- Email: [email protected]
Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study
Manual Therapy xxx (2016) 1e6
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Technical and measurement report
Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study Anitra Wilson a, Julie A. Hides b, Leigh Blizzard a, Michele Callisaya a, c, Andrew Cooper b, Velandai K. Srikanth c, a, Tania Winzenberg a, d, * a
Menzies Institute for Medical Research, University of Tasmania, 17 Liverpool St, Hobart, Tasmania 7000, Australia Centre for Musculoskeletal Research, Mary MacKillop Institute for Health Research, Australian Catholic University, 631 Stanley St, Woolloongabba, Queensland 4102, Australia c Department of Medicine, School of Clinical Sciences at Monash Health, Monash University, 246 Clayton Road, Clayton, Melbourne, Victoria 3168, Australia d Faculty of Health, University of Tasmania, 17 Liverpool St, Hobart, Tasmania 7000, Australia b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 August 2015 Received in revised form 8 January 2016 Accepted 8 January 2016
Measurement reliability of the L4/5 lumbar multifidus (MF) muscles is high in older adults, but few studies have investigated measurement reliability of the abdominal and upper lumbar MF muscles in this age group. Objectives: To determine measurement reliability of abdominal and lumbar MF muscles from a single ultrasound (US) image in older adults. Methods: Resting thickness of rectus abdominis and obliquus externus, resting and contracted thickness of obliquus internus, transversus abdominis and lumbar MF, and resting cross-sectional area (CSA) of MF levels (L2-5) were obtained from US images of 92 community-dwelling older adults (aged 65e89 years). Measurements of images were undertaken by an experienced rater and repeated 7e10 days later for intra-rater, and by a second expert rater for inter-rater calculations. Intra-rater reliability was estimated for all muscles. Inter-rater reliability was estimated for all abdominal muscles and for L5 multifidus. Reliability was estimated by intraclass correlation coefficients (ICC). Results: Intra-rater ICC(3,1) and inter-rater ICC(2,1) of resting thickness measures of all muscles and CSA of MF were 0.86. The ICCs for percentage thickness change were 0.76 for the abdominal muscles, and 0.42 for MF. Conclusions: Measurement reliability of US imaging for abdominal and MF muscle thickness and MF CSA was high, and consistent with previous findings for younger adults. Reliability of percentage thickness change was lower suggesting caution is needed when using this as an outcome measure or study factor among older adults. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Abdominal muscles Paraspinal muscles Ultrasound imaging Reliability Older adults
1. Introduction Ultrasound imaging (USI) is increasingly used in physiotherapy to quantify abdominal and lumbar muscle performance (Critchley and Coutts, 2002), assess clinical outcomes and provide biofeedback during functional re-education (Hides et al., 2001; Macedo et al., 2009). Reliability of ultrasound (US) measurements is well established for younger adults (Rankin et al., 2006; Hides et al., 2007; Norasteh et al., 2007; Wallwork et al., 2007; Teyhen et al.,
* Corresponding author. Menzies Institute for Medical Research, University of Tasmania, 17 Liverpool St, Hobart, Tasmania 7000, Australia. E-mail address: [email protected] (T. Winzenberg).
2011), but information for older adults remains sparse (Stetts et al., 2009; Sions et al., 2014). Sions examined L4/5 multifidus (MF) only, whilst Stetts examined the lateral abdominals e no paper to date has examined the rectus abdominis or the upper levels of lumbar MF in the older adult. Measuring older adults may be problematic due to age-related changes including increases in the subcutaneous fat layer, or fatty infiltrates between the muscle fibres, which may lead to an increase in echogenicity (Whittaker, 2007). For clinical and research purposes it is important to know if these measurements are sufficiently consistent to have low measurement error under these circumstances. This study aimed to determine the intra- and inter-rater reliability for the measurement of muscle thickness from stored US
http://dx.doi.org/10.1016/j.math.2016.01.004 1356-689X/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
2
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
images of the abdominal and lumbar MF muscles at rest and on contraction, and cross-sectional area (CSA) of the lumbar MF muscles in a sample of community-dwelling older adults. 2. Methods 2.1. Participants Participants, aged between 60 and 86 years in 2006, were recruited by random selection from the southern Tasmanian electoral roll for a longitudinal cohort study conducted between 2005 and 2011 examining associations between cognition and gait. The only exclusion criteria were living in residential care and presence of any contraindications to Magnetic Resonance Imaging as required by the larger study. The images of participants in the longitudinal study were randomly selected for assessment in this reliability sub-study until 40 complete sets for each muscle group were obtained, which required images from 92 participants. The study was approved by the Human Research Ethics Committee (Tasmania) Network. Participants provided informed consent. 2.2. Instrumentation and image capture The abdominal and lumbar MF muscles were imaged on both sides in brightness (B) mode using a Phillips HDI 5000 US machine (ATL Ultrasound, Washington, USA) with a hand held curved array transducer. For each participant, USI was performed by one of two operators, a medical specialist (obstetrician) and a senior physiotherapist, both of whom received specialised training in ultrasound imaging of older adults from JH. 2.2.1. Imaging the abdominal muscles Participants lay supine with pillows under their knees and hips flexed to 45 . Images for rectus abdominis (RA) and obliquus externus (OE) muscles were taken at rest, and images for obliquus internus (OI) and transversus abdominis (TrA) taken at rest and with contraction. Standardised instructions were given for muscle contraction asking participants to ‘take a relaxed breath in and out, hold the breath out and then draw in the lower abdomen without moving the spine” (Hides et al., 2007). Imaging of RA was performed over the anterior abdominal wall immediately above and slightly lateral to the umbilicus, with the muscle centred on the screen (Rankin et al., 2006). Imaging of the lateral abdominal muscles was performed along a line midway between the iliac crest and the inferior angle of the rib cage, and level with the umbilicus. The US transducer was aligned perpendicularly to the anterolateral muscles with the anterior fascial insertion of the TrA muscle aligned approximately 2.0 cm from the medial edge of the US image when the participant was relaxed (Ferreira et al., 2004; Hides et al., 2007). 2.2.2. Imaging the lumbar multifidus Participants lay in prone with pillows placed under the abdomen to minimise the lumbar lordosis. The lumbar spinous processes were palpated and marked with a pen prior to imaging and then confirmed by imaging them in the sagittal plane using the sacrum as a reference point. Participants were instructed how to perform an isometric contraction of the MF muscle after exhaling. To capture MF muscle thickness the transducer was placed in a longitudinal plane over the spinous process of the marked lumbar spine and moved laterally with the transducer angled medially to capture the zygoapophyseal joints in the lower part of the screen (Wallwork et al., 2009). Image capture for CSA of the relaxed lumbar MF muscles was taken in the transverse plane at each vertebral level from L2eL5.
2.3. Image analysis Images were saved and measured off-line using ImageJ software 1.36b (http://imagej.nih.gov/ij/) by research physiotherapists who were not involved in image capture. Following 20 h of supervised training by an experienced physiotherapist (JH) with more than 10 years of USI experience, three physiotherapists performed the intra-rater measurements with one rater per muscle grouping (abdominals, lumbar MF linear measurements and lumbar MF CSA). Each image was measured by the same rater on two occasions 7e10 days apart. The order of images was randomly assigned and the rater blinded to image identification and their previous result. For inter-rater reliability, measurements were performed by three expert physiotherapists each with more than 3 years of experience in USI. They were blinded to the original measurer's results. For logistical reasons, inter-rater reliability of lumbar MF measures was only performed at L5. Measurement of RA was made at the widest part of the muscle and perpendicular to the orientation of the fascia between its inner borders (Fig. 1). A split screen (with resting and contracted images side-by-side) was used to measure thickness of the lateral abdominals and lumbar MF muscles between the inner borders of the fascia at the widest point of the muscle belly (Figs. 2 and 3). CSA measurements of the lumbar MF muscles were taken by tracing the inner borders of the muscle at each vertebral level (Hides et al., 2008) (Fig. 4).
2.4. Data analysis Percentage thickness change was calculated as 100*(thickness of muscle contracted e thickness at rest)/thickness at rest. Measurement reliability was assessed from the mean difference between pair of measurements, intraclass correlation coefficients (ICC) using (3,1) form for intra-rater and (2,1) form for inter-rater reliability measurements (Shrout and Fleiss, 1979), with the standard error of measurement (SEM) (De Vet et al., 2006), and the minimal detectable change (MDC) (Weir, 2005) calculated. Bland and Altman plots were inspected to identify any systematic patterns in the differences related to the size of the measurements (Bland and Altman, 1986). Stata version 12 was used to analyse the data.
Fig. 1. Ultrasound image of Rectus Abdominis (RA) muscle. SST ¼ superficial soft tissue. Image taken at end of expiration, and measured between inner borders of fascia at the widest part.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
3
Fig. 2. Split screen image of lateral abdominal muscles. Left screen: muscle at rest, right screen: muscle contracted with ADIM at end of expiration. Abbreviations: SST: superficial soft tissue, EO: external oblique, IO: internal oblique, TrA: transversus abdominis.
3. Results
3.2. Inter-rater reliability
The 92 participants (62% males) had a mean age of 75.9 (6.9) years, mean body mass of 27.2 (3.9) kg/m2, and 18% reported current low back pain.
The ICCs for inter-rater reliability of all abdominal muscle thickness measures were at least 0.97. Those for L5 MF thickness and CSA were somewhat lower at 0.86 or greater (Table 3). The inter-rater ICCs for percentage thickness change of OE (0.76), OI (0.96), TrA (0.92) and L5 MF (0.42) were variable, with sizeable differences on average (and relatively high SEMs and MDCs) between the original and expert rater. Inspection of Bland and Altman plots of all muscle measurements revealed no systematic pattern of variability in measurement differences across the range of measurement (data not shown).
3.1. Intra-rater reliability Intra-rater reliability for thickness of the abdominal muscles (Table 1) and for MF thickness and CSA (Table 2) were all high with ICCs greater than 0.90 for MF measures and at least 0.96 for abdominal muscles. The intra-rater ICCs for abdominal and MF percentage thickness change were lower.
Fig. 3. Split screen image of muscle thickness of multifidus (L4 and L5). Left screen: muscle at rest, right screen: muscle contracted with voluntary isometric contraction at end of expiration.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
4
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
4. Discussion Despite the anticipated challenges of reading US images of older adults due to age-related changes in muscle morphology, we demonstrated high intra- and inter-rater reliability for measurements of abdominal (ICC 0.97) and MF muscle thickness (ICC 0.86), and MF CSA (ICC 0.92). These reliability indices are similar to those for measurements of younger populations (Hides et al., 2007; Hebert et al., 2009; Teyhen et al., 2011) suggesting that the use of USI of these muscles produces measurements in the older adult with reliability comparable to that in younger adults. Few studies have reported the reliability of measuring US images of older adults. One study (mean age 72 years) reported interrater reliability of TrA, OI and OE measurements (ICC(3,1) of 0.92, 0.93 and 0.94 respectively) that were slightly lower than ours (Stetts et al., 2009). Another study, (age range: 60e85 years), using an average of three measurements, reported a higher inter-rater ICC (0.98 compared to 0.86 in the current study) for L5 thickness (Sions et al., 2014). Very few studies have reported on resting thickness measurement of RA images, none in older adults. In younger adults, intra-rater ICC of 0.85 and >0.98 have been reported (Rankin et al., 2006; Norasteh et al., 2007), which are comparable to the current study.
Fig. 4. Bilateral image of multifidus at L4 level for cross-sectional area measurement. Right image has left multifidus muscle outlined with Image J tracing tool.
Table 1 Intra-rater measurement reliability of abdominal muscle thickness. Muscle
RA OE OI
TrA
n
Rest Rest Rest Contracted Percentage thickness change Rest Contracted Percentage thickness change
41 40 41 41 41 41 41 41
Measure
Remeasure
Difference
Mean (SD)
Mean (SD)
Mean (SD)
0.69 0.41 0.72 0.82 13.47 0.38 0.54 49.59
0.68 0.41 0.71 0.82 9.70 0.39 0.53 46.42
0.003 0.003 0.009 0.006 3.78 0.004 0.006 3.16
(0.19) (0.18) (0.30) (0.39) (37.10) (0.15) (0.19) (43.69)
(0.20) (0.18) (0.30) (0.39) (34.58) (0.15) (0.18) (42.99)
(0.032) (0.036) (0.035) (0.034) (21.63) (0.025) (0.036) (16.67)
SEM
MDC
Intraclass correlation ICC(3,1)
95% CI
0.02 0.03 0.02 0.02 6.75 0.02 0.03 11.79
0.06 0.07 0.07 0.07 18.71 0.05 0.07 32.88
0.99 0.98 0.99 0.99 0.96 0.99 0.98 0.93
(0.98e0.99) (0.96e0.99) (0.98e0.99) (0.99e0.99) (0.93e0.98) (0.97e0.99) (0.96e0.99) (0.87e0.96)
Abbreviations: RA: rectus abdominis, OE: obliquus externus, OI: obliquus internus; TrA: transversus abdominis; rest: resting thickness (cm); contracted: thickness (cm); percentage thickness change: 100*(contract e rest)/rest, reported as % (SD). SD: standard deviation; SEM: standard error of measurement (cm); MDC: minimal detectable change (cm); ICC: intraclass correlation coefficient (95% confidence interval). All measurements were taken on the right side.
Table 2 Intra-rater measurement reliability of lumbar multifidus (MF) measures. Muscle
MF muscle thickness (cm) L2/3 Rest Contracted Percentage thickness change L3/4 Rest Contracted Percentage thickness change L4/5 Rest Contracted Percentage thickness change L5/S1 Rest Contracted Percentage thickness change MF cross-sectional area (cm2) taken at rest L2 L3 L4 L5
n
Measure
Remeasure
Difference
SEM
Mean (SD)
Mean (SD)
Mean (SD)
40 40 40 40 40 40 40 40 40 40 40 40
2.80 2.83 1.40 2.54 2.59 2.25 3.26 3.28 0.42 2.91 3.00 2.53
(0.66) (0.67) (5.94) (0.61) (0.63) (4.28) (0.57) (0.58) (5.56) (0.55) (0.61) (6.02)
2.76 2.85 3.22 2.54 2.57 1.42 3.30 3.34 1.24 2.94 3.04 3.25
(0.65) (0.68) (6.17) (0.62) (0.62) (5.42) (0.59) (0.63) (6.17) (0.56) (0.62) (5.59)
0.038 0.016 1.83 0.003 0.022 0.82 0.032 0.063 0.83 0.026 0.047 0.72
(0.139) (0.27) (5.79) (0.105) (0.077) (4.13) (0.179) (0.272) (4.83) (0.156) (0.161) (4.63)
0.09 0.09 4.10 0.08 0.05 2.92 0.13 0.19 3.41 0.11 0.11 3.27
40 40 39 39
2.59 3.33 4.41 5.04
(1.18) (1.11) (1.56) (1.94)
2.65 3.33 4.26 4.94
(1.11) (0.99) (1.34) (1.68)
0.060 0.003 0.144 0.100
(0.297) (0.287) (0.526) (0.530)
0.21 0.20 0.37 0.37
MDC
Intraclass correlation ICC(3,1)
95% CI
0.27 0.25 11.36 0.21 0.15 8.10 0.35 0.53 9.46 0.31 0.32 9.07
0.98 0.98 0.54 0.99 0.99 0.64 0.95 0.90 0.66 0.96 0.97 0.68
(0.95e0.99) (0.97e0.99) (0.26e0.74) (0.97e0.99) (0.98e1.00) (0.39e0.80) (0.91e0.98) (0.81e0.95) (0.43e0.81) (0.92e0.98) (0.93e0.98) (0.46e0.83)
0.59 0.56 1.03 1.03
0.97 0.96 0.93 0.96
(0.94e0.98) (0.93e0.98) (0.88e0.96) (0.92e0.98)
Abbreviations: MF: multifidus; L2eL5 lumbar vertebral level; rest: resting thickness (cm); contracted: thickness (cm). Percentage thickness change: 100*(contract e rest)/rest, reported as % (SD). SD: standard deviation; SEM: standard error of measurement (cm); MDC: minimal detectable change (cm); ICC: intraclass correlation coefficient (95% confidence interval). All measurements were taken on the right side.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
5
Table 3 Inter-rater measurement reliability of abdominal and multifidus (MF) muscles. Muscle
Abdominal muscle thickness (cm) RA Rest OE Rest OI Rest Contracted Percentage thickness change TrA Rest Contracted Percentage thickness change L5/S1 MF muscle thickness (cm) L5/S1 Rest Contracted Percentage thickness change Cross-sectional area (cm2) of L5 MF at rest L5
n
40 20 40 40 40 40 40 40
Measure
Remeasure
Difference
Mean (SD)
Mean (SD)
Mean (SD)
0.69 0.38 0.70 0.80 15.58 0.38 0.53 50.18
0.69 0.38 0.70 0.80 17.61 0.38 0.53 47.26
0.012 0.007 0.007 0.007 2.03 0.003 0.007 2.92
(0.20) (0.20) (0.29) (0.39) (34.52) (0.14) (0.18) (44.06)
(0.20) (0.20) (0.30) (0.39) (35.44) (0.14) (0.18) (43.18)
SEM
MDC
Intraclass correlation ICC(3,1)
95% CI
(0.031) (0.043) (0.034) (0.035) (9.65) (0.023) (0.043) (16.80)
0.02 0.03 0.03 0.03 6.89 0.02 0.03 11.92
0.06 0.08 0.07 0.07 19.10 0.04 0.07 33.03
0.99 0.98 0.99 0.99 0.96 0.99 0.98 0.92
(0.98e0.99) (0.94e0.99) (0.99e1.00) (0.99e1.00) (0.93e0.98) (0.98e0.99) (0.96e0.99) (0.87e0.96)
40 40 40
2.89 (0.50) 2.96 (0.55) 2.49 (5.85)
2.84 (0.55) 2.94 (0.55) 3.58 (5.20)
0.049 (0.280) 0.027 (0.275) 1.10 (5.93)
0.20 0.20 4.21
0.55 0.54 11.67
0.86 0.87 0.42
(0.75e0.92) (0.77e0.93) (0.14e0.65)
40
5.04 (1.94)
5.05 (1.74)
0.008 (0.726)
0.51
1.40
0.92
(0.86e0.96)
Rater 1 ¼ original rater, Rater 2 ¼ expert rater. Percentage thickness change: 100*(contract e rest)/rest, reported as % (SD). Abbreviations: RA: rectus abdominis, OE: obliquus externus; OI: obliquus internus; TrA: transversus abdominis; MF: multifidus; L5: fifth lumbar vertebra; rest: resting thickness (cm); contracted: contracted thickness (cm). SD: standard deviation; SEM: standard error of measurement (cm); MDC: minimal detectable change (cm); ICC: intraclass correlation coefficient (95% confidence interval). All measurements were taken on the right side and only L5 was measured for the MF thickness and CSA.
Percentage thickness change between the contracted and resting states, which combines the variability of both measurements if the errors are not perfectly correlated, demonstrated lower reliability than the primary measurements. These measurements may also be confounded by difficulties in repeatable landmark location due to age-related or injury changes in the muscles. In view of this lower reliability, the use of percentage thickness change cannot be recommended. The current study has several limitations. We measured lumbar MF images of muscles whilst participants were lying prone, which may be a position difficult to attain for some older adults. We did so to be able to compare our results with other published results. The measurements were made from a single image, rather than from repeated multiple images, as time constraints in clinical practice may limit measurements to a single performance. Two studies of younger adults (Wallwork et al., 2007; Djordjevic et al., 2014) demonstrated that a single measurement can be as reliable as the average of three, but in another study, imaging muscles multiple times and averaging the results improved measurement precision (Koppenhaver et al., 2009). Nonetheless, the high reliability indices we obtained from one image compare favourably to results of other studies averaging multiple measurements. All muscle measurements were analysed on the right side only, but studies of younger adults have found no significant differences between sides (Rankin et al., 2006; Stetts et al., 2009). The inter-rater analysis was performed by the expert physiotherapist only on images from the L5 vertebral level and the results may not transfer to the higher levels (L2 to L4). In this study, standardised procedures were used for both image capture and measurement, and all raters had specific training. This may limit generalisability of the study findings to measurements made by untrained staff. 5. Conclusion Measurement reliability of US imaging for abdominal and MF muscle thickness and MF CSA was high, and consistent with previous findings for younger adults. Reliability of percentage thickness change was lower suggesting caution is needed when using this as an outcome measure or study factor among older adults.
Acknowledgements Thanks to the team of raters in Queensland, all TASCOG participants and the Royal Hobart Hospital Research Foundation (Grant No:12-024).
References Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;327:307e10. Critchley DJ, Coutts FJ. Abdominal muscle function in chronic low back pain patients: measurement with real-time ultrasound scanning. Physiotherapy 2002;88:322e32. De Vet HCW, Terwee CB, Knol DL, Bouter LM. When to use agreement versus reliability measures. J Clin Epidemiol 2006;59:1033e9. Djordjevic O, Djordjevic A, Konstantinovic L. Interrater and intrarater reliability of transverse abdominal and lumbar multifidus muscle thickness in subjects with and without low back pain. J Orthop Sports Phys Ther 2014;44:979e88. Ferreira P, Ferreira ML, Hodges P. Changes in recruitment of the abdominal muscles in people with low back pain: ultrasound measurement of muscle activity. Spine 2004;29:2560e6. Hebert JJ, Koppenhaver SL, Parent EC, Fritz JM. A systematic review of the reliability of rehabilitative ultrasound imaging for the quantitative assessment of the abdominal and lumbar trunk muscles. Spine 2009;34:E848. Hides J, Gilmore C, Stanton W, Bohlscheid E. Multifidus size and symmetry among chronic LBP and healthy asymptomatic subjects. Man Ther 2008;13:43e9. Hides J, Jull G, Richardson C. Long-term effects of specific stabilizing exercises for first-episode low back pain. Spine 2001;26:e243e8. Hides J, Miokovic T, Belavy D, Stanton W, Richardson C. Ultrasound imaging assessment of abdominal muscle function during drawing-in of the abdominal wall: an intrarater reliability study. J Orthop Sports Phys Ther 2007;37:480e6. Koppenhaver S, Parent E, Teyhen D, Hebert J, Fritz J. The effect of averaging multiple trials on measurement error during ultrasound imaging of transversus abdominis and lumbar multifidus muscles in individuals with low back pain. J Orthop Sports Phys Ther 2009;39:604e11. Macedo LG, Maher CG, Latimer J, McAuley JH. Motor control exercise for persistent, nonspecific low back pain: a systematic review. Phys Ther 2009;89:9e25. Norasteh A, Ebrahimi E, Salavati M, Rafiei J, Abbasnejad E. Reliability of B-mode ultrasonography for abdominal muscles in asymptomatic and patients with acute low back pain. J Bodyw Mov Ther 2007;11:17e20. Rankin G, Stokes M, Newham DJ. Abdominal muscle size and symmetry in normal subjects. Muscle Nerve 2006;34:320e6. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420. Sions JM, Velasco T, Teyhen D, Hicks GE. Ultrasound imaging: intraexaminer and interexaminer reliability for multifidus muscle thickness assessment in adults aged 60 to 85 years versus younger adults. J Orthop Sports Phys Ther 2014;44: 425e34. Stetts DM, Freund JE, Allison SC, Carpenter G. A rehabilitative ultrasound imaging investigation of lateral abdominal muscle thickness in healthy aging adults. J Geriatr Phys Ther 2009;32:60e6.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
6
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
Teyhen DS, George SZ, Dugan JL, Williamson J, Neilson BD, Childs JD. Inter-rater reliability of ultrasound imaging of the trunk musculature among novice raters. J Ultrasound Med 2011;30:347e56. Wallwork TL, Hides JA, Stanton WR. Intrarater and interrater reliability of assessment of lumbar multifidus muscle thickness using rehabilitative ultrasound imaging. J Orthop Sports Phys Ther 2007;37:608e12.
Wallwork TL, Stanton W, Freke M, Hides J. The effect of chronic low back pain on size and contraction of the lumbar multifidis muscle. Man Ther 2009;14:496e500. Weir J. Quantifying test-retest reliability using the intraclass correlation coefficient. J Strength Cond Res 2005;19:231e40. Whittaker J. Ultrasound imaging for rehabilitation of the lumbopelvic region: a clinical approach. London, UK: Elsevier Health Sciences; 2007. p. 9.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Technical and measurement report
Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study Anitra Wilson a, Julie A. Hides b, Leigh Blizzard a, Michele Callisaya a, c, Andrew Cooper b, Velandai K. Srikanth c, a, Tania Winzenberg a, d, * a
Menzies Institute for Medical Research, University of Tasmania, 17 Liverpool St, Hobart, Tasmania 7000, Australia Centre for Musculoskeletal Research, Mary MacKillop Institute for Health Research, Australian Catholic University, 631 Stanley St, Woolloongabba, Queensland 4102, Australia c Department of Medicine, School of Clinical Sciences at Monash Health, Monash University, 246 Clayton Road, Clayton, Melbourne, Victoria 3168, Australia d Faculty of Health, University of Tasmania, 17 Liverpool St, Hobart, Tasmania 7000, Australia b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 August 2015 Received in revised form 8 January 2016 Accepted 8 January 2016
Measurement reliability of the L4/5 lumbar multifidus (MF) muscles is high in older adults, but few studies have investigated measurement reliability of the abdominal and upper lumbar MF muscles in this age group. Objectives: To determine measurement reliability of abdominal and lumbar MF muscles from a single ultrasound (US) image in older adults. Methods: Resting thickness of rectus abdominis and obliquus externus, resting and contracted thickness of obliquus internus, transversus abdominis and lumbar MF, and resting cross-sectional area (CSA) of MF levels (L2-5) were obtained from US images of 92 community-dwelling older adults (aged 65e89 years). Measurements of images were undertaken by an experienced rater and repeated 7e10 days later for intra-rater, and by a second expert rater for inter-rater calculations. Intra-rater reliability was estimated for all muscles. Inter-rater reliability was estimated for all abdominal muscles and for L5 multifidus. Reliability was estimated by intraclass correlation coefficients (ICC). Results: Intra-rater ICC(3,1) and inter-rater ICC(2,1) of resting thickness measures of all muscles and CSA of MF were 0.86. The ICCs for percentage thickness change were 0.76 for the abdominal muscles, and 0.42 for MF. Conclusions: Measurement reliability of US imaging for abdominal and MF muscle thickness and MF CSA was high, and consistent with previous findings for younger adults. Reliability of percentage thickness change was lower suggesting caution is needed when using this as an outcome measure or study factor among older adults. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Abdominal muscles Paraspinal muscles Ultrasound imaging Reliability Older adults
1. Introduction Ultrasound imaging (USI) is increasingly used in physiotherapy to quantify abdominal and lumbar muscle performance (Critchley and Coutts, 2002), assess clinical outcomes and provide biofeedback during functional re-education (Hides et al., 2001; Macedo et al., 2009). Reliability of ultrasound (US) measurements is well established for younger adults (Rankin et al., 2006; Hides et al., 2007; Norasteh et al., 2007; Wallwork et al., 2007; Teyhen et al.,
* Corresponding author. Menzies Institute for Medical Research, University of Tasmania, 17 Liverpool St, Hobart, Tasmania 7000, Australia. E-mail address: [email protected] (T. Winzenberg).
2011), but information for older adults remains sparse (Stetts et al., 2009; Sions et al., 2014). Sions examined L4/5 multifidus (MF) only, whilst Stetts examined the lateral abdominals e no paper to date has examined the rectus abdominis or the upper levels of lumbar MF in the older adult. Measuring older adults may be problematic due to age-related changes including increases in the subcutaneous fat layer, or fatty infiltrates between the muscle fibres, which may lead to an increase in echogenicity (Whittaker, 2007). For clinical and research purposes it is important to know if these measurements are sufficiently consistent to have low measurement error under these circumstances. This study aimed to determine the intra- and inter-rater reliability for the measurement of muscle thickness from stored US
http://dx.doi.org/10.1016/j.math.2016.01.004 1356-689X/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
2
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
images of the abdominal and lumbar MF muscles at rest and on contraction, and cross-sectional area (CSA) of the lumbar MF muscles in a sample of community-dwelling older adults. 2. Methods 2.1. Participants Participants, aged between 60 and 86 years in 2006, were recruited by random selection from the southern Tasmanian electoral roll for a longitudinal cohort study conducted between 2005 and 2011 examining associations between cognition and gait. The only exclusion criteria were living in residential care and presence of any contraindications to Magnetic Resonance Imaging as required by the larger study. The images of participants in the longitudinal study were randomly selected for assessment in this reliability sub-study until 40 complete sets for each muscle group were obtained, which required images from 92 participants. The study was approved by the Human Research Ethics Committee (Tasmania) Network. Participants provided informed consent. 2.2. Instrumentation and image capture The abdominal and lumbar MF muscles were imaged on both sides in brightness (B) mode using a Phillips HDI 5000 US machine (ATL Ultrasound, Washington, USA) with a hand held curved array transducer. For each participant, USI was performed by one of two operators, a medical specialist (obstetrician) and a senior physiotherapist, both of whom received specialised training in ultrasound imaging of older adults from JH. 2.2.1. Imaging the abdominal muscles Participants lay supine with pillows under their knees and hips flexed to 45 . Images for rectus abdominis (RA) and obliquus externus (OE) muscles were taken at rest, and images for obliquus internus (OI) and transversus abdominis (TrA) taken at rest and with contraction. Standardised instructions were given for muscle contraction asking participants to ‘take a relaxed breath in and out, hold the breath out and then draw in the lower abdomen without moving the spine” (Hides et al., 2007). Imaging of RA was performed over the anterior abdominal wall immediately above and slightly lateral to the umbilicus, with the muscle centred on the screen (Rankin et al., 2006). Imaging of the lateral abdominal muscles was performed along a line midway between the iliac crest and the inferior angle of the rib cage, and level with the umbilicus. The US transducer was aligned perpendicularly to the anterolateral muscles with the anterior fascial insertion of the TrA muscle aligned approximately 2.0 cm from the medial edge of the US image when the participant was relaxed (Ferreira et al., 2004; Hides et al., 2007). 2.2.2. Imaging the lumbar multifidus Participants lay in prone with pillows placed under the abdomen to minimise the lumbar lordosis. The lumbar spinous processes were palpated and marked with a pen prior to imaging and then confirmed by imaging them in the sagittal plane using the sacrum as a reference point. Participants were instructed how to perform an isometric contraction of the MF muscle after exhaling. To capture MF muscle thickness the transducer was placed in a longitudinal plane over the spinous process of the marked lumbar spine and moved laterally with the transducer angled medially to capture the zygoapophyseal joints in the lower part of the screen (Wallwork et al., 2009). Image capture for CSA of the relaxed lumbar MF muscles was taken in the transverse plane at each vertebral level from L2eL5.
2.3. Image analysis Images were saved and measured off-line using ImageJ software 1.36b (http://imagej.nih.gov/ij/) by research physiotherapists who were not involved in image capture. Following 20 h of supervised training by an experienced physiotherapist (JH) with more than 10 years of USI experience, three physiotherapists performed the intra-rater measurements with one rater per muscle grouping (abdominals, lumbar MF linear measurements and lumbar MF CSA). Each image was measured by the same rater on two occasions 7e10 days apart. The order of images was randomly assigned and the rater blinded to image identification and their previous result. For inter-rater reliability, measurements were performed by three expert physiotherapists each with more than 3 years of experience in USI. They were blinded to the original measurer's results. For logistical reasons, inter-rater reliability of lumbar MF measures was only performed at L5. Measurement of RA was made at the widest part of the muscle and perpendicular to the orientation of the fascia between its inner borders (Fig. 1). A split screen (with resting and contracted images side-by-side) was used to measure thickness of the lateral abdominals and lumbar MF muscles between the inner borders of the fascia at the widest point of the muscle belly (Figs. 2 and 3). CSA measurements of the lumbar MF muscles were taken by tracing the inner borders of the muscle at each vertebral level (Hides et al., 2008) (Fig. 4).
2.4. Data analysis Percentage thickness change was calculated as 100*(thickness of muscle contracted e thickness at rest)/thickness at rest. Measurement reliability was assessed from the mean difference between pair of measurements, intraclass correlation coefficients (ICC) using (3,1) form for intra-rater and (2,1) form for inter-rater reliability measurements (Shrout and Fleiss, 1979), with the standard error of measurement (SEM) (De Vet et al., 2006), and the minimal detectable change (MDC) (Weir, 2005) calculated. Bland and Altman plots were inspected to identify any systematic patterns in the differences related to the size of the measurements (Bland and Altman, 1986). Stata version 12 was used to analyse the data.
Fig. 1. Ultrasound image of Rectus Abdominis (RA) muscle. SST ¼ superficial soft tissue. Image taken at end of expiration, and measured between inner borders of fascia at the widest part.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
3
Fig. 2. Split screen image of lateral abdominal muscles. Left screen: muscle at rest, right screen: muscle contracted with ADIM at end of expiration. Abbreviations: SST: superficial soft tissue, EO: external oblique, IO: internal oblique, TrA: transversus abdominis.
3. Results
3.2. Inter-rater reliability
The 92 participants (62% males) had a mean age of 75.9 (6.9) years, mean body mass of 27.2 (3.9) kg/m2, and 18% reported current low back pain.
The ICCs for inter-rater reliability of all abdominal muscle thickness measures were at least 0.97. Those for L5 MF thickness and CSA were somewhat lower at 0.86 or greater (Table 3). The inter-rater ICCs for percentage thickness change of OE (0.76), OI (0.96), TrA (0.92) and L5 MF (0.42) were variable, with sizeable differences on average (and relatively high SEMs and MDCs) between the original and expert rater. Inspection of Bland and Altman plots of all muscle measurements revealed no systematic pattern of variability in measurement differences across the range of measurement (data not shown).
3.1. Intra-rater reliability Intra-rater reliability for thickness of the abdominal muscles (Table 1) and for MF thickness and CSA (Table 2) were all high with ICCs greater than 0.90 for MF measures and at least 0.96 for abdominal muscles. The intra-rater ICCs for abdominal and MF percentage thickness change were lower.
Fig. 3. Split screen image of muscle thickness of multifidus (L4 and L5). Left screen: muscle at rest, right screen: muscle contracted with voluntary isometric contraction at end of expiration.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
4
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
4. Discussion Despite the anticipated challenges of reading US images of older adults due to age-related changes in muscle morphology, we demonstrated high intra- and inter-rater reliability for measurements of abdominal (ICC 0.97) and MF muscle thickness (ICC 0.86), and MF CSA (ICC 0.92). These reliability indices are similar to those for measurements of younger populations (Hides et al., 2007; Hebert et al., 2009; Teyhen et al., 2011) suggesting that the use of USI of these muscles produces measurements in the older adult with reliability comparable to that in younger adults. Few studies have reported the reliability of measuring US images of older adults. One study (mean age 72 years) reported interrater reliability of TrA, OI and OE measurements (ICC(3,1) of 0.92, 0.93 and 0.94 respectively) that were slightly lower than ours (Stetts et al., 2009). Another study, (age range: 60e85 years), using an average of three measurements, reported a higher inter-rater ICC (0.98 compared to 0.86 in the current study) for L5 thickness (Sions et al., 2014). Very few studies have reported on resting thickness measurement of RA images, none in older adults. In younger adults, intra-rater ICC of 0.85 and >0.98 have been reported (Rankin et al., 2006; Norasteh et al., 2007), which are comparable to the current study.
Fig. 4. Bilateral image of multifidus at L4 level for cross-sectional area measurement. Right image has left multifidus muscle outlined with Image J tracing tool.
Table 1 Intra-rater measurement reliability of abdominal muscle thickness. Muscle
RA OE OI
TrA
n
Rest Rest Rest Contracted Percentage thickness change Rest Contracted Percentage thickness change
41 40 41 41 41 41 41 41
Measure
Remeasure
Difference
Mean (SD)
Mean (SD)
Mean (SD)
0.69 0.41 0.72 0.82 13.47 0.38 0.54 49.59
0.68 0.41 0.71 0.82 9.70 0.39 0.53 46.42
0.003 0.003 0.009 0.006 3.78 0.004 0.006 3.16
(0.19) (0.18) (0.30) (0.39) (37.10) (0.15) (0.19) (43.69)
(0.20) (0.18) (0.30) (0.39) (34.58) (0.15) (0.18) (42.99)
(0.032) (0.036) (0.035) (0.034) (21.63) (0.025) (0.036) (16.67)
SEM
MDC
Intraclass correlation ICC(3,1)
95% CI
0.02 0.03 0.02 0.02 6.75 0.02 0.03 11.79
0.06 0.07 0.07 0.07 18.71 0.05 0.07 32.88
0.99 0.98 0.99 0.99 0.96 0.99 0.98 0.93
(0.98e0.99) (0.96e0.99) (0.98e0.99) (0.99e0.99) (0.93e0.98) (0.97e0.99) (0.96e0.99) (0.87e0.96)
Abbreviations: RA: rectus abdominis, OE: obliquus externus, OI: obliquus internus; TrA: transversus abdominis; rest: resting thickness (cm); contracted: thickness (cm); percentage thickness change: 100*(contract e rest)/rest, reported as % (SD). SD: standard deviation; SEM: standard error of measurement (cm); MDC: minimal detectable change (cm); ICC: intraclass correlation coefficient (95% confidence interval). All measurements were taken on the right side.
Table 2 Intra-rater measurement reliability of lumbar multifidus (MF) measures. Muscle
MF muscle thickness (cm) L2/3 Rest Contracted Percentage thickness change L3/4 Rest Contracted Percentage thickness change L4/5 Rest Contracted Percentage thickness change L5/S1 Rest Contracted Percentage thickness change MF cross-sectional area (cm2) taken at rest L2 L3 L4 L5
n
Measure
Remeasure
Difference
SEM
Mean (SD)
Mean (SD)
Mean (SD)
40 40 40 40 40 40 40 40 40 40 40 40
2.80 2.83 1.40 2.54 2.59 2.25 3.26 3.28 0.42 2.91 3.00 2.53
(0.66) (0.67) (5.94) (0.61) (0.63) (4.28) (0.57) (0.58) (5.56) (0.55) (0.61) (6.02)
2.76 2.85 3.22 2.54 2.57 1.42 3.30 3.34 1.24 2.94 3.04 3.25
(0.65) (0.68) (6.17) (0.62) (0.62) (5.42) (0.59) (0.63) (6.17) (0.56) (0.62) (5.59)
0.038 0.016 1.83 0.003 0.022 0.82 0.032 0.063 0.83 0.026 0.047 0.72
(0.139) (0.27) (5.79) (0.105) (0.077) (4.13) (0.179) (0.272) (4.83) (0.156) (0.161) (4.63)
0.09 0.09 4.10 0.08 0.05 2.92 0.13 0.19 3.41 0.11 0.11 3.27
40 40 39 39
2.59 3.33 4.41 5.04
(1.18) (1.11) (1.56) (1.94)
2.65 3.33 4.26 4.94
(1.11) (0.99) (1.34) (1.68)
0.060 0.003 0.144 0.100
(0.297) (0.287) (0.526) (0.530)
0.21 0.20 0.37 0.37
MDC
Intraclass correlation ICC(3,1)
95% CI
0.27 0.25 11.36 0.21 0.15 8.10 0.35 0.53 9.46 0.31 0.32 9.07
0.98 0.98 0.54 0.99 0.99 0.64 0.95 0.90 0.66 0.96 0.97 0.68
(0.95e0.99) (0.97e0.99) (0.26e0.74) (0.97e0.99) (0.98e1.00) (0.39e0.80) (0.91e0.98) (0.81e0.95) (0.43e0.81) (0.92e0.98) (0.93e0.98) (0.46e0.83)
0.59 0.56 1.03 1.03
0.97 0.96 0.93 0.96
(0.94e0.98) (0.93e0.98) (0.88e0.96) (0.92e0.98)
Abbreviations: MF: multifidus; L2eL5 lumbar vertebral level; rest: resting thickness (cm); contracted: thickness (cm). Percentage thickness change: 100*(contract e rest)/rest, reported as % (SD). SD: standard deviation; SEM: standard error of measurement (cm); MDC: minimal detectable change (cm); ICC: intraclass correlation coefficient (95% confidence interval). All measurements were taken on the right side.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
5
Table 3 Inter-rater measurement reliability of abdominal and multifidus (MF) muscles. Muscle
Abdominal muscle thickness (cm) RA Rest OE Rest OI Rest Contracted Percentage thickness change TrA Rest Contracted Percentage thickness change L5/S1 MF muscle thickness (cm) L5/S1 Rest Contracted Percentage thickness change Cross-sectional area (cm2) of L5 MF at rest L5
n
40 20 40 40 40 40 40 40
Measure
Remeasure
Difference
Mean (SD)
Mean (SD)
Mean (SD)
0.69 0.38 0.70 0.80 15.58 0.38 0.53 50.18
0.69 0.38 0.70 0.80 17.61 0.38 0.53 47.26
0.012 0.007 0.007 0.007 2.03 0.003 0.007 2.92
(0.20) (0.20) (0.29) (0.39) (34.52) (0.14) (0.18) (44.06)
(0.20) (0.20) (0.30) (0.39) (35.44) (0.14) (0.18) (43.18)
SEM
MDC
Intraclass correlation ICC(3,1)
95% CI
(0.031) (0.043) (0.034) (0.035) (9.65) (0.023) (0.043) (16.80)
0.02 0.03 0.03 0.03 6.89 0.02 0.03 11.92
0.06 0.08 0.07 0.07 19.10 0.04 0.07 33.03
0.99 0.98 0.99 0.99 0.96 0.99 0.98 0.92
(0.98e0.99) (0.94e0.99) (0.99e1.00) (0.99e1.00) (0.93e0.98) (0.98e0.99) (0.96e0.99) (0.87e0.96)
40 40 40
2.89 (0.50) 2.96 (0.55) 2.49 (5.85)
2.84 (0.55) 2.94 (0.55) 3.58 (5.20)
0.049 (0.280) 0.027 (0.275) 1.10 (5.93)
0.20 0.20 4.21
0.55 0.54 11.67
0.86 0.87 0.42
(0.75e0.92) (0.77e0.93) (0.14e0.65)
40
5.04 (1.94)
5.05 (1.74)
0.008 (0.726)
0.51
1.40
0.92
(0.86e0.96)
Rater 1 ¼ original rater, Rater 2 ¼ expert rater. Percentage thickness change: 100*(contract e rest)/rest, reported as % (SD). Abbreviations: RA: rectus abdominis, OE: obliquus externus; OI: obliquus internus; TrA: transversus abdominis; MF: multifidus; L5: fifth lumbar vertebra; rest: resting thickness (cm); contracted: contracted thickness (cm). SD: standard deviation; SEM: standard error of measurement (cm); MDC: minimal detectable change (cm); ICC: intraclass correlation coefficient (95% confidence interval). All measurements were taken on the right side and only L5 was measured for the MF thickness and CSA.
Percentage thickness change between the contracted and resting states, which combines the variability of both measurements if the errors are not perfectly correlated, demonstrated lower reliability than the primary measurements. These measurements may also be confounded by difficulties in repeatable landmark location due to age-related or injury changes in the muscles. In view of this lower reliability, the use of percentage thickness change cannot be recommended. The current study has several limitations. We measured lumbar MF images of muscles whilst participants were lying prone, which may be a position difficult to attain for some older adults. We did so to be able to compare our results with other published results. The measurements were made from a single image, rather than from repeated multiple images, as time constraints in clinical practice may limit measurements to a single performance. Two studies of younger adults (Wallwork et al., 2007; Djordjevic et al., 2014) demonstrated that a single measurement can be as reliable as the average of three, but in another study, imaging muscles multiple times and averaging the results improved measurement precision (Koppenhaver et al., 2009). Nonetheless, the high reliability indices we obtained from one image compare favourably to results of other studies averaging multiple measurements. All muscle measurements were analysed on the right side only, but studies of younger adults have found no significant differences between sides (Rankin et al., 2006; Stetts et al., 2009). The inter-rater analysis was performed by the expert physiotherapist only on images from the L5 vertebral level and the results may not transfer to the higher levels (L2 to L4). In this study, standardised procedures were used for both image capture and measurement, and all raters had specific training. This may limit generalisability of the study findings to measurements made by untrained staff. 5. Conclusion Measurement reliability of US imaging for abdominal and MF muscle thickness and MF CSA was high, and consistent with previous findings for younger adults. Reliability of percentage thickness change was lower suggesting caution is needed when using this as an outcome measure or study factor among older adults.
Acknowledgements Thanks to the team of raters in Queensland, all TASCOG participants and the Royal Hobart Hospital Research Foundation (Grant No:12-024).
References Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;327:307e10. Critchley DJ, Coutts FJ. Abdominal muscle function in chronic low back pain patients: measurement with real-time ultrasound scanning. Physiotherapy 2002;88:322e32. De Vet HCW, Terwee CB, Knol DL, Bouter LM. When to use agreement versus reliability measures. J Clin Epidemiol 2006;59:1033e9. Djordjevic O, Djordjevic A, Konstantinovic L. Interrater and intrarater reliability of transverse abdominal and lumbar multifidus muscle thickness in subjects with and without low back pain. J Orthop Sports Phys Ther 2014;44:979e88. Ferreira P, Ferreira ML, Hodges P. Changes in recruitment of the abdominal muscles in people with low back pain: ultrasound measurement of muscle activity. Spine 2004;29:2560e6. Hebert JJ, Koppenhaver SL, Parent EC, Fritz JM. A systematic review of the reliability of rehabilitative ultrasound imaging for the quantitative assessment of the abdominal and lumbar trunk muscles. Spine 2009;34:E848. Hides J, Gilmore C, Stanton W, Bohlscheid E. Multifidus size and symmetry among chronic LBP and healthy asymptomatic subjects. Man Ther 2008;13:43e9. Hides J, Jull G, Richardson C. Long-term effects of specific stabilizing exercises for first-episode low back pain. Spine 2001;26:e243e8. Hides J, Miokovic T, Belavy D, Stanton W, Richardson C. Ultrasound imaging assessment of abdominal muscle function during drawing-in of the abdominal wall: an intrarater reliability study. J Orthop Sports Phys Ther 2007;37:480e6. Koppenhaver S, Parent E, Teyhen D, Hebert J, Fritz J. The effect of averaging multiple trials on measurement error during ultrasound imaging of transversus abdominis and lumbar multifidus muscles in individuals with low back pain. J Orthop Sports Phys Ther 2009;39:604e11. Macedo LG, Maher CG, Latimer J, McAuley JH. Motor control exercise for persistent, nonspecific low back pain: a systematic review. Phys Ther 2009;89:9e25. Norasteh A, Ebrahimi E, Salavati M, Rafiei J, Abbasnejad E. Reliability of B-mode ultrasonography for abdominal muscles in asymptomatic and patients with acute low back pain. J Bodyw Mov Ther 2007;11:17e20. Rankin G, Stokes M, Newham DJ. Abdominal muscle size and symmetry in normal subjects. Muscle Nerve 2006;34:320e6. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420. Sions JM, Velasco T, Teyhen D, Hicks GE. Ultrasound imaging: intraexaminer and interexaminer reliability for multifidus muscle thickness assessment in adults aged 60 to 85 years versus younger adults. J Orthop Sports Phys Ther 2014;44: 425e34. Stetts DM, Freund JE, Allison SC, Carpenter G. A rehabilitative ultrasound imaging investigation of lateral abdominal muscle thickness in healthy aging adults. J Geriatr Phys Ther 2009;32:60e6.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004
6
A. Wilson et al. / Manual Therapy xxx (2016) 1e6
Teyhen DS, George SZ, Dugan JL, Williamson J, Neilson BD, Childs JD. Inter-rater reliability of ultrasound imaging of the trunk musculature among novice raters. J Ultrasound Med 2011;30:347e56. Wallwork TL, Hides JA, Stanton WR. Intrarater and interrater reliability of assessment of lumbar multifidus muscle thickness using rehabilitative ultrasound imaging. J Orthop Sports Phys Ther 2007;37:608e12.
Wallwork TL, Stanton W, Freke M, Hides J. The effect of chronic low back pain on size and contraction of the lumbar multifidis muscle. Man Ther 2009;14:496e500. Weir J. Quantifying test-retest reliability using the intraclass correlation coefficient. J Strength Cond Res 2005;19:231e40. Whittaker J. Ultrasound imaging for rehabilitation of the lumbopelvic region: a clinical approach. London, UK: Elsevier Health Sciences; 2007. p. 9.
Please cite this article in press as: Wilson A, et al., Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study, Manual Therapy (2016), http://dx.doi.org/10.1016/j.math.2016.01.004