Category: Central DXA: (DXA, TBS)

19 - Inter- and intra-observer variability in image analysis of dual energy x-ray absorptiometry in a large hospital

Introduction: A large hospital having several dual-energy X-ray absorptiometry (DXA) scanners and many technologists is usually expected to show an increase in the precision error because technologists may position the patient and analyze the image in a different manner. Assessment of inter- and intra-observer variability as well as training the skills of the technologist is important for quality assurance of DXA.

Objectives: The purpose of this study was to evaluate the inter- and intra-observer variability in image analysis by trained technologists for measuring the BMD.

Materials and
Methods:
Thirty patients (normal patients 16, patients with osteopenia or osteoporosis 14) underwent BMD measurement at the lumbar spine (L1-L4) and femur using DXA (GE Lunar Prodigy) by a single technologist. Five trained technologists (range of working periods: 12-132 months, median 54 months) analyzed the same images of the patients twice within one week. During image analysis, no side-by-side comparison was allowed. We compared the BMD values for each region analyzed by the five technologists. Inter-observer variability was calculated using intraclass correlation coefficient (ICC) and intra-observer variability of each technologist was calculated using coefficient of variation (%CV).


Results:
There was no significant difference in the BMD values analyzed by the five technologists at L1-L4 and femur total region in both normal and abnormal groups (p>0.05). ICC was 0.96. Intra-observer variability (%CV) was 0.21-0.29 % in the L1-L4 region and 0.07-0.15 % in the femur total in the normal group, and intra-observer variability (%CV) was 0.19-0.57 % in the L1-L4 region and 0.04-0.26 % in the femur total in the abnormal group. (Table1)


Conclusions:
Both inter- and intra-observer variability in image analysis of DXA by trained technologists were very low and they were within the acceptable range for measuring the BMD.

Jong sook Choi

Assistant manager
Department of Nuclear Medicine, Asan Medical center
Seoul, Seoul-t'ukpyolsi, Republic of Korea

Eun Hye Kim

Assistant manager
Department of Nuclear Medicine, Asan Medical center
Seoul, Seoul-t'ukpyolsi, Republic of Korea

Jae Seung Kim

Professor in deaprtment of Nuclear Medicine
Asan Medical center
Seoul, Seoul-t'ukpyolsi, Republic of Korea

Chang kyeong Namgung

Manager
Department of Nucaler Medicine, Asan Medical center
Seoul, Seoul-t'ukpyolsi, Republic of Korea

Soon Ki Park

Cheif
Department of Nuclear Medicine, Asan Medical center
Seoul, Seoul-t'ukpyolsi, Republic of Korea

Min Kyoung Cha

Assistant manager
Departmetn of Nuclear Medicine, Asan Medical center
Seoul, Seoul-t'ukpyolsi, Republic of Korea