Portal de Periódicos da CAPES

Does Subjective Sleep Affect Bone Mineral Density in Older People with Minimal Health Disorders? The PROOF Cohort

2016; American Academy of Sleep Medicine; Volume: 12; Issue: 11 Linguagem: Inglês

10.5664/jcsm.6266

1550-9397

Magali Saint Martin, Pierre Labeix, Martin Garet, Thierry Thomas, Jean Barthélemy, Philippe Collet, Fréderic Roche, Émilia Sforza,

Nutritional Studies and Diet

Free AccessSleep QualityDoes Subjective Sleep Affect Bone Mineral Density in Older People with Minimal Health Disorders? The PROOF Cohort Magali Saint Martin, PhD, Pierre Labeix, MSc, Martin Garet, PhD, Thierry Thomas, MD, Jean-Claude Barthélémy, MD, PhD, Philippe Collet, MD, Frédéric Roche, MD, PhD, Emilia Sforza, MD, PhD Magali Saint Martin, PhD Service de Physiologie Clinique et de l'Exercice, CHU Nord, Saint-Etienne, Faculté de Médecine Jacques Lisfranc, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France Service de Neurologie, réhabilitation cognitive, centre Hélio Marin, Hyères, France , Pierre Labeix, MSc Service de Physiologie Clinique et de l'Exercice, CHU Nord, Saint-Etienne, Faculté de Médecine Jacques Lisfranc, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France , Martin Garet, PhD Service de Physiologie Clinique et de l'Exercice, CHU Nord, Saint-Etienne, Faculté de Médecine Jacques Lisfranc, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France , Thierry Thomas, MD Service de Rhumatologie, CHU Saint-Etienne, Faculté de Médecine Jacques Lisfranc, Inserm U1059 LBTO, UJM, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France , Jean-Claude Barthélémy, MD, PhD Service de Physiologie Clinique et de l'Exercice, CHU Nord, Saint-Etienne, Faculté de Médecine Jacques Lisfranc, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France , Philippe Collet, MD Service de Rhumatologie, CHU Saint-Etienne, Faculté de Médecine Jacques Lisfranc, Inserm U1059 LBTO, UJM, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France , Frédéric Roche, MD, PhD Service de Physiologie Clinique et de l'Exercice, CHU Nord, Saint-Etienne, Faculté de Médecine Jacques Lisfranc, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France , Emilia Sforza, MD, PhD Address correspondence to: Sforza Emilia MD, PhD, Service de Physiologie Clinique et de l'Exercice, CHU Nord - Niveau 6, F - 42055 Saint-Etienne Cedex 2 France+33 4 77 82 83 00+33 4 77 82 84 47 E-mail Address: [email protected] Service de Physiologie Clinique et de l'Exercice, CHU Nord, Saint-Etienne, Faculté de Médecine Jacques Lisfranc, PRES de Lyon, Université Jean Monnet, Saint-Etienne, France Published Online:November 15, 2016https://doi.org/10.5664/jcsm.6266Cited by:17SectionsAbstractPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTStudy Objectives:Clinical and epidemiological studies suggest a relation between bone mineral density (BMD) and self-assessment of sleep with an effect on bone formation and osteoporosis (OS) risk in short and long sleepers. This study explores this association in a large sample of older subjects.Methods:We examined 500 participants without insomnia complaints aged 65.7 ± 0.8 y. Each participant had a full evaluation including anthropometric measurement, clinical examination and measurements of BMD at the lumbar spine and femoral sites by dual-energy X-ray absorptiometry. The daily energy expenditure (DEE) was measured by the Population Physical Activity Questionnaire. Sleep duration and quality were evaluated by the Pittsburgh Sleep Quality Index. The subjects were stratified into three groups according to sleep duration, i.e., short (< 6 h), normal (6–8 h), and long (≥ 8 h) sleepers.Results:Osteopenia was found in 40% of the subjects at the femoral level and 43% at the vertebral level. The prevalence of OS was lower both at femoral (8%) and vertebral (12%) levels. Short, normal, and long sleepers accounted for 29%, 40%, and 31% of subjects, respectively. After adjustments for metabolic, anthropometric, and DEE, multinomial logistic regression analysis indicated that long sleepers were more likely to have femoral neck OS with a slight effect of DEE at vertebral spine.Conclusions:In a sample of older subjects, self-reported long sleep was the best predictor of OS risk at the femoral level. This finding suggests an association between OS and self-reported sleep duration in older subjects.Clinical Trial Registration:NCT 00759304 and NCT 00766584Citation:Saint Martin M, Labeix P, Garet M, Thierry T, Barthélémy JC, Collet P, Roche F, Sforza E. Does subjective sleep affect bone mineral density in older people with minimal health disorders? The PROOF cohort. J Clin Sleep Med 2016;12(11):1461–1469.INTRODUCTIONOsteoporosis (OS) is a bone disorder characterized by reduced bone mass density and microarchitectural deterioration of bone tissue resulting in skeletal fragility and fracture susceptibility, both considered as significant risk factors for morbidity and mortality among elderly.1There is growing evidence suggesting that not only anthropometric factors, hypertension,2 diabetes,3 obesity,4 and physical activity5 but also sleep duration6,7,8 are implicated in decreased bone mineral density (BMD) in elderly. The association between sleep duration and osteoporosis has been recently studied with conflicting results,9,10 with some studies stressing the independent risk of short sleep for OS and others reporting an independent risk in long sleepers. In an epidemiological study of a population aged between 46–86 y, Tian et al.11 found that self-reported habitual sleep duration of 9 h or longer was associated with an increased prevalence of OS. However, in the NHANES12 study of 5,288 adults aged 50 y and older, a sleep duration < 6 h per night was associated with a significantly increased risk of OS, but only in the older age group (> 65 y).The discrepancies among studies stem from important limitations concerning the large range of age, the limited evaluation of sleep duration using just a single question, the incomplete assessment of BMD without strict criteria to define osteopenia (OP) and OS, and the lack of data concerning an individual's lifestyle and physical activity (Table 1).BRIEF SUMMARYCurrent Knowledge/Study Rationale: Osteoporosis is a risk factor for fracture in the elderly. The role of short or long sleep time may affect the risk of osteoporosis.Study Impact: In older people, long sleep duration might be a major factor of osteoporosis risk. Clinicians should consider the role of sleep on bone metabolism in order to educate patients to have not only an increased physical activity but also reduced sleep duration.Table 1 Reported literature studies on sleep duration and bone mineral density (BMD) or osteoporosis.Table 1 Reported literature studies on sleep duration and bone mineral density (BMD) or osteoporosis.The Prognostic Indicator of Cardiovascular and Cerebrovascular Events (PROOF) study13 provides an opportunity to examine the association between OS and self-reported sleep quality and duration in a larger sample of community-dwelling elderly with homogenous age, without insomnia complaints, and with no report of major health disease or previous diagnosis or treatment for OS. Sleep duration was assessed using a validated sleep questionnaire for self-reported sleep, i.e., the Pittsburgh Sleep Quality Index (PSQI),14 that assesses more accurately not only sleep duration but also sleep quality, which represents a somewhat novel approach to evaluate the effect of sleep on BMD. The aims of the current study were two-fold: (1) to determine whether sleep duration and sleep quality play a role on bone mineral density and on the presence of OP/OS in an older population with minimal health diseases; and (2) to assess whether in addition to physical activity sleep duration increase the OS risk in older people.METHODSPopulationThe participants consisted of a subset of individuals enrolled in a prospective study (the PROOF study) to investigate the influence of autonomic nervous system activity on cognitive functioning and cardiovascular and cerebrovascular morbidity. The details of the PROOF study have been previously described.13 Briefly, the subjects were recruited from Saint-Etienne, France, from 2001 to 2003 and were eligible if they were 65 y old at the inclusion date. The exclusion criteria were defined as follows: previous myocardial infarction, arrhythmia, cardiac pacemaker, stroke, neurological or psychiatric disease, insulin-dependent diabetes, chronic obstructive pulmonary disease, cerebral magnetic resonance results suggesting neurological diseases or initial dementia, and current residence in institutions. Of the 3,983 eligible participants, the final study included 1,011 volunteers. Two ancillary studies addressing the association between OSA and cardiovascular and cerebrovascular events (SYNAPSE study) and the change in cognitive function (SIEMPRE study) over time were proposed to the participants.Subjects under anti-OS medication (15%) and subjects without complete PSQI questionnaires were excluded from the analysis. The final sample consisted of 500 subjects, aged 65.7 ± 0.8 y (58% women).The PROOF study was approved by the University Hospital and the local Ethics Committee (CCPRB Rhone-Alpes Loire). The National Committee for Information and Liberty (CNIL) gave consent for data collection. All subjects gave their written consent prior to participation in the study.Clinical AssessmentDetailed self-reported clinical assessment was focused on cardiac and cerebrovascular disease, hypertension, diabetes, and respiratory, neurological and psychiatric disorders, alcohol consumption and smoking. Current medication was both self-reported by the subjects and confirmed by medical prescription with regard to antihypertensive, antidiabetic, antilipemic and hypnotic, anxiolytic and/or antidepressant therapy. None of the subjects had diagnosis of other sleep disorders except sleep disordered breathing, 42% of the sample having an apnea-hypopnea index > 15. For ethical reasons, subjects did not discontinue hypolipidemic, diabetic or hypertensive medication for the study.During the clinical evaluation, the height in stocking feet and weight in light clothing were measured, and the body mass index (BMI) was calculated as weight/height squared (kg/m2). Physical activity was assessed using the validated Population Physical Activity questionnaire (POPAQ).15 This questionnaire included a set of 82 questions investigating all items of physical activity over the previous week. The data collected allowed calculation of the daily energy expenditure (DEE, kJ/24 h).Measurement of Body Composition by Dual-Energy X-ray Absorptiometry (DXA)The subjects were referred to the Rheumatologic Research Group of the University Hospital of Saint-Etienne (France) for evaluation of bone metabolism. This was measured with a whole body DXA scanner (Hologic QDR-2000, software version V5.67A, Hologic Inc., Bedford, MA, USA). The standard procedures described in the literature for DXA measurement were applied.16 BMD was measured for all subjects at the proximal femur and lumbar spine (L1-L4) with a coefficient of variation of 0.8% and < 1.2%, respectively. It was expressed in g/cm2 and as peak bone mass percentage in normal subjects (T-score). BMD results for the femoral neck and lumbar spine were classified into 3 groups according to World Health Organization criteria: normal (T-score > −1.0 SD), OP (T-score −1.0 to −2.5 SD) and OS (T-score < −2.5 SD).17Self-Assessment of SleepSubjective sleep quality and duration were assessed by PSQI, which has been proven to have an excellent sensitivity and reliability in middle-aged subjects14 and in the elderly.18 The questionnaire assessed subjective sleep quality and quantitative sleep-wake parameters, i.e., sleep latency, sleep duration and sleep efficiency over the preceding month. A global PSQI score ranging from 0 to 21 is calculated, with higher scores indicating worse sleep. The responses to the 19 questions in the questionnaire were scaled into 7 component scores (subjective sleep quality [C1], sleep onset latency in minutes [C2], sleep duration in hours [C3], percentage sleep efficiency [time spent asleep divided by the time spent in bed; C4], sleep disturbances [such as awakenings from sleep due to discomfort or bad dreams; C5], use of sleep medication [C6], and daytime dysfunction [C7]). These 7 components were weighted equally on an ordinal scale of 0–3. Sleep duration was analyzed as a categorical variable and based on the reported sleep duration, three groups were defined: short sleepers, who slept < 6 h; normal sleepers, who slept between 6–8 h; and long sleepers, who slept ≥ 8 h.19 The cutoff values for sleep duration were chosen on the basis of previous studies showing that a sleep duration < 6 h20 indicated short sleep and a sleep duration of approximately 7 h conferred a low risk for cardiovascular consequences.21,22,23The sleep duration in hours, the subjective sleep quality and the percentage of sleep efficiency were quantified on the basis of the 4 possibilities for answers to items C1, C3 and C4.At-Home Respiratory RecordingAll subjects underwent a full night ambulatory polygraphic recording using a polygraphic system (HypnoPTT, Tyco Health-care, Puritan Bennett, US). The following parameters were included: sound measurement, electrocardiography, pulse transit time, R-R timing, nasal pressure, respiratory effort and body position. Oxygen saturation (SpO2) was measured by pulse oximetry. To minimize potential overestimation of sleep duration, subjects completed the St. Mary's Hospital questionnaire, while wakefulness before lights-off was excluded by the analysis. All examinations were visually validated and manually scored for respiratory events and nocturnal SpO2 according to the Chicago criteria24 by a single scorer (F.R.), with an intra-scorer reliability of 87%. Hypopnea was defined as ≥ 50%reduction in airflow from the baseline value lasting ≥ 10 s and associated with ≥ 3% oxygen desaturation. Apnea was defined as the absence of airflow in the nasal cannula lasting for ≥ 10 s. The absence of rib cage movements associated with apnea defined the event as central, while a progressive increase in pulse transit time and respiratory efforts allowed definition of the episode as obstructive. Apnea-hypopnea index (AHI) was defined as the ratio of the number of obstructive apneas and hypopneas per hour of reported sleeping time. The indices of nocturnal hypoxemia were as follows: mean SpO2; percentage of recording time with a SpO2≤ 90%; minimum SpO2 value recorded during sleep (minimum SpO2); and oxygen desaturation index (ODI), i.e., the number of episodes of oxyhemoglobin desaturation per hour of reported sleep time during which blood oxygen level fell by 3% or more. Pulse transit time was continuously monitored, and the autonomic respiratory-related and total autonomic arousal indices were calculated after visual correction.25 According to recent data in elderly subjects,26 an AHI ≥ 15 with at least 50% of events scored as obstructive was considered diagnostic of OSA.Statistical AnalysesThe characteristics of the subjects were summarized as the mean ± standard deviation (SD) for the continuous variables and as the frequency and percentages of the sample for the categorical variables. We examined the gaussian distribution of self-reported sleep quality and BMD using skew, kurtosis, and the Levene test for the equality of variances. Study population characteristics are reported as means ± SD. The differences between the men and the women, were analyzed using parametric tests (t test, univariate analysis of variance [ANOVA]) or nonparametric tests (the Mann-Whitney U test) for the continuous variables. The χ2 test was assessed for the categorical variables. Pearson correlation analysis was performed to establish the association of BMD and clinical, anthropometric and self-reported sleep data.Concerning the bone mineral data, we focused the analyses into 3 groups according to World Health Organization criteria: normal (T-score > −1.0 SD), OP (T-score −1.0 to −2.5 SD) and OS (T-score < −2.5 SD). First, differences between the subjects with normal BMD, with OP or OS were assessed by using the χ2 test for categorical variables and with Kruskall-Wallis test for continuous variables.Second, in order to identify predictors of individual class of the dependent variable (the group of the subjects considered (with OS or OP), we conducted multinomial logistic regressions analysis according for clustering at the reported sleep duration. The reference group was the normal group. The final models were adjusted for demographic and clinical data (gender, BMI, dyslipidemia, the DEE, the AHI and the ODI) according to the differences between the 3 groups.All statistical analyses were conducted using the SPSS statistical software package (SPSS for Windows, version 17.0, SPSS, Chicago, IL, USA). All reported p values are two-tailed, with significance set at p ≤ 0.05.RESULTSThe participants were mostly women (58%), with a mean age of 65.7 ± 0.8 y, a mean BMI of 25.3 ± 3.6 kg/m2 and a mean DEE of 10335.7 ± 1911 kJ/24 h. The mean AHI was 21.0 ± 14.4, with 42% of subjects having abnormal respiratory events. The means ± SD of the indices of nocturnal hypoxemia were 8.6 ± 9.0 for the ODI, 95.4% ± 1.8% for the mean SpO2 value, 89.9% ± 4.1% for the minimal SpO2 value, and 1.7 ± 5.1 for the SpO2 < 90%. As reported in Table 2 we found gender differences for BMI (p = 0.05), alcohol use (p < 0.001), smoking habits (p < 0.001), DEE (p < 0.001), sleep efficiency (p = 0.01), AHI (p < 0.001), ODI (p < 0.001), lumbar BMD (p < 0.001), and vertebral BMD (p < 0.001) greater in men than women. Women had higher PSQI score (p < 0.001) than men.Table 2 Clinical, anthropometric, polygraphic, self-reported sleep data and DXA measurements for the study population (mean ± SD).Table 2 Clinical, anthropometric, polygraphic, self-reported sleep data and DXA measurements for the study population (mean ± SD).The mean sleep duration was 6.9 ± 1.2 h; 29 % of the subjects were "short sleepers", 40% "normal sleepers" and 31% "long sleepers". Mean sleep efficiency was 81% ± 13% and mean sleep quality 1.01 ± 0.6, suggesting normal sleep quality. According to previously described threshold values,18 femoral neck OP was found in 42% of cases mostly in women (76%) while OS was found in 9% of the population without gender differences. When we considered the lumbar spine site, OP was found in 40% of subjects mostly in women (57%) and OS in 12% significantly higher in women (72%).Table 3 and Figure 1 show the percentage of subjects with normal BMD, OP, and OS. Femoral neck OP was found in 42% of cases mostly in women (76%), while OS was found in 9% of the population without gender differences. When we considered the lumbar spine site, OP was found in 40% of subjects, mostly in women (57%), and OS in 12%, significantly higher in women (72%).Table 3 Clinical, anthropometric and polygraphic data of subjects with normal BMD and subjects with osteopenia or osteoporosis at the femoral and vertebral levels (mean ± SD).Table 3 Clinical, anthropometric and polygraphic data of subjects with normal BMD and subjects with osteopenia or osteoporosis at the femoral and vertebral levels (mean ± SD).Figure 1: Percentage of subjects with normal BMD and subjects with osteopenia or osteoporosis at the femoral and vertebral levels.**p < 0.01Download FigureTable 3 reports the clinical and anthropometric data for subjects having normal BMD, OP and OS. At both sites, subjects with OS differ for lower BMI (p < 0.001) and DDE (p < 0001) and greater dyslipidemia. Considering subjective sleep data (Table 4), subjects with OS at the femoral level had a greater self-reported sleep duration (p = 0.03) and were more frequently long sleepers (p = 0.004) without differences for sleep quality (p = 0.34) and sleep efficiency (p = 0.80). No sleep data differences between groups were found at the lumbar spine.Table 4 Subjective sleep data in the total group and in subjects with normal BMD and osteopenia or osteoporosis at the femoral and vertebral levels (mean ± SD).Table 4 Subjective sleep data in the total group and in subjects with normal BMD and osteopenia or osteoporosis at the femoral and vertebral levels (mean ± SD).With correlation analysis, the only factors both at femoral and lumbar levels affecting BMD were BMI (femoral: r = 0.26, p < 0.001; lumbar spine: r = 0.24, p < 0.001) and DEE (femoral: r = 0.38, p < 0.001, lumbar spine: r = 0.33, p < 0.001), and dyslipidemia at the femoral level. No correlations were noted between BMD and other clinical, metabolic and subjective sleep data.Multiple logistic regression models (Table 5) were created to evaluate the risk of having OS or OP according to the presence of a normal, short or long sleep. After adjustment for gender, BMI, presence or absence of dyslipidemia and mean DEE, the long sleepers were 3.75 times more likely to have femoral neck OS (odds ratio [OR]: 6.26; 95% CI: 1.51– 15.92; p = 0.011). Similar results were found for lumbar spine long sleepers having increased risk to have OS (OR: 1.83; 95% CI: 0.91–4.14; p = 0.10) without however, statistical significance.Table 5 Multinomial logistic regression showing the association between osteopenia and osteoporosis at the two levels and explicative sleep, anthropometric and biological data.Table 5 Multinomial logistic regression showing the association between osteopenia and osteoporosis at the two levels and explicative sleep, anthropometric and biological data.DISCUSSIONThe aims of this study were to analyze if outside physical activity known to affect OS self-perceived sleep duration and quality might affect BMD in older subjects with minimal health disorders and if long sleep duration might increase the risk of the OS in an older sample. The first and most important finding of this study was that subjects with long sleep duration were at risk to develop OS compared to short and normal sleepers without interference of sleep quality. This association was lacking in the subjects with OP. The second finding was the lack of strong significant interference of DEE and anthropometric and metabolic factors, commonly considered as key factors implicated in OS risk at the femoral level. We can suggest that in older subjects without major health diseases and previous diagnosis of OS, long sleep duration increases OS and fracture risk, stressing the need to consider this effect on the diagnosis and follow-up of subjects with OS.BMD reaches a peak at approximately 40 y of age and then gradually decreases27 with a sharper drop-off after menopause and at more than 60 y of age, which would explain the reported increase of OS in elderly. Previous epidemiological data in elderly reported OS prevalence in 13–18% in the USA28 and 21% in Sweden.29 In contrast, in our sample OS was present only in 9% of cases, with a greater prevalence of subjects with OP. Although speculative, we can propose that diagnosis of OS based only on clinical self-report might induce an overestimation of the real prevalence of OS, with OP subjects probably included in the OS group.As shown in Table 1, differences in methodology surrounds the results of the association between sleep duration and quality and OS. In the primary studies done between 2006–2012, criteria to define OS were self-reported; and when assessed by DXA analysis, only two authors11,30 made assessments of T-score that allowed distinction between OP and OS. Moreover, the authors considered large populations from middle-aged to older subjects more frequently in women, with definition of sleep duration and sleep quality based on just one question, "how many hours you sleep," without specific assessment by a validated questionnaire. Finally, the majority of studies did not assess specific factors such as metabolic variables and DEE, which are known to affect BMD and consequently OS. In our population with homogenous age, we have more objective data on sleep quality and duration using PSQI, and a more detailed definition by DXA on OS and BMD at two levels, allowing a better definition of bone dysfunction. On the basis of these observations, we noted that subjects with long sleep duration were 3.75 times more likely to have femoral neck OS without significant association at the lumbar level. Interestingly, DEE and obesity did not affect OS risk, two factors considered as important factors in bone metabolism. Despite methodological differences, our results are in line with the results of Chen and coworkers,25 who examined 8,688 participants and found an increased OS risk in subjects sleeping more than 7 h. Similar data where obtained by Kobayashi et al.8 in a large population of 19,321 healthy individuals, which found that individuals with a sleep duration > 8 h had a high risk of OS. Finally, in a recent study,11 Tian and coworkers examined 31,769 participants aged 45–86 y and they found that participants having long sleep duration (> 9 h) and early sleep timing were independently associated with an increased risk of OS in older subjects.The mechanisms of the association between sleep duration and increased risk of OS in elderly are still not clearly defined. The association between short sleep and OS31 may be explained by the bad sleep quality related to associated disorders such as depression,32 rheumatism/arthritis/arthroses, changes in cortisol26 and estrogen levels related to bad sleep and menopause,33 and to over activity of inflammatory processes,34 all factors favoring bone resorption. Considering associated sleep disorders, controversy surrounds the association between sleep apnea and OS. This effect could be excluded in this sample in that, as previously reported,35 any association between OS and apnea severity was found.When we consider the link between self-reported long sleep and OS, the mechanisms are still obscure. It is unclear whether this association is causal or reflecting the role of some factors that influence sleep. Depression, physical inactivity, low socioeconomic status, poor health status, chronic disease7,36 and diabetes37 may contribute to increased OS risk. In contrast to this hypothesis, our population was free of chronic health diseases such as depression or severe diabetes. An alternative hypothesis is that long sleep reduces the wake time, which may induce a reduction of exercise time that is the most important factor to maintain a good bone metabolism.38 However, even though in our subjects with OS had lower energy expenditure, this effect acts only at the vertebral level. A second hypothesis is the role of obesity, which is considered as a preventive factor of OS. As depicted in Table 3, individuals with OS had a significantly low BMI, suggesting a reduced gravitational and mechanical load in these subjects.39,40,41Strengths and LimitationThe strengths of our study are the following: (1) the analysis of a sample of older subjects with similar age and without chronic disease; (2) the inclusion of a near homogenous group of women and men with similar age for whom no significant differences were found; and (3) the use of validated sleep and physical activity questionnaires. Some limitations, however, need to be considered. Firstly, the cross-sectional nature of our study is unable to assess the causative relationship between sleep duration and BMD. Secondly, since we analyzed a homogenous group of older subjects, with minimal health disease, that precludes the application of these results to clinical populations or to very old subjects.CONCLUSIONSThe present study on older people reveals that reported long sleep time may increase OS risk. Despite reduced physical activity and body mass which might contribute to OS, the greater factor affecting OS was the reported sleep duration. Further epidemiological and clinical prospective studies are necessary to confirm our findings in order to educate patients to have a reduced sleep time and increased physical activity.DISCLOSURE STATEMENTThis was not an industry supported study. The authors have indicated no financial conflicts of interest.ABBREVIATIONSAHIapnea-hypopnea indexBMDbone mineral densityBMIbody mass indexDEEdaily energy expenditureDXAdual-energy X-ray absorptiometryODIoxygen desaturation indexOPosteopeniaOSosteoporosisPOPAQPopulation Physical Activity questionnairePROOFPrognostic Indicator of Cardiovascular and Cerebrovas cular Events studyPSQIPittsburgh Sleep Quality IndexREFERENCES1 Consensus development conference: prophylaxis and treatment of osteoporosis. Am J Med; 1991;90:107-10, 1986575. CrossrefGoogle Scholar2 Varenna M, Manara M, Galli L, Binelli I, Zucchi F, Spinigaglia LThe association between osteoporosis and hypertension: the role of low dairy intake. Calcif Tissue Int; 2013;92:217-27, 23192372. Google Scholar3 Dominiquez LJ, Muratore M, Quarta E, Zagone G, Barbagallo MOsteoporosis and diabetes. Reumatismo; 2004;56:235-41, 15643475. Google Scholar4 Migliaccio S, Greco EA, Fornari R, Donini LM, Renzi AIs obesity in women protective against osteoporosis?Diab Metab Syndr Obes Targets Ther; 2011;4:273-82. CrossrefGoogle Scholar5 Johansson J, Nordström A, Nordström PObjectively measured physical activity is associated with parameters of bone in 70-year-old men and women. Bone; 2015;81:72-9, 26151120. CrossrefGoogle Scholar6 Fu X, Zhao X, Lu H, Fan J, Ma X, Zhu SAssociation between sleep duration and bone mineral density in Chinese population. Bone; 2011;49:1062-6, 21864732. CrossrefGoogle Scholar7 Lima MG, Bergamo Franscisco PMS, de Azevedo Barro BertiSleep duration pattern and chronic diseases in Brazilian adults (ISACAMP, 2008/09). Sleep Med; 2012;13:139-44, 22137111. CrossrefGoogle Scholar8 Kobayashi D, Takahashi O, Deshpande GA, Shimbo T, Fukui TAssociation between osteoporosis and sleep duration in healthy middle-aged and elderly adults: a large-scale, cross-sectional study in Japan. Sleep Breath; 2012;16:579-83, 21688188. CrossrefGoogle Scholar9 Specker BL, Binkley M, Vukovich M, Beare TVolumetric bone mineral density and bone size in sleep-deprived indivi