Introduction: The combined influence of risk factors on osteoporosis being known, the problem whether it is correctly identified and quantified arises, aiming to the improvement of populational screening. The correct judgment related to osteoporotic pathology refers to fracture risk assessment.
Materials and methods: We investigated 2149 women aged 2091, without treatment for osteoporosis, for anthropometric, anamnestic and bone densitometry features. Relations between fracture history, clinical factors, anamnesis, bone mineral density (BMD) at different sites, were computed using bivariate analysis (chi-square or ROC curve method) and stepwise logistic regression, in order to assess the probability of clinical osteoporotic fracture.
Results: Of 271 women had a history of frailty fractures, mostly nonvertebral. They are statistically negatively associated to lumbar spine BMD, total femur, femoral neck, whole body and distal radius BMD; also, there are associations to numerous clinical factors. We retained, as independent predictors, function of the site taken into the logistic equation: fractures in first-degree relatives, renal lithiasis, age, body height and weight. We computed the probability of existing clinical osteoporotic fractures and drawn risk maps for each densitometric region of interest. These maps can be used as quick screening methods.
Discussion: Describing fracture risk groups, even in nonosteoporotic patients, by the means of a tool like the fracture risk map, adapted to regional populational features, is a step towards improving therapeutic protocols in osteoporosis.
Conclusions: BMD stays as the most powerful predictor of osteoporotic fractures, but it should be combined with clinical risk factors in order to improve the strategy of diagnosis. The presence of certain described risk factors is sufficient for recommending bone densitometry in individual cases.