A 5-Year Long-term Follow-up Study of a Cross-Sectional Cohort Study (PK-VF) For the Examination of the Association of Vitamin D/Bone Turnover/Bone Mineral Density With an Incident Fracture in Chinese Postmenopausal Women

1. In 2013, 5 years after PK-VF, the same 2070 subjects were contacted by the original sites. Among them 1242 subjects were able to come for the follow-up assessment.
2. Clinical assessments: The participant's bio-information, physical examination and medical history were collected;Questionnaire including social/life style and medical evaluations (years since menopause (YSM), fracture history, milk/yoghourt/coffee/wine intake, calcium intake, or smoking history) were collected by PK-VF investigators. Non-vertebral fracture history evaluation: specific non-vertebral fracture sites include rib or clavicle, forearm, upper arm, hand (including wrist), pelvis, hip, thigh (not including hip), leg, and foot (including ankle). When non-vertebral fractures are suspected, questions were raised to the participant to eliminate possible biases (How did you get these fractures, a slight fall at home, fell from a high place, hit by someone, broken during a car accident or an operation? Did you see a doctor to confirm these fractures?) A fracture occurred in regular daily activities or due to mild trauma was defined as fragile non-vertebral fracture.
3. Biochemical measurements: Fasting blood sample (~5 ml) was collected from each participant at participating sites; In 2007-2008 study, blood samples were collected during April-July, while in the 5-year follow-up; samples were collected in the same period of time. C-terminal telopeptide of type I collagen (β-CTX), N-aminoterminal prepeptide of type I procollagen (P1NP), and 25 (OH) D will be determined by a laboratory method of electrochemiluminescence (E170; Roche Diagnostics, Basel, Switzerland) in the institute (Peking Union); Chemistry including alkaline phosphatase (ALP), calcium (Ca), creatinine (Cr), and glucose, will be measured by using automated techniques in the institute (Peking Union);
4. BMD measurements: Lumbar spine (LS) and femoral neck (FN) BMDs by dual-energy x-ray absorptiometry (DXA) (Lunar or Norland) at PK-VF sites. BMD calibration: The participant's BMD were evaluated by the same type of DXA as previous. The coefficients of variation of the seven hospitals were 0.75% to 1.7% for LS and 0.56% to 1.0% for FN. Cross-calibration equations between machines are: LS BMD (g/cm^2) Lunar = 1.012 × Norland + 0.0137 and, FN BMD (g/cm^2) Lunar = 1.0377 × Norland + 0.00026
5. Vertebral fracture diagnosis: Lateral radiographs of the thoracolumbar spine (T4-L5) were taken at PK-VF sites. Vertebral fractures will be assessed using Genant's semi-quantitative visual criteria. Two specialist radiologists will independently evaluate and diagnose vertebral fracture. A worsened existing vertebral fracture will be regarded as a new vertebral fracture.

In 2007-2008 study(Published article about this study could be found in Pubmed, PMID: 24760246), 2070 participants were recruited in this cohort, and 837 subjects (40%) were diagnosed as osteoporosis. After 5 years, 1242 subjects agreed to be re-evaluated in 2013. Questionnaires and blood samples were collected, and BMD and spine x-ray were obtained at the 5-year follow up. We estimate that around 625 subjects would be diagnosed as osteoporosis. The remaining works include blood sample test (25(OH)D, CTX and P1NP),spine x-ray films reading, data input and statistical analysis, paper writing and publication.