Associations of Components of Sarcopenia with Risk of Fracture in the Osteoporotic Fractures in Men (MrOS) Study

Abstract

Introduction:

The association between the components of sarcopenia and fractures have not been clearly elucidated and has hindered the development of appropriate therapeutic interventions. Our aim was to evaluate the associations between the individual components of sarcopenia, specifically lean mass, strength, and physical performance and fracture (any fracture, hip fracture, major osteoporotic fracture) in the Osteoporotic Fractures in Men (MrOS) study.

Methods:

The Osteoporotic Fractures in Men study (MrOS) recruited 5,995 men ≥ 65 years of age. We measured appendicular lean mass (ALM) by dual-energy x-ray absorptiometry (low as residual value <20^th percentile for the cohort), walking speed (fastest trial of usual pace, values <0.8 m/s were low), and grip strength (max score of 2 trials, values <30 kg were low). Information on fractures was assessed tri-annually over an average follow-up of 12 years and centrally adjudicated. Cox proportional hazards models estimated the hazard ratio (HR) (95% Confidence Intervals) for slow walking speed, low grip strength and low lean mass.

Results:

Overall, 1,413 men had a fracture during follow-up. Slow walking speed was associated with an increased risk for any HR=1.39, 1.05-1.84; hip HR= 2.37, 1.54-3.63; and major osteoporotic, HR= 1.89, 1.34-2.67 in multi-variate adjusted models. Low lean mass and low grip strength were not significantly associated with fracture.

Conclusions:

In this cohort of older adult men, the risk of experiencing any, hip, or major osteoporotic fracture is greater in men with slow walking speed in comparison to men with normal walking speed, but low grip strength and low lean mass were not associated with fracture.

Mini-Abstract

Our aim was to evaluate the associations between the individual components of sarcopenia and fracture types. In this cohort, the risk of experiencing any clinical, hip, or major osteoporotic fracture is greater in men with slow walking speed in comparison to normal walking speed.

Introduction

The lack of a consensus definition of sarcopenia has limited its use clinically and has hindered the development of appropriate therapeutic interventions.[1] Sarcopenia was initially defined as the loss of lean mass associated with aging.[2] Lean mass, measured by dual energy x-ray absorptiometry (DXA), has variable associations with adverse health outcomes including fractures. [3-6] Current definitions of sarcopenia include varying cut-points for body composition, strength, and functional performance and scaling by height or body mass index further creating challenges for clinical utility. [1, 7] These variations in the definition of sarcopenia result in discrepancies in the prevalence of the condition and likelihood of adverse outcomes. [8, 9] It is not clear whether all the components of sarcopenia definitions (slow walking speed, low lean mass and low grip strength) each individually predict adverse outcomes, particularly fractures, in older adults. The Sarcopenia Definitions and Outcomes Consortium (SDOC) did not define specific cut points for determining low lean mass but did suggest cut-points to examine grip strength and gait speed. Disentangling which components are meaningful predictors of outcomes will elucidate which of these measures should be included in a composite definition.

Previous work in MrOS has demonstrated that once BMD is accounted for, appendicular lean mass divided by height square (ALM/ht²) mass is not an independent risk factor for fracture. It is unknown if another definition of lean mass, such as the Newman method, is an independent risk factor for fracture. The Newman method has been associated with lower function in older adults. [4]Worse performance on functional measures and weakness has been associated with increased risk of fracture and are recommended components of sarcopenia definitions by the Sarcopenia Definitions and Outcomes Consortium. [1, 10-12] The association between this alternative measure to assess low lean mass and various fracture types has not been explored in MrOS.

Therefore, we evaluated the associations between the individual components, low lean mass, low strength, and slow gait speed using 3 fracture outcomes (any fracture, hip fracture, and major osteoporotic fracture) in the Osteoporotic Fractures in Men (MrOS) study.

Methods

The Osteoporotic Fractures in Men study (MrOS) is a multicenter prospective study of aging with a particular focus on risk factors for fractures. There are 6 clinic sites included in the MrOS US cohort: Birmingham, AL; Minneapolis, MN; Palo Alto, CA; Monongahela Valley near Pittsburgh ,PA; Portland, OR; San Diego, CA. The design, measures, and recruitment have been previously described.[13] In brief, from March 2000- April 2002, 5,995 men ≥ 65 years of age were recruited and enrolled from 6 centers across the United States: Birmingham, Alabama; Minneapolis, Minnesota; Palo Alto, California; the Monongahela Valley near Pittsburgh, Pennsylvania; Portland, Oregon; and San Diego, California.[13] Men were excluded from the study if they could not walk without the assistance of another or had bilateral hip replacement. Approval of the conduct of MrOS was obtained from institutional review boards of participating institutions, and written informed consent was obtained from all participants before data collection, study information can be located at https://mrosonline.ucsf.edu.[13]

Body Composition

Whole body lean mass, including appendicular lean mass, and total body fat were obtained using Hologic QDR 4500 dual energy x-ray absorptiometry machines. Central training, quality control, and standardized procedures were used to insure reproducibility of measurements. Appendicular lean mass (ALM) was calculated as the sum of soft-tissue lean mass in the arms and legs. Men were considered to have low lean mass if their ALM was below the 20^th percentile of the regression residuals derived from the cohort (Newman definition). The Newman residuals method to define low lean mass regresses both height and fat mass on total ALM and has been shown to be a strong predictor of physical function. [4] This method allows the residuals of the regression to identify those whose ALM was higher or lower than the predicted value. Definitions of sarcopenia that utilize only lean mass and height have primarily identified thin people as sarcopenic but not obese as this method is highly correlated with body mass index (BMI).[14] The Newman-residuals method to define sarcopenia adjusts for fat mass and height to the expected lean mass, to be able to identify individuals who do not have adequate lean mass for their size, which can encompass individuals across the range of BMI.[14] This method attempts to capture the effects of both low lean mass and high fat mass together. A higher value would indicate a more muscular individual.[4] (aLM (kg) = −23.53 + 25.34 x height (m) + 0.17 total fat mass (kg), 20^th percentile cut point = −2.17).

Gait speed

Gait speed was assessed by usual gait speed over a 6-meter course. Two trials were performed, and the fastest value was used in this analysis. Slow gait speed was defined as usual walking speed < 0.8 m/s. This cut-point has been used in definitions of sarcopenia (European Working Group on Sarcopenia and the Foundation for NIH Sarcopenia Project) and shown to be associated with physical performance and clinical relevance. [15-17] This cut-point was additionally recommended by the SDOC. [1]

Low muscle strength

Grip strength was measured using a Jamar hand dynamometer with adjustable handgrip. Two trials were performed on each hand with highest value used in the analysis. Low muscle strength was defined as grip strength <30 kg which is used in definitions of sarcopenia [15] and shown to be associated with increased risk of non-spine fractures in men. [18]

Bone Mineral Density

Total body, lumbar spine (L1-L4), and total hip areal BMD were measured at baseline using dual-energy x-ray absorptiometry (DXA). The same scanner model was used at all 6 sites (QDR 4500 W, Hologic Inc.; Bedford, MA, USA). Standardized procedures for positioning and scan analysis were followed for all scans. All DXA operators were centrally certified based on an evaluation of scanning and analysis techniques.

Fractures Ascertainment

Men were followed for incident fractures by completing and returning a questionnaire every 4 months that was administered by mail or telephone; >95% complete follow-up over a mean of 12 years.[19] Fractures were verified by centralized physician adjudication of the medical records. Pathological fractures were excluded. Exclusion of fractures resulting from excess trauma underestimates the contribution of osteoporosis to fractures, we included fractures regardless of trauma level.[20, 21]

Covariates

At baseline, information regarding demographic, anthropometric, personal and family medical history, lifestyle, functional status, visual acuity, and cognitive data were obtained through self-report, interview, or examination by trained and certified staff. [13] Data on age and race were collected at baseline. Physical activity was assessed with the Physical Activity Scale for the Elderly (PASE).[22] Additional questions included self-report for physician diagnosis of specific common medical conditions. Participants were asked about falls over the previous 12 months. General health status was self-rated and categorized as either excellent/good versus fair/poor. Mood was assessed using the Geriatric Depression Scale. [23] Cognitive assessment was completed using the Teng Modified Mini-Mental Status Examination (3MS). [24]

Body weight (kg, indoor clothing without shoes) was recorded with a calibrated balance beam or electronic scale. Height (cm) was measured using a wall mounted Harpenden stadiometer (DyFed). Body mass index was calculated by dividing weight by height squared (kg/m²).[25] Participants were asked to bring in all prescription and non-prescription medications taken regularly during the previous 30 days to their clinic visit. All medications were recorded by the clinics were stored in an electronic medication inventory database (San Francisco Coordinating Center, San Francisco, CA, USA). Each medication was matched to its ingredients based on the Iowa Drug Information Services Drug Vocabulary (College of Pharmacy, University of Iowa, Iowa City, IA). [26]

Statistical Analysis

Participant characteristics were summarized using means and standard deviation, median and interquartile range, or frequencies and percentages as appropriate. Baseline characteristics of the cohort were assessed as a whole cohort and based on fracture status. To test for differences across groups, analysis of variance was used for continuous variables and chi square tests for categorical variables.

Separate Cox proportional hazards models were used to estimate the risk of incident fracture by low lean mass, slow walking speed, and weakness as independent variables, hazard ratios (HR) and 95% confidence intervals (CI) were calculated per standard deviation (SD). All models were adjusted for age and clinic site. To construct multi-variable models, additional variables were selected for inclusion in the model based on identification from previous literature as a risk factor for fracture, including each of the independent variables low lean mass, slow walking speed, and weakness. These variables include age, race, history of diabetes, fall, and arthritis, health rating, physical activity, smoking status, alcohol consumption, education, symptoms of depression. Interaction between independent variables, low lean mass, slow walking speed, and grip strength were assessed in each model. Statistical significance indicated by p <0.05. All analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC).

Results

The average age of the cohort was 74 years. Men who experienced any clinical fracture were older than those who did not experience a fracture over follow-up. Nearly 23% of the men who experienced a fracture had low lean mass compared to 19% of men who did not fracture. Slow walking speed and weak grip strength did not differ by fracture status. Those who experienced a fracture were more likely to have a history of falls in comparison to men with no fracture. Most men reported good to excellent health although the proportion of men with good to excellent health was slightly higher among men who fractured.No differences were seen in physical activity, smoking status, alcohol intake, and living alone across the groups. Participant characteristics are presented in Table 1.

Table 1.

Baseline Characteristics

	Whole Cohort (N= 5994)	No Fracture (n= 4581)	Fracture (n=1413)	p-value
Age ($\overline{\mathrm{x}}\pm \sigma$)	73.7 ± 5.9	73.5 ± 5.9	74.2 ± 5.9	<0.0001
Caucasian (n,%)	5362 (89.5)	4047 (88.3)	1315 (93.1)	<0.0001
Low Lean Mass (n,%)	1191 (20)	876 (19.2)	315 (22.6)	0.0063
Slow Walking Speed <0.8 m/s (n,%)	273 (4.6)	208 (4.5)	65 (4.6)	0.92
Grip Strength <30 kg (n,%)	380 (6.4)	286 (6.3)	94 (6.7)	0.84
BMD T-score	−0.62 ± 1.07	−0.51 ± 1.06	−0.96 ± 1.01	<0.0001
History of Diabetes (n,%)	653 (10.9)	504 (11.0)	149 (10.5)	0.63
History of Arthritis/gout (n,%)	2847 (47.5)	2183 (47.7)	664 (47.0)	0.66
History of a fall (n,%)	1268 (21.2)	908 (19.8)	360 (25.5)	<0.0001
Good/Excellent Health Rating (n,%)	5135 (85.7)	3896 (85.1)	1239 (87.7)	0.0146
Use of Benzodiazepiene (n,%)	205 (3.6)	157 (3.6)	48 (3.6)	0.99
Use of Selective Serotonin Reuptake Inhibitors (n,%)	163 (2.8)	117 (2.7)	46 (3.4)	0.15
Visual Acuity 20/50 (n,%)	114 (1.9)	91 (2.0)	23 (1.6)	0.39
3MSa ($\overline{\mathrm{x}}\pm \sigma$)	93.3 ± 5.9	93.1 ± 6.1	93.7 ± 5.1	0.0007
College Education (n,%)	3188 (53.2)	2378 (51.9)	810 (57.3)	0.0004
Depressive feelings (n,%)	178 (3.0)	130 (2.8)	48 (3.4)	0.28
PASE score (median + IQR)	142 (100 – 186.1)	142.1 (101- 185.9)	141.2 (95.7 – 189.7)	0.63
Current Smoker (n, %)	206 (3.4)	153 (3.3)	53 (3.8)	0.46
Zero Alcoholic Drinks per week (n,%)	2121 (35.4)	1639 (35.8)	482 (34.2)	0.42
Live Alone (n,%)	840 (14.0)	642 (14.0)	198 (14.0)	0.99

^*Low Lean Mass defined by the Newman Residual Method

^a3MS: Modified Mini-Mental Status Examination; p-scores indicate difference between no fracture and fracture groups, Bolded values indicate statistical significance

In minimally adjusted (base) models, accounting for age and clinic site, low lean mass was associated with 25% increased risk of any fracture and 35% increased risk of major osteoporotic fracture (table 2) but not hip fracture. These associations were attenuated and no longer significant after adjustment for covariates including BMD

Table 2.

Association between Low Lean Mass, Slow Walking Speed, and Weakness with Any Fracture, Hip Fracture, and Major Osteoporotic Fractures

	Any Fracture HR (95% CI)		Hip Fracture HR (95% CI)		Major Osteoporotic Fracture HR (95% CI)
	Base	Full	Base	Full	Base	Full
Low Lean Mass	1.25 (1.10 −1.42) *	1.10 (0.96 −1.26)	1.23 (0.95 – 1.59)	0.91 (0.69 – 1.20)	1.35 (1.12 – 1.62) *	1.16 (0.95 – 1.40)
Slow Walking Speed	1.70 (1.32 – 2.20) *	1.39 (1.05 – 1.84) *	3.44 (2.33 – 5.07) *	2.37 (1.54 – 3.63) *	2.39 (1.73 – 3.29) *	1.89 (1.34– 2.67) *
Low muscle strength	1.31 (1.06 – 1.63) *	1.20 (0.96 – 1.50)	1.01 (0.65 – 1.57)	0.81 (0.51 – 1.30)	1.12 (0.82 – 1.55)	0.97 (0.70 – 1.35)

Base (minimally adjusted) models adjusted for clinic site & age. Fully adjusted models include base model plus BMD T-score, history of diabetes, history of arthritis/gout history of falls, self-reported health rating, depressive feelings, PASE score, smoking status, alcoholic drinks per week, living alone, education status, visual acuity, use of benzodiazepines, use of selective serotonin reuptake inhibitors, 3MS score, education status. Bold* indicates statistical significance

In minimally adjusted (base) models, accounting for age and clinic site low muscle strength was only associated with any fracture (but not hip or major osteoporotic fracture). In fully adjusted models, low muscle strength was no longer significantly associated with any fracture, hip or major osteoporotic fracture.

Slow walking speed was associated with all fracture outcomes. Although the association was attenuated in the fully adjustment models, the relationship between slow walking speed and fracture remained strong, particularly for hip fractures. Slow walking speed showed a 39% increased risk for any clinical fracture, 89% increased risk for osteoporotic fracture and 137% increased risk for hip fracture in our multivariate adjusted models (HR_any= 1.39, 95% CI =1.05-1.84; HR_oSteoporotic= 1.89, 95% CI= 1.34-2.67; HR_hip= 2.37, 95% CI= 1.54-3.63).

No interaction terms between low lean mass, slow walking speed, and weakness were significant in any of the models. Model details and results are presented in Table 2.

Discussion

In this cohort of older adult men, the risk of experiencing any, hip, or a major osteoporotic fracture was greater in men with slow walking speed in comparison to men with normal walking speed independent of many covariates. In contrast there was no association of grip strength or lean mass with any fracture outcome after accounting for multiple confounding variables.

Low lean mass, when defined by the Newman-residuals methods, was associated with an increased risk of any fracture and major osteoporotic fractures in minimally adjusted models, but this association was attenuated after adjustment for confounders including BMD. This finding is in line with the previous literature that demonstrates no association between DXA ALM/ht² and fracture after accounting for BMD. These results are also consistent with the updated guidelines from the Sarcopenia Definition and Outcomes Consortium (SDOC) that lean mass measured by DXA is not a strong predictor of adverse health outcomes, including fracture. [1]

Low muscle strength was not an independent risk factor for fracture in this analysis. While low muscle strength measured using the maximum score on a handheld dynamometer is easy and reliable. [27, 28] Although grip strength has been shown to be cross-sectionally associated with lower-extremity strength, [27] assessment of lower extremity power may be a stronger predictor of fracture. [29-32] Our results vary from other analyses within the MrOS cohort. Harvey et al reported using the MrOS cohorts from US, Sweden, and Hong Kong that performance and strength measures, including grip strength, is an independent risk factor for incident factures over 9 years of follow-up above BMD, prior falls, and Fracture Risk Assessment Tool (FRAX) probability. [33] Our results may vary as we utilized a cut-point of <30 kg to define low muscle strength versus use of grip strength as a continuous measure and only included results from the US cohort. Grip strength may not be the most robust measure of function to assess in terms of fracture risk. In fact, previous work in MrOS has demonstrated that a measure of lower extremity function – chair stands performance – is strongly related to hip fractures and varies greatly from grip strength assessments. [10, 34] Chair stands are a weight bearing activity that stands as a proxy for power. Another potential limitation of grip strength is that other factors may also have impacted the measurement including arthritis, pain, depression, and motivation.

Walking speed has become a vital measure across many clinical areas as it can provide insight into an older person’s health A gait speed less than 0.8 m/s has been consistently used in definitions of sarcopenia and has been associated with independent community ambulation. [1, 35] Gait speed is a task that combines the effort of numerous body systems neuromuscular, vestibular, cardiopulmonary, and musculoskeletal. In this analysis, gait speed itself was an independent predictor of any fracture, hip fracture, and major osteoporotic fractures in this cohort. Routine measurement of gait speed and identifying the cause of slow gait speed would be important to consider with clinical care for this risk factor, given it may be addressed in rehabilitative treatment by addressing lower extremity strength and power, efficiency of gait, and biomechanical factors. [36-38].

Limitations of this study include the limited generalizability of the cohort, given it is predominately white race and limitation to men only. A second limitation may be the inclusion of high-energy trauma fractures as there may be high probability of fracture regardless of sarcopenia status. An additional limitation may be the use of DXA to approximate lean mass as it may not provide adequate assessment of muscle quality. Other methods to identify low lean mass, such as D₃-Creatinine may provide more information about the quality of the muscle, which may impact strength and function.[39] Cawthon et al have reported associations between low lean mass, measured by this novel method, and self-reported limitations in ADLs and mobility tasks.[39] Strengths of this study include use of a well characterized cohort of community dwelling older men who are at risk for adverse health outcomes, with ascertainment of fractures over an extended follow up time.

Overall, slow walking speed was associated with an increased risk of all fracture types in older men. In contrast there was no association with grip strength or appendicular lean mass. Thus, all components within definitions of sarcopenia do not similarly increase the risk of fracture. Our findings suggest that slow gait speed may be a simple measure to identify those at risk of poor outcomes and may be most appropriate in the context of fracture risk consideration in addition to the Fracture Risk Assessment (FRAX) tool.