Huihui Xu ^1,2 Hong Liu ^3* MEIJIE LIU ³ Yan Li ^3* Jinghua Pan ^3* Shaojun Wang ^3* Guowei Wang ⁴ Xin Liu ^3* Ying Liu ^5* Xiaoqin Hou ^1* Zhao Hongyan ^6*

• ¹ Shenzhen Traditional Chinese Medicine Hospital, Shenzhen ，Guangdong, China
• ² The Fourth Clinical Medical College, Guangzhou University of Chinese Medicine, Shenzhen, China
• ³ Beijing Key Laboratory of Traditional Chinese Medicine Basic Research on Prevention and Treatment of Major Diseases, Experimental Research Center, China Academy of Chinese Medical Sciences, Beijing, China
• ⁴ Institute of Basic Theory for Chinese Medicine, China Academy of Chinese Medical Science, Dongcheng, China
• ⁵ Fangta Hospital of Traditional Chinese Medicine, Shanghai, China
• ⁶ Department of Clinic No.1 office, Shenzhen Traditional Chinese Medicine Hospital, The Fourth Clinical 14 Medical College of Guangzhou University of Chinese Medicine, Shenzhen, China

Introduction: Selection of regions of interest (ROI) is essential to obtain accurate data in analysis of Microcomputed tomography (micro-CT). There are distinct ROI selection methods for lumbar vertebrae in current osteoporotic animal model research. However, whether different ROI selection methods directly affect the results of quantitative micro-CT-based microarchitectural data is still unknown. This study aimed to compare diagnostic accuracy of central ROI (a convenient method but accuracy is unclear) with overall ROI (a wellrecognized but complicated method) micro-CT for measurement of lumbar vertebral trabecular bone microstructure in rats at different time points after ovariectomy (OVX).Methods: Rats were randomly divided into the baseline group (n=6), the sham-or OVX-operated groups, with the 3 months or 6 months sham-operated (Sham) group as controls (n=6), and other rats as the 3 months or 6 months OVX group (n=6). The fifth lumbar vertebrae were collected after ovariectomy and scanned by micro-CT. Quantitative analyses were performed by two ROI methods (central and overall ROIs) and the bone microarchitecture parameters were investigated.The results showed that the Tb. N of 3-month-old rats by the central ROI method was significantly lower than that by the overall ROI method. the Tb. Th of 6-month-old rats and 9-month-old rats was significantly increased compared with that of the 3-month-old rats Using the overall ROI analysis. The bone mineral density (BMD) and percentage bone volume (BV/TV) of the lumbar vertebrae was significantly lower by the central ROI than by the overall ROI method at all different groups of rats. OVX rats had significantly reduced lumbar vertebrae bone mass using both two ROI method. Moreover, in both normal and OVX rats, the BMD, BV/TV, and trabecular bone number (Tb. N) were significantly lower in the analysis by the central ROI than by the overall ROI.The consistency between the statistical results of the central ROI and the overall ROI are excellent for quantitative bone microstructure assessment. Moreover, this study suggested the potential value of the central ROI for evaluating osteoporosis of the lumbar vertebrae in OVX rats and provides a basis for analysing the morphological changes of lumbar trabecular.