The main finding in this study is that age was a significant predictor of vertebral deformities in both women and men with a prevalence increasing from approximately 3% in the age group below 60 years to approximately 20% in the age group 70+ in women, and from approximately 7.5% to 20% in men, respectively.
The rates of hip and forearm fractures in Norway are among the highest in the world. Because of this, one would expect the occurrence of vertebral fractures to be high as well. A surprising finding from this study is that this is not the case. For women, it can even be regarded as rather low compared to other studies [25
], reported from Vietnam to be from 17.1% in the age group 50-59 to 39.2% in the age group 70+ (overall prevalence 23%) and in Spain from 7.2% in the age group 55-59 and 46.3% in the age group 75+ (overall 21.4%). In men, our results are more similar to those reported by others [24
], prevalence being 4.7% in the age group 60-69, 10% in the 70-79 group, 14.6% in the 80+ group in Australia, and among Mexican men, 2% in age group 50-59 rising to 21.4% in the 80+ group, with an overall prevalence score of 9.7%. A multinational, European study from 1996 [45
] found the overall prevalence to be 12% both in women and men, which is very much the same as in our study, but the Norwegian rates reported in that study were 19.2% in women, 15.7% in men, along with Sweden the highest rates in Europe. These Norwegian data were, however, extracted from a small sample (289 men, 298 women), mean age 65 years. In addition, another technology was used, making comparison difficult.
As reported by others, we also find the prevalence of vertebral deformities to be highest in the midthoracic region (5th
thoracic) and thoracolumbar transition [42
]. Wedge deformities were mostly found in the higher thoracic and the biconcave in the lower thoracic and lumbar region. It has been reported that fracture related disability may be greater among patient with lumbar fractures [42
]. This could not be verified in the present study, but biconcave deformities were associated with lower BMD at the femoral sites in both sexes compared to the wedge deformities, suggesting a higher degree of severity. However, there is no consensus in the literature concerning type of vertebral fracture and severity [9
]. The finding that prevalent radiographic vertebral fractures, of any type, are associated with low BMD measured at the femoral sites is reported by others [42
]. As no X-rays were available in our study, we were unfortunately unable to assess whether the observed vertebral deformities are related to osteoporosis or other causes.
The Tromsø Study is a population-based, longitudinal study with a high participation rate. The present study is a cross-sectional survey within the framework of the Tromsø Study, where vertebral fracture assessments (VFA) were done for the first time. The intra-class correlation coefficient showed good reproducibility, indicating high methodological precision. Limitations of this study are that only prevalence data on vertebral deformities are presently available, also vertebral deformities were identified by DXA scanning only. Quality control of our data with x-rays on a sub-group was not possible within the scope of the survey. It is, however, reported that DXA scans are more precise in measuring moderate and severe than mild deformities [20
]. Because of the methodological uncertainty concerning detection of mild deformities, the prevalence reported from our study may therefore be under-estimations. To address the issue of selection bias, we compared central characteristics between women and men who were randomly selected to either total body (TB) measurements (960 persons) or to the VFA (2894). In the VFA group, 58% were female compared to 62% in the TB group, with an OR of 1.21 (95% CI 1.04, 1.41), adjusted for age 1.22 (95% CI 1.05, 1.42). In the VFA group, both women and men were younger (65.4 versus 67.5 years in women, 65.3 versus 68.6 years in men), taller (162.2 versus 161.2 cm in women, 175.3 versus 174.4 cm in men), and men in the VFA group were also heavier (84.3 versus 82.3 kg) compared to men in the TB group. BMD levels at the total hip and femoral neck, health status, educational level and physical activity level did not differ between the groups. Despite the random selection, the VFA group was younger with a slightly higher proportion of women. However, when we compare the VFA group with the remaining phase 2 participants of the Tromsø VI survey, whom to our best knowledge should be a representative sample [37
], the VFA sample of women and men was slightly older (3 years) and shorter (2 cm), but did not differ significantly in any other way. To summarize: we believe that the representativity of our sample is fair.
Throughout the study, we have deliberately used the term "vertebral deformity", though regarding these deformities as vertebral fractures [12
]. Interestingly, the prevalence of vertebral fractures in the Tromsø population, which is considered a representative Norwegian population [38
], does not follow the trend reported for non-vertebral fractures [35
]. Difference in fracture mechanisms may possibly explain the discrepancy in prevalence, as non-vertebral fractures are connected to falls [47
], whereas vertebral fractures are not [5
]. One possible interpretation of the findings from this study is that the prevalence of vertebral fractures was low because our population was generally healthy and because of possible underestimations of mild deformities with the technology used. It has been reported that a large amount of vertebral fractures are asymptomatic [11
]. Further studies should elaborate if physical function, pain and self-perceived health, as well as comorbidities, differ between persons with and without vertebral fractures.