Abstract

Background.

Accumulated evidence shows how important spinal posture is for aged populations in maintaining independence in everyday life. However, the cross-sectional designs of most previous studies prevent elucidation of the relationship between spinal posture and future dependence in activities of daily living (ADL). We tried to clarify the association by measuring spinal posture noninvasively in a community-based prospective cohort study of older adults, paying particular attention to thoracic curvature, lumbar curvature, sacral hip angle, and inclination to determine which parameter is most strongly associated with dependence in ADL.

Methods.

Spinal posture was evaluated in 804 participants (338 men, 466 women, age range: 65–94 years) who were independent in ADL at baseline. We defined dependence in ADL as admission to a nursing home or need of home assistance. During the 4.5-year follow-up period, 126 (15.7%) participants became dependent in ADL. The relationship between the spinal posture parameters and outcome was assessed by dividing the participants into sex-specific quartiles of the parameters.

Results.

Only inclination (angle subtended between the vertical and a line joining C7 to the sacrum) was associated with outcome, although lumbar curvature also showed a marginal association. The age- and sex-adjusted odds ratio for a 1 unit increase in the quartiles of inclination was 1.79 (confidence interval: 1.44, 2.23). After mutual adjustment for the 4 parameters, statistical significance for inclination still remained, with no substantial changes in the association estimates.

Conclusions.

This study indicates that spinal inclination is associated with future dependence in ADL among older adults.

POPULATIONS are aging rapidly worldwide. This trend is particularly evident in Japan, whose population has the world’s longest life expectancy (79.4 years for men and 85.9 years for women [1]). The already rapid pace at which society is aging is expected to accelerate further, meaning that fewer young people will be available to take care of the elderly people, and thus making it even more important for the elderly people to be able to live independent and active lives.

Spinal posture changes with age, but accumulated evidence shows that continued good spinal posture is important in allowing the aged to maintain independent lives (2,3). Hirose and colleagues (4) reported that the posture of the trunk in the sagittal plane is associated not only with the distance and time parameters of gait, but also with functional performance in the elderly population. In a study by Takahashi and colleagues (5), participant groups with trunk deformities tended to score lower than the control group on subjective healthiness and life satisfaction measures. However, the cross-sectional designs of most studies to date prevent conclusions being drawn about the relationship between spinal posture and future dependence in activities of daily living (ADL). The participants of these studies were patients with spinal deformities and diagnoses of osteoporosis, and evidence is lacking from community-based studies.

Determination of spinal posture requires the examination of multiple elements, including the cervical vertebrae, thoracic vertebrae, lumbar vertebrae, and pelvis. Because such examinations have generally been done with x-ray equipment, they have not been carried out at local health facilities due to the lack of specialized equipment. In recent years, however, a computer-assisted and easily operated, noninvasive, portable device to measure spinal shape has been developed (6,7). With this device, sagittal spinal curve divided into thoracic curvature, lumbar curvature, sacral hip angle, and inclination can be examined. We used the device in this study to examine spinal posture noninvasively in older adults and tried to clarify the association between spinal posture and future dependence in ADL through a community-based prospective cohort study design. We paid particular attention to thoracic curvature, lumbar curvature, sacral hip angle, and inclination to determine which of these four parameters is most strongly associated with dependence in ADL.

Methods

Study Population

The Kurabuchi Study, a community-based prospective cohort study of aging involving functional assessment of an older population, was launched in 2005 (8,9). Briefly, the study population included all residents aged 65 years or older of Kurabuchi Town, Gunma Prefecture (approximately 100 km north of Tokyo, Japan). Excluding those who were hospitalized or institutionalized, a total of 1,294 residents, were eligible for inclusion in the study. Of these, 834 participated in the baseline examination (participant proportion = 64.5%) and gave written informed consent. For the purposes of this study, we excluded those who were dependent in ADL at the baseline (n = 29) and those who did not undergo spinal curvature measurements (n = 1). Thus, a total of 804 participants (338 men and 466 women) were subject to the study. The study protocol was approved by the Ethics Committee of the School of Medicine, Keio University (Tokyo, Japan) and by that of Toho University (Tokyo, Japan).

Assessment of Spinal Posture

The participants were asked to stand in a relaxed position wearing one layer of clothing, and spinal posture was evaluated with a Spinal Mouse (Indiag, Volkerswill, Switzerland), a computer-assisted, noninvasive device for measuring spinal shape. The device is guided along the midline of the spine, starting at the spinous process at C7 and finishing at the top of the anal crease (approximately S3). Measurements were repeated 3 times, and the best two values were averaged. The relevant parameters recorded with the Spinal Mouse were thoracic curvature (Th1-2 to Th11-12), lumbar curvature (Th12–L1 to the sacrum), sacral hip angle (angle between a straight line from S1 to S3 and true vertical), and trunk angle of inclination (angle between a straight line from Th1 to S1 and true vertical [6]), as shown Figure 1. The larger the figures for thoracic and lumbar curvature measurements were, the greater the degree of kyphosis. The sacral hip angle and inclination measurements reflected forward pelvic tilt and forward stooped posture.

The intraexaminer reliability and interexaminer reliability of the Spinal Mouse were high in terms of intraclass correlation coefficients: 0.82–0.95 (6,7) and 0.81–0.86 (7), respectively.

Outcome Measurements

We defined dependence in ADL as either admission to a nursing home or need of assistance at home during the follow-up period (10). The latter was defined as long-term care (LTC) eligibility or a need for help in any of the six basic ADL items in the Katz Index of independence in ADL (11). LTC eligibility is a requirement for receiving LTC insurance services in Japan, which began in 2000. In this study, any of the seven levels of LTC insurance services was considered LTC eligible. Information on death, nursing home admission, and LTC eligibility was obtained from the Kurabuchi Branch Office of Takasaki City Hall. Information on Katz ADL was obtained from repeat face-to-face home interviews conducted every year until 2010 by public health nurses and local welfare commissioners, and occurrence of ADL dependence in any year was defined as dependence in ADL.

Covariates

We collected information on age, sex, smoking status (current vs former or never), alcohol drinking (current vs former or never), educational level (junior high vs high school or higher), back pain, including low back pain (yes or no in the past year), knee joint pain (never vs occasionally vs often vs always), and current or past history of life-threatening diseases, including stroke, myocardial infarction/angina, diabetes mellitus, and cancer (summary answer of yes or no). Body mass index was calculated as weight (kg) divided by the square of height (m) predicted by demi-span (12) and then categorized (<18.5 vs 18.5–24.9 vs 25≤ kg/m2). Estimated bone mineral density was assessed from calcaneal quantitative ultrasound measurements made with an A-1000 Express (GE Yokogawa Medical Systems, Tokyo, Japan) and expressed as a stiffness index. All of the above covariates have been reported to be involved in ADL dependence outcomes.

Statistical Analysis

All analyses were performed with STATA version 12 (STATA Corporation, College Station, Texas).

Distributions of the four parameters of spinal posture were calculated according to age and sex. Trends by age category were examined with logistic regression, with consecutive integers given to each category. The relationships between the spinal posture parameters and outcomes were assessed by dividing the participants into sex-specific quartiles. First, age-category (5-year increments) and sex-adjusted analyses were carried out with logistic regression models. Then, other covariates (education, current/past history of life-threatening diseases, knee joint pain, and body mass index category) were included in the models (Model 1). Smoking and drinking status were not included because they were not associated with outcomes in this study. Second, a model mutually adjusted for all four parameters of spinal posture (Model 2) was applied. Additionally, models including back pain and stiffness were applied to Model 2 (Model 3). Trends across increasing quartiles of the parameters were also calculated by treating the quartiles as an integral value. Because there was no interaction by sex, all analyses were carried out with combined data for men and women. This analytic method was repeated for dependence in ADL and for the composite outcome of dependence in ADL and death. Participants who died during the follow-up period were excluded from the analysis of dependence in ADL. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were used to describe the strengths of associations.

Results

During the 4.5-year follow-up period, 126 (16.4%) participants became dependent in ADL, 61 (7.6%) died, and 6 (0.7%) moved out of the town. Table 1 shows the characteristics of the study participants. Those in their seventies constituted the majority and women made up to 58%. The distributions of the four parameters of spinal posture are presented in Table 2. The effect of age on thoracic curvature seemed to vary between men and women. In men, thoracic kyphosis decreased with age, whereas it appeared to increase with age in women. However, the trend was not statistically significant. Lumbar lordosis decreased and inclination increased with age in both men and women.

Table 1.

Characteristics of the Study Participants (n = 804; Kurabuchi Study 2005)

  n (%)* 
Age category (y) 65–69 174 (21.7) 
70–74 237 (29.5) 
75–79 193 (24.0) 
80–84 137 (17.0) 
85 63 (7.8) 
Sex Men 338 (42.0) 
Women 466 (58.0) 
Current smoking Yes 101 (12.9) 
No 680 (87.1) 
Current drinking Yes 245 (31.6) 
No 530 (68.4) 
Education High school or higher 182 (23.5) 
Junior high school or below 593 (76.5) 
History of life-threatening diseases Yes 189 (24.5) 
No 582 (75.5) 
Back pain Yes 480 (59.7) 
No 324 (40.3) 
Knee joint pain Never 419 (54.1) 
Occasionally 170 (22.0) 
Often 63 (8.1) 
Always 122 (15.8) 
BMI category (kg/m2) <18.5 87 (10.8) 
18.5–24.9 511 (63.6) 
≥25 205 (25.5) 
Stiffness Mean (SD71.4 (16.8) 
  n (%)* 
Age category (y) 65–69 174 (21.7) 
70–74 237 (29.5) 
75–79 193 (24.0) 
80–84 137 (17.0) 
85 63 (7.8) 
Sex Men 338 (42.0) 
Women 466 (58.0) 
Current smoking Yes 101 (12.9) 
No 680 (87.1) 
Current drinking Yes 245 (31.6) 
No 530 (68.4) 
Education High school or higher 182 (23.5) 
Junior high school or below 593 (76.5) 
History of life-threatening diseases Yes 189 (24.5) 
No 582 (75.5) 
Back pain Yes 480 (59.7) 
No 324 (40.3) 
Knee joint pain Never 419 (54.1) 
Occasionally 170 (22.0) 
Often 63 (8.1) 
Always 122 (15.8) 
BMI category (kg/m2) <18.5 87 (10.8) 
18.5–24.9 511 (63.6) 
≥25 205 (25.5) 
Stiffness Mean (SD71.4 (16.8) 

Notes: SD = standard deviation; BMI = body mass index.

*Due to missing values, the totals for the stratified subgroups are not equal.

Stroke, myocardial infarction or angina, diabetes mellitus, and cancer were included.

BMI was calculated as weight (kg) divided by the square of height (m) predicted by demi-span (12).

Table 2.

Distributions of the parameters of spinal curvature by age and sex

  Thoracic Curvature Lumbar Curvature Sacral hip angle Inclination 
Age Category (y) n Median (25, 75 percentiles) Median (25, 75 percentiles) Median (25, 75 percentiles) Median (25, 75 percentiles) 
Men & women (n = 804)      
    65−69 174 44 (37, 50) –13.5 (–22, −4) 7 (1, 13) 7 (5, 11) 
    70−74 237 41 (32, 49) –10 (–18, 0) 7 (0, 12) 9 (6, 11) 
    75−79 193 43 (35, 50) –5 (–15, 6) 6 (0, 12) 12 (8, 17) 
    80−84 137 40 (30, 49) 0 (–11, 10) 4 (–2, 12) 13 (9, 22) 
    85− 63 44 (30, 54) –2 (–14, 15) 5 (–1, 14) 16 (10, 29) 
    trend*  p = 0.234 p < 0.001 p = 0.083 p < 0.001 
Men (n = 338)      
    65−69 74 47.5 (40, 54) –14.5 (–24, –4) 4.5 (1, 14) 8 (6, 12) 
    70−74 103 41 (31, 49) –10 (–17, –1) 6 (0, 13) 9 (7, 12) 
    75−79 82 43.5 (36, 49) –8 (–17.5, 3) 7 (0, 12) 12 (7, 16) 
    80−84 54 40 (29, 48) 0 (–9, 9) 3.5 (–2, 12) 11 (8, 20) 
    85− 25 39 (29, 51) –2 (–14, 17) 4 (1, 12) 10 (9, 21) 
    trend*  p < 0.001 p < 0.001 p = 0.172 p < 0.001 
Women (n = 466)      
    65−69 100 41 (35, 48.5) –13 (–21.5, –3.5) 8.5 (3, 12.5) 7 (4, 11) 
    70−74 134 41 (32, 50) –10 (–18, 3) 7 (0, 11) 9 (6, 13) 
    75−79 111 43 (32, 50) –4 (–14, 8) 5 (0, 13) 12 (9, 18) 
    80−84 83 42 (34, 51) 1 (–14, 12) 4 (–1, 13) 14 (9, 23) 
    85− 38 45 (33, 58) –2.5 (–16, 14) 5.5 (1, 14) 17 (13, 29) 
    trend*  p = 0.173 p < 0.001 p = 0.277 p < 0.001 
  Thoracic Curvature Lumbar Curvature Sacral hip angle Inclination 
Age Category (y) n Median (25, 75 percentiles) Median (25, 75 percentiles) Median (25, 75 percentiles) Median (25, 75 percentiles) 
Men & women (n = 804)      
    65−69 174 44 (37, 50) –13.5 (–22, −4) 7 (1, 13) 7 (5, 11) 
    70−74 237 41 (32, 49) –10 (–18, 0) 7 (0, 12) 9 (6, 11) 
    75−79 193 43 (35, 50) –5 (–15, 6) 6 (0, 12) 12 (8, 17) 
    80−84 137 40 (30, 49) 0 (–11, 10) 4 (–2, 12) 13 (9, 22) 
    85− 63 44 (30, 54) –2 (–14, 15) 5 (–1, 14) 16 (10, 29) 
    trend*  p = 0.234 p < 0.001 p = 0.083 p < 0.001 
Men (n = 338)      
    65−69 74 47.5 (40, 54) –14.5 (–24, –4) 4.5 (1, 14) 8 (6, 12) 
    70−74 103 41 (31, 49) –10 (–17, –1) 6 (0, 13) 9 (7, 12) 
    75−79 82 43.5 (36, 49) –8 (–17.5, 3) 7 (0, 12) 12 (7, 16) 
    80−84 54 40 (29, 48) 0 (–9, 9) 3.5 (–2, 12) 11 (8, 20) 
    85− 25 39 (29, 51) –2 (–14, 17) 4 (1, 12) 10 (9, 21) 
    trend*  p < 0.001 p < 0.001 p = 0.172 p < 0.001 
Women (n = 466)      
    65−69 100 41 (35, 48.5) –13 (–21.5, –3.5) 8.5 (3, 12.5) 7 (4, 11) 
    70−74 134 41 (32, 50) –10 (–18, 3) 7 (0, 11) 9 (6, 13) 
    75−79 111 43 (32, 50) –4 (–14, 8) 5 (0, 13) 12 (9, 18) 
    80−84 83 42 (34, 51) 1 (–14, 12) 4 (–1, 13) 14 (9, 23) 
    85− 38 45 (33, 58) –2.5 (–16, 14) 5.5 (1, 14) 17 (13, 29) 
    trend*  p = 0.173 p < 0.001 p = 0.277 p < 0.001 

* Trend was examined by scoring the age category (1 to 5) as a continuous term in the regression analysis.

The associations of the four parameters of spinal posture with future dependence in ADL are summarized in Table 3. In Model 1, only inclination was associated with outcome, although lumbar curvature and sacral hip angle showed marginal associations. When the lowest quartile of inclination was used as a reference, the adjusted ORs (95% CI) for the second, third, and highest quartiles were 1.46 (0.60, 3.59), 3.90 (1.76, 8.63), and 4.93 (2.23, 10.91), respectively. The adjusted OR for a 1 unit increase in the quartiles of inclination was 1.75 (1.39, 2.20). Even when inclination was included in the model as a continuous variable, the adjusted OR for a 1 unit increase in inclination was 1.04 (1.02–1.06). In the model where the four parameters were mutually adjusted (Model 2), statistical significance for inclination was maintained, and the association estimates did not change substantially. When back pain and stiffness were added as covariates, the ORs were essentially the same as those in Model 2. When the analysis was repeated for the composite outcome of dependence in ADL and death, the association of inclination with outcome was only slightly attenuated and remained statistically significant. The adjusted ORs for 1 unit increases in the quartiles of inclination were 1.52 (1.19, 1.93) in Model 2 and 1.50 (1.18, 1.91) in Model 3 for the composite outcome of dependence in ADL and death.

Table 3.

Associations of the parameters with dependence in activities of daily living

  Median (25, 75 percentiles) n/N (%) Age- & Sex-Adjusted OR (95% CI) Adjusted OR (95% CI) Model 1* Adjusted OR (95% CI) Model 2** Adjusted OR (95% CI) Model 3*** 
Thoracic curvature Q1 27 (22, 30) 29/168 (17.3) 1.00 1.00 1.00 1.00 
 Q2 38 (36, 40) 38/183 (20.8) 1.73 (0.96−3.10) 1.81 (0.98−3.34) 2.04 (1.05−3.96) 2.05 (1.05−4.00) 
 Q3 45 (44, 48) 30/188 (16.0) 1.17 (0.64−2.14) 1.17 (0.62−2.20) 1.37 (0.68−2.76) 1.42 (0.70−2.85) 
 Q4 56 (52, 61) 29/198 (14.7) 0.94 (0.51−1.71) 0.75 (0.39−1.44) 0.92 (0.43−1.96) 0.89 (0.41−1.92) 
 one unit increase in the quartiles   0.94 (0.79−1.13) 0.89 (0.73−1.08) 0.98 (0.77−1.23) 0.97 (0.77−1.23) 
Lumbar curvature Q1 −24 (−28, −20) 20/188 (10.6) 1.00 1.00 1.00 1.00 
 Q2 −12 (−15, −10) 24/194 (12.4) 1.07 (0.55−2.08) 1.20 (0.60−2.42) 1.07 (0.47−2.44) 1.08 (0.47−2.46) 
 Q3 −2 (−4, 0) 39/186 (21.0) 1.77 (0.95−3.30) 1.72 (0.89−3.33) 1.50 (0.57−3.98) 1.47 (0.55−3.93) 
 Q4 13 (8, 28) 43/169 (25.4) 1.65 (0.88−3.07) 1.71 (0.88−3.33) 1.52 (0.47−4.92) 1.38 (0.42−4.50) 
 one unit increase in the quartiles   1.22 (1.00−1.48) 1.21 (0.99−1.48) 1.20 (0.81−1.76) 1.17 (0.79−1.72) 
Sacral hip angle Q1 −5 (−9, −2) 28/165 (17.0) 1.00 1.00 1.00 1.00 
 Q2 3 (1, 4) 32/189 (16.9) 1.10 (0.61−2.00) 1.20 (0.65−2.24) 1.27 (0.62−2.61) 1.28 (0.62−2.62) 
 Q3 9 (7, 11) 27/198 (13.6) 1.14 (0.61−2.11) 1.17 (0.60−2.26) 1.32 (0.56−3.14) 1.22 (0.51−2.93) 
 Q4 16 (15, 20) 39/185 (21.1) 1.63 (0.91−2.92) 1.83 (0.98−3.42) 2.26 (0.87−5.85) 2.20 (0.84−5.73) 
 one unit increase in the quartiles   1.17 (0.97−1.41) 1.20 (0.98−1.47) 1.30 (0.95−1.78) 1.29 (0.94−1.77) 
Inclination Q1 4 (3, 6) 10/185 (5.4) 1.00 1.00 1.00 1.00 
 Q2 8 (7, 9) 15/180 (8.3) 1.48 (0.63−3.48) 1.46 (0.60−3.59) 1.34 (0.54−3.33) 1.43 (0.57−3.57) 
 Q3 12 (11, 13) 42/199 (21.1) 3.71 (1.75−7.85) 3.90 (1.76−8.63) 3.32 (1.43−7.69) 3.28 (1.41−7.62) 
 Q4 20 (16, 27) 59/173 (34.1) 5.30 (2.51−11.18) 4.93 (2.23−10.91) 3.65 (1.43−9.37) 3.47 (1.35−8.93) 
 one unit increase in the quartiles   1.79 (1.44−2.23) 1.75 (1.39−2.20) 1.67 (1.25−2.23) 1.62 (1.21−2.17) 
  Median (25, 75 percentiles) n/N (%) Age- & Sex-Adjusted OR (95% CI) Adjusted OR (95% CI) Model 1* Adjusted OR (95% CI) Model 2** Adjusted OR (95% CI) Model 3*** 
Thoracic curvature Q1 27 (22, 30) 29/168 (17.3) 1.00 1.00 1.00 1.00 
 Q2 38 (36, 40) 38/183 (20.8) 1.73 (0.96−3.10) 1.81 (0.98−3.34) 2.04 (1.05−3.96) 2.05 (1.05−4.00) 
 Q3 45 (44, 48) 30/188 (16.0) 1.17 (0.64−2.14) 1.17 (0.62−2.20) 1.37 (0.68−2.76) 1.42 (0.70−2.85) 
 Q4 56 (52, 61) 29/198 (14.7) 0.94 (0.51−1.71) 0.75 (0.39−1.44) 0.92 (0.43−1.96) 0.89 (0.41−1.92) 
 one unit increase in the quartiles   0.94 (0.79−1.13) 0.89 (0.73−1.08) 0.98 (0.77−1.23) 0.97 (0.77−1.23) 
Lumbar curvature Q1 −24 (−28, −20) 20/188 (10.6) 1.00 1.00 1.00 1.00 
 Q2 −12 (−15, −10) 24/194 (12.4) 1.07 (0.55−2.08) 1.20 (0.60−2.42) 1.07 (0.47−2.44) 1.08 (0.47−2.46) 
 Q3 −2 (−4, 0) 39/186 (21.0) 1.77 (0.95−3.30) 1.72 (0.89−3.33) 1.50 (0.57−3.98) 1.47 (0.55−3.93) 
 Q4 13 (8, 28) 43/169 (25.4) 1.65 (0.88−3.07) 1.71 (0.88−3.33) 1.52 (0.47−4.92) 1.38 (0.42−4.50) 
 one unit increase in the quartiles   1.22 (1.00−1.48) 1.21 (0.99−1.48) 1.20 (0.81−1.76) 1.17 (0.79−1.72) 
Sacral hip angle Q1 −5 (−9, −2) 28/165 (17.0) 1.00 1.00 1.00 1.00 
 Q2 3 (1, 4) 32/189 (16.9) 1.10 (0.61−2.00) 1.20 (0.65−2.24) 1.27 (0.62−2.61) 1.28 (0.62−2.62) 
 Q3 9 (7, 11) 27/198 (13.6) 1.14 (0.61−2.11) 1.17 (0.60−2.26) 1.32 (0.56−3.14) 1.22 (0.51−2.93) 
 Q4 16 (15, 20) 39/185 (21.1) 1.63 (0.91−2.92) 1.83 (0.98−3.42) 2.26 (0.87−5.85) 2.20 (0.84−5.73) 
 one unit increase in the quartiles   1.17 (0.97−1.41) 1.20 (0.98−1.47) 1.30 (0.95−1.78) 1.29 (0.94−1.77) 
Inclination Q1 4 (3, 6) 10/185 (5.4) 1.00 1.00 1.00 1.00 
 Q2 8 (7, 9) 15/180 (8.3) 1.48 (0.63−3.48) 1.46 (0.60−3.59) 1.34 (0.54−3.33) 1.43 (0.57−3.57) 
 Q3 12 (11, 13) 42/199 (21.1) 3.71 (1.75−7.85) 3.90 (1.76−8.63) 3.32 (1.43−7.69) 3.28 (1.41−7.62) 
 Q4 20 (16, 27) 59/173 (34.1) 5.30 (2.51−11.18) 4.93 (2.23−10.91) 3.65 (1.43−9.37) 3.47 (1.35−8.93) 
 one unit increase in the quartiles   1.79 (1.44−2.23) 1.75 (1.39−2.20) 1.67 (1.25−2.23) 1.62 (1.21−2.17) 

Notes: OR: odds ratio, CI: confidence interval.

In this analysis, residents who died during the follow-up period (n = 61) were excluded.

*Age category, sex, educational category, history of life-threatening diseases (stroke, myocardial infarction/angina, diabetes mellitus, and cancer), knee joint pain and body mass index category were adjusted for.

**In addition to the variables included in Model 1, all parameters (thoracic curvature, lumbar curvature, sacral hip angle, inclination) were mutually adjusted.

***Back pain and stiffness were added to Model 2.

Discussion

We evaluated four parameters of spinal posture (thoracic curvature, lumbar curvature, sacral hip angle, and inclination) in older adults and demonstrated for the first time, after 4.5 years of follow-up, that of the four parameters, inclination has the greatest effect on dependence in ADL with no clear threshold. We showed too that this association was independent of back pain and estimated bone mineral density. Our results indicate that attention needs to be paid to inclination in spinal posture to identify elderly people at high risk of becoming dependent in ADL.

Many reports have indicated that posture of the trunk in the sagittal plane is associated with body function and dependence in ADL. Various methods of measuring spinal posture were used in these studies. Leech and colleagues (13) and Lombardi and colleagues (14) measured the Cobb angle to assess kyphosis, and reported that hyperkyphosis might be associated with pulmonary function. In another study, participants lay supine with the neck in a neutral position and occiput to table distance was measured with 1.7-cm blocks placed between the head and the examination table; moderate hyperkyphotic posture was found to indicate an increased risk of injurious falls in older men, with a less pronounced association in older women (15). Ryan and colleagues (16) evaluated kyphosis through measurement of the distance between the occiput and a wall, finding that kyphosis is associated with ADL decline. Takahashi and colleagues (5) examined sagittal spinal posture using lateral-view photographs of the participants and found that groups with trunk deformities tended to score lower than the control group on subjective measures of healthiness and life satisfaction. The abovementioned methods of kyphosis assessment are easy and useful, requiring no medical information from local health centers. However, they evaluate spinal posture as a whole, making assessment of each composite parameter of spinal posture impossible. We overcame this disadvantage by using a noninvasive device for measuring spinal shape that has been developed in recent years, the Spinal Mouse.

In our study, mutual adjustment for the four parameters of spinal posture showed that only inclination is associated with future dependence in ADL. Other cross-sectional studies using the Spinal Mouse may help explain this association. Sakamitsu and colleagues (17) used the Spinal Mouse to measure balance (1 ft standing with eyes open) and gait (walking speed in 10-m walk and walking distance in 3min) in 28 elderly people. They concluded that the larger the anterior inclination of the trunk, the greater the decline in balance and gait skills was. A study by Ishikawa and colleagues (18) of 93 osteoporotic patients with a mean age of 70 years showed that forward spinal inclination with a forward stooped posture affected postural balance. It is reasonable to suppose, therefore, that declines in balance and gait skills caused by inclination lead to falls and fractures, and that these negative outcomes in turn lead to dependence in ADL among elderly people. In fact, data exist to show that community-dwelling women with osteoporosis and hyperkyphosis have weaker back extensor strength, weaker lower extremity strength, slower gait, poorer balance, and greater body sway, which as a result gives them a propensity to fall (19).

The line of gravity line moves naturally with changes in spinal alignment (20–22). Arita and colleagues (23) reported that the center of gravity runs anterior to L4-5 and 6mm anterior to the hip joint in most elderly people, and another study showed that the center of gravity runs anterior to L4 in healthy elderly people (24). The gravity line moves further anterior as inclination of the trunk increases.

In one study, full-length, free-standing radiographs of the spine and pelvis were examined in 125 adult patients with spinal deformities. The study demonstrated that the T1 spinopelvic inclination (the angle between the T1-hip axis and vertical) correlated with Health-Related Quality of Life measures (25). In another study, 752 patients with spinal deformities were enrolled from a multicenter prospective database. Positive sagittal balance was defined as an anterior deviation of the plumb line from the seventh cervical spinous process. This study showed that although even mildly positive sagittal balance is somewhat detrimental, the decline in health status increases in a linear fashion with progressive sagittal imbalance (26).

Some reports in Japanese indicated that lumbar lordosis decreased with increase in age (27,28) and our results supported these earlier studies. Lumbar lordosis is also reported to be associated with decline in walking ability and in muscle power of lower limb (28–30). In this study, however, lumbar curvature only showed a marginal association with future dependence in ADL. If we included inclination in the model, this association disappeared. Whereas, thoracic kyphosis decreased with age among men. Among women, on the contrary, thoracic kyphosis tended to increase with age as reported in other studies (2,29,31). Although we could not explain this sex difference, thoracic kyphosis might decrease as a compensation for the decrease in lumbar lordosis in men. These potential difference between Japanese and people in other countries and between men and women needs further study.

Although we examined only spinal posture in this study, we recognize that examination of lower limb alignment is also important because changes in spinal posture can influence the alignment of the legs: burdens requiring more than the normal compensatory reactions can lead to joint diseases such as osteoarthritis, which in turn lead to declines in ADL. Such being the case, future studies including evaluations of lower limb alignment are necessary. Although the Spinal Mouse method was easy and useful for evaluating sagittal spinal posture as a whole, it does not seem to be a reliable tool yet for measuring intersegmental spinal range of motion (7). Furthermore, the Spinal Mouse method does not measure spinopelvic alignment. When sagittal unbalance is detected with Spinal Mouse method, therefore, we recommend full x-ray investigation for evaluating spine and pelvis. Another limitation of our study is that we focused only on sagittal spinal posture. Differences in body sizes and lifestyles also mean that caution is necessary in applying our results to other populations. However, we believe our conclusion that inclination is associated with future dependence in ADL among older adults warrants wide attention.

Figure 1.

A schema illustrating the four parameters. Thoracic curvature: thoracic kyphosis (corresponds to Cobb angle between Th1 and Th12). Lumbar curvature: lumbar kyphosis (corresponds to Cobb angle between Th12 and S1). Sacral hip angle: the angle between a straight line from S1 to S3 and true vertical. Inclination: the angle between a straight line from Th1 to S1 and true vertical.

Figure 1.

A schema illustrating the four parameters. Thoracic curvature: thoracic kyphosis (corresponds to Cobb angle between Th1 and Th12). Lumbar curvature: lumbar kyphosis (corresponds to Cobb angle between Th12 and S1). Sacral hip angle: the angle between a straight line from S1 to S3 and true vertical. Inclination: the angle between a straight line from Th1 to S1 and true vertical.

Funding

This work was supported by a grant in aid from the Ministry of Health, Labour and Welfare, Japan (H20-Choujyu-009).

Acknowledgments

The authors would like to thank the staff of the City person & Health Division, Kurabuchi Branch Office, Takasaki City Hall, Gunma prefecture, Japan for their valuable help.

References

1.
Ministry of Health, Labour and Welfare, Japan. 2011
. http://www.mhlw.go.jp/toukei/saikin/hw/life/life11/dl/life11-14.pdf. Accessed July 26, 2012.
2.
Nishiwaki
Y
Kikuchi
Y
Araya
K
et al
.
Association of thoracic kyphosis with subjective poor health, functional activity and blood pressure in the community-dwelling elderly
.
Environ Health Prev Med
 .
2007
;
12
:
246
250
.
3.
Kado
DM
Prenovost
K
Crandall
C
.
Narrative review: hyperkyphosis in older persons
.
Ann Intern Med
 .
2007
;
147
:
330
338
.
4.
Hirose
D
Ishida
K
Nagano
Y
Takahashi
T
Yamamoto
H
.
Posture of the trunk in the sagittal plane is associated with gait in community-dwelling elderly population
.
Clin Biomech (Bristol, Avon)
 .
2004
;
19
:
57
63
.
5.
Takahashi
T
Ishida
K
Hirose
D
et al
.
Trunk deformity is associated with a reduction in outdoor activities of daily living and life satisfaction in community-dwelling older people
.
Osteoporos Int
 .
2005
;
16
:
273
279
.
6.
Mannion
AF
Knecht
K
Balaban
G
Dvorak
J
Grob
D
.
A new skin-surface device for measuring the curvature and global and segmental ranges of motion of the spine: reliability of measurements and comparison with data reviewed from the literature
.
Eur Spine J
 .
2004
;
13
:
122
136
.
7.
Post
RB
Leferink
VJ
.
Spinal mobility: sagittal range of motion measured with the SpinalMouse, a new non-invasive device
.
Arch Orthop Trauma Surg
 .
2004
;
124
:
187
192
.
8.
Michikawa
T
Nishiwaki
Y
Kikuchi
Y
et al
.
Gender-specific associations of vision and hearing impairments with adverse health outcomes in older Japanese: a population-based cohort study
.
BMC Geriatr
 .
2009
;
9
:
50
9.
Michikawa
T
Nishiwaki
Y
Asakura
K
et al
.
Sunlight exposure may be a risk factor of hearing impairment: a community-based study in Japanese older men and women
.
J Gerontol A Biol Sci Med Sci
 .
2012
;
68
:
96
103
.
10.
Nishiwaki
Y
Michikawa
T
Yamada
M
Eto
N
Takebayashi
T
; Kurabuchi Study Group
Knee pain and future self-reliance in older adults: evidence from a community-based 3-year cohort study in Japan
.
J Epidemiol
 .
2011
;
21
:
184
190
.
11.
Katz
S
Ford
AB
Moskowitz
RW
Jackson
BA
Jaffe
MW
.
Studies of illness in the aged. The index of adl: a standardized measure of biological and psychosocial function
.
JAMA
 .
1963
;
185
:
914
919
.
12.
Nishiwaki
Y
Michikawa
T
Eto
N
Takebayashi
T
.
Body mass index misclassification due to kyphotic posture in Japanese community-dwelling adults aged 65 years and older
.
J Gerontol A Biol Sci Med Sci
 .
2011
;
66
:
326
331
.
13.
Leech
JA
Dulberg
C
Kellie
S
Pattee
L
Gay
J
.
Relationship of lung function to severity of osteoporosis in women
.
Am Rev Respir Dis
 .
1990
;
141
:
68
71
.
14.
Lombardi
I
Jr
Oliveira
LM
Mayer
AF
Jardim
JR
Natour
J
.
Evaluation of pulmonary function and quality of life in women with osteoporosis
.
Osteoporos Int
 .
2005
;
16
:
1247
1253
.
15.
Kado
DM
Huang
MH
Nguyen
CB
Barrett-Connor
E
Greendale
GA
.
Hyperkyphotic posture and risk of injurious falls in older persons: the Rancho Bernardo Study
.
J Gerontol A Biol Sci Med Sci
 .
2007
;
62
:
652
657
.
16.
Ryan
SD
Fried
LP
.
The impact of kyphosis on daily functioning
.
J Am Geriatr Soc
 .
1997
;
45
:
1479
1486
.
17.
Sakamitsu
T
Urabe
Y
Yamamoto
T
.
Relationship of kyphosis with balance and walking ability in the elderly
.
Rigakuryoho Kagaku
 .
2007
;
22
:
489
494
(in Japanese with English abstract).
18.
Ishikawa
Y
Miyakoshi
N
Kasukawa
Y
Hongo
M
Shimada
Y
.
Spinal curvature and postural balance in patients with osteoporosis
.
Osteoporos Int
 .
2009
;
20
:
2049
2053
.
19.
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
:
1004
1010
.
20.
Thomas
DP
Whitney
RJ
.
Postural movements during normal standing in man
.
J Anat
 .
1959
;
93
:
524
539
.
21.
Asmussen
E
.
The weight-carrying function of the human spine
.
Acta Orthop Scand
 .
1960
;
29
:
276
290
.
22.
Woodhull
AM
Maltrud
K
Mello
BL
.
Alignment of the human body in standing
.
Eur J Appl Physiol Occup Physiol
 .
1985
;
54
:
109
115
.
23.
Arita
C
Kobayashi
I
.
The analysis of the posture of the senile
.
Rinshouseikeigeka
 .
1980
;
15
:
115
122
(in Japanese).
24.
Harada
T
Motegi
M
Okazima
Y
Hasegawa
K
Abe
H
Tsuruoka
H
.
Posture of the aged
.
Sougouriha
 .
1994
;
22
:
133
136
(in Japanese).
25.
Lafage
V
Schwab
F
Patel
A
Hawkinson
N
Farcy
JP
.
Pelvic tilt and truncal inclination: two key radiographic parameters in the setting of adults with spinal deformity
.
Spine
 .
2009
;
34
:
E599
E606
.
26.
Glassman
SD
Bridwell
K
Dimar
JR
Horton
W
Berven
S
Schwab
F
.
The impact of positive sagittal balance in adult spinal deformity
.
Spine
 .
2005
;
30
:
2024
2029
.
27.
Hatake
K
Nakano
T
Ochi
T
Inaba
D
Yasuoka
H
.
Comparison of spinal alignment in healthy women and old women with vertebral compression fracture –Consideration of compensated spinal alignment
.
Therap Res
 .
2008
;
29
:
597
600
(in Japanese).
28.
Miyazaki
J
Murata
S
Otao
H
Horie
J
Murata
J
Suzuki
S
.
Relationship between physical function and spinal sagittal plane alignment of elderly men
.
Rigakuryoho Kagaku
 .
2009
;
24
:
907
911
(in Japanese with English abstract).
29.
Iwaya
T
Shirakihara
N
Tobimatu
Y
.
The effects of low back pain on physical and social function in elderly people living in rural area
.
J Jpn Soc Lumbar Spine Disord
 .
2005
;
11
:
27
34
(in Japanese).
30.
Kai
Y
Murata
S
Otao
H
et al
.
A relationship between spinal deformity and lower limb muscle strength in community-dwelling elderly females
.
Rigakuryoho Kagaku
 .
2009
;
24
:
45
48
(in Japanese with English abstract).
31.
Mori
S
.
Guaging the impact of vertebral compression fractures and spinal deformity on the level of daily activity in patients with osteoporosis
.
J Jpn Soc Lumbar Spine Disord
 .
2002
;
8
:
58
63
(in Japanese with English abstract).

Author notes

Decision Editor: Stephen Kritchevsky, PhD