Osteoporosis, by far the most common metabolic disease in western countries, is a systemic skeletal disease with quantitative abnormality of bone whereas, in rickets/osteomalacia, there is qualitative abnormality of bone (Fig. 1). Osteoporosis is characterized by reduction in bone mass (amount of mineralized bone per volume unit) and by altered trabecular structure due to a loss of trabeculae interconnectivity with a consequent increase in bone fragility (decrease in biomechanical strength) and susceptibility to fracture (insufficiency fractures) [3] (Fig. 2).

In the western world, osteoporosis is reported to affect 1 in 2 women and 1 in 5 men older than the age of 50 years in their life time [4]. The risk of fracture increases with advancing age and progressive loss of bone mass and varies with the population being considered. The incidence of hip fracture has doubled over the past three decades and is predicted to continue to grow beyond what one would predicted from increased longevity. At 1 year after hip fracture, 40% patients are unable to walk independently, 60% have difficulty with one essential activity of life, 80% are restricted in other leaving activity, and 27% will be admitted to a nursing home for the first time [4].

Generalized osteoporosis is the end-stage of several diseases and can be either primary or secondary (Tables 2 and 3). Clinical significance of osteoporosis is limited until fractures develop. Vertebral fractures are the most commonly occurring osteoporotic fractures (Fig. 2). They occur as an acute event related to minor trauma or spontaneously. Vertebral fractures cause disability and limited spinal mobility and are associated within increased morbidity but symptoms generally resolve spontaneously over 6–8 weeks. They are powerful predictors of future fracture with a 12% increased risk of a future vertebral fracture within 12 months if a single vertebral fracture is present (22% increased risk in the presence of multiple fractures) [2]. Questionnaires are also available to evaluate fracture risk (FRAX). Consequently, the accurate identification and clear reporting of vertebral fracture by radiologists have a vital role to play in the diagnosis and appropriate management of patient with, or at risk of, osteoporosis.

Radiography is relatively insensitive in detecting early bone loss and 30–40% loss of bone tissue usually remain occult on radiographs. Radiographic bone density is also affected by patient characteristics and technical parameters used to obtain the radiographs. The subjectivity of visual judgement of bone density on conventional radiographs supports the value of modern quantitative techniques such as bone densitometry.

The main radiographic features of generalized osteoporosis include cortical thinning and disappearance of the trabecular network. Resorption and thinning of trabeculae initially affect secondary trabeculae that are parallel to the biomechanical stresses and the primary trabeculae that are perpendicular to the biomechanical stresses may appear more prominent because they are affected at a later stage (Fig. 3).

Cortical thinning occurs as a result of endosteal, periosteal, or intra-cortical (cortical tunnelling) bone resorption. Endosteal resorption is the least specific radiographic finding because it may be evident in metabolic disorders, including osteoporosis but also in marrow disorders. Intra-cortical tunnelling is more specific occurring mainly in disorders with rapid bone turnover such as diffuse osteoporosis and reflex sympathetic osteodystrophy. Subperiosteal resorption is the most specific finding, being diagnostic of hyperparathyroidism.

Osteoporosis remains occult on MR images (Fig. 3) although a relationship between trabecular bone density and marrow fat has been reported. The presence at spine MRI of multiple vertebral fractures including acute (marrow edema) and healed (marrow fat) fractures suggests increased bone fragility and, hence, osteoporosis.

Several quantitative techniques including Dual energy X-ray absorptiometry (DXA), quantitative computed tomography (QCT), and quantitative ultrasonography (QUS) have been developed to enable accurate and precise assessment of mineral bone density [2]. Several methods have been used in the past to standardize measurements of cortical thickness (radiogrametry), trabecular pattern (Singh index), and vertebral deformity (morphometry) from radiographs (Table 4). These measurements lack accuracy or predictive value but they remain worth reporting when present on radiographs.