Comparison of Reproducibility in Volume Measurement with Plain Radiograph and Magnetic Resonance Imaging of Osteosarcoma.
- Author:
Seong Hwan MOON
1
;
Kyoo Ho SHIN
;
Jin Suck SUH
;
Soo Bong HAHN
Author Information
1. Department of Orthopaedic Surgery, Yonsei University College of Medicine, Seoul, Korea.
- Publication Type:Original Article
- Keywords:
Osteosarcoma;
Volume measurement;
Reproducibility;
MRI
- MeSH:
Chemotherapy, Adjuvant;
Cisplatin;
Doxorubicin;
Drug Therapy;
Extremities;
Gadolinium DTPA;
Humans;
Magnetic Resonance Imaging*;
Observer Variation;
Osteosarcoma*;
Radiography;
Tumor Burden
- From:The Journal of the Korean Orthopaedic Association
1999;34(6):1035-1039
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
PURPOSE: The volumetric change of osteosarcoma after preoperative adjuvant chemotherapy is a significant prognostic factor. It is absolutely necessary that all volumetric measurements must be based on acceptable reproducibility. The purpose of this study was to investigate intra- and inter-observer variability of volumetric measurements, including plain radiography and magnetic resonance image (MRI). MATERIALS AND METHODS: Forty-one patients with osteosarcoma of the limb were included in this study. All patients underwent plain radiographs and MRIs before and after preoperative adjuvant chemotherapy (intraarterial Cisplatin and intravenous Adriamycin, 3 cycles). Volume measurement, using ellipsoid formula from maximal length, width and depth of the tumor on plain radiograph and MRI, were calculated. Three dimensional summation of the tumor volume from MRI, with digitized scanner and software (MatLab. MatWork Inc, Natick, MA, USA) was also performed. Coefficient of variation (CV) of each method was compared. RESULTS: CVs for intra- and inter-observer variability in plain radiograph using ellipsoid formula were 9.4% and 11.7% in prechemotherapy and 8.1% and 9.3% in post chemotherapy. CVs for intra- and inter-observer variability in MRI using ellipsoid formula were 7.3% and 7.9% in prechemotherapy, and 7.6% and 8.2% in postchemotherapy in T1 weighted image (T1WI) , 6.7% and 7.1% in prechemotherapy and 6.5% and 7.6% in postchemotherapy in T2 weighted image (T2WI), and 7.8% and 8.5% in prechemotherapy and 8.2% and 8.7% in postchemotherapy in Gd-DTPA enhanced image (GdEI) . CVs for intra- and inter-observer variability in MRI using three dimensional summation of tumor volume were 2.4% and 3.2% in prechemotherpay and 2.7% and 3.4% in postchemotherapy in T1WI, 1.2% and 1.4% in prechemotherapy and 1.3% and 1.4% in postchemotherapy in T2WI, and 3.8% and 4.5% in prechemotherapy and 3.1% and 4.6% in postchemotherapy in GdEI. CONCLUSION: There was higher reproducibility, that is lower CV, in three dimensional summation of tumor volume in T2WI. The statistically significant volume change after chemotherapy was 4.2% decrease or increase in volume compared with its original tumor volume in T2WI.
