Antonelli-Incalzi et al. (2007) |
Italy |
Cross-sectional |
Random sample. Subjects of both sexes from two different towns were randomly invited to participate in the InCHIANTI study. Subjects were divided into three groups according to the OWD distribution (1st quartile: short OWD; 2nd and 3rd quartiles: medium OWD; 4th quartile: long OWD). |
783 participants, 55% women, (75 ± 6,85 years) and 45% men (73.8 ± 6.34 years). |
Flexed posture: measured using the occiput-to-wall distance (OWD) Tool: rigid tape measure - subject standing with the head in the neutral position. |
Balance: feet-together stance, semitandem standing and tandem standing for 10 seconds. |
Balance: There were significant differences between groups. Subjects with short OWD performed better in the balance test, p<0.009 for men and p<0.001 for women. After adjustment for age and other variables, there was a correlation between balance and OWD, but only in women (p=0.002). |
Kado et al. (2007) |
USA |
Cohort. Rancho Bernardo Study |
Convenience sample. Subjects of both sexes who participated in the Rancho Bernardo Study (1988 -1992). |
1883 participants, women (72.7 ± 9.0 years) and men (73.6 ± 8.9 years) |
Thoracic hyperkyphosis: blocks method, 1.7 cm blocks. Occiput-to-table distance while in the supine position. Hyperkyphosis was defined as requiring the use of ≥ 1 blocks. |
Self-reported falls in the previous 12 months. |
Thoracic hyperkyphosis: n= 430 ± 30.2 (nonfallers) vs n=165 ± 36.2 (fallers) p=0.015; Participants were divided according to gender: Men (n= 342) OR 1.86 (1.22-2.85) 95% CI, p=0.004; Women (n=252) OR 1.50 (1.08-2.07) 95% CI, p=0.01; When adjusted for age: men OR 1.50 (0.96-2.36) 95% CI, p=0.08; women OR 1.30 (0.92-1.83) 95% CI, p=0.24. |
Kasukawa et al.(2010) |
Japan |
Cross-sectional |
Convenience sample. Subjects of both sexes, divided into three groups: 1- subjects without a history of falls or fear of falls (nonfallers); 2 - subjects with a history of fear of falls or requiring any support when walking; and 3 - subjects with a history of falls (fallers). |
92 participants: Group 1, n=40 (72.9 ± 8.1 years), Group 2, n=36 (74.2 ± 9.4 years), and Group 3, n=16 (77.3 ± 6.5 years) |
Spinal curvature: Angles of thoracic (T1-T-12) and lumbar kyphosis (L1-S1): Spinal inclination (T1-S1) Tool: SpinalMouse |
History of falls in the previous 12 months: questionnaire. Balance: Stabilometry. Subject standing unaided in the upright position for 20 s with eyes open. Horizontal (X direction) and vertical (Y direction) movements of center of gravity (COG) during measurement were calculated as deviation of COG to origin. Tool: Force plate JK-101. |
Thoracic kyphosis: 32.6 ± 15.6o (fallers) vs 3.2 ± 12.6 o (nonfallers) NS Lumbar kyphosis: 8.3 ± 16.8 o (fallers) vs -12.5 ± 13.7 o (nonfallers) p<0.0001 Spinal inclination: 17.1 ± 13.8 o (fallers) vs 5.7 ± 6.6 o (nonfallers) p=0.0007 Lumbar mobility: 18.8 ± 15.7° (fallers) vs 33.6 ± 15.2° (nonfallers) p=0.004 Stabilometry: 435.1 ± 199.9 mm (fallers) vs 278.1 ± 104.7 mm (nonfallers), p=0.0021 |
Regolin & Carvalho (2010) |
Brazil |
Cross-sectional |
Convenience sample. Women divided into three groups: 1 - loss of bone mass and increase in thoracic kyphosis; 2 - loss of bone mass without increase in thoracic kyphosis; 3 - without loss of bone mass and without increase in thoracic kyphosis; 4 - without loss of bone mass and with increase in thoracic kyphosis |
95 (67.20 ± 5.01 years). |
Thoracic kyphosis angle Tool: flexicurve method |
Balance: Stabilometry. Subject standing for 10 seconds, first with eyes open, then with eyes closed. One-minute rest between measurements. Tool: force plate, F-Scan system version 4.2 (Tekscan, Inc., South Boston, MA, USA) with 100 Hz sampling frequency. |
Stabilometry, anteroposterior (AP) direction: Eyes open: 1.68 ± 0.56 cm (Group 1) vs 1.32 ± 0.58 cm (Group 3), p=0.0124 Eyes closed: 1.77 ± 1.17 cm (Group 1) vs 1.27 ± 0.44 cm (Group 3), p=0.0263 |
Eum et al. (2013) |
USA |
MOBILIZE Boston Study cohort. |
Convenience sample. Subjects of both sexes. |
620 (79.2 ± 5.4 years) |
Kyphosis index (KI) Tool: flexicurve ruler |
Balance: Berg Balance Scale (BBS) Falls: FES (Falls Efficacy Scale), a cutoff of 90 was used to separate likely fallers from nonfallers. |
There was no significant association between KI and BBS (p=0.23) or FES p=0.527 |
Ishikawa et al. (2013) |
Japan |
Cross-sectional |
Convenience sample. Subjects of both sexes, divided into: Fallers (n=29) and nonfallers (n=184) |
213 (70.1 ± 7.9 years) |
Thoracic kyphosis (T1-T12) and lumbar lordosis angles (T12-S1), and spinal inclination (T1-S1) Tool: SpinalMouse |
History of falls in the previous 12 months: questionnaire. Balance: Stabilometry. Tool: Force plate (JK-101) |
Thoracic kyphosis: 33.3 ± 14.7° (fallers) vs 34.4 ± 14.4° (nonfallers) p=0.396; Lumbar lordosis: 3.8 ± 20.5° (fallers) vs 11.9 ± 14.3° (nonfallers) p=0.035; Spinal inclination: 10.9 ± 12.2° (fallers) vs 5.8 ± 7.9° (nonfallers), p=0.017; Stabilometry: AP direction 336.9 ± 243.9 mm (fallers) vs 214.5 ± 127.0 mm (nonfallers), p=0.046; ML direction 198.8 ± 116.2 mm (fallers) vs 150.1 ± 70,5 mm (nonfallers), p=0.017. |
Abe et al. (2013) |
Japan |
Cross-sectional |
Convenience sample. Community-dwelling women from Nagasaki, Japan, who were invited to participate in periodic health examinations in 2006. |
107 (66.2 ± 10.7 years) |
Spinal inclination (T1-S1) Tool: SpinalMouse |
Balance: Timed Up and Go (TUG) test and Functional reach (FR). |
Spinal inclination = 4.6 ± 5.2°; TUG = 8.2 ± 1.6 s; FR= 27.5 ± 6.5 cm. There was a positive correlation between spinal inclination and TUG (r=0.37 p<0.001) and a negative correlation between spinal inclination and FR (r=- 0.37 p<0.001) |
Miyazaki et al (2013) |
Japan |
Cross-sectional |
Convenience sample. Community-dwelling males. |
124 (73 ± 7.2 years) |
Thoracic kyphosis angle (TKA) (T1-T12) and lumbar lordosis angle (LLA) (L1-L5). Tool: SpinalMouse |
Balance: TUG and one-leg stance with eyes open for both lower limbs. |
TKA (35.8 ± 9.7°) showed a negative correlation with LLA (-13.1 ± 9.3°, r=- 0.36, p<0.01); LLA was correlated with TUG (r=0.36, p<0.01) and one-leg stance (r=-0.31, p<0.05); TKA was not correlated with TUG nor with one-leg stance (r=0.17 and r=-0.07, respectively, p>0.05) |
Imagama et al. (2013) |
Japan |
prospective-longitudinal cohort |
Convenience sample. Males who underwent a basic health checkup organized by the local government in 2007. |
100 (70.2 ± 7.1 years) |
Thoracic kyphosis angle (T1-T12) and lumbar lordosis angle (T12-L5). Tool: SpinalMouse. |
History of falls: recorded in a journal Balance: Stabilometry. Tool: Force plate (G-620). |
Correlation between thoracic kyphosis and balance: r=-0.08 (eyes open) vs r=0.054 (eyes closed), p>0.05 Correlation between lumbar lordosis and balance: r=-0.465 (eyes open) vs r=-0.398 (eyes closed), p<0.0001. Correlation between lumbar lordosis and falls: 8.0 ± 10.6 (multiple falls) vs 17.7 ± 10.3 (one or no falls), p=0.0280. |
Groot et al (2014) |
Netherlands |
Cross-sectional |
Convenience sample. Subjects of both sexes, divided into two groups: flexed posture (FP, n=25) and normal posture (NP, n=31) |
56 (80 ± 5.2 years) |
Thoracic kyphosis: Cobb angle (hyperkyphosis > 50 degrees), assessment of lateral X-rays. Flexed posture: defined as an occiput-to-wall distance (OWD) of 0.5 cm or more. |
Risk of falls: Pluijm score |
Thoracic hyperkyphosis: 44.5 ± 12.1° (NP group) vs 58.6 ± 11.9° (FP group), p<0.01; Risk of falls: 4 (0-19 scores) for the NP group vs 4 (0-20 scores) for the FP group, p= 0.73 |
Drzał-Grabiec et al (2014) |
Poland |
- |
Female |
90 (70 ± 8.01 years) |
Thoracic kyphosis and lumbar lordosis: Photogrammetry based on the Moiré method |
Balance: Stabilometry. Tool: CQstab platform. |
All of the assessed balance parameters showed a significant (p<0.05), low to moderate correlation with spinal curvature r=0.20-0.40. Lumbar lordosis and thoracic kyphosis showed stronger positive correlations with balance, with r=0.40 and r=0.44, respectively, (p<0.001) |
Ota et al. (2015) |
Japan |
Cross-sectional |
Convenience sample. Women recruited from day care service users in Toyohashi. |
53 (83.7 ± 6.3 years) |
Body posture: spherical, colored-reflective markers attached to specific anatomic landmarks of participants’ bodies in the standing position. Tool: Digital video camera (GR-D850), right lateral view. |
TUG; one-leg stance with eyes open test (OLST) |
There was a positive correlation between TUG and forward head position (r=0.30); and between TUG and lower lumbar spine angles (r=0.34), p<0.05. OLST was not correlated with any of the variables assessed (p>0.05). |
Van der Jagt-Willems et al. (2015) |
Netherlands |
Prospective cohort |
Convenience sample. Subjects of both sexes who visited an outpatient clinic in Amsterdam between October 2010 and April 2012. |
51 (76 ± 4.8 years) |
Thoracic hyperkyphosis: Cobb angle: hyperkyphosis was defined as Cobb angle ≥50°; Flexed posture: occiput-to-wall distance (OWD) > 5cm. |
Falls: registered by monthly phone contact for the duration of 12 months. |
Thoracic hyperkyphosis: 59 ± 16° (fallers) vs 49 ± 13° (nonfallers) p=0.04; Thoracic hyperkyphosis and future falls: OR 2.13 (1.10-4.51) 95% CI p=0.04; Flexed posture (OWD): 6.2 ± 4.1 cm (fallers) vs 4.2 ± 4.5 cm (nonfallers), p=0.18 |
Suzuki et al. (2015) |
Japan |
Cohort. |
Convenience sample. Subjects of both sexes who participated in the Itabashi Cohort Study 2011. |
834 participants: 484 women (72.7 ± 4.9 years) and 350 men (73.7 ± 5.5 years) |
Inclination angle of the trunk: the digital inclinometer measured the inclination of the sternum. Tool: Digital inclinometer. |
Static balance (one-leg stance with eyes open, dominant side); and dynamic balance (TUG) |
In both sexes, the sternum inclination angle was not associated with static balance (p>0.05); the sternum inclination angle was associated with TUG in men, r=0.26 (p<0.01), but not in women, r=0.02 (p=0.61). |