Handgrip Strength but not Appendicular Lean Mass is an Independent Predictor of Functional Outcome in Hip-Fracture Women: A Short-Term Prospective Study

Handgrip Strength but not Appendicular Lean Mass is an Independent Predictor of Functional Outcome in Hip-Fracture Women: A Short-Term Prospective Study

Accepted Manuscript Handgrip Strength but not Appendicular Lean Mass is an Independent Predictor of Functional Outcome in Hip-Fracture Women: A Short-...

119KB Sizes 0 Downloads 0 Views

Recommend Documents

No documents
Accepted Manuscript Handgrip Strength but not Appendicular Lean Mass is an Independent Predictor of Functional Outcome in Hip-Fracture Women: A Short-Term Prospective Study. Marco Di Monaco, MD Carlotta Castiglioni, MD Elena De Toma, OT Luisa Gardin, OT Silvia Giordano, OT Roberto Di Monaco, PhD Rosa Tappero, MD PII:

S0003-9993(14)00286-X

DOI:

10.1016/j.apmr.2014.04.003

Reference:

YAPMR 55806

To appear in:

ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION

Received Date: 7 January 2014 Revised Date:

19 March 2014

Accepted Date: 4 April 2014

Please cite this article as: Di Monaco M, Castiglioni C, De Toma E, Gardin L, Giordano S, Di Monaco R, Tappero R, Handgrip Strength but not Appendicular Lean Mass is an Independent Predictor of Functional Outcome in Hip-Fracture Women: A Short-Term Prospective Study., ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION (2014), doi: 10.1016/j.apmr.2014.04.003. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

RUNNING HEAD:

2

STRENGTH AND RECOVERY AFTER HIP FRACTURE

3

TITLE:

5

HANDGRIP STRENGTH BUT NOT APPENDICULAR LEAN MASS IS AN INDEPENDENT

6

PREDICTOR OF FUNCTIONAL OUTCOME IN HIP-FRACTURE WOMEN: A SHORT-

7

TERM PROSPECTIVE STUDY.

SC

RI PT

4

9

AUTHORS AND INSTITUTIONS:

M AN U

8

Marco Di Monaco MD1, Carlotta Castiglioni1 MD, Elena De Toma2 OT, Luisa Gardin2 OT,

11

Silvia Giordano2 OT, Roberto Di Monaco3 PhD, Rosa Tappero1 MD

12

From the 1Osteoporosis Research Center, Division of Physical Medicine and Rehabilitation,

13

Presidio Sanitario San Camillo; 2Service of Occupational Therapy, Presidio Sanitario San

14

Camillo, 3 Department of Social Science, University; Torino, Italy

TE D

10

15

All the authors have no conflicts of interest.

EP

16 17

CORRESPONDING AUTHOR:

19

Dr. Marco Di Monaco

20

Osteoporosis Research Center, Division of Physical Medicine and Rehabilitation, Presidio

21

Sanitario San Camillo, Strada Santa Margherita 136, 10131, Torino, Italy

22

Tel. 0039 011 8199411; Fax 0039 011 8193012; Email [email protected]; [email protected]

23

sancamillo.to.it

AC C

18

1

ACCEPTED MANUSCRIPT

1

HANDGRIP STRENGTH BUT NOT APPENDICULAR LEAN MASS IS AN INDEPENDENT

2

PREDICTOR OF FUNCTIONAL OUTCOME IN HIP-FRACTURE WOMEN: A SHORT-

3

TERM PROSPECTIVE STUDY

RI PT

4

ABSTRACT

6

OBJECTIVE. To investigate the contribution of muscle mass and handgrip strength in predicting

7

the functional outcome after hip fracture in women.

8

DESIGN. Observational study.

9

SETTING. Rehabilitation hospital in Italy.

M AN U

SC

5

PARTICIPANTS. We investigated white women (N=123) of 149 who were consecutively

11

admitted to a rehabilitation hospital because of their first fracture of the hip.

12

INTERVENTIONS. Not applicable.

13

MAIN OUTCOME MEASURE. We measured appendicular lean mass (aLM) by dual-energy x-

14

ray absorptiometry (DXA), 21.1±8.7 (mean ± SD) days after hip fracture occurrence in the 123

15

women. On the same day, we assessed grip strength at the non-dominant arm with a Jamar

16

dynamometer. At the end of acute inpatient rehabilitation we measured ability to function in

17

activities of daily living by the Barthel Index and lower-limb performance by the Timed Up and

18

Go (TUG) test.

19

RESULTS. We found significant correlations between handgrip strength measured before

20

rehabilitation and Barthel index scores after rehabilitation (ρ=0.50; p<0.001), Barthel index

21

effectiveness (ρ=0.45; p<0.001), and TUG test (ρ=-0.41; p<0.001). Conversely, we found no

22

significant correlations between aLM/height2 and Barthel index scores after rehabilitation

23

(ρ=0.075; p=0.41), Barthel index effectiveness (ρ=0.06; p=0.53), or TUG test (ρ=0.005; p=0.96).

AC C

EP

TE D

10

1

ACCEPTED MANUSCRIPT

Significant associations between grip strength and all the outcome measures persisted after

25

adjustment for eight potential confounders, including Barthel index scores before rehabilitation,

26

age, number of medications, number of comorbidities, pressure ulcers, concomitant infections,

27

time between fracture occurrence and assessment, and aLM/height2.

28

CONCLUSIONS. Grip strength, but not DXA-assessed aLM, significantly predicted short-term

29

functional outcome in women following a hip fracture.

SC

30

RI PT

24

KEY WORDS

32

Body Composition, Hand Strength, DXA, Hip Fracture, Skeletal Muscle.

33

M AN U

31

LIST OF ABBREVIATIONS

35

Dual-Energy X-Ray Absorptiometry (DXA); Timed Up and Go test (TUG test); Appendicular

36

Lean Mass (aLM); Appendicular Lean Mass divided by height squared (aLM/ht2).

TE D

34

AC C

EP

37

2

ACCEPTED MANUSCRIPT

INTRODUCTION

39

A decrease in muscle mass accompanies aging, as shown by several longitudinal studies 1,2.

40

Muscle loss is associated with mobility disorders, increased risk of falls, reduced ability to

41

function in activities of daily living, loss of independence, frailty, and reduced life expectancy 3-7 .

42

However, the reduction in muscle mass found in older persons does not fully account for the loss

43

in muscle function. In particular, muscle strength does not depend solely on muscle mass, and the

44

relationship between mass and strength is not linear 8,9. This is why it is recommended to take

45

into account both muscle mass and function for the diagnosis of the geriatric syndrome named

46

sarcopenia, which is strongly associated with unfavourable clinical outcomes in older people 10,11.

47

Hip fracture often occurs in frail persons 12,13 and it is associated with a high risk of both death

48

and disability: after hip fracture there is a 8% to 36% excess mortality within one year 14 and

49

approximately 20% of hip fracture survivors require long-term nursing home care, whereas only

50

40% fully regain their pre-injury level of independence 15 .

51

Sarcopenia is thought to play a role in the genesis of hip fracture, because it enhances the risk of

52

falling

53

strength 15. Furthermore, the loss of muscle tissue is strictly linked with the loss of bone mass and

54

strength 17-19, and a high prevalence of low muscle mass together with its association with low

55

bone mass has actually been shown in hip fracture women 20,21.

56

Besides enhancing fracture risk, sarcopenia may increase the risk of poor functional outcome

57

following fracture occurrence, but the relationship between muscle mass, muscle strength, and

58

functional recovery has not been fully elucidated in this group of frail subjects.

59

Our aim was to investigate the relative contribution of muscle mass and handgrip strength in

60

predicting the functional outcome after hip fracture in women. We hypothesized that muscle

and nearly all hip fractures occur as a result of a fall in individuals with reduced bone

AC C

EP

16

TE D

M AN U

SC

RI PT

38

3

ACCEPTED MANUSCRIPT

61

strength but not mass could influence both ability to function in activities of daily living and

62

lower-limb performance following a hip fracture.

63

METHODS

65

Patients

66

The study was performed in a city with about one-million inhabitants. We evaluated 149 white

67

women without cognitive impairment (Mini Mental State Examination Test score > 23) and

68

without prevalent motor impairment due to neurologic diseases, consecutively admitted to our

69

physical medicine and rehabilitation division because of their first hip fracture during a 18-month

70

period (between January 2012 and June 2013). We focused on white patients because few non-

71

white elderly subjects live in our country. The women came from several orthopedic wards from

72

various hospitals and were referred for acute inpatient rehabilitation by the consultant physiatrists

73

of the orthopedic wards. The criteria agreed upon for selecting hip-fracture women to undergo

74

acute inpatient rehabilitation were: health conditions allowing a total of three hours of physical

75

therapy and/or occupational therapy daily; weight-bearing to tolerance on the fractured hip; a

76

potential high increase in ability to function in activities of daily living due to an intensive

77

rehabilitation regimen.

78

Six of the 149 patients we evaluated were excluded from our study because their hip fractures

79

resulted from either major trauma or cancer affecting bone. The remaining 143 women sustained

80

fractures that either were spontaneous or resulted from minimal trauma (trauma equal to or less

81

than a fall from a standing position). Two of these women were excluded from our study because

82

they had hip or knee arthroplasties that could alter dual-energy x-ray absorptiometry (DXA)

83

assessment. The remaining 141 subjects were asked to undergo a DXA scan. One of these 141

84

refused to undergo DXA assessment and was excluded from the study. The 140 remaining

AC C

EP

TE D

M AN U

SC

RI PT

64

4

ACCEPTED MANUSCRIPT

women gave their informed consent to participate in the study. Two women could not complete

86

inpatient rehabilitation because of acute concomitant diseases, whereas 15 women were not able

87

to walk independently at discharge and could not perform the Timed Up and Go (TUG) test. The

88

final study sample included 123 women whose data were included in the main analyses. Data

89

from the fifteen patients who were not able to ambulate independently at the end of the

90

rehabilitation course were recorded, and we included them in additional analyses on the available

91

outcome measures.

92

Our rehabilitation protocol included three hours a day for five days a week of physical exercise to

93

improve strength and balance, advice and training on the use of assistive devices, training in

94

mobility tasks and activities of daily living conducted by physical therapists and occupational

95

therapists. At least three hours during the stay in the rehabilitation hospital were dedicated by a

96

skilled occupational therapist to suggest targeted modifications of home environment and

97

behavioral changes to prevent falls. The criterion for discharge from rehabilitation was the

98

achievement of the highest possible Barthel index score (as judged by the responsible physiatrist)

99

in the three following items: dressing, transfers, and walking. Institution Revision Board

SC

M AN U

TE D

approval was obtained for the study protocol.

101

EP

100

RI PT

85

Outcome measures

103

DXA (Discovery Wi a) was used to measure whole and regional body composition. Appendicular

104

lean mass (aLM) was calculated as the sum of lean mass (LM) in arms and legs. Because metal

105

implants (prostheses, plates, screws, and nails) were reported to affect the regional assessment of

106

body composition with overestimation of LM

107

postoperative edema, we corrected aLM by substituting LM in unfractured leg for LM in

AC C

102

22

and to avoid the confounding role of

5

ACCEPTED MANUSCRIPT

fractured leg, as previously described 23,24 : corrected aLM = (LM in unfractured leg x 2) + LM in

109

arms.

110

LM cannot be interpreted without some indexing to body size: it is necessary to account for

111

height when comparisons are performed among different subjects. Height was assessed by a

112

standard method (with the patients standing) in the majority of the patients, whereas four

113

subjects, who could not keep the standing position, were measured supine. We accounted for

114

body size by dividing corrected aLM by height squared (aLM/ht2) 3,4,10,11.

115

Handgrip strength was measured with a Jamar hand dynamometer b on the same day of DXA scan

116

in each subject. Testing was performed with the participant in sitting position and with her

117

shoulder adducted and neutrally rotated, elbow flexed at 90° with the forearm in neutral position,

118

and wrists between 0° and 30° of flexion and between 0° and 15° of ulnar deviation. The best

119

recorded of three attempts of maximal voluntary contraction, performed at 1-minute intervals at

120

non-dominant arm, was considered for analyses.

121

In each patient we recorded age, number of medications in use, presence of pressure ulcers (stage

122

2 or higher according to the classification from the National Pressure Ulcer Advisory Panel),

123

number of concomitant diseases (all the prevalent diseases judged clinically relevant during the

124

length of stay), infections (at least one infection needing antibiotic therapy during the stay in

125

hospital), and time interval between fracture occurrence and DXA scan as potential confounders.

126

Functional evaluation, both at rehabilitation admission and at discharge from the rehabilitation

127

hospital, was assessed by skilled physiatrists by using the Barthel index (original version

128

unchanged). The functional index assesses basic activities of daily living; its score ranges from 0

129

(total dependence) to 100 (total independence). The physiatrists were not aware of the results of

130

DXA assessment and handgrip strength measure at the time of Barthel index score evaluation.

131

Barthel Index effectiveness was calculated using the following formula: [change in Barthel Index

AC C

EP

TE D

M AN U

SC

RI PT

108

6

ACCEPTED MANUSCRIPT

score during rehabilitation / (100 – Barthel Index score at admission to rehabilitation)] x 100. It

133

represented the proportion of potential improvement actually achieved by each subject.

134

A TUG test was performed at the end of the rehabilitation course to determine the amount of time

135

required for the subject to rise from a standard armchair, walk 3 meters away, turn, return, and sit

136

down again. One practice and one test trial were performed for each participant. Before testing, a

137

trained evaluator provided standardized verbal instructions regarding the test procedures. Time

138

was recorded with a stopwatch started on the command “ready-set-go” and stopped as the

139

participant sat down.

SC

RI PT

132

M AN U

140

Data analysis

142

We assessed the linear correlation between handgrip strength and Barthel index scores, Barthel

143

index effectiveness and TUG test by using a Spearman rank test, taking into account the non-

144

normal distribution of the outcome variables in our sample as shown by a Shapiro-Wilk test. The

145

same correlations were explored after substituting aLM/height2 for handgrip strength.

146

Additionally, both handgrip strength and aLM/height2 were included in a linear multiple

147

regression model as independent variables together with seven potential confounders: Barthel

148

index scores assessed on admission to rehabilitation, age, number of medications in use, presence

149

of pressure ulcers, number of comorbidities, presence of concomitant infections, and time

150

between fracture occurrence and DXA scan. The dependent variable in the regression model was

151

the Barthel index score on discharge. As the Barthel index score was non-normally distributed,

152

area transformation was performed, using the formula (r-1/2)/w, where w is the number of

153

observations and r is the rank 23,24. Linear multiple regression was also performed after

154

substituting either the Barthel index effectiveness or the TUG test (both normalized by area

155

transformation) for the Barthel index score. Following area transformation of the dependent

AC C

EP

TE D

141

7

ACCEPTED MANUSCRIPT

variables, the residuals were normally distributed in the regression models. Homoscedasticity was

157

verified by plotting the residuals against the predicted values: the variance of the residuals looked

158

homogeneous across levels of the predicted values. Collinearity diagnostics showed that the

159

percent of variance in each predictor that could not be accounted for by the other predictors was

160

always greater than 75% (no redundant predictors were found). We had no missing data.

161

The statistical package used was SPSS, version 14 c.

RI PT

156

SC

162

RESULTS

164

Descriptives for the 123 women are shown in Table I. At the end of the rehabilitation course the

165

median Barthel index score was 95 (interquartile range from 90 to 100) and the median Barthel

166

index effectiveness was 87.5 (interquartile range from 75 to 100).

167

We found a significant positive correlation between handgrip strength measured at the non-

168

dominant arm before rehabilitation and both the Barthel index score on discharge (ρ=0.50;

169

p<0.001) and the Barthel index effectiveness (ρ=0.45; p<0.001). Conversely, we found no

170

significant correlations between aLM/height2 assessed before rehabilitation and the Barthel index

171

score after rehabilitation (ρ=0.075; p=0.41) or the Barthel index effectiveness (ρ=0.06; p=0.53).

172

The significant association between grip strength (but not aLM/height2) and the functional score

173

or its effectiveness persisted after multiple adjustments (Table 2 and Table 3). The results did not

174

materially change when the fifteen patients who were not able to ambulate independently at the

175

end of the course of rehabilitation were included in the analyses (data not shown).

176

The median time to complete the TUG test at the end of inpatient rehabilitation was 32”

177

(interquartile range from 25” to 41.5”). We found a significant negative correlation between

178

handgrip strength measured at the non-dominant arm before rehabilitation and the time to

AC C

EP

TE D

M AN U

163

8

ACCEPTED MANUSCRIPT

perform the TUG test on discharge (ρ=-0.41; p<0.001). Conversely, we found no significant

180

correlations between aLM/height2 assessed before rehabilitation and the time to complete the

181

TUG test after rehabilitation (ρ=0.005; p=0.96). The significant association between grip strength

182

(but not aLM/height2) and the TUG test persisted after multiple adjustments (Table 4).

RI PT

179

183

DISCUSSION

185

Data shows that handgrip strength assessed before rehabilitation, but not aLM/height2 , was

186

significantly associated both with ability to function in activities of daily living and TUG test at

187

the end of inpatient rehabilitation in hip-fracture women.

188

Only two previous studies (both prospective with a one-year follow-up) assessed together the

189

prognostic roles of grip strength and aLM following a hip fracture. Visser et al. showed that the

190

changes in grip strength but not in aLM were significantly associated with the changes in a five-

191

item mobility score in 90 hip-fracture women 25.Wehren et al. found a significant positive

192

correlation between grip strength and ability to function in activities of daily living both at

193

concurrent measures and when grip strength assessment preceded the functional evaluation in 205

194

hip-fracture women, whereas the correlation between aLM and the functional score was very

195

weak 26.

196

The originality of our data rests on three aspects. Firstly, we included aLM and grip strength as

197

independent variables in regression models together with seven potential confounders and we

198

show that the prognostic role of grip strength was independent of the covariates. The second

199

original aspect is the focus on post-acute inpatient rehabilitation: we show that grip strength on

200

admission predicted recovery on discharge. Thirdly, the outcome measures were assessed by

201

health professionals, and not self-reported as in the previous studies 25,26. Overall, our results are

AC C

EP

TE D

M AN U

SC

184

9

ACCEPTED MANUSCRIPT

202

in line with those from the two previous reports, and emphasize the prognostic role of grip

203

strength but not of aLM on functional recovery in hip-fracture women. This finding is consistent

204

with data from four previous studies that examined either aLM or grip strength (not both of them

205

in the same sample): aLM did not predict ability to function in activities of daily living

206

whereas handgrip strength did 27,28 . In particular, Di Monaco et al. showed no significant

207

associations between aLM/height2 and short-term Barthel index scores in two samples of hip-

208

fracture women 23,24. Beloosesky et al. showed that grip strength measured 7-10 days

209

postoperatively in 105 hip-fracture women was significantly associated with the Functional

210

Independence Measure score assessed six months later 27, and Savino et al. showed that handgrip

211

strength measured at hospital admission in 504 hip-fracture patients significantly predicted

212

walking recovery at a one-year follow-up 28.

213

The prognostic role of grip strength we show after hip fracture is consistent with data from

214

several prospective studies performed in the general population on various outcome measures

RI PT

,

M AN U

SC

29,30

TE D

215

23,24

. Also, the predictive capability of grip strength is consistent with few reports on the 31

prognostic role of lower-limb strength and power assessed following a fracture of the hip

217

given the well-established strict correlation between grip and total-body muscle strength 10. These

218

findings, together with the well-known loss of strength following fractures with mobility

219

reduction

220

fracture patients. Indeed, favorable effects on several outcome measures were reported for

221

strengthening exercises performed either alone 33-37 or as a part of multi-dimensional interventions 38

, supply the rationale for interventions aiming at increasing muscle strength in hip-

AC C

222

32

,

EP

216

after hip fracture, although some negative results on the effectiveness of resistance training are 39

223

available

, and several issues still need to be addressed to define optimal treatment regimens

224

and settings 40,41 .

10

ACCEPTED MANUSCRIPT

An explanation for the lack of significant associations between aLM and function in hip-fracture

226

women rests on DXA failure in capturing the qualitative alterations of the aging muscle,

227

including fat infiltration, a decreased proportion of type II fibers, increased connective tissue,

228

metabolic changes, and variations to the muscle spindle 42-44. DXA is the measure of choice for

229

the assessment of aLM 45,46, but it simply assesses mass, without taking into account quality,

230

which may play a key role in affecting the functional properties of the aging muscle. Indeed, two

231

factors that are thought to contribute to muscle impairment in older people, i.e. inflammation and

232

vitamin D depletion, were actually shown to be associated with function, but not with DXA-

233

assessed muscle mass following a fracture of the hip 24,47 . The changes in muscle quality and

234

their confounding role may be greater in women than in men 48, as suggested by the discrepancy

235

between the negative studies in women 23-26 and a single study of hip-fracture men that showed a

236

significant association between aLM assessed by DXA and ability to function in activities of

237

daily living 48. Further studies in men with the concomitant assessment of muscle mass and

238

strength are needed to confirm this hypothesis.

TE D

M AN U

SC

RI PT

225

239

Study limitations

241

We evaluated white women without cognitive impairment admitted to a single rehabilitation

242

hospital in Italy, who were surgically operated on, who agreed to be studied, who could undergo a

243

DXA scan, and who regained independent ambulation. As a consequence, our data is not

244

generalizable to the overall population of hip-fracture patients. LM rapidly decreases after hip

245

fracture, and this change may affect the relationship between LM assessed post-operatively and

246

function. In a prospective study 49, the percentage of decrease in LM was 6.4% two months after

247

fracture occurrence. In our study, the loss of LM is expected to be lower because we performed

248

DXA assessment about three weeks after fracture occurrence. Anyway, we included time

AC C

EP

240

11

ACCEPTED MANUSCRIPT

between fracture and DXA scan in multiple regression analysis as an independent variable to

250

adjust our results for this potential confounder. Overall, we adjusted our results for seven

251

prognostic factors, but we did not collect data on other potential confounders, including pre-

252

injury level of independence, nutritional state, prevalent vertebral fractures, depression, pain at

253

various times during rehabilitation and when the outcome measures were performed, social

254

support, fear of falling, and balance confidence 31,50. Furthermore, we did not take into account

255

adherence to rehabilitation including the actual number of hours each patient was active in

256

rehabilitation every day. We performed outcome measures at the end of acute inpatient

257

rehabilitation. It would be useful to verify the long-term role of grip strength and aLM as

258

outcome predictors by follow-up assessments for a period from 6 months to 2 years. However,

259

acute inpatient rehabilitation accounts for most functional recovery in improving patients’

260

mobility and activities of daily living 51,52.

261

There is wide variability in the protocols used for grip strength assessment and variation in

262

approach can affect the values measured 53. One potential confounder is dominance, which is not

263

accounted for when either right or left hand is assessed. We took into account this source of bias

264

by assessing the non-dominant arm in all the patients.

265

Conclusions

266

Despite limitations of our study, data supports the prognostic role of grip strength at admission to

267

inpatient rehabilitation after a hip fracture. Models aimed at predicting the functional outcome in

268

hip-fracture survivors may benefit from grip-strength assessment, although a number of factors

269

have been shown to play a prognostic role

270

further examined. Together with other prognostic factors, grip strength may be helpful in

271

selecting patients for proper rehabilitation protocols and settings. Comprehensive evaluations of

272

all the known prognostic factors, including grip strength, in large samples of hip-fracture patients

AC C

EP

TE D

M AN U

SC

RI PT

249

31

, and independency of grip strength should be

12

ACCEPTED MANUSCRIPT

should be performed in prospective studies to define the best predictive models. The unfavorable

274

prognostic role of low muscle strength, emphasizes the relevancy of resistance training to

275

optimize recovery. Conversely, the role of low DXA-assessed aLM in hip fracture women is not

276

defined at present, and its assessment cannot be recommended in everyday practice. Further

277

research studies are needed to elucidate the role of muscle mass and its dramatic changes in

278

affecting the functional recovery following a hip fracture.

RI PT

273

AC C

EP

TE D

M AN U

SC

279

13

ACCEPTED MANUSCRIPT

280 281

REFERENCES 1. Visser M, Pahor M, Tylavsky F, Kritchevsky SB, Cauley JA, Newman AB, Blunt BA,

283

Harris TB. One- and two-year change in body composition as measured by DXA in a

284

population-based cohort of older men and women. J Appl Physiol 2003;94:2368-74.

RI PT

282

2. Fantin F, Di Francesco V, Fontana G, Zivelonghi A, Bissoli L, Zoico E, Rossi A,

286

Micciolo R, Bosello O, Zamboni M. Longitudinal body composition changes in old men

287

and women: interrelationship with worsening disability. J Gerontol A Biol Sci Med Sci

288

2007;62:1375-81.

M AN U

SC

285

289

3. Baumgartner RN, Koehler KM, Gallagher D, Romero L, Heymsfield SB, Ross RR, Garry

290

PJ, Lindeman RD. Epidemiology of sarcopenia among the elderly in New Mexico. Am J

291

Epidemiol 1998;147:755-63.

295 296 297

TE D

294

Epidemiology of sarcopenia. J Am Geriatr Soc 2000;48:625-30. 5. Janssen I. Influence of sarcopenia on the development of physical disability: the Cardiovascular Health Study. J Am Geriatr Soc 2006;54:56-62.

EP

293

4. Melton LJ 3rd, Khosla S, Crowson CS, O'Connor MK, O'Fallon WM, Riggs BL.

6. Bunout D, De la Maza MP, Barrera G, Leiva L, Hirsch S. Association between sarcopenia and mortality in healthy older people. Australas J Ageing 2011;30:89–92.

AC C

292

298

7. Frisoli A Jr, Chaves PH, Ingham SJ, Fried LP. Severe osteopenia and osteoporosis,

299

sarcopenia, and frailty status in community-dwelling older women: results from the

300

Women's Health and Aging Study (WHAS) II. Bone 2011;48:952-7.

301 302

8. Goodpaster BH, Park SW, Harris TB, Kritchevsky SB, Nevitt M, Schwartz AV, Simonsick EM, Tylavsky FA, Visser M, Newman AB. The loss of skeletal muscle

14

ACCEPTED MANUSCRIPT

303

strength, mass, and quality in older adults: the health, aging and body composition study. J

304

Gerontol A Biol Sci Med Sci 2006;61:1059-64.

305

9. Janssen I, Baumgartner RN, Ross R, Rosenberg IH, Roubenoff R. Skeletal muscle cutpoints associated with elevated physical disability risk in older men and women. Am J

307

Epidemiol 2004;159:413-21.

308

RI PT

306

10. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, Boirie Y, Cederholm T, Landi F, Martin FC, Michel JP, Rolland Y, Schneider SM, Topinková E, Vandewoude M, Zamboni

310

M; European Working Group on Sarcopenia in Older People. Sarcopenia: European

311

consensus on definition and diagnosis: Report of the European Working Group on

312

Sarcopenia in Older People. Age Ageing 2010;39:412-23.

M AN U

SC

309

11. Lang T, Streeper T, Cawthon P, Baldwin K, Taaffe DR, Harris TB. Sarcopenia: etiology,

314

clinical consequences, intervention, and assessment. Osteoporos Int 2010;21:543-59.

315

12. Hershkovitz A, Polatov I, Beloosesky Y, Brill S. Factors affecting mortality of frail hip-

318 319 320 321 322 323 324

13. Di Monaco M. Rehabilitation after hip fracture in older people. Eur J Phys Rehabil Med 2011;47:253-5.

EP

317

fractured elderly patients. Arch Gerontol Geriatr 2010;51:113-6 .

14. Abrahamsen B, van Staa T, Ariely R, Olson M, Cooper C. Excess mortality following hip fracture: a systematic epidemiological review. Osteoporos Int 2009;20:1633-50.

AC C

316

TE D

313

15. Physician’s Guide to Prevention and Treatment of Osteoporosis. Washington, DC, the National Osteoporosis Foundation, 2013. 16. Montero-Fernandez N, Serra-Rexach JA. Role of exercise on sarcopenia in the elderly. Eur J Phys Rehabil Med 2013;49:131-43.

15

ACCEPTED MANUSCRIPT

325

17. Invernizzi M, Cisari C. Sarcopenia and muscular modifications in disabling pathologies of

326

the elderly from the physical and rehabilitation medicine: point of view. Eur J Phys

327

Rehabil Med 2013;49:107-9. 18. Polidoulis I, Beyene J, Cheung AM. The effect of exercise on pQCT parameters of bone

RI PT

328 329

structure and strength in postmenopausal women—a systematic review and meta-analysis

330

of randomized controlled trials. Osteoporos Int 2012;23:39-51.

333

SC

332

19. Cederholm T, Cruz-Jentoft AJ, Maggi S. Sarcopenia and fragility fractures. Eur J Phys Rehabil Med 2013;49:111-7.

20. Di Monaco M, Vallero F, Di Monaco R, Tappero R. Prevalence of sarcopenia and its

M AN U

331

334

association with osteoporosis in 313 older women following a hip fracture. Arch Gerontol

335

Geriatr 2011;52:71-4.

336

21. Hida T, Ishiguro N, Shimokata H, Sakai Y, Matsui Y, Takemura M, Terabe Y, Harada A. High prevalence of sarcopenia and reduced leg muscle mass in Japanese patients

338

immediately after a hip fracture. Geriatr Gerontol Int 2013;13:413-20.

TE D

337

22. Giangregorio LM, Webber CE. Effects of metal implants on whole-body dual-energy x-

340

ray absorptiometry measurements of bone mineral content and body composition. Can

341

Assoc Radiol J 2003;54:305-9.

343 344

23. Di Monaco M, Vallero F, Di Monaco R, Tappero R, Cavanna A. Muscle mass and

AC C

342

EP

339

functional recovery in women with hip fracture. Am J Phys Med Rehabil 2006;85:209-15. 24. Di Monaco M, Castiglioni C, Vallero F, Di Monaco R, Tappero R. Appendicular lean

345

mass does not mediate the significant association between vitamin D status and functional

346

outcome in hip-fracture women. Arch Phys Med Rehabil 2011; 92:271-6.

16

ACCEPTED MANUSCRIPT

347

25. Visser M, Harris TB, Fox KM, Hawkes W, Hebel JR, Yahiro JY, Michael R, Zimmerman

348

SI, Magaziner J. Change in muscle mass and muscle strength after a hip fracture:

349

relationship to mobility recovery. J Gerontol A Biol Sci Med Sci 2000;55:M434-40. 26. Wehren L, Hawkes W, Hebel JR, Orwig DL, Magaziner J. Bone mineral density, soft

351

tissue body composition, strength, and functioning after hip fracture. J Gerontol A Biol

352

Sci Med Sci 2005;60:80-4.

355

SC

354

27. Beloosesky Y, Weiss A, Manasian M, Salai M. Handgrip strength of the elderly after hip fracture repair correlates with functional outcome. Disabil Rehabil 2010;32:367-73. 28. Savino

E, Martini

E, Lauretani

F, Pioli

G, Zagatti

M AN U

353

RI PT

350

AM, Frondini

C, Pellicciotti

356

F, Giordano A, Ferrari A, Nardelli A, Davoli ML, Zurlo A, Lunardelli ML, Volpato S.

357

Handgrip strength predicts persistent walking recovery after hip fracture surgery. Am J

358

Med 2013;126:1068-1075.

360

29. Rantanen T, Guralnik JM, Foley D, Masaki K, Leveille S, Curb JD, White L. Midlife

TE D

359

hand grip strength as a predictor of old age disability. JAMA 1999;281:558-60. 30. Ling CH, Taekema D, de Craen AJ, Gussekloo J, Westendorp RG, Maier AB. Handgrip

362

strength and mortality in the oldest old population: the Leiden 85-plus study. CMAJ

363

2010;182:429-35.

365 366

31. Kristensen MT. Factors affecting functional prognosis of patients with hip fracture. Eur J

AC C

364

EP

361

Phys Rehabil Med 2011;47:257-64. 32. Kubota M, Uchida K, Kokubo Y, Shimada S, Matsuo H, Yayama T, Miyazaki T, Takeura

367

N, Yoshida A, Baba H. Changes in gait pattern and hip muscle strength after open

368

reduction and internal fixation of acetabular fracture. Arch Phys Med Rehabil

369

2012;93:2015-21.

17

ACCEPTED MANUSCRIPT

370

33. Binder EF, Brown M, Sinacore DR, Steger-May K, Yarasheski KE, Schechtman KB.

371

Effects of extended outpatient rehabilitation after hip fracture: a randomized controlled

372

trial. JAMA 2004;292:837-46. 34. Mangione KK, Craik RL, Palombaro KM, Tomlinson SS, Hofmann MT. Home-based leg-

RI PT

373 374

strengthening exercise improves function 1 year after hip fracture: a randomized

375

controlled study. J Am Geriatr Soc 2010;58:1911-7.

35. Sylliaas H, Brovold T, Wyller TB, Bergland A. Progressive strength training in older

377

patients after hip fracture: a randomised controlled trial. Age Ageing. 2011;40:221-7.

379 380

36. Sylliaas H, Brovold T, Wyller TB, Bergland A. Prolonged strength training in older

M AN U

378

SC

376

patients after hip fracture: a randomised controlled trial. Age Ageing. 2012;41:206-12. 37. Edgren J, Rantanen T, Heinonen A, Portegijs E, Alen M, Kiviranta I, Kallinen M, Sipila S. Effects of progressive resistance training on physical disability among older

382

community-dwelling people with history of hip fracture. Aging Clin Exp Res

383

2012;24:171-5.

TE D

381

38. Singh NA, Quine S, Clemson LM, Williams EJ, Williamson DA, Stavrinos TM, Grady

385

JN, Perry TJ, Lloyd BD, Smith EU, Singh MA. Effects of high-intensity progressive

386

resistance training and targeted multidisciplinary treatment of frailty on mortality and

387

nursing home admissions after hip fracture: a randomized controlled trial. J Am Med Dir

388

Assoc 2012;13:24-30.

AC C

389

EP

384

39. Orwig DL, Hochberg M, Yu-Yahiro J, Resnick B, Hawkes WG, Shardell M, Hebel

390

JR, Colvin P, Miller RR, Golden J, Zimmerman S, Magaziner J. Delivery and outcomes

391

of a yearlong home exercise program after hip fracture: a randomized controlled trial.

392

Arch Intern Med 2011;171:323-31.

18

ACCEPTED MANUSCRIPT

393

40. Mallinson T, Deutsch A, Bateman J, Tseng HY, Manheim L, Almagor O, Heinemann AW. Comparison of Discharge Functional Status After Rehabilitation in Skilled Nursing,

395

Home Health, and Medical Rehabilitation Settings for Patients After Hip Fracture Repair.

396

Arch Phys Med Rehabil 2013; doi: 10.1016/j.apmr.2013.05.031.

397 398

RI PT

394

41. Sherrington C, Tiedemann A, Cameron I. Physical exercise after hip fracture: an evidence overview. Eur J Phys Rehabil Med 2011;47:297-307.

42. Goodpaster BH, Carlson CL, Visser M, Kelley DE, Scherzinger A, Harris TB, Stamm E,

400

Newman AB. Attenuation of skeletal muscle and strength in the elderly: the Health ABC

401

study. J Appl Physiol 2001;90:2157-65.

M AN U

402

SC

399

43. Visser M, Kritchevsky SB, Goodpaster BH, Newman AB, Nevitt M, Stamm E, Harris TB. Leg muscle mass and composition in relation to lower extremity performance in men and

404

women aged 70 to 79: the Health, Aging and Body Composition study. J Am Geriatr Soc

405

2002; 50:897-904.

406

TE D

403

44. Liu JX, Eriksson PO, Thornell LE, Pedrosa-Domellof F. Fiber content and myosin heavy chain composition of muscle spindles in aged human biceps brachii. J Histochem

408

Cytochem 2005;53:445-54.

410 411

45. Albanese CV, Diessel E, Genant HK. Clinical applications of body composition measurements using DXA. J Clin Densitom 2003;6:75-85.

AC C

409

EP

407

46. Chumlea WC, Cesari M, Evans WJ, Ferrucci L, Fielding RA, Pahor M, Studenski

412

S, Vellas B; International Working Group on Sarcopenia Task Force Members.

413

Sarcopenia: designing phase IIB trials. J Nutr Health Aging 2011;15:450-5.

414

47. Miller RR, Shardell MD, Hicks GE, Cappola AR, Hawkes WG, Yu-Yahiro JA, Magaziner

415

J. Association between interleukin-6 and lower extremity function after hip fracture--the

416

role of muscle mass and strength. J Am Geriatr Soc 2008;56:1050-6. 19

ACCEPTED MANUSCRIPT

417

48. Di Monaco M, Vallero F, Di Monaco R, Tappero R, Cavanna A. Muscle mass and

418

functional recovery in hip-fracture men. Am J Phys Med Rehabil 2007;86:818-25. 49. Fox KM, Magaziner J, Hawkes WG, Yu-Yahiro J, Hebel JR, Zimmerman SI, Holder L,

420

Michael R. Loss of bone density and lean body mass after hip fracture. Osteoporos Int

421

2000;11:31-5.

422

RI PT

419

50. Portegijs E, Edgren J, Salpakoski A, Kallinen M, Rantanen T, Alen M, Kiviranta

I, Sihvonen S, Sipila S. Balance confidence was associated with mobility and balance

424

performance in older people with fall-related hip fracture: a cross-sectional study. Arch

425

Phys Med Rehabil 2012;93:2340-6. 51.

Fiedler RC, Granger CV, Post LA. The Uniform Data System for Medical Rehabilitation: report of first admissions for 1998. Am J Phys Med Rehabil 2000;79:87-92.

427 428

M AN U

426

SC

423

52.

Di Monaco M, Vallero F, De Lauso L, De Toma E, Tappero R, Cavanna A. Incident falls impair ability to function in hip-fracture survivors: a prospective study of 95 elderly

430

women. Arch Gerontol Geriatr 2009;48:397-400.

TE D

429

53. Roberts HC, Denison HJ, Martin HJ, Patel HP, Syddall H, Cooper C, Sayer AA. A review

432

of the measurement of grip strength in clinical and epidemiological studies: towards a

433

standardised approach. Age Ageing 2011;40:423-9.

AC C

434

EP

431

20

ACCEPTED MANUSCRIPT

54. SUPPLIERS’ LIST

436

a. Hologic, Inc., Bedford, MA, USA

437

b. Patterson Medical, Inc., Bolingbrook, IL, USA

438

c. SPSS, Inc., Chicago, IL, USA

AC C

EP

TE D

M AN U

SC

RI PT

435

21

ACCEPTED MANUSCRIPT

Table 1: characteristics of the 123 women included in the study.

79.2 ± 7.4

Body weight, kg (mean ± SD)

60.4 ± 12.6

RI PT

Age: years, (mean ± SD)

Body height, cm (mean ± SD)

157.9 ± 6.3 54/46

Pressure ulcers, (%) Infections during the stay in hospital, (%)

Number of medications in use, (mean ± SD)

M AN U

Number of concomitant diseases, (mean ± SD)

SC

Hip-fracture type: trochanteric/cervical, (%)

Barthel index score at admission to inpatient rehabilitation, (median; IQR) Length of stay in hospital, days (median; IQR)

11% 56

2.8 ± 1.5 4.0 ± 2.4 60; from 45 to 65 34.7; from 30 to 40 15.3 ± 5.7

Corrected appendicular lean mass / height2, g/m2 (mean ± SD)

5052 ± 742

Time between fracture occurrence and DXA scan, days (mean ± SD)

21.1 ± 8.7

EP

TE D

Handgrip strength at non-dominant arm, kg (mean ± SD)

4 (from 2 to 6)

AC C

Time between fracture occurrence and surgery, days (median; IQR)

SD = Standard Deviation IQR = Interquartile

1

ACCEPTED MANUSCRIPT

Table 2: multiple regression analysis model. Association between handgrip strength measured

Partial correlation

P

Grip strength assessed on admission

0.22

0.016

Appendicular lean mass / height2 assessed on admission

0.01

Barthel index on admission

0.35

Age

-0.31

Number of medications in use

-0.25

0.006

Pressure ulcers

-0.04

0.67

0.01

0.89

-0.11

0.22

-0.09

0.31

Number of concomitant diseases Infections during the stay in hospital

M AN U

SC

Variables

RI PT

before rehabilitation and Barthel index scores assessed at the end of inpatient rehabilitation.

<0.001 <0.001

TE D

Days between fracture occurrence and basal assessment

0.92

The dependent variable was the Barthel index score assessed at the end of inpatient rehabilitation (after normalization by area transformation). The independent variables were those listed in the

EP

Table. Pressure ulcers, and infections during the stay in hospital were conventionally attributed a value of one (the absence of the two conditions was conventionally attributed a value of 0). R2 =

AC C

0.43; F = 22.13; p<0.001.

2

ACCEPTED MANUSCRIPT

Table 3: multiple regression analysis model. Association between handgrip strength measured before rehabilitation and Barthel index effectiveness. P

Grip strength assessed on admission

0.23

0.011

Appendicular lean mass / height2 assessed on admission

0.05

0.62

Barthel index on admission

0.12

0.18

Age

-0.31

0.001

-0.26

0.003

-0.01

0.88

-0.04

0.67

-0.07

0.43

-0.06

0.51

RI PT

Partial correlation

SC

Variables

Pressure ulcers Number of concomitant diseases Infections during the stay in hospital

M AN U

Number of medications in use

TE D

Days between fracture occurrence and basal assessment

The dependent variable was the Barthel index effectiveness (after normalization by area transformation). The independent variables were those listed in the Table. Pressure ulcers, and

EP

infections during the stay in hospital were conventionally attributed a value of one (the absence of

AC C

the two conditions was conventionally attributed a value of 0). R2 = 0.30; F = 16.92; p<0.001.

3

ACCEPTED MANUSCRIPT

Table 4: multiple regression analysis model. Association between handgrip strength measured before rehabilitation and time to complete Up and Go Test at the end of inpatient rehabilitation. Partial correlation

P

Grip strength assessed on admission

-0.25

0.005

Appendicular lean mass / height2 assessed on admission

0.09

Barthel index on admission

-0.11

Age

0.32

Number of medications in use

0.12

0.20

Pressure ulcers

0.50

0.58

0.06

0.49

-0.30

0.74

-0.13

0.15

Number of concomitant diseases Infections during the stay in hospital

M AN U

SC

RI PT

Variables

0.22

<0.001

TE D

Days between fracture occurrence and basal assessment

0.30

The dependent variable was the time needed to complete the Up and Go Test assessed at the end of inpatient rehabilitation (after normalization by area transformation). The independent variables

EP

were those listed in the Table. Pressure ulcers, and infections during the stay in hospital were conventionally attributed a value of one (the absence of the two conditions was conventionally

AC C

attributed a value of 0). R2 = 0.25; F = 19.8; p<0.001.

4