Statistics from Altmetric.com
We read with interest the article by Nolla et al, which demonstrates that only 3% of patients with symptomatic non-traumatic vertebral fracture have normal bone mineral density (BMD).1 We agree with their conclusions that in this clinical setting measurement of BMD is not required to confirm a diagnosis of osteoporosis before starting treatment.
A large number of studies have shown that a previous history of vertebral fracture increases the risk of future vertebral and non-vertebral fracture, independently of BMD.2 Vertebral fractures are also associated with significant morbidity, leading to an impaired quality of life3 and increased mortality.4 A recent study by Lindsay et al demonstrated the speed of disease progression in osteoporosis, with 20% of patients experiencing a new incident vertebral fracture within 12 months after a vertebral fracture.5 These data suggest that osteoporosis treatment should be started as soon as possible after a fracture has been diagnosed, as any delay in initiating treatment while waiting for bone densitometry may put the patient at risk of further fractures. The availability of dual energy absorptiometry (DXA) is poor in the United Kingdom in comparison with some other European countries. The Advisory Group Report on Osteoporosis noted that in the UK there were 1.6 DXA machines per million population, compared with 2.9 in the USA and 6.6 in France.6 The limitation of DXA machine provision in the UK compared with the clinical demand has led to long waiting lists for BMD measurements and a potential delay in starting osteoporosis treatment.
Under these circumstances, what is the evidence that patients can be treated solely on the basis of vertebral fracture without the need for BMD measurement? The majority of studies have evaluated drug treatment in patients with low BMD alone, or with low BMD and prevalent vertebral fractures. Studies of risedronate,7,8 raloxifene,9 and parathyroid hormone10 have, however, included patients with two or more asymptomatic vertebral fractures in the absence of BMD readings.
In the study by Harris et al 80% of patients had two or more vertebral fractures, and analysis of this subgroup showed that patients treated with risedronate had a 43% reduction in new vertebral fractures at three years compared with those receiving placebo.7 A further study of risedronate recruited patients solely on the basis of vertebral fracture history (≥2) irrespective of BMD and demonstrated that active treatment reduced the risk of new vertebral fractures by 49% and of new non-vertebral fractures by 33% over three years compared with placebo.8
Studies of raloxifene9 and parathyroid hormone10 have also included patients with a vertebral fracture history alone. Although the results of these studies showed an overall reduction in fracture risk, subgroup analysis of the patients with two or more vertebral fractures and no BMD measurement was not performed. It is therefore not possible to determine accurately the effect of treatment in this group.
We feel that the evidence suggests that patients presenting with two or more non-traumatic vertebral fractures should be considered for treatment of osteoporosis without the need for measurement of BMD, after a metabolic or secondary cause of fracture has been excluded. This is reflected in some of the recent guidelines for the management of osteoporosis.
We thank Dr Moss and Dr Keen for their interest in our article and for their comments, especially relevant for clinical practice. We agree that whenever the availability of DXA is limited, treatment for osteoporosis in postmenopausal women presenting with non-traumatic vertebral fractures can be started without the measurement of BMD.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.