215 / 225
DOS Kongressen 2017 ·
215
Update and external validation of the SPRING score
for prediction of survival in patients having surgery
for metastatic bone disease the appendicular skel-
eton
Michala Skovlund Sørensen, Thomas A. Gerds, Klaus Hindsø, Michael Mørk Pe-
tersen
Department of Orthopedic Surgery, Rigshospitalet, University of Copenhagen,
Musculoskeletal Tumor Section; Department of Biostatistics, University of Co-
penhagen; Department of Orthopedic Surgery, Rigshospitalet, University of
Copenhagen, Section of pediactrics; Department of Orthopedic Surgery, Rig-
shospitalet, University of Copenhagen, Musculoskeletal Tumor Section
Background:
The SPRING (Sørensen, PeteRsen, hINdsø, Gerds) score was in-
troduced in 2016 as tool to assist the surgeon in the decision making of a sur-
gical implant that will outlive the patient in the treatment of metastatic bone
disease in the appendicular skeleton (MBDA). The scores performance in a pop-
ulation based cohort is pending.
Purpose / Aim of Study:
Refit and external validate the SPRING score in a
population based cohort.
Materials and Methods:
270 patients having surgery with total joint replace-
ment for MBDA from January 2003 – December 2013 was used to refit the
model and an independent cohort of 165 patients having surgery in the Capital
Region of Denmark from May 19, 2014 to May 18, 2016 was used as a valida-
tion cohort. Survival outcome was predicted at 3, 6 and 12 months after sur-
gery using a logistic regression model fitted with primary cancer, pre-operative
hemoglobin, fracture versus impending fracture, Karnofsky score, visceral me-
tastases, multiple bony metastases and American Society of Anaesthesiologist’s
score. We evaluated using Brier score, AUC of ROC and calibration plots.
Findings / Results:
The predictive scores obtained showed AUC values of
81.9 % (C.I.: 72.5%-91.2%),84.5% (C.I.: 75.6%-93.3%) and 85.8 (C.I.: 76.7-
94.9%) at 3, 6 and 12 months respectively. Brier score was 0.155, 0.162 and
0.152 at 3, 6, 12 months respectively. The model showed good calibration at
all three end- points.
Conclusions:
The SPRING score is applicable to a general population to esti-
mate residual life expectancy after surgery for MBDA and can assist the surgeon
in decision making in regard to surgical solution.
No conflicts of interest reported
167.