- Email: [email protected]
The relationship between body composition and femoral neck osteoporosis or osteopenia in adults with previous poliomyelitis
ARTICLE IN PRESS
Disability and Health Journal
-
(2014)
-
www.disabilityandhealthjnl.com
Brief Report
The relationship between body composition and femoral neck osteoporosis or osteopenia in adults with previous poliomyelitis Kwang-Hwa Chang, M.D.a,b, Sung-Hui Tseng, M.D., Ph.D.c, Yu-Ching Lin, M.D., M.Sc.d, Chien-Hung Lai, M.D., Ph.D.c,e, Wen-Tien Hsiao, M.Sc.f, and Shih-Ching Chen, M.D., Ph.D.c,e,* a Department of Physical Medicine and Rehabilitation, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan Graduate Institute of Injury Prevention and Control, College of Public Health and Nutrition, Taipei Medical University, Taipei, Taiwan c Department of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei, Taiwan d Department of Physical Medicine and Rehabilitation, National Cheng Kung University Hospital, Tainan, Taiwan e Department of Physical Medicine and Rehabilitation, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan f Department of Diagnostic Radiology, Taipei Medical University Hospital, Taipei, Taiwan b
Abstract Background: Articles in the literature describing the association between body composition and osteoporosis in subjects with poliomyelitis are scarce. Objective: To assess the relationship between body composition and femoral neck osteoporosis or osteopenia in adults with previous polio. Method: After excluding postmenopausal women, 44 polio (mean age 6 standard deviation, 46.1 6 3.3 years) and 44 able-bodied control volunteers (47.0 6 4.0 years) participated in the study. Each participant’s femoral neck bone mineral density (FNBMD) and whole body composition were measured using dual-energy X-ray absorptiometry. With local reference BMD values of normal young adults installed in the instrument, we obtained T-score values that depended on each FNBMD value. A T-score value of <1.0 indicated decreased T-score, including osteoporosis (T-score < 2.5) and osteopenia (1.0 to 2.5). This study conducted logistic regression analyses to find factors associated with osteoporosis and osteopenia. Results: Based on the FNBMD T-score values, 60.0% of middle-aged men with polio had osteoporosis. In adjusted logistic regression analyses, total lean tissue mass (Adjusted odds ratio [95% confidence interval], 0.74 [0.56e0.99], P ! 0.05) and male gender (947.16 [6.02e148,926.16], P ! 0.01) were important factors associated with decreased T-score in polio group. Conclusions: Osteoporosis or osteopenia is a common medical problem for middle-aged men with polio. Reduced total lean tissue mass seems to be one of the important factors associated with osteoporosis or osteopenia among subjects with polio. Further research for a clinical tool to assess lean tissue mass for subjects with polio is needed. Ó 2014 Elsevier Inc. All rights reserved. Keywords: Osteoporosis; Poliomyelitis; Femoral neck; Body composition
Osteoporosis is an important global medical issue, and bony fracture is one of its most serious complications. Among the various types of bony fractures, hip fractures often have compromised outcomes. A recent study shows
that approximately 1 year following a hip fracture, 20e 30% of the individuals affected had died,1 40% were dependent in walking, and 60% had some difficulty with the activities of daily living. Thus, estimating the rate of
Abbreviations: BFM, body fat mass; BMD, bone mineral density; DXA, dual-energy X-ray absorptiometry; FNBMD, femoral neck bone mineral density; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LTM, lean tissue mass. Disclosure of financial interests: The authors report nothing to be disclosed. This disclosure includes direct or indirect financial or personal relationships, interests, and affiliations relevant to the subject matter of the manuscript that have occurred over the last two years, or that are expected in the foreseeable future. This disclosure includes, but is not limited to, grants or funding, employment, affiliations, patents (in preparation, filed,
or granted), inventions, honoraria, consultancies, royalties, stock options/ ownership, or expert testimony. Conflicts of interest: The authors report no conflicts of interest. The authors disclose that no prior presentation of abstract at meeting or posting any part or all of the content on a website, even in draft form. * Corresponding author. Taipei Medical University Hospital, 252 Wu-Hsing Street, Taipei 11031, Taiwan. Tel.: þ886 2 27372181x1236; fax: þ886 2 55589880. E-mail address: [email protected] or [email protected] (S.-C. Chen).
1936-6574/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.dhjo.2014.09.011
ARTICLE IN PRESS 2
K.-H. Chang et al. / Disability and Health Journal
experiencing osteoporosis of the femoral neck and identifying the risk factors of osteoporosis are essential tasks in caring for those people at risk. Bone mineral density (BMD) is a major predictor for the risk of osteoporotic fracture.2 However, this continuous measurement is subject to ethnic differences.3 BMD values of Caucasian people are often higher than those of Chinese.4 A BMD cutoff value to differentiate a high risk of developing fractures from low risk is lacking. For clinical practice and the assessment of fracture risk,5 BMD values are typically expressed as a T-score or the standard deviation (SD) from the mean BMD value of normal young adults aged 20e40 years of the same sex and race.6 According to the WHO criteria,5 a BMD T-score of 1.0 to 2.5 is classified as being osteopenic, and a T-score of <2.5 is osteoporotic. Combined with age and sex, T-scores offer a useful way to estimate the long-term risk of osteoporotic fracture.3 Polio was an epidemic in children aged <3 years in Taiwan from 1950 to 1970.7,8 Many of those survivors lived an active adult life of mild to moderate degree of physical activity.9 Because subjects with polio often have physical disabilities and low endurance of knee extensor muscles,10 they are susceptible to bony fractures caused by falls. A questionnaire survey study showed that two-thirds of polio subjects had experienced a fall from a standing height in the previous 1-year period.11 Bony fractures were the most common injury. The literature also shows that fractures with low-impact trauma are usually osteoporosis-related.3,12 Compared to able-bodied controls, men with polio have 13e23% lower values for femoral neck BMD (FNBMD).13 Smeltzer et al showed that O40% of adult women with previous polio had osteoporosis at the os calcis.14 However, the rate of osteoporosis at the femoral neck in polio subjects of different sexes remains unknown. In addition, BMD is related to body composition, including body fat mass (BFM), lean tissue mass (LTM), and bone mass. Recent articles show a positive association between LTM and BMD.15e17 Body fat mass is also related to BMD,18 and this is especially true for women.19 Compared to able-bodied controls, polio subjects have a decreased LTM and increased BFM for the whole body and the leg region.20 Thus, changes in body composition could influence the occurrence of osteoporosis among subjects with prior polio. To our knowledge, previous literature describing the association between body composition and osteoporosis in subjects with polio is scarce. The aim of this study was to assess the relationship between body composition and the presence of osteopenia or osteoporosis among subjects with previous polio.
Methods Participants This was a cross-sectional study with subjects who experienced acute paralytic polio at the age of 3 years or
-
(2014)
-
younger (polio group). All subjects experienced flaccid paralysis of one or both lower extremities. Having a leglength discrepancy, each subject with polio often had a shorter leg, which was shorter than his/her other leg (the longer leg). This study also recruited able-bodied persons without a history of any neuromuscular disease as controls (control group). Both groups’ participants had a body mass index range of 16e30 kg/m2.21 To focus on the association between polio and early bone loss, postmenopausal women were excluded from the study. Considering that rapid bone loss often begins after the age of 40 years12 and age of O65 years is an important risk factor for osteoporosis in the general population,22 this study recruited people of both groups between the ages of 40 and 65 years. This study also excluded those who had a body weight of !40 kg,23 those who had a metal implant or a history of a leg fracture, those who had used glucocorticoids or other bone metabolismrelated medications in the most recent 3 months, and those with a history of other neuromuscular diseases. This study used advertisements to recruit participants from the community of urban areas in northern Taiwan. A total of 51 subjects with polio and 44 able-bodied controls were recruited and assessed by an experienced physiatrist. Of the subjects with polio, three had a body weight of !40 kg, three had a history of exclusion criteria (one oophorectomy, one hip fracture, and one stroke), and one failed in measuring FNBMD. Thus, 44 subjects with polio (response rate: 86%) and 44 able-bodied controls (response rate: 100%) completed this study. The Institutional Review Board of our university-affiliated hospital approved this study. Each participant signed an informed consent form before entering this study. Measurements This study used dual-energy X-ray absorptiometry (DXA; Norland XR-36, version 2.5.3 software, Norland Corp., Fort Atkinson, Wisconsin, USA) to measure the FNBMD and whole-body composition of all participants at a university-affiliated hospital. This study measured the shorter leg’s FNBMD because, on average, the shorter leg has lower FNBMD in patients with polio.13 This study arbitrarily chose the control participants’ right leg as the reference leg for measuring FNBMD. A densitometry technologist certified by the International Society for Clinical Densitometry performed these measurements. The long-term coefficient of variation of the instrument was 1.2% for FNBMD. With local reference BMD values of normal young adults installed in the instrument,6 this study obtained T-score values that depended on each FNBMD value. Because both osteopenia and osteoporosis are risk factors for low-trauma fractures,24 legs with a T-score of <1.0 were identified as having ‘‘decreased T-score.’’ This study obtained weights of the body compositional variables of the whole body and both legs from a whole-body DXA scan, and identified the weights of the LTM and BFM distal
ARTICLE IN PRESS K.-H. Chang et al. / Disability and Health Journal
to the femoral trochanters of both legs as the leg LTM and BFM, respectively. The demarcation between the legs and trunk was based on the Norland user’s instructions. According to Norland Corp., the reported error for total BFM estimates using this device is !2%. Using the methods by Chang et al,20 this study measured the body weight, body height (from the vertex to the heel of the longer leg), and blood pressure of all participants, and took an overnight fasting blood sample from each participant. This study measured the biochemical profile of each blood sample, including fasting blood glucose, serum alanine transaminase, blood urea nitrogen, total cholesterol, low- (LDL) and high-density lipoprotein (HDL), triglyceride, and hemoglobin. By comparing these data to reference ranges, this study obtained the frequencies of participants with potential risk factors for osteoporosis,22 including systolic hypertension (systolic blood pressure O130 mm Hg),25 diastolic hypertension (diastolic blood pressure O85 mm Hg),25 hyperglycemia (fasting blood glucose O100 mg/dL),26 hypercholesterolemia (total serum cholesterol O200 mg/dL),27 hypertriglycemia (serum triglyceride O150 mg/dL),28,29 high LDL (serum LDL O130 mg/ dL),28 low HDL (serum HDL !40 mg/dL),29 liver disease (serum alanine transaminase O40 IU/L),22 renal disease (blood urea nitrogen O18 mg/dL),22 and anemia (hemoglobin !12 g/dL).30 This study used a quasi-structured questionnaire to assess the participants’ lifestyles, including daily cigarette smoking, daily alcohol consumption, and daily amount of exercise. This questionnaire comprised three open- and eight close-ended questions. Because cigarette smoking increases the risk of hip fracture,31 this study assessed each participant’s long-term amount of smoking and calculated the cumulative lifetime cigarette use by multiplying the average number of cigarettes smoked daily by the lifetime duration of smoking. Those who exercised less than 90 min on a weekly basis in the most recent 3 months were identified as physically inactive participants, whereas the others were identified as physically active. Data analysis This study compared the differences in clinical characteristics between the polio and control groups using a Pearson chi-squared test or Fisher’s exact test, and compared the age, body weight, body height, body mass index, total LTM, total BFM, leg LTM, leg BFM, and cumulative lifetime cigarette use between groups using two-tailed Student’s t tests. This study also compared the FNBMD T-score distributions of polio subjects with the controls by using Pearson’s chi-squared test. This study conducted logistic regression analyses with a forward stepwise method to find predictors with decreased T-score (T-score < 1.0). A stratified analysis based on the polio status was conducted. This study entered body compositional variables and any variables significantly associated with decreased
-
(2014)
-
3
T-score in the univariate analyses into each model. Using 2 log likelihood, we tested the goodness-of-fit of each logistic regression model. Each variable was entered into each model at P < 0.05, and was removed from each model at P O 0.10. This study analyzed the data using the Statistical Package for the Social Sciences (SPSS, version 13.0, Chicago, Illinois, USA). Differences between the polio and control groups with and without a decrease in T-score were considered significant if P values were !0.05.
Results A total of 44 subjects with polio and 44 able-bodied controls completed this study. The mean age of the male polio participants (6SD, range) was 46.6 (63.4, 41e57) years, which was non-significantly higher than 44.3 (61.8, 42e48) years for the female polio participants (t 5 1.890, P 5 0.07). The mean age of the male control participants was 47.7 (64.1, 40e58) years, which was significantly higher than 45.0 (62.3, 41e49) years for the female control participants (t 5 2.041, P 5 0.048). Table 1 shows comparisons of the clinical characteristics of the polio and control groups. All participants with polio
Table 1 Comparisons of clinical characteristics between subjects with poliomyelitis and able-bodied controls Poliomyelitis Control Variable, unit (n 5 44) (n 5 44) Age, years: mean 6 SD Body weight, kg: mean 6 SD Body height, cm: mean 6 SD Body mass index, kg/m2: mean 6 SD Total bone mass, kg: mean 6 SD Total lean tissue mass, kg: mean 6 SD Total body fat mass, kg: mean 6 SD Leg lean tissue mass, kg: mean 6 SD Leg body fat mass, kg: mean 6 SD Leg area, cm2: mean 6 SD Cumulative lifetime cigarette use, 1000-pack years: mean 6 SD Gender, male: n (%) Physical inactivity, yes: n (%) Systolic hypertension, yes: n (%) Diastolic hypertension, yes: n (%) Alcohol consumption, yes: n (%) Anemia, yes: n (%) Hypercholesterolemia, yes: n (%) Hypertriglycemia, yes: n (%) Low HDL level, yes: n (%) High LDL level, yes: n (%) Hyperglycemia, yes: n (%) Liver disease, yes: n (%) Renal disease, yes: n (%)
46.1 6 3.3 59.3 6 8.8* 160.5 6 7.3 23.0 6 3.3 2.4 6 0.3y 33.2 6 6.4z 23.2 6 6.5z 7.2 6 2.6z 6.8 6 1.9z 681.9 6 118.6z 1.4 6 3.1*
47.0 6 4.0 63.4 6 8.6 163.7 6 7.8 23.7 6 2.7 2.6 6 0.3 44.3 6 7.7 16.1 6 5.2 14.5 6 2.5 4.6 6 1.5 844.6 6 79.4 5.2 6 8.9
35 33 9 17 26 3 21 9 8 26 9 8 0
33 28 14 19 35 4 16 7 8 14 10 8 3
(79.5) (75.0) (20.5) (38.6) (60.5) (6.8) (47.7) (20.5) (19.0) (61.9)y (20.5) (18.2) (0.0)
(75.0) (63.6) (31.8) (43.2) (79.5) (9.1) (36.4) (15.9) (18.6) (32.6) (22.7) (18.2) (6.8)
HDL, high-density lipoprotein; LDL, low-density lipoprotein; SD, standard deviation. *P ! 0.05, yP ! 0.01, zP ! 0.001 (vs the control, by independent t-test for continuous variables; Pearson chi-squared test or Fisher’s exact test for categorical variables).
ARTICLE IN PRESS 4
K.-H. Chang et al. / Disability and Health Journal
Table 2 Comparison of femoral neck bone mineral density T-score distribution between subjects with poliomyelitis (n 5 44, 35 men) and able-bodied controls (n 5 44, 33 men) Differences between T-score classificationa Poliomyelitis Control groupsb Male: n (%) Normal Osteopenia Osteoporosis Female: n (%) Normal Osteopenia Osteoporosis Differences between gendersb
25.2y 2 (5.7) 12 (34.3) 21 (60.0)
14 (42.4) 18 (54.6) 1 (3.0)
5 (55.6) 4 (44.4) 0 (0) 10.3*
10 (90.9) 1 (9.1) 0 (0) 0.3
3.3
*P ! 0.005, yP ! 0.001 (Pearson chi-squared test or Fisher’s exact test). a Femoral neck bone mineral density T-score classification of polio subjects’ shorter legs and the controls’ right legs. Based on WHO criteria,3 Normal: T-score O 1.0; Osteopenia: 1.0 > T-score O 2.5; Osteoporosis: T-score < 2.5. b Pearson chi-squared value.
had a leg-length discrepancy and had no problem with their arms in the study. Two subjects with polio needed a wheelchair for locomotion (decreased T-score: two subjects) and the other 42 subjects could walk with a walking aid (decreased T-score: 35 subjects). The frequency difference between polio subjects needing different aids for locomotion was non-significant (X2 5 0.396, P 5 0.53). Those subjects who were physically active participated in recreational exercises, such as taking a walk (nine polio and one control subjects), swimming (one polio and eight control subjects), jogging (three control subjects), matted exercise (one polio subject), and ball sports (three control subjects). Table 2 shows the T-score distributions of polio subjects and the controls. Compared to subjects having an FNBMD T-score of O1 (n 5 31), subjects having an FNBMD T-score of
-
(2014)
-
<1 (n 5 57) had lower leg LTM (mean 6 SD, 9.9 6 4.4 kg vs 12.6 6 4.0 kg, P ! 0.01), had higher frequencies of experiencing polio (64.9% vs 22.6%, P ! 0.001) and hypertriglycemia (24.6% vs 6.5%, P ! 0.05), were male (91.2% vs 51.6%, P ! 0.001) and physically inactive (78.9% vs 51.6%, P ! 0.01). According to logistic regression analyses, men, LTM, and physical inactivity were important factors associated with decreased T-score (Table 3). The interaction of LTM by sex was insignificant (total LTM, b 5 0.07, P 5 0.66; leg LTM, b 5 0.02, P 5 0.94).
Discussion Based on the FNBMD T-score values measured with DXA, middle-aged subjects who previously had polio experienced a higher rate of osteopenia or osteoporosis, which was associated with the changed body composition. More than half of men with polio met the WHO criteria for osteoporosis. Reduced total LTM and men seemed to be an important factor related to a decrease in T-score among polio subjects. Compared to the prevalence of osteoporosis among Chinese people aged 40e60 years,32 polio participants had a 9e22 times higher rate of osteoporosis in Taiwan. Hsu et al showed that healthy Chinese people with higher BFM% experience an increased risk of low FNBMD.28 A high BFM% could result from either increased BFM, decreased LTM, or both. In this study, total BFM and leg BFM did not significantly differ between participants with and without decreased T-score. Instead, the significant decline in T-score among middle-aged polio subjects might be secondary to the reduction of total LTM (Table 3). The influence of LTM on BMD may depend on muscle mass. Not all of the LTM is muscle mass. Including the visceral mass in the trunk of the body, the total LTM has a higher
Table 3 Logistic regression analysis of potential predictive factors for a decrease in femoral neck bone mineral density with T-scores of <1a Predictors bb (standard error) Adjusted odds ratio (95% confidence interval) Goodness-of-fit statisticsc For polio group (n 5 44) Gender, male Total lean tissue mass, kg For control group (n 5 44) Gender, male For both groups (n 5 88) Gender, male Physical inactivity, yes Leg lean tissue mass, kg
6.85 (2.58)y 0.30 (0.15)*
21.9 (42%) 947.16 (6.02e148,926.16) 0.74 (0.56e0.99) 44.8 (24%)
2.75 (1.25)* 4.70 (1.11)z 1.83 (0.68)y 0.39 (0.11)z
15.61 (1.35e181.01) 48.8 (43%) 110.30 (12.56e968.39) 6.22 (1.64e23.54) 0.68 (0.55e0.84)
*P ! 0.05, yP ! 0.01, zP ! 0.001. a Independent variables included body compositional variables and those significant variables in the univariate analyses. Body compositional variables included total and leg body fat mass (kg) as well as total and leg lean tissue mass (kg). Those significant variables included gender (male 5 1, female 5 0) for polio group; gender, physical inactivity (yes 5 1, no 5 0), high low-density lipoprotein (yes 5 1, no 5 0) and hypertriglycemia (yes 5 1, no 5 0) for control group; and poliomyelitis (yes 5 1, no 5 0), gender, physical inactivity, and hypertriglycemia for the combination of both groups. b Standardized regression coefficient. c Using 2 log likelihoodintercept-and-covariates value (change %) for goodness-of-fit statistics. The 2 log likelihoodintercept-and-covariates change % 5 [(2 log likelihoodintercept-only) (2 log likelihoodintercept-and-covariates)] O (2 log likelihoodintercept-only).
ARTICLE IN PRESS K.-H. Chang et al. / Disability and Health Journal
proportion of water and soft tissue other than muscle mass. Thus, the total LTM has a potential bias of overestimating muscular mass and was not associated with decreased T-score for the group combining both polio and control subjects. Male gender was also a factor associated with decreased T-score (Table 3). Middle-aged men with polio were more likely to have osteoporosis than premenopausal women with polio (Table 2). The sex difference in decreased T-score among polio subjects was in accord with Haziza et al’s study.33 Coinciding with healthy people aged 40e50 years in Taiwan,6 the sex difference in FNBMD of the controls was non-significant. These findings suggest that men with polio might suffer early bone loss. Although 60.0% of men with polio met the WHO criteria for osteoporosis in this study, none of the premenopausal women with polio had osteoporosis. The antiresorptive effect of estrogen on bony metabolism30 might protect women with polio from premature osteoporosis. Possibly because of the narrow age range (40e58 y) of the participants, age is not a predictor for decreased T-score in this study. This study has four limitations. First, compared to the controls, polio participants had a lower body weight, which is a risk factor for osteoporosis. Thus, the contribution of polio to the decrease in T-score might have been overestimated. However, the mean body weight (SD) of 59.3 (8.8) kg among polio subjects exceeded the recommended cutoff value of low body weight of 57.6 kg for osteoporosis.22 Besides, the influence of body weight on the T-score of polio subjects might be not as significant as that for able-bodied persons because of changes in body composition.20 Those subjects having a diet to keep bones health were not recognized in this study either. Second, polio subjects had a lower leg area than controls (Table 1). Lowerextremity hypoplasia is a common feature of polio subjects, and is often accompanied by small bones.34 The T-score value may be underestimated with small bones.35 Again, the decrease in T-score in the polio group might have been overestimated. Third, a few female participants completed the study. Considering the negative log likelihood change % of 42% (Table 3), the model’s goodness-of-fit for polio group could be acceptable. Based on the limited data of this study, osteoporosis might not be a common morbidity of premenopausal women with polio. Fourth, those subjects who were less likely to participate in social activities (selection bias) and those who had had a history of an osteoporotic leg fracture (survival bias) might have less chance to be recruited in this cross-sectional study. The selection bias might not be great because many polio survivors had a job and lived an active life in Taiwan.9
Conclusion Osteoporosis or osteopenia is a common medical problem among middle-aged subjects who previously had polio.
-
(2014)
-
5
Within this group, men are more vulnerable to early osteoporosis than premenopausal women. The significant decrease in T-score among polio subjects might be related to the reduction of total LTM. Further study to discover a clinical tool to assess LTM is needed.
Acknowledgments We thank Chien-Hua Wu, PhD, for his review of the statistical methods we used for data analysis.
References 1. Johnell O, Kanis J. Epidemiology of osteoporotic fractures. Osteoporos Int. 2005;16:S3eS7. 2. Johnell O, Kanis JA, Oden A, et al. Predictive value of BMD for hip and other fractures. J Bone Miner Res. 2005;20:1185e1194. 3. Brown JP, Josse RG. 2002 clinical practice guidelines for the diagnosis and management of osteoporosis in Canada. CMAJ. 2002;167: S1eS34. 4. Pollitzer WS, Anderson JJB. Ethnic and genetic differences in bone mass: a review with a hereditary vs environmental perspective. Am J Clin Nutr. 1989;50:1244e1259. 5. Kanis JA, Melton LJ III, Christiansen C, et al. The diagnosis of osteoporosis. J Bone Miner Res. 1994;9:1137e1141. 6. Tsai KS, Cheng WC, Sanchez TV, et al. Bone densitometry of proximal femur in Chinese subjects: gender differences in bone mass and bone areas. Bone. 1997;20:365e369. 7. Wu Chien-Hua, Liou Tsan-Hon, Chen Hung-Hsin, et al. Stroke risk in poliomyelitis survivors: a nationwide population-based study. Arch Phys Med Rehabil. 2012;93(12):2184e2188. 8. Chang CW, Huang SF. Varied clinical patterns, physical activities, muscle enzymes, electromyographic and histologic findings in patients with post-polio syndrome in Taiwan. Spinal Cord. 2001;39: 526e531. 9. Tsai HC, Hung TH, Chen CC, et al. Prevalence and risk factors for upper extremity entrapment neuropathies in polio survivors. J Rehabil Med. 2009;41:26e31. 10. Grabljevec K, Burger H, Kersevan K, et al. Strength and endurance of knee extensors in subjects after paralytic poliomyelitis. Disabil Rehabil. 2005;27:791e799. 11. Silver JK, Aiello DD. Polio survivors: falls and subsequent injuries. Am J Phys Med Rehabil. 2002;81:567e570. 12. Cooper C. Osteoporosis: disease severity and consequent fracture management. Osteoporos Int. 2010;21:S425eS429. 13. Chang KH, Lai CH, Chen SC, et al. Femoral neck bone mineral density in ambulatory men with poliomyelitis. Osteoporos Int. 2011;22: 195e200. 14. Smeltzer SC, Zimmerman V, Capriotti T. Osteoporosis risk and low bone mineral density in women with physical disabilities. Arch Phys Med Rehabil. 2005;86:582e586. 15. Chiu GR, Araujo AB, Travison TG, et al. Relative contributions of multiple determinants to bone mineral density in men. Osteoporos Int. 2009;20:2035e2047. 16. Li S, Wagner R, Holm K, et al. Relationship between soft tissue body composition and bone mass in perimenopausal women. Maturitas. 2004;47:99e105. 17. Lee N, Radford-Smith GL, Forwood M, et al. Body composition and muscle strength as predictors of bone mineral density in Crohn’s disease. J Bone Miner Metab. 2009;27:456e463. 18. Liu HY, Wu AT, Tsai CY, et al. The balance between adipogenesis and osteogenesis in bone regeneration by platelet-rich plasma for agerelated osteoporosis. Biomaterials. 2011;32:6773e6780.
ARTICLE IN PRESS 6
K.-H. Chang et al. / Disability and Health Journal
19. Gjesdal CG, Halse JI, Eide GE, et al. Impact of lean mass and fat mass on bone mineral density: the Hordaland Health Study. Maturitas. 2008;59:191e200. 20. Chang KH, Lai CH, Chen SC, et al. Body composition assessment in Taiwanese individuals with poliomyelitis. Arch Phys Med Rehabil. 2011;92:1092e1097. 21. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet; 2004:157e163. 22. Gronholz MJ. Prevention, diagnosis, and management of osteoporosisrelated fracture: a multifactoral osteopathic approach. J Am Osteopath Assoc. 2008;108:575e585. 23. Beltran-Lagunes L, Munguıa-Lozano S, Lopez-Hernandez D. The effect on bone mineral density in patients with osteoporosis and obesity of once-weekly treatment with Risedronate/Vitamin D3 combined in a single pill for 12 months: a post-marketing study. Austin J Endocrinol Diabetes. 2014;1(4):8. 24. Pasco JA, Seeman E, Henry MJ, et al. The population burden of fractures originates in women with osteopenia, not osteoporosis. Osteoporos Int. 2006;17:1404e1409. 25. Cappuccio FP, Meilahn E, Zmuda JM, et al. High blood pressure and bone-mineral loss in elderly white women: a prospective study. Study of Osteoporotic Fractures Research Group. Lancet. 1999;354:971e975. 26. Vestergaard P. Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes e a meta-analysis. Osteoporos Int. 2007;18:427e444.
-
(2014)
-
27. Trimpou P, Oden A, Simonsson T, et al. High serum total cholesterol is a long-term cause of osteoporotic fracture. Osteoporos Int. 2011;22: 1615e1620. 28. Hsu YH, Venners SA, Terwedow HA, et al. Relation of body composition, fat mass, and serum lipids to osteoporotic fractures and bone mineral density in Chinese men and women. Am J Clin Nutr. 2006;83:146e154. 29. Cui LH, Shin MH, Chung EK, et al. Association between bone mineral densities and serum lipid profiles of pre- and postmenopausal rural women in South Korea. Osteoporos Int. 2005;16: 1975e1981. 30. Lane NE. Epidemiology, etiology, and diagnosis of osteoporosis. Am J Obstet Gynecol. 2006;194:S3eS11. 31. Law MR, Hackshaw AK. A meta-analysis of cigarette smoking, bone mineral density and risk of hip fracture: recognition of a major effect. BMJ. 1997;315(7112):841e846. 32. Wang Y, Tao Y, Hyman ME, et al. Osteoporosis in china. Osteoporos Int. 2009;20:1651e1662. 33. Haziza M, Kremer R, Benedetti A, et al. Osteoporosis in a postpolio clinic population. Arch Phys Med Rehabil. 2007;88:1030e1035. 34. Marafi FA, Al-Said Ali A, Esmail AA, et al. Baseline patterns of bone scintigraphy in patients with established post-poliomyelitis paralysis. Skeletal Radiol. 2010;39:891e895. 35. Wren T, Liu X, Pitukcheewanont P, et al. Bone densitometry in pediatric populations: discrepancies in the diagnosis of osteoporosis by DXA and CT. J Pediatr. 2005;146:776e779.
Disability and Health Journal
-
(2014)
-
www.disabilityandhealthjnl.com
Brief Report
The relationship between body composition and femoral neck osteoporosis or osteopenia in adults with previous poliomyelitis Kwang-Hwa Chang, M.D.a,b, Sung-Hui Tseng, M.D., Ph.D.c, Yu-Ching Lin, M.D., M.Sc.d, Chien-Hung Lai, M.D., Ph.D.c,e, Wen-Tien Hsiao, M.Sc.f, and Shih-Ching Chen, M.D., Ph.D.c,e,* a Department of Physical Medicine and Rehabilitation, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan Graduate Institute of Injury Prevention and Control, College of Public Health and Nutrition, Taipei Medical University, Taipei, Taiwan c Department of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei, Taiwan d Department of Physical Medicine and Rehabilitation, National Cheng Kung University Hospital, Tainan, Taiwan e Department of Physical Medicine and Rehabilitation, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan f Department of Diagnostic Radiology, Taipei Medical University Hospital, Taipei, Taiwan b
Abstract Background: Articles in the literature describing the association between body composition and osteoporosis in subjects with poliomyelitis are scarce. Objective: To assess the relationship between body composition and femoral neck osteoporosis or osteopenia in adults with previous polio. Method: After excluding postmenopausal women, 44 polio (mean age 6 standard deviation, 46.1 6 3.3 years) and 44 able-bodied control volunteers (47.0 6 4.0 years) participated in the study. Each participant’s femoral neck bone mineral density (FNBMD) and whole body composition were measured using dual-energy X-ray absorptiometry. With local reference BMD values of normal young adults installed in the instrument, we obtained T-score values that depended on each FNBMD value. A T-score value of <1.0 indicated decreased T-score, including osteoporosis (T-score < 2.5) and osteopenia (1.0 to 2.5). This study conducted logistic regression analyses to find factors associated with osteoporosis and osteopenia. Results: Based on the FNBMD T-score values, 60.0% of middle-aged men with polio had osteoporosis. In adjusted logistic regression analyses, total lean tissue mass (Adjusted odds ratio [95% confidence interval], 0.74 [0.56e0.99], P ! 0.05) and male gender (947.16 [6.02e148,926.16], P ! 0.01) were important factors associated with decreased T-score in polio group. Conclusions: Osteoporosis or osteopenia is a common medical problem for middle-aged men with polio. Reduced total lean tissue mass seems to be one of the important factors associated with osteoporosis or osteopenia among subjects with polio. Further research for a clinical tool to assess lean tissue mass for subjects with polio is needed. Ó 2014 Elsevier Inc. All rights reserved. Keywords: Osteoporosis; Poliomyelitis; Femoral neck; Body composition
Osteoporosis is an important global medical issue, and bony fracture is one of its most serious complications. Among the various types of bony fractures, hip fractures often have compromised outcomes. A recent study shows
that approximately 1 year following a hip fracture, 20e 30% of the individuals affected had died,1 40% were dependent in walking, and 60% had some difficulty with the activities of daily living. Thus, estimating the rate of
Abbreviations: BFM, body fat mass; BMD, bone mineral density; DXA, dual-energy X-ray absorptiometry; FNBMD, femoral neck bone mineral density; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LTM, lean tissue mass. Disclosure of financial interests: The authors report nothing to be disclosed. This disclosure includes direct or indirect financial or personal relationships, interests, and affiliations relevant to the subject matter of the manuscript that have occurred over the last two years, or that are expected in the foreseeable future. This disclosure includes, but is not limited to, grants or funding, employment, affiliations, patents (in preparation, filed,
or granted), inventions, honoraria, consultancies, royalties, stock options/ ownership, or expert testimony. Conflicts of interest: The authors report no conflicts of interest. The authors disclose that no prior presentation of abstract at meeting or posting any part or all of the content on a website, even in draft form. * Corresponding author. Taipei Medical University Hospital, 252 Wu-Hsing Street, Taipei 11031, Taiwan. Tel.: þ886 2 27372181x1236; fax: þ886 2 55589880. E-mail address: [email protected] or [email protected] (S.-C. Chen).
1936-6574/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.dhjo.2014.09.011
ARTICLE IN PRESS 2
K.-H. Chang et al. / Disability and Health Journal
experiencing osteoporosis of the femoral neck and identifying the risk factors of osteoporosis are essential tasks in caring for those people at risk. Bone mineral density (BMD) is a major predictor for the risk of osteoporotic fracture.2 However, this continuous measurement is subject to ethnic differences.3 BMD values of Caucasian people are often higher than those of Chinese.4 A BMD cutoff value to differentiate a high risk of developing fractures from low risk is lacking. For clinical practice and the assessment of fracture risk,5 BMD values are typically expressed as a T-score or the standard deviation (SD) from the mean BMD value of normal young adults aged 20e40 years of the same sex and race.6 According to the WHO criteria,5 a BMD T-score of 1.0 to 2.5 is classified as being osteopenic, and a T-score of <2.5 is osteoporotic. Combined with age and sex, T-scores offer a useful way to estimate the long-term risk of osteoporotic fracture.3 Polio was an epidemic in children aged <3 years in Taiwan from 1950 to 1970.7,8 Many of those survivors lived an active adult life of mild to moderate degree of physical activity.9 Because subjects with polio often have physical disabilities and low endurance of knee extensor muscles,10 they are susceptible to bony fractures caused by falls. A questionnaire survey study showed that two-thirds of polio subjects had experienced a fall from a standing height in the previous 1-year period.11 Bony fractures were the most common injury. The literature also shows that fractures with low-impact trauma are usually osteoporosis-related.3,12 Compared to able-bodied controls, men with polio have 13e23% lower values for femoral neck BMD (FNBMD).13 Smeltzer et al showed that O40% of adult women with previous polio had osteoporosis at the os calcis.14 However, the rate of osteoporosis at the femoral neck in polio subjects of different sexes remains unknown. In addition, BMD is related to body composition, including body fat mass (BFM), lean tissue mass (LTM), and bone mass. Recent articles show a positive association between LTM and BMD.15e17 Body fat mass is also related to BMD,18 and this is especially true for women.19 Compared to able-bodied controls, polio subjects have a decreased LTM and increased BFM for the whole body and the leg region.20 Thus, changes in body composition could influence the occurrence of osteoporosis among subjects with prior polio. To our knowledge, previous literature describing the association between body composition and osteoporosis in subjects with polio is scarce. The aim of this study was to assess the relationship between body composition and the presence of osteopenia or osteoporosis among subjects with previous polio.
Methods Participants This was a cross-sectional study with subjects who experienced acute paralytic polio at the age of 3 years or
-
(2014)
-
younger (polio group). All subjects experienced flaccid paralysis of one or both lower extremities. Having a leglength discrepancy, each subject with polio often had a shorter leg, which was shorter than his/her other leg (the longer leg). This study also recruited able-bodied persons without a history of any neuromuscular disease as controls (control group). Both groups’ participants had a body mass index range of 16e30 kg/m2.21 To focus on the association between polio and early bone loss, postmenopausal women were excluded from the study. Considering that rapid bone loss often begins after the age of 40 years12 and age of O65 years is an important risk factor for osteoporosis in the general population,22 this study recruited people of both groups between the ages of 40 and 65 years. This study also excluded those who had a body weight of !40 kg,23 those who had a metal implant or a history of a leg fracture, those who had used glucocorticoids or other bone metabolismrelated medications in the most recent 3 months, and those with a history of other neuromuscular diseases. This study used advertisements to recruit participants from the community of urban areas in northern Taiwan. A total of 51 subjects with polio and 44 able-bodied controls were recruited and assessed by an experienced physiatrist. Of the subjects with polio, three had a body weight of !40 kg, three had a history of exclusion criteria (one oophorectomy, one hip fracture, and one stroke), and one failed in measuring FNBMD. Thus, 44 subjects with polio (response rate: 86%) and 44 able-bodied controls (response rate: 100%) completed this study. The Institutional Review Board of our university-affiliated hospital approved this study. Each participant signed an informed consent form before entering this study. Measurements This study used dual-energy X-ray absorptiometry (DXA; Norland XR-36, version 2.5.3 software, Norland Corp., Fort Atkinson, Wisconsin, USA) to measure the FNBMD and whole-body composition of all participants at a university-affiliated hospital. This study measured the shorter leg’s FNBMD because, on average, the shorter leg has lower FNBMD in patients with polio.13 This study arbitrarily chose the control participants’ right leg as the reference leg for measuring FNBMD. A densitometry technologist certified by the International Society for Clinical Densitometry performed these measurements. The long-term coefficient of variation of the instrument was 1.2% for FNBMD. With local reference BMD values of normal young adults installed in the instrument,6 this study obtained T-score values that depended on each FNBMD value. Because both osteopenia and osteoporosis are risk factors for low-trauma fractures,24 legs with a T-score of <1.0 were identified as having ‘‘decreased T-score.’’ This study obtained weights of the body compositional variables of the whole body and both legs from a whole-body DXA scan, and identified the weights of the LTM and BFM distal
ARTICLE IN PRESS K.-H. Chang et al. / Disability and Health Journal
to the femoral trochanters of both legs as the leg LTM and BFM, respectively. The demarcation between the legs and trunk was based on the Norland user’s instructions. According to Norland Corp., the reported error for total BFM estimates using this device is !2%. Using the methods by Chang et al,20 this study measured the body weight, body height (from the vertex to the heel of the longer leg), and blood pressure of all participants, and took an overnight fasting blood sample from each participant. This study measured the biochemical profile of each blood sample, including fasting blood glucose, serum alanine transaminase, blood urea nitrogen, total cholesterol, low- (LDL) and high-density lipoprotein (HDL), triglyceride, and hemoglobin. By comparing these data to reference ranges, this study obtained the frequencies of participants with potential risk factors for osteoporosis,22 including systolic hypertension (systolic blood pressure O130 mm Hg),25 diastolic hypertension (diastolic blood pressure O85 mm Hg),25 hyperglycemia (fasting blood glucose O100 mg/dL),26 hypercholesterolemia (total serum cholesterol O200 mg/dL),27 hypertriglycemia (serum triglyceride O150 mg/dL),28,29 high LDL (serum LDL O130 mg/ dL),28 low HDL (serum HDL !40 mg/dL),29 liver disease (serum alanine transaminase O40 IU/L),22 renal disease (blood urea nitrogen O18 mg/dL),22 and anemia (hemoglobin !12 g/dL).30 This study used a quasi-structured questionnaire to assess the participants’ lifestyles, including daily cigarette smoking, daily alcohol consumption, and daily amount of exercise. This questionnaire comprised three open- and eight close-ended questions. Because cigarette smoking increases the risk of hip fracture,31 this study assessed each participant’s long-term amount of smoking and calculated the cumulative lifetime cigarette use by multiplying the average number of cigarettes smoked daily by the lifetime duration of smoking. Those who exercised less than 90 min on a weekly basis in the most recent 3 months were identified as physically inactive participants, whereas the others were identified as physically active. Data analysis This study compared the differences in clinical characteristics between the polio and control groups using a Pearson chi-squared test or Fisher’s exact test, and compared the age, body weight, body height, body mass index, total LTM, total BFM, leg LTM, leg BFM, and cumulative lifetime cigarette use between groups using two-tailed Student’s t tests. This study also compared the FNBMD T-score distributions of polio subjects with the controls by using Pearson’s chi-squared test. This study conducted logistic regression analyses with a forward stepwise method to find predictors with decreased T-score (T-score < 1.0). A stratified analysis based on the polio status was conducted. This study entered body compositional variables and any variables significantly associated with decreased
-
(2014)
-
3
T-score in the univariate analyses into each model. Using 2 log likelihood, we tested the goodness-of-fit of each logistic regression model. Each variable was entered into each model at P < 0.05, and was removed from each model at P O 0.10. This study analyzed the data using the Statistical Package for the Social Sciences (SPSS, version 13.0, Chicago, Illinois, USA). Differences between the polio and control groups with and without a decrease in T-score were considered significant if P values were !0.05.
Results A total of 44 subjects with polio and 44 able-bodied controls completed this study. The mean age of the male polio participants (6SD, range) was 46.6 (63.4, 41e57) years, which was non-significantly higher than 44.3 (61.8, 42e48) years for the female polio participants (t 5 1.890, P 5 0.07). The mean age of the male control participants was 47.7 (64.1, 40e58) years, which was significantly higher than 45.0 (62.3, 41e49) years for the female control participants (t 5 2.041, P 5 0.048). Table 1 shows comparisons of the clinical characteristics of the polio and control groups. All participants with polio
Table 1 Comparisons of clinical characteristics between subjects with poliomyelitis and able-bodied controls Poliomyelitis Control Variable, unit (n 5 44) (n 5 44) Age, years: mean 6 SD Body weight, kg: mean 6 SD Body height, cm: mean 6 SD Body mass index, kg/m2: mean 6 SD Total bone mass, kg: mean 6 SD Total lean tissue mass, kg: mean 6 SD Total body fat mass, kg: mean 6 SD Leg lean tissue mass, kg: mean 6 SD Leg body fat mass, kg: mean 6 SD Leg area, cm2: mean 6 SD Cumulative lifetime cigarette use, 1000-pack years: mean 6 SD Gender, male: n (%) Physical inactivity, yes: n (%) Systolic hypertension, yes: n (%) Diastolic hypertension, yes: n (%) Alcohol consumption, yes: n (%) Anemia, yes: n (%) Hypercholesterolemia, yes: n (%) Hypertriglycemia, yes: n (%) Low HDL level, yes: n (%) High LDL level, yes: n (%) Hyperglycemia, yes: n (%) Liver disease, yes: n (%) Renal disease, yes: n (%)
46.1 6 3.3 59.3 6 8.8* 160.5 6 7.3 23.0 6 3.3 2.4 6 0.3y 33.2 6 6.4z 23.2 6 6.5z 7.2 6 2.6z 6.8 6 1.9z 681.9 6 118.6z 1.4 6 3.1*
47.0 6 4.0 63.4 6 8.6 163.7 6 7.8 23.7 6 2.7 2.6 6 0.3 44.3 6 7.7 16.1 6 5.2 14.5 6 2.5 4.6 6 1.5 844.6 6 79.4 5.2 6 8.9
35 33 9 17 26 3 21 9 8 26 9 8 0
33 28 14 19 35 4 16 7 8 14 10 8 3
(79.5) (75.0) (20.5) (38.6) (60.5) (6.8) (47.7) (20.5) (19.0) (61.9)y (20.5) (18.2) (0.0)
(75.0) (63.6) (31.8) (43.2) (79.5) (9.1) (36.4) (15.9) (18.6) (32.6) (22.7) (18.2) (6.8)
HDL, high-density lipoprotein; LDL, low-density lipoprotein; SD, standard deviation. *P ! 0.05, yP ! 0.01, zP ! 0.001 (vs the control, by independent t-test for continuous variables; Pearson chi-squared test or Fisher’s exact test for categorical variables).
ARTICLE IN PRESS 4
K.-H. Chang et al. / Disability and Health Journal
Table 2 Comparison of femoral neck bone mineral density T-score distribution between subjects with poliomyelitis (n 5 44, 35 men) and able-bodied controls (n 5 44, 33 men) Differences between T-score classificationa Poliomyelitis Control groupsb Male: n (%) Normal Osteopenia Osteoporosis Female: n (%) Normal Osteopenia Osteoporosis Differences between gendersb
25.2y 2 (5.7) 12 (34.3) 21 (60.0)
14 (42.4) 18 (54.6) 1 (3.0)
5 (55.6) 4 (44.4) 0 (0) 10.3*
10 (90.9) 1 (9.1) 0 (0) 0.3
3.3
*P ! 0.005, yP ! 0.001 (Pearson chi-squared test or Fisher’s exact test). a Femoral neck bone mineral density T-score classification of polio subjects’ shorter legs and the controls’ right legs. Based on WHO criteria,3 Normal: T-score O 1.0; Osteopenia: 1.0 > T-score O 2.5; Osteoporosis: T-score < 2.5. b Pearson chi-squared value.
had a leg-length discrepancy and had no problem with their arms in the study. Two subjects with polio needed a wheelchair for locomotion (decreased T-score: two subjects) and the other 42 subjects could walk with a walking aid (decreased T-score: 35 subjects). The frequency difference between polio subjects needing different aids for locomotion was non-significant (X2 5 0.396, P 5 0.53). Those subjects who were physically active participated in recreational exercises, such as taking a walk (nine polio and one control subjects), swimming (one polio and eight control subjects), jogging (three control subjects), matted exercise (one polio subject), and ball sports (three control subjects). Table 2 shows the T-score distributions of polio subjects and the controls. Compared to subjects having an FNBMD T-score of O1 (n 5 31), subjects having an FNBMD T-score of
-
(2014)
-
<1 (n 5 57) had lower leg LTM (mean 6 SD, 9.9 6 4.4 kg vs 12.6 6 4.0 kg, P ! 0.01), had higher frequencies of experiencing polio (64.9% vs 22.6%, P ! 0.001) and hypertriglycemia (24.6% vs 6.5%, P ! 0.05), were male (91.2% vs 51.6%, P ! 0.001) and physically inactive (78.9% vs 51.6%, P ! 0.01). According to logistic regression analyses, men, LTM, and physical inactivity were important factors associated with decreased T-score (Table 3). The interaction of LTM by sex was insignificant (total LTM, b 5 0.07, P 5 0.66; leg LTM, b 5 0.02, P 5 0.94).
Discussion Based on the FNBMD T-score values measured with DXA, middle-aged subjects who previously had polio experienced a higher rate of osteopenia or osteoporosis, which was associated with the changed body composition. More than half of men with polio met the WHO criteria for osteoporosis. Reduced total LTM and men seemed to be an important factor related to a decrease in T-score among polio subjects. Compared to the prevalence of osteoporosis among Chinese people aged 40e60 years,32 polio participants had a 9e22 times higher rate of osteoporosis in Taiwan. Hsu et al showed that healthy Chinese people with higher BFM% experience an increased risk of low FNBMD.28 A high BFM% could result from either increased BFM, decreased LTM, or both. In this study, total BFM and leg BFM did not significantly differ between participants with and without decreased T-score. Instead, the significant decline in T-score among middle-aged polio subjects might be secondary to the reduction of total LTM (Table 3). The influence of LTM on BMD may depend on muscle mass. Not all of the LTM is muscle mass. Including the visceral mass in the trunk of the body, the total LTM has a higher
Table 3 Logistic regression analysis of potential predictive factors for a decrease in femoral neck bone mineral density with T-scores of <1a Predictors bb (standard error) Adjusted odds ratio (95% confidence interval) Goodness-of-fit statisticsc For polio group (n 5 44) Gender, male Total lean tissue mass, kg For control group (n 5 44) Gender, male For both groups (n 5 88) Gender, male Physical inactivity, yes Leg lean tissue mass, kg
6.85 (2.58)y 0.30 (0.15)*
21.9 (42%) 947.16 (6.02e148,926.16) 0.74 (0.56e0.99) 44.8 (24%)
2.75 (1.25)* 4.70 (1.11)z 1.83 (0.68)y 0.39 (0.11)z
15.61 (1.35e181.01) 48.8 (43%) 110.30 (12.56e968.39) 6.22 (1.64e23.54) 0.68 (0.55e0.84)
*P ! 0.05, yP ! 0.01, zP ! 0.001. a Independent variables included body compositional variables and those significant variables in the univariate analyses. Body compositional variables included total and leg body fat mass (kg) as well as total and leg lean tissue mass (kg). Those significant variables included gender (male 5 1, female 5 0) for polio group; gender, physical inactivity (yes 5 1, no 5 0), high low-density lipoprotein (yes 5 1, no 5 0) and hypertriglycemia (yes 5 1, no 5 0) for control group; and poliomyelitis (yes 5 1, no 5 0), gender, physical inactivity, and hypertriglycemia for the combination of both groups. b Standardized regression coefficient. c Using 2 log likelihoodintercept-and-covariates value (change %) for goodness-of-fit statistics. The 2 log likelihoodintercept-and-covariates change % 5 [(2 log likelihoodintercept-only) (2 log likelihoodintercept-and-covariates)] O (2 log likelihoodintercept-only).
ARTICLE IN PRESS K.-H. Chang et al. / Disability and Health Journal
proportion of water and soft tissue other than muscle mass. Thus, the total LTM has a potential bias of overestimating muscular mass and was not associated with decreased T-score for the group combining both polio and control subjects. Male gender was also a factor associated with decreased T-score (Table 3). Middle-aged men with polio were more likely to have osteoporosis than premenopausal women with polio (Table 2). The sex difference in decreased T-score among polio subjects was in accord with Haziza et al’s study.33 Coinciding with healthy people aged 40e50 years in Taiwan,6 the sex difference in FNBMD of the controls was non-significant. These findings suggest that men with polio might suffer early bone loss. Although 60.0% of men with polio met the WHO criteria for osteoporosis in this study, none of the premenopausal women with polio had osteoporosis. The antiresorptive effect of estrogen on bony metabolism30 might protect women with polio from premature osteoporosis. Possibly because of the narrow age range (40e58 y) of the participants, age is not a predictor for decreased T-score in this study. This study has four limitations. First, compared to the controls, polio participants had a lower body weight, which is a risk factor for osteoporosis. Thus, the contribution of polio to the decrease in T-score might have been overestimated. However, the mean body weight (SD) of 59.3 (8.8) kg among polio subjects exceeded the recommended cutoff value of low body weight of 57.6 kg for osteoporosis.22 Besides, the influence of body weight on the T-score of polio subjects might be not as significant as that for able-bodied persons because of changes in body composition.20 Those subjects having a diet to keep bones health were not recognized in this study either. Second, polio subjects had a lower leg area than controls (Table 1). Lowerextremity hypoplasia is a common feature of polio subjects, and is often accompanied by small bones.34 The T-score value may be underestimated with small bones.35 Again, the decrease in T-score in the polio group might have been overestimated. Third, a few female participants completed the study. Considering the negative log likelihood change % of 42% (Table 3), the model’s goodness-of-fit for polio group could be acceptable. Based on the limited data of this study, osteoporosis might not be a common morbidity of premenopausal women with polio. Fourth, those subjects who were less likely to participate in social activities (selection bias) and those who had had a history of an osteoporotic leg fracture (survival bias) might have less chance to be recruited in this cross-sectional study. The selection bias might not be great because many polio survivors had a job and lived an active life in Taiwan.9
Conclusion Osteoporosis or osteopenia is a common medical problem among middle-aged subjects who previously had polio.
-
(2014)
-
5
Within this group, men are more vulnerable to early osteoporosis than premenopausal women. The significant decrease in T-score among polio subjects might be related to the reduction of total LTM. Further study to discover a clinical tool to assess LTM is needed.
Acknowledgments We thank Chien-Hua Wu, PhD, for his review of the statistical methods we used for data analysis.
References 1. Johnell O, Kanis J. Epidemiology of osteoporotic fractures. Osteoporos Int. 2005;16:S3eS7. 2. Johnell O, Kanis JA, Oden A, et al. Predictive value of BMD for hip and other fractures. J Bone Miner Res. 2005;20:1185e1194. 3. Brown JP, Josse RG. 2002 clinical practice guidelines for the diagnosis and management of osteoporosis in Canada. CMAJ. 2002;167: S1eS34. 4. Pollitzer WS, Anderson JJB. Ethnic and genetic differences in bone mass: a review with a hereditary vs environmental perspective. Am J Clin Nutr. 1989;50:1244e1259. 5. Kanis JA, Melton LJ III, Christiansen C, et al. The diagnosis of osteoporosis. J Bone Miner Res. 1994;9:1137e1141. 6. Tsai KS, Cheng WC, Sanchez TV, et al. Bone densitometry of proximal femur in Chinese subjects: gender differences in bone mass and bone areas. Bone. 1997;20:365e369. 7. Wu Chien-Hua, Liou Tsan-Hon, Chen Hung-Hsin, et al. Stroke risk in poliomyelitis survivors: a nationwide population-based study. Arch Phys Med Rehabil. 2012;93(12):2184e2188. 8. Chang CW, Huang SF. Varied clinical patterns, physical activities, muscle enzymes, electromyographic and histologic findings in patients with post-polio syndrome in Taiwan. Spinal Cord. 2001;39: 526e531. 9. Tsai HC, Hung TH, Chen CC, et al. Prevalence and risk factors for upper extremity entrapment neuropathies in polio survivors. J Rehabil Med. 2009;41:26e31. 10. Grabljevec K, Burger H, Kersevan K, et al. Strength and endurance of knee extensors in subjects after paralytic poliomyelitis. Disabil Rehabil. 2005;27:791e799. 11. Silver JK, Aiello DD. Polio survivors: falls and subsequent injuries. Am J Phys Med Rehabil. 2002;81:567e570. 12. Cooper C. Osteoporosis: disease severity and consequent fracture management. Osteoporos Int. 2010;21:S425eS429. 13. Chang KH, Lai CH, Chen SC, et al. Femoral neck bone mineral density in ambulatory men with poliomyelitis. Osteoporos Int. 2011;22: 195e200. 14. Smeltzer SC, Zimmerman V, Capriotti T. Osteoporosis risk and low bone mineral density in women with physical disabilities. Arch Phys Med Rehabil. 2005;86:582e586. 15. Chiu GR, Araujo AB, Travison TG, et al. Relative contributions of multiple determinants to bone mineral density in men. Osteoporos Int. 2009;20:2035e2047. 16. Li S, Wagner R, Holm K, et al. Relationship between soft tissue body composition and bone mass in perimenopausal women. Maturitas. 2004;47:99e105. 17. Lee N, Radford-Smith GL, Forwood M, et al. Body composition and muscle strength as predictors of bone mineral density in Crohn’s disease. J Bone Miner Metab. 2009;27:456e463. 18. Liu HY, Wu AT, Tsai CY, et al. The balance between adipogenesis and osteogenesis in bone regeneration by platelet-rich plasma for agerelated osteoporosis. Biomaterials. 2011;32:6773e6780.
ARTICLE IN PRESS 6
K.-H. Chang et al. / Disability and Health Journal
19. Gjesdal CG, Halse JI, Eide GE, et al. Impact of lean mass and fat mass on bone mineral density: the Hordaland Health Study. Maturitas. 2008;59:191e200. 20. Chang KH, Lai CH, Chen SC, et al. Body composition assessment in Taiwanese individuals with poliomyelitis. Arch Phys Med Rehabil. 2011;92:1092e1097. 21. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet; 2004:157e163. 22. Gronholz MJ. Prevention, diagnosis, and management of osteoporosisrelated fracture: a multifactoral osteopathic approach. J Am Osteopath Assoc. 2008;108:575e585. 23. Beltran-Lagunes L, Munguıa-Lozano S, Lopez-Hernandez D. The effect on bone mineral density in patients with osteoporosis and obesity of once-weekly treatment with Risedronate/Vitamin D3 combined in a single pill for 12 months: a post-marketing study. Austin J Endocrinol Diabetes. 2014;1(4):8. 24. Pasco JA, Seeman E, Henry MJ, et al. The population burden of fractures originates in women with osteopenia, not osteoporosis. Osteoporos Int. 2006;17:1404e1409. 25. Cappuccio FP, Meilahn E, Zmuda JM, et al. High blood pressure and bone-mineral loss in elderly white women: a prospective study. Study of Osteoporotic Fractures Research Group. Lancet. 1999;354:971e975. 26. Vestergaard P. Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes e a meta-analysis. Osteoporos Int. 2007;18:427e444.
-
(2014)
-
27. Trimpou P, Oden A, Simonsson T, et al. High serum total cholesterol is a long-term cause of osteoporotic fracture. Osteoporos Int. 2011;22: 1615e1620. 28. Hsu YH, Venners SA, Terwedow HA, et al. Relation of body composition, fat mass, and serum lipids to osteoporotic fractures and bone mineral density in Chinese men and women. Am J Clin Nutr. 2006;83:146e154. 29. Cui LH, Shin MH, Chung EK, et al. Association between bone mineral densities and serum lipid profiles of pre- and postmenopausal rural women in South Korea. Osteoporos Int. 2005;16: 1975e1981. 30. Lane NE. Epidemiology, etiology, and diagnosis of osteoporosis. Am J Obstet Gynecol. 2006;194:S3eS11. 31. Law MR, Hackshaw AK. A meta-analysis of cigarette smoking, bone mineral density and risk of hip fracture: recognition of a major effect. BMJ. 1997;315(7112):841e846. 32. Wang Y, Tao Y, Hyman ME, et al. Osteoporosis in china. Osteoporos Int. 2009;20:1651e1662. 33. Haziza M, Kremer R, Benedetti A, et al. Osteoporosis in a postpolio clinic population. Arch Phys Med Rehabil. 2007;88:1030e1035. 34. Marafi FA, Al-Said Ali A, Esmail AA, et al. Baseline patterns of bone scintigraphy in patients with established post-poliomyelitis paralysis. Skeletal Radiol. 2010;39:891e895. 35. Wren T, Liu X, Pitukcheewanont P, et al. Bone densitometry in pediatric populations: discrepancies in the diagnosis of osteoporosis by DXA and CT. J Pediatr. 2005;146:776e779.