SEARCH

SEARCH BY CITATION

Keywords:

  • thoracic kyphosis;
  • postmenopausal osteoporosis;
  • vertebral fracture risk;
  • quality of life;
  • strontium ranelate

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Disclosures
  9. References

We attempt to assess quantitatively thoracic kyphosis and its influence on incident fractures and quality of life over three years in postmenopausal women with osteoporosis and the effect of strontium ranelate on thoracic kyphosis progression. This study was performed on women with postmenopausal osteoporosis from the Spinal Osteoporosis Therapeutic Intervention (SOTI) and Treatment of Peripheral Osteoporosis (TROPOS) studies. Vertebral fractures were assessed on lateral thoracic radiographs performed at baseline and at three years according to standardized procedure. Kyphosis index (KI, %), was defined as the percentage ratio between the maximum depth of thoracic curvature and the height measured from the T4 to the T12 vertebrae. Baseline characteristics of the 3218 patients (1594 strontium ranelate, 1624 placebo) were mean age 73.3 years, spine bone mineral density (BMD) T-score (L2–4) −3.1, femoral neck T-score −3.0, and KI 25.4%. In the placebo group, patients with the highest baseline KI experienced significantly more vertebral fractures than those with medium KIs [relative risk (RR) = 1.53; 95% confidence interval (CI) 1.19–1.96, p < .001) or the lowest KIs (RR = 1.70, 95%CI 1.32–2.21, p < .001), even after adjusting for the presence of prevalent fractures, age, body mass index (BMI), and BMD. There was no difference in the risk of nonvertebral fractures according to baseline KI. Three-year changes in quality-of-life physical scores reflected significantly better status for patients in the lowest tertile of KI compared with those in the highest at baseline. Over three years, the KI increased for all patients, indicating worsening of thoracic kyphosis, whatever the presence of prevalent or incident vertebral fractures. This KI progression was lower in the strontium ranelate group than in the placebo group. Thoracic kyphosis is a risk factor for vertebral fractures over three years and influences physical capacity changes in postmenopausal women with osteoporosis. Thoracic kyphosis progression over three years is lower in a subgroup of strontium ranelate–treated patients compared with placebo-treated patients. © 2010 American Society for Bone and Mineral Research


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Disclosures
  9. References

Exaggerated thoracic kyphosis, that is, a forward curvature of the thoracic spine, is a frequent feature in the elderly and in postmenopausal women with osteoporosis. An increase in thoracic kyphosis has been associated with height loss, upper and middle back pain, decreased physical function,1, 2 increased body sway and risk of falls,3 impairments in pulmonary function,4 esophagial hiatal hernia,5 and an increased risk of mortality in older women.6–8 Limited prospective data exist on the natural history of thoracic kyphosis in the elderly.9

Thoracic kyphosis can be caused by vertebral fractures, frequently at the midthoracic spine,10 and is characterized by a decrease in vertebral height predominantly on the middle and anterior parts of the vertebral bodies.11 However, at least half of all patients having hyperkyphosis with clinical complications do not have vertebral fractures,7, 10, 12, 13 and the exaggerated curvature is also related to degenerative changes of the spine, including intervertebral disk space narrowing, deformities of the anterior part of the vertebrae, and reduced spinal muscle strength.3, 10 Whether or not patients have prevalent vertebral fractures, hyperkyphosis has been shown to be a risk factor for sustaining fractures.14 Therefore, thoracic kyphosis deserves clinical attention. Several treatments have shown an efficacy in decreasing the risk of vertebral fractures in patients with osteoporosis, but their effect on spinal curvature, including kyphosis, has not been assessed.

The aims of this study were to assess prospectively in postmenopausal women with osteoporosis (1) the predictive value of baseline hyperkyphosis on the risk for subsequent fractures, (2) the three-year change of thoracic kyphosis, and (3) the effect of strontium ranelate on this change.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Disclosures
  9. References

Study subjects

Data from Spinal Osteoporosis Therapeutic Intervention (SOTI)15 and Treatment of Peripheral Osteoporosis (TROPOS)16 were used for this study. These randomized, double-blind, placebo-controlled clinical trials have been presented in detail previously.15, 16 In SOTI, 1649 patients were enrolled [50 years old or more and postmenopausal for at least five years, at least one vertebral fracture confirmed by spinal X-rays, and lumbar spine bone mineral density (BMD) of 0.840 g/cm2 or less]. In TROPOS, 5091 patients were enrolled [≥74 years of age or between 70 and 74 years with one additional fracture risk factor, postmenopausal for at least five years, and femoral neck BMD of 0.600 g/cm2 or less (Hologic devices)]. Since study design, centers, BMD, and X-ray central reading centers were common to both studies, the data were pooled. Throughout both studies, subjects received daily calcium supplements and vitamin D according to their needs. They were randomly assigned to receive 2 g/day of strontium ranelate or placebo.

Assessment of outcomes

The assessment of vertebral fractures and BMD was similar in both studies. Three lateral radiographs of the spine (thoracic and lumbar radiographs and an image of the thoracolumbar junction) were obtained with subjects in the left lateral decubitus position at baseline and annually over three years according to standardized procedures of acquisition. All radiographs were assessed at a central facility; central readers were blinded to the treatment assignment. The semiquantitative visual assessment of each vertebra was performed by the same reader throughout the study using a grading scale from 0 to 3.17 A new vertebral fracture was defined as a change in the score of a vertebra from grade 0 at baseline to a subsequent grade of 1 or more. The deformities of nonosteoporotic origin were not graded.

A quantitative assessment was performed on all X-rays placed on a digitizing table using a cursor-like device equiped with fine crosshairs to allow an exact placement of points. Because of the plexiglass viewing area, it was possible to choose points at the intersection between horizontal and vertical lines to best describe the area of the vertebral body. A six-point digitization method was used. The four corner points of the vertebral body from T4 to L4 were marked, as well as an additional point in the middle of the upper and lower endplates.18 In case the outer contours of the endplate were not perfectly superimposed, the middle points were chosen in the center between the upper and lower contours. The method ignores spurs and osteophytes; when present, unaffected levels above and below the affected level were used to guide point placement. The uncinate processes at the posterosuperior corner of the vertebral body were excluded.

All these measurements were performed at the time of reception of X-rays in the central facility, and the coordonates of the points were stored in the database. After the end of the studies, we used these points to calculate a kyphosis index as the ratio BD/AC × 100 (Fig. 1), AC being a line going from the anterosuperior edge of T4 (A) to the anteroinferior edge of T12 (C) and BD being a perpendicular line from the furthest posterosuperior or posteroinferior point of the T7, T8, or T9 vertebra to the line AC. We calculated the length of AC using the coordinates of the points A and C. We then calculated the area of the triangle ABC from the lengths of its three sides. The value of BD was calculated as the height of the triangle ABC. The KI was used to assess thoracic kyphosis at baseline and at the end of follow-up (three years). The higher the KI value, the worse was the thoracic kyphosis. KI was not available when either A,B, or C coordinates were lacking because of poor legibility. The reproducibility of KI was calculated using an intraclass correlation coefficient from the data from baseline X-rays of 22 patients. Intra- and interreader reproducibilities were 0.996 [95% confidence interval (CI) 0.990–0.998] and 0.989 (95% CI 0.956–0.996), respectively.

thumbnail image

Figure 1. Kyphosis index (KI), defined as the BD/AC ratio.

Download figure to PowerPoint

BMD at lumbar spine and proximal femur sites was measured by DXA at baseline and at six-month intervals (Hologic densitometers). All the scans were analyzed centrally, and a quality control of all the devices was conducted throughout the study. For patient diagnostic categorization, baseline lumbar spine and femoral neck BMD T-scores were used (calculated using a reference database from DO Slosman, Geneva, Switzerland).

An assessment of quality of life was performed for all patients with the 36-item Short Form health survey (SF-36) 19 using physical and psychological scores.

Statistical Analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Disclosures
  9. References

Data from the first three years of SOTI and TROPOS have been studied. The flowchart of patients is shown in Fig. 2.

thumbnail image

Figure 2. Flowchart showing the number of patients with missing or non-missing KI at baseline and M36, according to the treament group: PI (placebo) or SR (Strontium Ranelate).

Download figure to PowerPoint

KI and risk of fractures

The predictive value of baseline KI on fracture risk over three years was assessed on a subpopulation of 1624 patients of the placebo group having a baseline KI (see Fig. 2). Three groups were defined using the tertiles of the distribution of KI. Baseline characteristics of patients in these subgroups were described and compared using chi-square tests or analysis of variance according to the nature of the criteria. The incidence of patients experiencing a first new vertebral fracture was estimated based on to the Kaplan-Meier method. A Cox regression model was used to estimate the relative risk (and its 95% CI) of a new fracture in subjects in the highest tertile of KI at baseline as compared with those in the medium tertile and then with those in the lowest tertile. To assess the predictive value of the KI whether vertebral fractures were present or not, the same analysis was performed, adjusting on this latter covariate. The impact of KI on the occurrence of fractures also was assessed after adjustment on potential risk factors [e.g., age, body mass index (BMI), baseline lumbar BMD, and presence of vertebral fractures] in the Cox models. The same analysis also was performed on nonvertebral fractures.

KI and quality of life

The predictive value of KI on quality of life over three years was studied in a subpopulation of 1226 patients (among the 1624 with a KI at baseline) having completed the SF-36 questionnaire at baseline and at month 36 (M36). The changes in quality of life in subjects in the highest tertile of KI were compared with those in the medium and lowest tertiles using generalized linear models adjusted on the baseline quality-of-life scale value.

Three-year changes in KI

The assessment of the relative change in KI over three years was performed in a subpopulation of 2019 patients (of 4055) having a baseline and M36 values for KI (see Fig. 2) and, in a second step, in a group of patients with no prevalent or incident thoracic fractures (n = 1193, of the 2019 patients mentioned earlier). Baseline characteristics (e.g., age, BMI, lumbar spine T-score, prevalence of vertebral fractures, and KI) were similar in patients included for the three-year changes analysis, as in studied patients (data not shown). A Student's t test was used to compare the relative changes in KI over three years in placebo- and strontium ranelate–treated groups. In each treatment group, the baseline KI value was compared with the M36 value using a paired Student's t test. Similar analyses were performed to compare the sum of anterior heights of thoracic vertebrae between groups of KI.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Disclosures
  9. References

KI and risk of fractures

A population of 1624 patients had an assessable KI at baseline (see Fig. 2). The characteristics of these 1624 patients were not different from those of the initial group (Table 1). The population was divided in three subgroups according to the KI tertiles (tertile 1: baseline KI ≤ 22.83; tertile 2: KI > 22.83 and ≤ 27.16; tertile 3: baseline KI > 27.16). Baseline characteristics of the patients according to baseline KI are shown in Table 1.

Table 1. Kyphosis Index and Three-Year Risk of Fracture: Baseline Characteristics of the Patients
 Initial placebo groupKI tertiles, placebo group with KI assessment at baseline
20171 (KI ≤ 22.83)2 (22.83 < KI ≤ 27.16)3 (K > 27.16)
N 543540541
Age (years)73.4 ± 6.172.1 ± 6.073.2 ± 6.274.3 ± 5.7
Body mass index (BMI, kg/m2)25.6 ± 4.025.6 ± 3.725.6 ± 4.025.5 ± 4.0
Lumbar spine T-score−3.1 ± 1.5−2.8 ± 1.6−3.1 ± 1.4−3.4 ± 1.4
Femoral neck T-score−3.0 ± 0.7−2.9 ± 0.7−3.0 ± 0.6−3.1 ± 0.6
Prevalent vertebral fractures (%)    
0 fracture1025 (50.8)294 (54.2)290 (53.7)245 (45.3)
1 fracture462 (22.9)124 (22.8)135 (25.0)123 (22.7)
≥2 fractures530 (26.3)125 (23.0)115 (21.3)173 (32.0)
Prevalent thoracic vertebral fractures (%)    
0 fracture1266 (62.8)387 (71.3)354 (65.5)286 (52.9)
1 fracture412 (20.4)104 (19.1)116 (21.5)118 (21.8)
≥2 fractures339 (16.8)52 (9.6)70 (13.0)137 (25.3)
Baseline KI25.4 ± 5.320.0 ± 2.124.9 ± 1.231.3 ± 3.7
Quality-of-life parameters (SF-36)    
Physical score39.3 ± 10.339.2 ± 10.040.5 ± 10.639.0 ± 10.3
Psychological score47.2 ± 11.445.9 ± 11.247.7 ± 11.147.7 ± 11.7

In the highest tertile, patients were older, and T-scores were lower (p < .0001). The number of prevalent vertebral (whole spine) or thoracic vertebral fractures was significantly different between tertiles. Patients in the lowest KI tertile had fewer prevalent fractures than patients in the highest KI tertile, where the number of patients with two or more prevalent vertebral fractures was the highest (see Table 1). On average, half the patients had no prevalent vertebral fracture at baseline, even in the highest tertile of KI. Moreover, in this subgroup, 52.9% of patients did not have thoracic vertebral fractures.

Patients with the highest KI experienced significantly more new vertebral fractures over the three-year study period (incidence 27.36%) compared with those in the medium tertile [fracture incidence 19.07%, relative risk (RR) = 1.5, 95% CI 1.19–1.96, p < .001] or to those in the lowest tertile (incidence 17.31%, RR = 1.70, 95% CI 1.32–2.21, p < .001). When adjusting on the presence of prevalent fractures, the difference in vertebral fracture occurrence between patients in the highest KI tertile remained statistically significant compared with those in the medium tertile (RR = 1.43, 95% CI 1.11–1.84, p = .005) or in the lowest tertile (RR = 1.58, 95% CI 1.22–2.05, p < 0.001). A further adjustment on additional risk factors (i.e., age, BMI, baseline lumbar BMD, and presence of prevalent vertebral fractures) led to similar results (Table 2). There was no effect of baseline KI values on the risk of nonvertebral fractures (data not shown).

Table 2. Relative Risks of Subsequent Vertebral Fractures in Postmenopausal Women with Osteoporosis, According to Tertiles of Baseline KI
TertilesUnadjustedAdjustedaAdjustedb
  • a

    Prevalent vertebral fractures.

  • b

    Prevalent vertebral fractures, age, BMI, spine BMD.

High-low1.70 (1.32–2.21) p < .0011.58 (1.22–2.05) p < .0011.42 (1.08–1.86) p < .011
High-medium1.50 (1.19–1.96) p < 0.0011.43 (1.11–1.84) p = .0051.30 (1.00–1.68) p = .045

KI and quality of life

There was no relationship between baseline KI and baseline values of SF-36 scores. A significant difference was found for changes over three years in the physical dimensions for SF-36 questionnaires in patients according to their baseline KI value. The three-year change difference between the highest tertile of KI and the two other tertiles was −0.96 ± 0.56 (p = 0.08), and the difference between the highest and lowest tertiles was −1.36 ± 0.56 (p = .015). There was no difference in the psychological score changes according to baseline KI.

Three-year changes in KI

Data were obtained in 2019 patients having KI data at both baseline and M36 evaluations (see Fig. 2). Characteristics of these patients are listed in Table 3; they were not different from those in the initial group.

Table 3. Three-Year Changes in KI: Baseline Characteristics of the Patients
 Placebo (n = 981)Strontium ranelate (n = 1038)Initial population (n = 4055)
Age (years)72.7 ± 6.073.4 ± 5.973.5 ± 6.1
Body mass index (BMI, kg/m2)25.2 ± 3.625.5 ± 4.025.7 ± 4.0
Lumbar spine T-score−3.1 ± 1.4−3.1 ± 1.5−3.1 ± 1.5
Femoral neck T-score−3.0 ± 0.7−3.0 ± 0.6−3.0 ± 0.6
Prevalent vertebral fractures (%)461 (47.0)487 (46.9)1967 (48.5)
Prevalent thoracic vertebral fractures (%)342 (34.9)362 (34.9)1500 (37.0)
Prevalent lumbar fractures (%)262 (26.7)287 (27.7)1150 (28.4)

Over three years, there was a significant increase in KI for postmenopausal patients in the placebo group: +4.70 ± 7.32% (p < .001 versus baseline) (Fig. 3). Strontium ranelate limited the progression of KI (+3.71 ± 7.69%, p < .001 vs. baseline) by comparison with placebo (p = .003) (see Fig. 3).

thumbnail image

Figure 3. Relative changes versus baseline in KI in postmenopausal women with osteoporosis receiving strontium ranelate or placebo over three years (in %),

Download figure to PowerPoint

The same calculations were repeated after exclusion of patients having either prevalent or incident thoracic vertebral fractures, leading to 1193 patients included in this analysis (see Fig. 2). The increase in KI was lower (p < .001) in the strontium ranelate–treated groups by comparison with placebo (+2.72 ± 7.17% and +4.34 ± 6.54%, respectively, both p < .001 vs. baseline values). In these patients without vertebral fractures, we assessed the sum of anterior heights of thoracic vertebrae (T4 to T12) at baseline and at the end of three-year-follow-up using the quantitative assessment database. There were no differences in baseline values and changes of the heights (−3.79 ± 8.44 and −4.01 ± 8.11 mm in the placebo group and strontium group, respectively), indicating that changes in the anterior heights of vertebral bodies do not explain this difference in KI changes.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Disclosures
  9. References

This study shows that thoracic kyphosis is a determinant of incident vertebral fractures and deterioration of physical capacities in postmenopausal osteoporotic women. Moreover, this prospective study suggests that there is a worsening of the kyphosis in this population over three years.

Previous data on thoracic kyphosis have been obtained with clinical measurements such as the distance from the occiput to the table in patients unable to lie flat without neck hyperextension14 or in patients in the upright position using a ruler pressed against the back.20, 21 Our study is the first to use prospective radiologic assessment of KI. It confirms that hyperkyphosis is a risk factor for vertebral fractures in patients with postmenopausal osteoporosis. In the Rancho Bernardo study, conducted in community-dwelling women, older women with hyperkyphosis had a 1.7-fold increased risk of having a future fracture,14 which is in line with the risk calculated with a different kyphosis assessment method for women in the highest tertile in our population. Actually, as the kyphosis increases, the pressure on the anterior parts of the vertebral bodies is increased, thus increasing the probability of a fracture occurrence. However, our result is still significant after adjustment on prevalent vertebral fractures, age, BMI, and BMD, confirming data shown previously.14 This indicates that both osseous and nonosseous parameters must be assessed to interpret cross-sectional and prospective KI results appropriately.

Limited data exist on the natural history of thoracic kyphosis in the elderly.9, 10 In cross-sectional studies, the increase in kyphosis angle can be assessed in every decade.12 Our prospective data show that changes in thoracic spine curvature can occur in the short term, that is, over three years, in osteoporotic elderly women. This worsening is slight, but it has been shown that small increments of kyphosis have strong clinical implications.14

From our data, KI increases over three years in osteoporotic postmenopausal women without prevalent and incident vertebral fractures. This may be related to the wedging of the vertebrae. This deformity may be of osteoporotic origin, responsible for a progressive vertebral deformity. However, our methodology took care to qualify appropriately any vertebral deformity.11 The fracture assessment was conducted using standardized procedures of acquisition in the context of a clinical trial, and the centralized reading of X-rays was performed by a single experienced investigator throughout the study, including comparison with patient's previous X-rays. Before diagnosis of fracture, a nonosteoporotic origin was considered for each deformity. Attention was paid to osteoporotic depressions of the central endplates of the vertebrae because osteoarthritic changes occur only on the anterior part of the vertebrae.22–24 The interpretation of isolated short anterior vertebral heights at the midthoracic spine took into account degenerative changes of adjacent disks and the presence of deformities of similar appearance on contiguous vertebrae, both signs being more in favor of a nonosteoporotic origin of the deformity. Thus deformities of osteoporotic origin cannot explain all the results of KI changes. Degeneration of the intervertebral disks occurs mainly in the anterior fibers of the annulus fibrosus, and a decrease in the anterior parts of the disks is a common feature in the elderly. Interestingly, there was a difference in KI changes in patients receiving placebo or strontium ranelate. This observation also has been made in patients without any prevalent or incident vertebral fractures. Moreover, using morphometric data, we checked that the mean sum of thoracic vertebrae anterior heights at baseline and their changes over the study were similar in both treatment groups. Strontium ranelate, besides its antifracture efficacy, has been associated with demonstrated clinical benefits in back pain and quality of life.25 Strontium ranelate and alendronate as well have been shown to be associated with less lumbar disk space narrowing,26, 27 suggesting that these antiosteoporotic treatments can alter the pathologic process of osteoarthritis. Whether or not this effect could have an implication for thoracic kyphosis warrants further study. Kyphosis changes can be related to reduced thoracic extensor muscular strength, as shown in the elderly and in osteoporotic women.3, 10, 28 Because X-rays were performed in the lying position, we cannot assess the role of decreased muscle strength in the spinal curvature.

Our results show that KI has an impact on changes of physical dimensions and quality of life in osteoporotic postmenopausal women. Controversies have been raised in the literature on the relationship between thoracic kyphosis and quality of life. Ettinger20 stated that kyphosis does not cause subtantial disability in older women, and Martin29 found that kyphosis was associated with increased physical difficulty in daily life activities. In a cross-sectional study conducted in the elderly, women with greater degrees of kyphosis were only slightly more likely to report back-related disability.2 Differences in results may be explained by differences in the populations studied, locations of fractures, and means of kyphosis assessment. Moreover, the link between symptoms and anatomic lesions is not easy to be established in a cross-sectional assessment because symptoms are mainly related to structural changes (i.e., incident vertebral fracture or disk narrowing). Indeed, we did not find any difference in quality-of-life parameters at baseline among tertiles of KI, but in contrast, we found a difference between tertiles on changes in these parameters assessed prospectively. No effect of KI was found on psychological dimensions of the quality-of-life scores. This may be a result of disease acceptance by patients as an inescapable effect of aging or to the context of this assessment because participating in a clinical study may decrease anxiety.

Our study has several limitations. Data were obtained in osteoporotic women and thus cannot be applied directly to other populations, including men. Thoracic kyphosis assessment was indirect, using coordinates of points from a database and not direct measurement on X-rays. Moreover, it is possible that KI changes were underestimated because patients were not in a standing position for spinal X-ray acquisition, and thus our results can be considered conservative. Results on strontium ranelate effect were obtained in a post hoc analysis, and other determinants of hyperkyphosis, such as muscle strength, were not available.

Our data show that thoracic kyphosis is a risk factor for vertebral fractures, even after adjustment on the main determinants of these fractures. Progression of thoracic kyphosis can be observed over three years and may be decreased by strontium ranelate, according to a post hoc analysis. Thoracic kyphosis deserves clinical attention in an optimal management of postmenopausal women with osteoporosis.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Statistical Analysis
  6. Results
  7. Discussion
  8. Disclosures
  9. References
  • 1
    Kado DM, Huang MH, Barrett-Connor E, Greendale GA. Kyphotic posture and poor physical functional ability in older community dwelling men and women: the Rancho Bernardo Study. J Gerontol Med Sci. 2005; 60: 633637.
  • 2
    Ensrud KE, Black DM, Harris F, Ettinger B, Cummings SR. Correlates of kyphosis in older women: the Fracture Intervention Trial Research Group. J Am Geriatr Soc. 1997; 45: 682687.
  • 3
    Sinaki M, Brey RH, Hughes CA, Larson DR, Kaufman KR. Balance disorder and increased risk of falls in osteoporosis and kyphosis: significance of kyphotic posture and muscle strength. Osteoporos Int. 2005; 16: 10041010.
  • 4
    Harrison RA, Kiminoski K, Vethanayagam D, Majumdar SR. Osteoporosis-related kyphosis and impairments in pulmonary function: a systematic review. J Bone Miner Res. 2007; 22: 447457.
  • 5
    Yamaguchi T, Sugimoto T, Yamada H, et al. The presence and severity of vertebral fractures is associated with the presence of esophageal hiatal hernia in postmenopausal women. Osteoporos Int. 2002; 13: 331336.
  • 6
    Kado DM, Huang MH, Karlamangla A, Barrett-Connor E, Greendale GA. Hyperkyphotic posture predicts mortality in older community dwelling men and women: a prospective study. J Am Geriatr Soc. 2004; 62: 16621667.
  • 7
    Kado DM, Greendale GA, Lui L, et al. Hyperkyphosis predicts mortality independent of vertebral osteoporosis in older women. J Bone Miner Res. 2003; 18: S249.
  • 8
    Kado DM, Browner WS, Palermo L, Nevitt MC, Genant HK, Cummings SR. Vertebral fractures and mortality in older women: a prospective study. Study of Osteoporotic Fractures Research Group. Arch Intern Med. 1999; 159: 12151220.
  • 9
    Milne JS, Williamson J. A longitudinal study of kyphosis in older people. Age Ageing. 1983; 12: 225233.
  • 10
    Bartynski WS, Heller MT, Grahovac SZ, Rothfus WE, Kurs-Lasky M. Severe thoracic kyphosis in the older patient in the absence of vertebral fracture: association of extreme curve with age. Am J Neuroradiol. 2005; 26: 20772085.
  • 11
    Roux C, Fechtenbaum J, Kolta S, Briot K, Girard M. Mild prevalent and incident vertebral fractures are risk factors for new fractures. Osteoporos Int. 2007; 18: 16171624.
  • 12
    Schneider DL, von Mühlen DG, Barrett-Connor E, Sartoris D. Kyphosis does not equal vertebral fractures: the Rancho Bernard Study. J Rheumatol. 2004; 31: 747752.
  • 13
    Manns RA, Haddaway MJ, McCall IW, Cassar PV, Davie MW. The relative contribution of disc and vertebral morphometry to the angle of kyphosis in asymptomatic subjects. Clin Radiol. 1996; 51: 258262.
  • 14
    Huang MH, Barrett-Connor E, Greendale GA, Kado DM. Hyperkyphotic posture and risk of future osteoporotic fractures: the Rancho Bernardo Study. J Bone Miner Res. 2006; 21: 419423.
  • 15
    Meunier PJ, Roux C, Seeman E, et al. The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis. N Engl J Med. 2004; 350: 459468.
  • 16
    Reginster JY, Seeman E, de Vernejoul MC, et al. Strontium ranelate reduces the risk of nonvertebral fractures in postmenopausal women with osteoporosis: Treatment of Peripheral Osteoporosis (TROPOS) study. J Clin Endocrinol Metab. 2005; 90: 28162822.
  • 17
    Genant HK, Wu CY, Van Kuijk C, Nevitt MC. Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res. 1991; 6: 2534.
  • 18
    Smith-Bindman R, Cummings SR, Steiger P, Genant HK. A comparison of morphometric definitions of vertebral fracture. J Bone Miner Res. 1991; 6: 2534.
  • 19
    Ware JE, Snow KK, Kosinski MK, et al., SF-36 Health Survey, Manual and Interpretation Guide. Boston: The Health Institute, New England Medical Center.
  • 20
    Ettinger B, Black DM, Palermo L, Nevitt MC, Melnikoff S, Cummings SR. Kyphosis in older women and its relation to back pain, disability and osteopenia: the Study of Osteoporotic Fractures. Osteoporos Int. 1994; 4: 5560.
  • 21
    Prince RL, Devine A, Dick IM. The clinical utility of measured kyphosis as a predictor of the presence of vertebral deformities. Osteoporos Int. 2007; 18: 621627.
  • 22
    Ferrar L, Jiang G, Adams J, Eastell R. Identification of vertebral fractures: an update. Osteoporos Int. 2005; 16: 717728.
  • 23
    Jiang G, Barrington NA, Eastell R, Ferrar L. Visual identification of prevalent vertebral fracture: comparison between methods. Osteoporos Int. 2004; 15: 887896.
  • 24
    Abdul-Hamid Osman A, Bassionni H, Koutri R, Nus J, Geusens P, Dequeker J. Aging of the thoracic spine: distinction between wedging in osteoarthritis and fracture in osteoporosis—a cross-sectional and longitudinal study. Bone. 1994; 15: 437442.
  • 25
    Marquis P, Roux C, de la Loge C, et al. Strontium ranelate prevents quality of life impairment in post-menopausal women with established vertebral osteoporosis. Osteoporos Int. 2008; 19: 503510.
  • 26
    Bruyere O, Delferriere D, Roux C, et al. Effects of strontium ranelate on spinal osteoarthritis progression. Ann Rheum Dis. 2008; 67: 335339.
  • 27
    Neogi T, Nevitt MC, Ensrud KE, Bauer D, Felson DT. The effect of alendronate on progression of spinal osteophytes and disk space narrowing. Ann Rheum Dis. 2008; 67: 14271430.
  • 28
    Mika A, Unnithan VB, Mika P. Differences in thoracic kyphosis and in back muscle strength in women with bone loss due to osteoporosis. Spine. 2005; 30: 241246.
  • 29
    Martin AR, Sornay-Rendu E, Chandler JM, Duboeuf F, Girman CJ, Delmas PD. The impact of osteoporosis on quality of life: the OFELY cohort. Bone. 2002; 31: 3236.