Vertebral Fractures Identified by Lateral Spine Dual Energy X-Ray Absorptiometry Scans and Mortality Risk in Adults

Article information

J Bone Metab. 2025;32(2):133-142

Publication date (electronic) : 2025 May 31

doi : https://doi.org/10.11005/jbm.24.831

Carlos H. Orces

Department of Medicine, Laredo Medical Center, Laredo, TX, USA

Corresponding author: Carlos H. Orces, Department of Medicine, Laredo Medical Center, 1700 East Saunders Street, Laredo, TX 78041, USA, Tel: +1-956-712-4077, Fax: +1-956-712-3555, E-mail: corces07@yahoo.com

Received 2024 December 27; Revised 2025 March 29; Accepted 2025 April 8.

Abstract

GRAPHICAL ABSTRACT

INTRODUCTION

Vertebral fractures, the most common type of osteoporotic fractures, are associated with considerable pain, functional limitations, and impaired quality of life. [1,2] Fractures detected by spinal radiographs also predict future vertebral and non-vertebral fractures, independent of bone mineral density (BMD).[3,4] In general, vertebral fractures are similarly distributed by gender, increase with age, and differ according to geographic regions.[2,5] Notably, vertebral fractures may be asymptomatic and frequently underreported on radiographs.[6,7] Several studies have previously demonstrated that vertebral fractures identified by radiographic morphometry are associated with mortality risk, particularly in older women with osteoporosis.[8–11]

Vertebral fracture assessment (VFA) with dual energy X-ray absorptiometry (DXA) has been proposed as an alternative approach for the identification of vertebral fractures owing to low-dose radiation exposure, high positive predictive value, significant agreement as compared with spinal radiographs, and simultaneous measurements of BMD. Moreover, the detection of vertebral fractures by VFA has important clinical implications regarding BMD monitoring and treatment of osteoporosis.[2,12,13]

Despite these facts, there is scarce data with contradictory results regarding the association between vertebral fractures identified by VFA and mortality risk.[14,15] Schousboe et al. [14] in a large cohort of older adults meeting criteria for VFA described a weak association between vertebral fractures and mortality risk. In contrast, participants in the Tromsø study with severe vertebral fractures or ≥3 vertebral fractures detected by VFA, and after a mean follow-up of 11.2 years, had 1.6- and 2.4-fold greater all-cause mortality risk than their counterparts without fractures, respectively.[ 15] Thus, the present study aimed to examine the association between vertebral fractures identified by VFA and all-cause mortality risk, in a nationally representative sample of adults.

METHODS

RESULTS

The study sample comprised 3,219 participants with a mean age of 57.2 years (standard error, 0.2). Overall, the prevalence of vertebral fractures identified by VFA was 5.5% (95% CI, 4.7–6.5) among adults aged 40 years and older. In addition, the prevalence of moderate to severe vertebral fractures and ≥2 fractures occurred in 3.2% (95% CI, 2.4–4.3) and 1.8% (95% CI, 1.2–2.6) of participants, respectively.

As shown in Table 1, the prevalence of vertebral fractures was significantly increased among older adults, non-Hispanic whites, former smokers, subjects with hypertension, and particularly those with osteoporosis. During a median follow-up of 71.0 months, 277 participants died. Of those, 18.2% (95% CI, 12.2–26.3) with vertebral fractures and 6.3% (95% CI, 5.0–8.0) without fractures (P<0.0001). In general, malignant neoplasms and diseases of the heart were reported as the leading underlying causes of death in 27.6% (95% CI, 21.4–34.7) and 23.3% (95% CI, 17.3–30.5) of the decedents, respectively.

Table 1

Prevalence of vertebral fractures according to characteristics of participants, National Health and Nutrition Examination Survey 2013–2014

Kaplan-Meier curves showed a decrease in survival probability among participants with prevalent vertebral fractures (Fig. 1). Also, shorter survival occurred in subjects with ≥2 vertebral fractures and particularly among those with moderate to severe vertebral fractures, which appear to diverge about 40 months into the study period (Fig. 2, 3).

Fig. 1

Prevalence of vertebral fractures and all-cause mortality.

Fig. 2

Number of vertebral fractures and all-cause mortality.

Fig. 3

Vertebral fracture severity and mortality risk.

As shown in Table 2, Cox regression analysis demonstrated that participants with baseline prevalent vertebral fractures and those with ≥2 vertebral fractures had 40% and 49% greater all-cause mortality risk compared with their counterparts without fractures, respectively. However, these associations did not reach statistical significance. In contrast, after adjusting for potential confounders including lumbar spine and femoral neck BMD, participants with moderate to severe vertebral fractures (≥25% height loss) were 1.7 times more likely to die than those without fractures (HR, 1.79; 95% CI, 1.05–3.04).

Table 2

The association between vertebral fractures identified by vertebral fracture assessment and mortality risk in adults

In a subgroup analysis, older adults with any vertebral fracture had 2-fold greater all-cause mortality risk than those without (HR, 2.16; 95% CI, 1.41–3.30). Similarly, men with prevalent vertebral fractures were 2.2 times more likely to die than their counterparts without (HR, 2.28; 95% CI, 1.49–3.47). In contrast, a baseline vertebral fracture identified by VFA did not significantly increase the risk of mortality among younger participants, women, or according to BMD status (Table 3).

Table 3

Baseline prevalence of vertebral fractures and mortality risk according to age, gender, and bone mineral density

DISCUSSION

In a nationally representative sample, adults with baseline moderate to severe vertebral fractures identified by VFA, and during a median follow-up of 71 months, were 1.7 times more likely to die than their counterparts without vertebral fractures. Likewise, participants with any vertebral fracture had 1.5-fold higher risk of dying. However, after adjusting for behavioral characteristics, comorbidities, and BMD, this association was attenuated and did not reach statistical significance. Notably, in a subgroup analysis, older adults and men with any prevalent vertebral fracture had 2-fold greater mortality risk than those who did not.

In agreement with the present results, participants in the Tromsø study with severe vertebral fractures identified by VFA, and after a mean follow-up time of 12.2 years, had 1.6-fold higher mortality risk than those without fractures. However, in that study, the prevalence of any vertebral fracture did not significantly increase the risk of mortality. [15] A large cohort study among older adults with a T-score of ≤ −1.5 and meeting criteria for VFA also reported a weak association between prevalent vertebral fractures and all-cause mortality risk.[14] Prior studies using radiographic morphometry have demonstrated a non-significant association between prevalent vertebral fractures and mortality risk.[10,11,22] Possible explanations for differences in study results might be related to the characteristics of participants, methods used to define vertebral fractures, and follow-up time.

Although a small number of participants had ≥2 vertebral fractures identified by VFA, the mortality rate was significantly higher than those without fractures. However, after adjusting for possible confounders, the HR did not reach statistical significance. The present findings differed from those in the Tromsø study in which participants with ≥3 vertebral fractures had 2.4-fold greater mortality risk than their counterparts without fractures, even after adjusting for total hip BMD.[15] Likewise, in the Study of Osteoporotic Fractures, women with multiple vertebral fractures or severe fractures had the greatest increase in mortality risk over time, regardless of whether they were aware of their fractures.[8,9] Moreover, in the Rancho Bernardo study, women with ≥2 vertebral fractures had a 56% increased risk of all-cause mortality.[22]

Of relevance, in a subgroup analysis, men with any baseline prevalent vertebral fracture had 2-fold higher risk of all-cause mortality than those who did not. Similarly, Son et al. [23] using data from a cohort study in South Korea reported that older men with prevalent vertebral fractures, and after a mean follow-up period of 5.8 years, were 1.8 times more likely to die as compared with women. Previously, Edidin et al. [24] described among Medicare beneficiaries that mortality risk was higher in men than in women for operated and non-operated vertebral compression fractures, even after controlling for comorbidities, which is consistent with the present study results.

The association between vertebral fractures and mortality risk has not been completely elucidated. Nevertheless, excess mortality does not appear to be exclusively explained by poor health status.[8] In fact, in a large cohort of patients hospitalized for vertebral fractures only 28% of the deaths were attributed to the vertebral fracture itself.[25] Schousboe [26] postulated that increased mortality risk may be linked to loss of functional capacity and physical performance following vertebral fractures, which may subsequently reduce the physiological reserve to withstand the stress of comorbidities. Kado et al. [27] also reported that older women with incident vertebral fractures may have an increased risk of mortality that may be explained by weight loss and physical frailty. Notably, in the Dubbo Osteoporosis Epidemiology Study, among participants with vertebral fractures, those with quadriceps muscle weakness had a greater mortality risk than their counterparts who did not. [28] Similarly, in the MrOs cohort study, older men with fractures who were unable to complete physical performance measurements had a greater risk of post-fracture mortality than those who did not.[29] Although prevalent vertebral fractures are associated with greater thoracic kyphosis, women with prevalent vertebral fractures and increased kyphosis are at greater risk of mortality than those without fractures.[30,31] More recently, severe kyphosis in older men was also reported to increase mortality risk, independent of prior vertebral fractures.[32] This association has been explained by impaired pulmonary function, poor physical function, incident fractures and falls, which may subsequently increase the risk for mortality.[31]

Although the cause-specific mortality associated with vertebral fractures has not been consistently described across studies, older women with vertebral fractures were 2 to 3 times more likely to die of pulmonary causes.[9] Likewise, a long-term population-based study in Finland demonstrated that men and women with vertebral fractures of the thoracic spine had 3- and 8-fold higher risk of dying from respiratory diseases and injuries than their counterparts without fractures, respectively.[10] Choi et al. [33] using data from the Korean National Health Insurance Service similarly described higher mortality rates for diseases of the musculoskeletal, digestive, and respiratory systems among subjects with vertebral fractures.

Clinically relevant, pharmacological treatments and minimally invasive surgery for vertebral fractures have been shown to potentially decrease the risk of mortality. Iida et al. [34] reported that among older Japanese hospitalized for vertebral fractures, the use bisphosphonates improved survival by 41%. Likewise, a secondary data analysis of the Taiwanese Health Insurance also demonstrated that older adults with vertebral fractures who received treatment for osteoporosis had 26% greater survival than those who did not. Furthermore, survival rate improved by 66% among those treated ≥3 years.[35] Despite this evidence, the use of osteoporosis medication in U.S. postmenopausal women has decreased over time and it is suboptimal even in women with osteoporosis.[36] Among Medicare beneficiaries diagnosed with vertebral fractures identified from inpatient and outpatient claims, and after a follow-up period of 4 years, patients who underwent vertebroplasty or kyphoplasty were 37% less likely to die than non-operated vertebral fracture patients, which was attributed to improvement in their pulmonary function.[24]

The present study has several limitations that should be mentioned. First, as an observational study, the causality of the association between vertebral fractures and all-cause mortality risk may not be established. Second, the mortality risk was limited to baseline vertebral fractures identified by VFA. Thus, comorbidities or events leading to mortality during the follow-up period could not be determined. Third, the present study has limited power to examine the association between vertebral fractures and cause-specific mortality. Despite these limitations, the present findings may be generalized to U.S. middle-aged and older adults.

In conclusion, vertebral fracture severity identified by VFA was significantly associated with greater all-cause mortality risk. This association was also demonstrated among older adults, including men, with any baseline vertebral fractures. These findings underscore the importance of broader implementation of VFA during routine DXA scans and highlight the need for further research to elucidate the sequence of events leading to increased mortality risk following vertebral fractures.

Article information Continued