBM model demonstrated 75% (95%CI: 70%-80%), overall model performance by Brier score was 0.09 (95%CI: 0.077-0.11) and DCA demonstrated a positive net- benefit for threshold probabilities above 0.5. External validation of the developed GBM model demonstrated AUC ROC of 63% (95%CI: 57%-68%) and the Brier score was 0.14 (95%CI: 0.12-0.16).
Conclusions: External validation of the 1-year Bayesian belief network survival model yielded poor outcome and is hence not recommendable for clinical usage. The developed Gradient Boosting Machine 1-year survival model did not outperform the prior Bayesian belief network model, and modernization is pending.

72. Work ability and physical activities in patients with tumour prosthesis in hip or knee following bone sarcoma. A cross-sectional study comparing patients with healthy controls.
Linda Fernandes, Mette Kreutzfeldt Zebis, Allan Villadsen, Christina Encisio Holm, Michala Skovlund Sørensen, Lars L Andersen, Michael Mørk Peteresen
Faculty of Health and Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, University College Copenhagen and Rigshospitalet; Faculty of Health, University College Copenhagen; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet; , National Research Centre for the Working Environment; Musculoskeletal Tumour Section, Department of Orthopaedic Surgery, Rigshospitalet


Background: While most patients going through limb-sparing surgery (LSS) due to bone sarcoma are of the working-age population, limited knowledge exist about their work ability.
Purpose / Aim of Study: The study aim was to assess work ability and physical activity (PA) in these patients.
Materials and Methods: This cross-sectional study compared employed patients, receiving LSS and reconstruction in proximal (n=9) or distal femur (n=7) or proximal tibia (n=4) between 2006 and 2016, with employed healthy controls (n=20). The Musculoskeletal Tumour Society Score (MSTS) was used as descriptive information. The Work Ability Index (WAI), financial difficulties, the Patient Specific Functional Scale (PSFS), activity monitor and the International Physical Activity Questionnaire (IPAQ) were used to evaluate work ability and physical activities. Between group differences were assessed using unadjusted and adjusted (sex, age, BMI, educational level, type of work) general linear models.
Findings / Results: The patients were seen on average 7 (±2.9) years post-surgery, had a mean age of 43 (±13.6) years, BMI 27 (±3.7) and MSTS of 69 (±14.6)%. Six patients (30%) experienced that their physical condition had caused financial difficulties. The adjusted analyses showed differences in general work ability (7.8 vs. 9.1 points, p=0.002) and work ability due to physical demands (3.2 vs. 4.6 points, p<0.001) between patients and controls. There were between group differences in PSFS mean score (2.8 vs. 9.5 points, p<0.001). No differences were seen in step counts/day (10588 vs. 12239 steps, p=0.144) or the IPAQ (4107 vs. 4035 METs/week, p=0.942).
Conclusions: Most patients experienced difficulties in performing tasks requiring physical demands at work. Furthermore, 1/3 reported financial difficulties caused by their condition. Although we found no differences in PA, patients reported great difficulties in performing activities that they scored as important. Assessments of work demands and collaboration with workplaces should be considered in rehabilitation following LSS and reconstruction with tumour prosthesis.

73. Quantitative measurements of adaptive bone remodeling around the Cemented Zimmer® Segmental stem after tumor resection arthroplasty using dual-energy X-ray absorptiometry
Christina Holm, Peter Horstmann, Michala Skovlund Sørensen, Karen Dyreborg, and Michael Mørk Petersen
Musculoskeletal Tumor Section, Department of Orthopedic Surgery, Rigshospitalet, University of Copenhagen, Denmark


Background: Limb salvage surgery is currently offered to more than 90% of patients with bone or soft tissue sarcomas and to a greater extent also to patients with metastatic bone disease.
Purpose / Aim of Study: The aim of the present study was to evaluate the adaptive remodeling of the periprosthetic cortical bone after insertion of a tumor prosthesis with a cemented stem.
Materials and Methods: A prospective study of 21 patients (F/M=12/9), mean age 55 years (range 15-81) with metastatic bone disease (n=9), sarcomas (n=8) or aggressive benign tumors (n=4) who underwent bone tumor resection and reconstruction with a tumor prosthesis (Zimmer® Segmental 130 mm straight fluted cemented stem with trabecular metal (TM) collars) in the proximal femur (n=10), distal femur (n=9) or proximal tibia (n=2) . Measurements of bone mineral density (BMD) were done postoperatively and after 3, 6, and 12 months of the periprosthetic bone and in both ankles by using dual-energy X-ray absorptiometry. BMD (/g/cm2) was measured in 4 regions of interest around the cemented stem and in one region of interest 1 cm proximal from the ankle joint. Repeated measures ANOVA and students paired t-test was used to evaluate BMD changes over time
Findings / Results: At 1-year follow-up, BMD compared to the postoperative value was seen in all 4 regions of interest with a statistically significant bone loss of 8- 15%. The bone loss was most pronounced (14-15%) in the 2 regions of interest closest to the TM collar and lowest (8%) adjacent to the tip of the stem. After 1 year the decrease in BMD of the ankle on the affected extremity was 9% and the ankle on the contralateral extremity was close to baseline.
Conclusions: The periprosthetic BMD around the cemented 130 mm Segmental stem decreased significantly during the first postoperative year and is considered caused by a combination of stress shielding and immobilization.

74. Reconstruction of the humerus using the Comprehensive Segmental Revision System in patients suffering from metastatic bone disease.
Claus Lindkær Jensen, Michala Skovlund Sørensen, Michael Mørk Petersen
Musculoskeletal Tumor Section, Department of Orthopedics, Rigshospitalet, University of Copenhagen; Musculoskeletal Tumor Section, Department of Orthopedics, Rigshospitalet, University of Copenhagen; Musculoskeletal Tumor Section, Department of Orthopedics, Rigshospitalet, University of Copenhagen


Background: Surgical treatment of malignant bone tumors of the humerus is often followed by reconstruction with a tumor prosthesis. The Comprehensive Segmental Revision System® (Zimmer-Biomet, Warsaw, IN, USA) (SRS) offers a new option for such reconstructions. Biomet, Warsaw, IN, USA) (SRS) offers a new option for such reconstructions.
Purpose / Aim of Study: The aim og this study was to evaluate implant failure incidence, surgical complications, and clinical results.
Materials and Methods: A study of 22 consecutive patients (F/M= 9/13, mean age = 67 (51-83) years) suffering from metastatic bone disease having surgery with bone resection of the humerus (18 proximal humerus replacements (hemiarthroplasty/total-reverse joint = 4 4/14) and 4 distal humerus r replacements) and reconstruction u using the SRS prosthesis from May 2 2014 to January 2017. Statistics: Kaplan-Meier survival analysis (patient survival) and Aalen-Johansson estimate (incidence of implant failure) presented with 95% Johansson estimate (incidence of implant failure) presented with 95%- -confidence intervals (CI). Results are g given as mean (range).
Findings / Results: Two-year overall patient survival was 41% (CI: 20-62%). Five patients suffered from surgical complications: radial nerve palsy (n=2) and superficial postoperative infection (n=3). Two patients experienced revision surgery: soft tissue revision (n=1) and a hemiarthroplasty revised to a reverse total shoulder replacement (n=1). Two-year implant failure incidence was 5% (CI:0 year implant failure incidence was 5% (CI:0-13%). The mean MSTS score (n=6) was 16 (11-25), 221 (95-360) days postoperatively.
Conclusions: The use of the SRS prosthesis in orthopedic oncology patients resulted in low incidence of implant failure. Since the introduction of the SRS prosthesis in our department represented at shift from using hemiarthroplasty to prefer total-reverse joint implants, the fact that we observed no shoulder dislocations was a positive short-term result.

75. Ewing´s sarcoma of the calcaneus treated by limb sparing surgery with calcanectomy and reconstruction with a composite of an allograft and a vascularized osteocutaneous fibula graft.
Michael Mørk Petersen, Lisa Toft Jensen, Christian Bonde, Werner Herbert Hettwer
Department of Orthopedics (Musculoskeletal Tumor Section), Rigshospitalet, University of Copenhagen, Denmark; Department of Plastic Surgery, Rigshospitalet, University of Copenhagen, Denmark; Department of Plastic Surgery, Rigshospitalet, University of Copenhagen, Denmark; Department of Orthopedics (Musculoskeletal Tumor Section), Rigshospitalet, University of Copenhagen, Denmark


Background: Primary malignant bone tumors of the calcaneus are very seldom, and due to poor possibilities to do surgery with wide margins in this region and limited options for reconstruction after calcanectomy many orthopedic oncologists use amputation as the preferred surgical treatment in such cases.
Purpose / Aim of Study: We present two cases of Ewing´s sarcoma of the calcaneus treated with calcanectomy and reconstruction with a composite of an allograft and a vascularized osteocutaneous fibula graft.
Materials and Methods: The medial cases of 2 girls suffering from Ewing´s sarcoma of the calcaneus that were almost 6 years old (case 1) and 16 years old (case 2) at the time of surgical treatment with calcanectomy in respectively August 2012 and October 2013 are presented. Both patients were without metastatic disease and received pre- and post-operative chemotherapy.
Findings / Results: In both cases removal of the calcaneus was performed using a combined medial and lateral incision. In case 1 a femoral head allograft was fitted to replace the removed calcaneus, and in case 2 a calcaneus allograft was used. In both cases, with the aim of obtaining arthrodesis, the allograft was fixed to the talus and cuboid bone with screws. A distally pedicled osteocutaneous flap was use for reconstruction of soft tissue, and a 5-6 cm piece of vascularized fibula bone was fitted into the allograft and fixed using staples. Arthrodesis between talus and the graft healed and full weight- bearing was allowed in both cases 8-9 months postoperatively. At follow-up 6½ and 4½ years after surgery both patients were without local recurrence or metastases, and they were both pain-free and able to walk using normal footwear without any walking-aids.
Conclusions: Biological reconstruction is possible after removal of the calcaneus because of primary malignant bone tumors and long-term good functional results can be achieved.