主 办:北 京 中 医 药 大 学
ISSN 1006-2157 CN 11-3574/R

JOURNAL OF BEIJING UNIVERSITY OF TRADITIONAL CHINESE MEDICINE ›› 2020, Vol. 43 ›› Issue (6): 516-521.doi: 10.3969/j.issn.1006-2157.2020.06.012

• Clinical Studies • Previous Articles     Next Articles

Preliminary study on diagnostic criteria for polycystic ovary syndrome in terms of pattern elements based on latent variable analysis*

Xing Yu1, Lu Qiudan1, Shen Lingyu2, Liu Haitao3, Tong Qing1, Huang Haitao1, Yang Yan1, Zheng Lingqi1, Li Hongbo1, Xiao Shuangshuang4, Xiao Hui1, Sun Junjian5, Liu Yanxia1#   

  1. 1 Dongfang Hospital, Beijing University of Chinese Medicine, Beijing 100078, China;
    2 Department of Gynecology, Beijing Hospital of Traditional Chinese Medicine in Shunyi, Beijing 101300, China;
    3 Department of Gynecology, Xuanwu TCM Hospital Beijing, Beijing100050, China;
    4 Department of Gynecology, Beijing Gulou Hospital of Chinese Medicine, Beijing 100009, China;
    5 Department of Gynecology, Beijing First Hospital of Integrated Chinese and Western Medicine, Beijing 100026, China
  • Received:2019-10-16 Published:2020-07-06
  • Contact: Prof. Liu Yanxia, M.D., Chief Physician, Doctoral Supervisor. Dongfang Hospital, Beijing University of Chinese Medicine. No. 6 District 1, Fangxingyuan, Fangzhuang, Fengtai District, Beijing 100078. E-mail:lyx7028@sina.com
  • Supported by:
    National Natural Science Foundation of China (No. 81904241)

Abstract: Objective To explore the criteria for identifying pattern elements of polycystic ovary syndrome (PCOS), and to provide evidence for setting TCM clinical pattern differentiation criteria for PCOS. Methods Clinical data of 518 PCOS patients were collected. Latent variable analysis was made to establish TCM diagnostic criterion model of PCOS based on pattern element, after which, diagnoses made using the model were compared with clinical diagnoses made by TCM physicians. Results Both goodness of fit index (GFI) and GFI adjusted for degrees of freedom (AGFI) were close to 1, suggesting that the model was well-fitting. The root mean square error of approximation (RMSEA) was 0.063 7, which is≤0.08, indicating that the model was acceptable. The results of consistency test were as follows. TCM diagnosis of the pattern element of disease location or affected zang-fu organ (i.e. kidney, liver, and spleen) could be made when just one symptom in the model manifests itself (i.e. lumbosacral pain, frequent urination at night, and tinnitus for kidney; depression, impatience and irascibility, and frequent or excessive sighing for liver; and poor appetite and digestion, abdominal distention after eating, and diarrhea for spleen, respectively). The rates of consistency between the model diagnosis of disease location and pattern differentiation made by TCM physicians were 71.6%, 88%, and 77% for kidney, liver and spleen, respectively. TCM diagnosis of the pattern element of disease nature (i.e. qi deficiency, qi stagnation, and phlegm-damp) could be made when just two symptoms in one category in the model manifest themselves (i.e. mental fatigue, lack of strength, and spontaneous sweating for qi deficiency; chest tightness, chest and rib-side distending pain, and distending pain of the breasts for qi stagnation; and somnolence, heaviness sensation in the head or dizziness as if the head is wrapped up, and expectoration of phlegm for phlegm-damp, respectively). The rates of consistency between the model diagnosis of disease nature and pattern differentiation made by TCM physicians were 86.3%, 83.2%, and 80.5% for qi deficiency, qi stagnation, and phlegm-damp, respectively. Conclusion The preliminary diagnostic criteria for PCOS based on pattern elements have been set. After meeting basic diagnostic criteria for PCOS, disease locations such as kidney, liver, and spleen can be diagnosed based on just one symptom in the model, and disease nature like qi deficiency, qi stagnation, and phlegm-damp can be diagnosed based on just two symptoms in one category in the model.

Key words: polycystic ovary syndrome, pattern elements, cross-sectional study, latent variable analysis, structural equation modelling

CLC Number: 

  • R271.1