Advertisement

Inter- and Intrarater Reliability of Isokinetic Thigh Muscle Strength Tests in Postmenopausal Women With Osteopenia

Published:January 24, 2012DOI:https://doi.org/10.1016/j.apmr.2011.10.001

      Abstract

      Eitzen I, Hakestad KA, Risberg MA. Inter- and intrarater reliability of isokinetic thigh muscle strength tests in postmenopausal women with osteopenia.

      Objective

      To evaluate inter- and intrarater reliability of isokinetic muscle strength measurements during knee extension and flexion in postmenopausal women with osteopenia.

      Design

      Reliability study assessing inter- and intrarater reliability.

      Setting

      General community.

      Participants

      A convenience sample of 27 postmenopausal women (mean age ± SD, 68.2±7.3y) with defined osteopenia from a bone mineral density T score of less than 1.5 and a wrist fracture within the last 2 years.

      Interventions

      Not applicable.

      Main Outcome Measures

      Isokinetic concentric muscle strength during knee extension and flexion was measured for 2 test conditions: 5 repetitions at 60°/s, and 25 repetitions at 180°/s. Agreement between tests was evaluated with the intraclass correlation coefficient (ICC2,1). Mean difference between tests, standard error of measurement (SEM and SEM%), and smallest real difference (SRD and SRD%) were calculated with 95% confidence intervals. SRD% and SEM% are emphasized in the results to allow congruent comparisons between the different test conditions.

      Results

      ICC2,1 reflected high agreement both for inter- and intrarater reliability, with most of the values .90 or greater. There were no significant differences between the left and the right leg at any of the 3 tests. Some differences were apparent between the test sessions, but these were not systematic. Agreements were overall higher for assessments during knee extension than knee flexion. The SEM% was between 3.5% and 10.2% for knee extension, and 7.0% and 17.7% for knee flexion. SRD% was suggested to be between 15% and 20% for knee extension, and 25% and 30% for knee flexion.

      Conclusions

      Isokinetic assessments of thigh muscle strength in postmenopausal women with osteopenia are of high reliability, with a level of agreement comparable to the levels found in previous reliability studies concerning both the healthy elderly and elderly with different health conditions. The measurement errors are small to moderate. The established SRD% provides thresholds for whether observed changes in strength in this patient group represent true change, which allows evaluations of minimal clinical importance in future studies.

      Key Words

      List of Abbreviations:

      BMD (bone mineral density), CI (confidence interval), CV (coefficient of variance), DXA (dual-energy x-ray absorptiometry), ICC (intraclass correlation coefficient), MIC (minimally important change), PASE (Physical Activity Scale for the Elderly), RCT (randomized controlled trial), ROM (range of motion), SEM (standard error of measurement), SRD (smallest real difference)
      POSTMENOPAUSAL WOMEN ARE particularly exposed to reductions in muscle strength and bone mineral density (BMD), which increase their risk of fractures.
      • Doherty T.J.
      The influence of aging and sex on skeletal muscle mass and strength.
      • Lindle R.S.
      • Metter E.J.
      • Lynch N.A.
      • et al.
      Age and gender comparisons of muscle strength in 654 women and men aged 20-93 yr.
      Since maintenance or improvement of muscle strength will decrease the loss of BMD, muscle strength assessments are of high relevance for elderly women. Isokinetic dynamometry is today an established method for evaluation of muscle strength during knee extension and flexion for different conditions and diagnoses,
      • Dvir Z.
      Isokinetics muscle testing, interpretation and clinical applications.
      • Gaines J.M.
      • Talbot L.A.
      Isokinetic strength testing in research and practice.
      and the reliability of the method has been investigated in a number of studies concerning both the healthy elderly,
      • Hartmann A.
      • Knols R.
      • Murer K.
      • de Bruin E.D.
      Reproducibility of an isokinetic strength-testing protocol of the knee and ankle in older adults.
      • Schroeder E.T.
      • Wang Y.
      • Castaneda-Sceppa C.
      • et al.
      Reliability of maximal voluntary muscle strength and power testing in older men.
      • Symons T.B.
      • Vandervoort A.A.
      • Rice C.L.
      • Overend T.J.
      • Marsh G.D.
      Reliability of isokinetic and isometric knee-extensor force in older women.
      • Symons T.B.
      • Vandervoort A.A.
      • Rice C.L.
      • Overend T.J.
      • Marsh G.D.
      Reliability of a single-session isokinetic and isometric strength measurement protocol in older men.
      as well as elderly persons with knee osteoarthritis,
      • Carpenter M.R.
      • Carpenter R.L.
      • Peel J.
      • et al.
      The reliability of isokinetic and isometric leg strength measures among individuals with symptoms of mild osteoarthritis.
      • Kean C.O.
      • Birmingham T.B.
      • Garland S.J.
      • Bryant D.M.
      • Giffin J.R.
      Minimal detectable change in quadriceps strength and voluntary muscle activation in patients with knee osteoarthritis.
      knee arthroplasty,
      • de Amorim A.M.
      • Leme L.E.
      Isokinetic dynamometry in elderly women undergoing total knee arthroplasty: a comparative study.
      stroke,
      • Cetin N.
      • Dilek A.
      • Aytar A.
      • Akman M.N.
      Reproducibility of isokinetic test findings for assessment of wrist spasticity in stroke patients.
      • Hsu A.L.
      • Tang P.F.
      • Jan M.H.
      Test-retest reliability of isokinetic muscle strength of the lower extremities in patients with stroke.
      • Pohl P.S.
      • Startzell J.K.
      • Duncan P.W.
      • Wallace D.
      Reliability of lower extremity isokinetic strength testing in adults with stroke.
      • Tripp E.J.
      • Harris S.R.
      Test-retest reliability of isokinetic knee extension and flexion torque measurements in persons with spastic hemiparesis.
      • Noorizadeh D.S.
      • Talebian S.
      • Olyaei G.
      • Montazeri A.
      Reliability of isokinetic normalized peak torque assessments for knee muscles in post-stroke hemiparesis.
      late effects of polio,
      • Flansbjer U.B.
      • Lexell J.
      Reliability of knee extensor and flexor muscle strength measurements in persons with late effects of polio.
      different neurologic disorders,
      • Tiffreau V.
      • Ledoux I.
      • Eymard B.
      • Thevenon A.
      • Hogrel J.Y.
      Isokinetic muscle testing for weak patients suffering from neuromuscular disorders: a reliability study.
      and heart failure.
      • Qittan M.
      • Wiesinger G.F.
      • Crevenna R.
      • et al.
      Isokinetic strength testing in patients with chronic heart failure: a reliability study.
      In total, these studies have found isokinetic dynamometry to be an adequate test method for thigh muscle strength for elderly people, with or without dysfunctions. However, to our knowledge, no studies have evaluated the reliability of isokinetic dynamometry in postmenopausal osteopenic or osteoporotic women. Because osteopenia and osteoporosis might be associated with an inherent fear of fractures during physical performance, it could be spurious to rely on reliability studies investigating dissimilar populations. Further, to evaluate strength development after exercise therapy interventions, it is important to know the standard error of measurement (SEM) and the threshold for the smallest change in score that must be surpassed to regard the change as real—that is, the smallest real difference (SRD).
      • Lexell J.E.
      • Downham D.Y.
      How to assess the reliability of measurements in rehabilitation.
      • Scholtes V.A.
      • Terwee C.B.
      • Poolman R.W.
      What makes a measurement instrument valid and reliable?.
      • Weir J.P.
      Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM.
      • Turner D.
      • Schunemann H.J.
      • Griffith L.E.
      • et al.
      The minimal detectable change cannot reliably replace the minimal important difference.
      The purpose of the present study was, therefore, to evaluate the inter- and intrarater reliability of selected isokinetic muscle strength outcomes during knee extension and flexion in postmenopausal women with osteopenia, and to establish the SEM and SRD for these outcomes.

      Methods

      Twenty-seven postmenopausal women with osteopenia who were already attending an ongoing randomized controlled trial (RCT) (www.clinicaltrials.gov; reference number NCT01357278) were included. The inclusion criteria were osteopenia defined from a proven low BMD (T score <–1.5).
      • Kanis J.A.
      WHO Study Group
      Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report.
      BMD was measured by means of dual-energy x-ray absorptiometry (DXA).
      • Kahn K.
      • McKay H.
      • Kannus R.
      • Bailey D.
      • Wark J.
      • Bennel K.
      Physical activity and bone health.
      • Blake G.M.
      • Fogelman I.
      The role of DXA bone density scans in the diagnosis and treatment of osteoporosis.
      To establish the level of physical activity, we used the self-reported form Physical Activity Scale for the Elderly (PASE),
      • Dinger M.K.
      • Oman R.F.
      • Taylor E.L.
      • Vesely S.K.
      • Able J.
      Stability and convergent validity of the Physical Activity Scale for the Elderly (PASE).
      • Washburn R.A.
      • Smith K.W.
      • Jette A.M.
      • Janney C.A.
      The Physical Activity Scale for the Elderly (PASE): development and evaluation.
      with a range in scores from 0 to 315, where higher scores reflect a higher activity level. Since the subjects included in this study also are participating in an RCT, it was essential to recruit a homogeneous sample population. Therefore, we included a wrist fracture within the last 2 years as an inclusion criterion, and excluded patients with a previous hip or vertebral fracture, a history of more than 3 osteoporotic fractures, and/or a high level of physical activity (participation in moderate or hard intensity >4 hours weekly). Subject characteristics of the included subjects are presented in table 1.
      Table 1Characteristics of the Included Subjects (n=27)
      CharacteristicValues
      Age (y)68.2±7.3
      Height (cm)163.9±6.2
      Weight (kg)66.1±8.3
      BMI (kg/m2)24.6±3.6
      Body fat (%)36.8±6.1
      Lumbar spine (L1-4) (T score)–2.1±0.7
      Hip total (T score)–1.6±0.8
      PASE (score)112.5±54.8
      NOTE. Values are mean ± SD.
      Abbreviation: BMI, body mass index.
      The study was approved by the Regional Committee for Medical Research Ethics for South-Eastern Norway. Before inclusion, all participants received oral and written information about the study, and signed an informed consent. The data collection was carried out in accordance with the directives given in the Declaration of Helsinki. Participants were offered information about their test results after completion of the study.

      Outcome Measurements

      The main outcome measurement for this study was the inter- and intrarater reliability of isokinetic muscle strength during knee extension and flexion. Data were collected with an isokinetic dynamometer (Biodex 6000a) at 2 different conditions: 5 repetitions at 60°/s and 25 repetitions at 180°/s. Lower velocities are believed to primarily reflect muscle strength, whereas higher velocities represent muscle endurance.
      • Dvir Z.
      Isokinetics muscle testing, interpretation and clinical applications.
      Because many existing studies evaluating isokinetic assessments include different testing velocities, it was of interest to investigate potential differences in reliability of the 2 conditions. Furthermore, we analyzed the left and the right limb separately because we wanted to reveal whether limb dominance, starting order, or both, could affect the reliability of the test. As secondary outcomes, we registered the maintained range of motion (ROM) for each test in degrees, and the coefficient of variance (CV) in percent, expressing the consistency of the repetitions within each test. Finally, the level of physical activity was registered with the PASE questionnaire both the day before the baseline test and the day before the intrarater retest, to evaluate whether subjects had changed their activity during the week between test day 1 and test day 2.

      Procedures

      Data collection was done during February and March 2011. At test day 1, subjects performed 2 strength tests where the test leaders (I.E. and K.A.H.) were responsible for 1 test each to assess interrater reliability. All tests were performed first by I.E. (test 1), thereafter by K.A.H. (test 2). Between test 1 and test 2, subjects had a standardized rest period of 30 minutes. Test day 2 was scheduled at the same time of day, 7 days after test day 1. All tests (test 3) on test day 2 were performed by I.E., to assess intrarater reliability. BMD was measured 1 to 2 weeks before test day 1, while PASE was completed by the subjects 1 day before both test days 1 and 2.

      Isokinetic Muscle Strength Test Protocol

      All participants performed a standardized warm-up on a stationary bicycle for 5 minutes. The participants were then seated in the chair of the isokinetic dynamometer with their hips flexed at 100° and knees flexed at 90°. Whether the left or the right limb was tested first was decided from a computer-generated randomized order. The same test order was kept for all 3 tests. The position with regard to the depth and height of the test chair, the side-to-side placement of the dynamometer, and the length of the attachment arm was adjusted for each subject before test 1 to ensure correct alignment of the anatomical axis of the knee joint and the rotational axis of the lever arm. The numbers describing the positioning were noted to ensure the exact same positioning for tests 2 and 3. Effects of gravity were corrected with the knee joint stabilized at 10° knee flexion. To minimize changes in positioning caused by forward sliding during the test, participants were stabilized with straps placed firmly around the chest, pelvis, thigh, and ankle. Adequate positioning in the chair was confirmed between testing of the 2 limbs.
      The ROM was set from 90° knee flexion to full extension (defined as 0°), which is the established ROM for evaluation of isokinetic thigh muscle strength.
      • Reichard L.B.
      • Croisier J.L.
      • Malnati M.
      • Katz-Leurer M.
      • Dvir Z.
      Testing knee extension and flexion strength at different ranges of motion: an isokinetic and electromyographic study.
      The importance of moving through the complete ROM was emphasized, and subjects performed 1 full extension and flexion movement after positioning to ensure that they understood their instructions. Before initiating the tests, subjects were told the principles of isokinetic strength testing. It was emphasized that they had to put maximum effort into the movement. To familiarize the subjects with the dynamometer, the subjects performed a standardized trial session of 3 repetitions with submaximal effort before the 60°/s test condition, and a standardized trial session of 5 repetitions before the 180°/s test condition for each leg. For the trial session before the test at 60°/s, subjects were told to gradually increase their effort, whereas they were told to give their maximum effort throughout all 5 repetitions for the test. For the trial session before the test at 180°/s, subjects were instructed to move as fast as possible. For the test they were told explicitly to start at the highest speed possible, and then try to maintain this speed throughout the 25 repetitions. Between the trial session and the test session there was a standardized 1-minute pause. Standardized verbal encouragement was given from the test leaders. When the test at 60°/s was initiated, the present test leader shouted, “Start, and give your best effort.” Further verbal cues consisted of counting from 1 to 5. For the 180°/s test condition, the present test leader shouted, “Start, move as fast as you can,” when the test was initiated; thereafter, the test leader counted to 13 and then said, “You are halfway through, try to maintain speed,” before finally counting down from 5 to 1.

      Analyses

      Muscle strength was expressed as peak torque and average peak torque in newton meters, and total work in joules. Peak torque reflects the value of the single point during the ROM where the individual is able to produce the highest force
      • Eitzen I.
      • Eitzen T.J.
      • Holm I.
      • Snyder-Mackler L.
      • Risberg M.A.
      Anterior cruciate ligament-deficient potential copers and noncopers reveal different isokinetic quadriceps strength profiles in the early stage after injury.
      • Tsepis E.
      • Giakas G.
      • Vagenas G.F.
      • Georgoulis A.
      Frequency content asymmetry of the isokinetic curve between ACL deficient and healthy knee.
      ; average peak torque, the mean value of the peak torques attained for each repetition within each test; and total work, the ability of the muscle to produce force throughout the ROM.
      • Eitzen I.
      • Eitzen T.J.
      • Holm I.
      • Snyder-Mackler L.
      • Risberg M.A.
      Anterior cruciate ligament-deficient potential copers and noncopers reveal different isokinetic quadriceps strength profiles in the early stage after injury.
      • Holm I.
      Quantification of muscle strength by isokinetic performance.
      • Pincivero D.M.
      • Heller B.M.
      • Hou S.I.
      The effects of ACL injury on quadriceps and hamstring torque, work and power.
      Descriptive statistics (means and SDs) were calculated to describe the characteristics of the included subjects. Differences in absolute strength values between the left and right limb were computed with a 2-sided paired Student t test. The reliability analyses included assessments of both inter- and intrarater reliability, and because the data of interest were continuous, the intraclass correlation coefficient (ICC) for absolute agreement (ICC2,1) was chosen.
      • Scholtes V.A.
      • Terwee C.B.
      • Poolman R.W.
      What makes a measurement instrument valid and reliable?.
      The mean difference between the tests and the ICC2,1 was calculated with 95% confidence intervals (95% CIs). To assess errors of measurements, the SEM and the SEM% were calculated from the following formulas:
      SEM=totalvariance


      SEM%=SEMMeanoftheabsolutestrengthvaluesfromthetwotests×100


      To estimate the SRD, SRD%, and the 95% SRD we used the following formulas:
      SRD=1.96×2×SEM


      SRD%=SRDMeanoftheabsolutestrengthvaluesfromthetwotests×100


      95% SRD = mean difference ± SRD
      All data analyses were conducted with the SPSS version 17.0 for Windows.b The level of statistical significance was set at a P value of less than .05.

      Results

      The subjects included in the study had a mean age ± SD of 68.2±7.3 years and were all residents in the Oslo region in Norway. All 27 subjects completed tests 1 and 2, but 1 subject got sick and was restricted from participating in test 3. For the 26 subjects completing all 3 tests, the interval between test day 1 and test day 2 was 7 days, except for 1 subject who showed up 1 day early, and therefore had 6 days between test days 1 and 2. DXA testing was performed a mean ± SD of 10.6±2.3 days before test 1. There was no significant difference (P=.98) in PASE score before test day 1 (mean score, 112.5±54.8) and test day 2 (mean score, 112.6±62.3) 1 week later.
      Absolute strength values for peak torque, average peak torque, and total work for the different test conditions are shown for tests 1 and 2 in table 2, and for tests 1 and 3 in table 3. No significant differences (P>.05) were established between the left and right limb at either test, but some significant differences were found between the test sessions. From test 1 to test 2, the left knee extension total work at 180°/s decreased (P=.010), whereas knee flexion peak torque, average peak torque, and total work increased at 60°/s for both limbs (P values between <.001 and .017). From test 1 to test 3, a significant increase was found for peak torque (P=.011), average peak torque (P=.022), and total work (P=.011) knee flexion at 60°/s, but only for the right limb. In addition, a significant increase was established for knee flexion peak torque at 180°/s (P=.016).
      Table 2Interrater Absolute Values of Isokinetic Muscle Strength During Knee Extension and Flexion: Peak Torque, Average Peak Torque, and Total Work (n=27)
      ExtensionFlexion
      MeasureRightLeftRightLeft
      Peak torque 60° (Nm)
       Rater 1 (test 1)99.1±23.499.7±24.848.9±15.548.1±14.1
       Rater 2 (test 2)97.6±21.597.1±21.753.7±15.5
      Significant difference between rater 1 and rater 2 (P<.05).
      50.9±13.9
      Significant difference between rater 1 and rater 2 (P<.05).
      Average peak torque 60° (Nm)
       Rater 1 (test 1)90.7±21.992.2±22.844.6±14.843.6±13.0
       Rater 2 (test 2)89.8±20.789.8±20.848.9±14.2
      Significant difference between rater 1 and rater 2 (P<.05).
      46.3±13.0
      Significant difference between rater 1 and rater 2 (P<.05).
      Total work 60° (J)
       Rater 1 (test 1)500.3±121.2490.2±130.2254.7±88.2249.9±77.2
       Rater 2 (test 2)483.5±123.0475.3±110.5277.9±90.5
      Significant difference between rater 1 and rater 2 (P<.05).
      265.0±75.6
      Significant difference between rater 1 and rater 2 (P<.05).
      Peak torque 180° (Nm)
       Rater 1 (test 1)64.8±20.864.9±15.034.1±11.334.6±10.5
       Rater 2 (test 2)65.3±13.563.5±13.635.7±10.835.9±10.8
      Average peak torque 180° (Nm)
       Rater 1 (test 1)50.0±10.650.8±10.526.9±9.327.3±8.8
       Rater 2 (test 2)50.2±11.050.5±10.128.2±9.528.5±8.7
      Total work 180° (J)
       Rater 1 (test 1)1317.0±299.21306.1±294.1655.5±252.8655.5±257.5
       Rater 2 (test 2)1298.4±285.51248.0±269.7
      Significant difference between rater 1 and rater 2 (P<.05).
      668.5±225.0686.2±225.0
      NOTE. The isokinetic knee flexion and extension torques, resulting from the force applied perpendicularly to the moment arm of the dynamometer, is expressed in Newton-meters (Nm). The isokinetic knee flexion and extension total work, where the force acts in the same direction as the distance travelled, is expressed in Joule (J). Values are mean ± SD.
      low asterisk Significant difference between rater 1 and rater 2 (P<.05).
      Table 3Intrarater Absolute Values of Isokinetic Muscle Strength During Knee Extension and Flexion: Peak Torque, Average Peak Torque, and Total Work (n=26)
      ExtensionFlexion
      MeasureRightLeftRightLeft
      Peak torque 60° (Nm)
       Rater 1
        (test 1)98.3±23.599.0±25.048.1±15.347.3±13.7
        (test 3)97.7±24.496.3±22.251.2±15.8
      Significant difference for rater 1 between test 1 and test 3 (P<.05).
      48.9±13.8
      Average peak torque 60° (Nm)
       Rater 1
        (test 1)90.2±22.291.8±23.244.1±14.842.9±12.7
        (test 3)88.9±23.987.4±22.547.2±14.9
      Significant difference for rater 1 between test 1 and test 3 (P<.05).
      44.6±12.3
      Total work 60° (J)
       Rater 1
        (test 1)497.4±122.6487.1±131.8252.2±88.9247.1±77.3
        (test 3)493.4±133.1477.6±130.9281.6±108.9
      Significant difference for rater 1 between test 1 and test 3 (P<.05).
      268.5±96.2
      Peak torque 180° (Nm)
       Rater 1
        (test 1)64.6±14.664.5±15.133.4±10.934.0±10.3
        (test 3)64.8±15.863.7±16.335.6±12.0
      Significant difference for rater 1 between test 1 and test 3 (P<.05).
      33.8±10.8
      Average peak torque 180° (Nm)
       Rater 1
        (test 1)50.0±10.850.8±10.726.6±9.327.0±8.8
        (test 3)50.9±11.750.2±11.728.1±10.227.7±9.1
      Total work 180° (J)
       Rater 1
        (test 1)1313.9±304.71299.6±297.9650.9±256.7649.3±260.6
        (test 3)1331.8±326.21293.7±319.8689.1±281.7686.0±251.2
      NOTE. The isokinetic knee flexion and extension torques, resulting from the force applied perpendicularly to the moment arm of the dynamometer, is expressed in Newton-meters (Nm). The isokinetic knee flexion and extension total work, where the force acts in the same direction as the distance travelled, is expressed in Joule (J). Values are mean ± SD.
      low asterisk Significant difference for rater 1 between test 1 and test 3 (P<.05).
      As shown in table 4, the interrater ICC2,1 values between tests 1 and 2 ranged from .88 to .95 for knee extension and .86 to .93 for knee flexion. The 95% CIs of the ICC2,1 were in general wider for knee flexion than for knee extension. The SEM% between tests 1 and 2 ranged from 5.0 to 8.3 for knee extension and 8.1 to 12.0 for knee flexion, whereas the SRD% ranged from 8.8 to 15.2 for knee extension and 15.2 to 22.2 for knee flexion. For intrarater reliability between tests 1 and 3 (table 5), ICC2,1 ranged from .84 to .98 for knee extension and from .78 to .95 for knee flexion. The 95% CIs of the ICC2,1 were also wider for knee flexion than for knee extension. The SEM% of knee extension between tests 1 and 3 ranged from 3.5 to 10.2, whereas corresponding values for knee flexion were from 7.0 to 17.7. The SRD% ranged from 6.4 to 18.9 for knee extension and 13.0 to 33.4 for knee flexion. In accordance with the interrater reliability, but even more pronounced, both the SEM% and the SRD% were in general larger for flexion than for extension.
      Table 4Interrater Reliability (Test 1 and Test 2) of Isokinetic Muscle Strength During Knee Extension and Flexion; Peak Torque, Average Peak Torque, and Total Work (n=27)
      95% CI for ICC2,195% CI for Mean Difference95% SRD
      MeasureICC2,1LowerUpperMean DifferenceLowerUpperSEMSEM%SRDSRD%LowerUpper
      Peak torque extension 60°/s
       Right limb.93.86.971.49–1.824.805.96.016.411.1–14.9117.89
       Left limb.93.85.972.63−0.776.036.06.417.311.7−14.6719.93
      Average peak torque extension 60°/s
       Right limb.93.86.970.87−2.233.975.56.115.211.2−14.3316.07
       Left limb.93.86.972.35−0.725.425.66.215.611.4−13.2517.95
      Total work extension 60°/s
       Right limb.92.83.9616.80−1.6635.2634.57.095.612.9−78.80112.40
       Left limb.89.78.9514.83−7.2036.8640.08.3110.915.2−96.07125.76
      Peak torque extension 180°/s
       Right limb.95.87.98−0.47−2.301.373.25.09.09.2−9.478.53
       Left limb.93.86.971.43−0.603.463.75.810.310.7−8.8711.73
      Average peak torque extension 180°/s
       Right limb.88.75.94−0.19−2.311.933.77.410.313.7−10.4910.11
       Left limb.91.82.960.34−1.382.073.06.08.48.8−8.069.14
      Total work extension 180°/s
       Right limb.94.88.9718.59−20.4857.6169.75.3193.39.8−147.71211.89
       Left limb.91.77.9658.1514.91101.3986.36.8239.1112.4−180.96297.26
      Peak torque flexion 60°/s
       Right limb.90.52.97−4.84−6.93−2.755.09.813.918.4−18.749.06
       Left limb.92.78.97−2.85−4.83−0.864.08.111.215.2−14.058.35
      Average peak torque flexion 60°/s
       Right limb.86.61.96−4.27−6.54−1.995.010.713.920.1−18.179.63
       Left limb.90.76.96−2.62−4.72−0.524.19.211.417.1−14.028.78
      Total work flexion 60°/s
       Right limb.91.68.96−23.17−36.01−10.3327.910.577.219.6−100.3754.03
       Left limb.91.78.96−15.15−26.81−3.4823.19.064.016.7−79.1548.85
      Peak torque flexion 180°/s
       Right limb.92.82.96−1.65−3.330.023.29.18.816.9−10.457.15
       Left limb.92.83.96−1.34−3.000.313.18.78.516.2−9.847.16
      Average peak torque flexion 180°/s
       Right limb.92.83.96−1.23−2.690.232.79.87.518.3−8.736.27
       Left limb.92.83.91−1.18−2.550.162.58.96.916.6−8.085.72
      Total work flexion 180°/s
       Right limb.91.80.96−13.27−58.1931.6379.312.0219.422.2−232.67206.13
       Left limb.93.85.97−30.72−65.604.1564.99.17179.918.0−210.62149.18
      Table 5Intrarater Reliability (Test 1 and Test 3) of Isokinetic Muscle Strength During Knee Extension and Flexion: Peak Torque, Average Peak Torque, and Total Work (n=26)
      95% CI for ICC2,195% CI for Mean Difference95% SRD
      MeasureICC2,1LowerUpperMean DifferenceLowerUpperSEMSEM%SRDSRD%LowerUpper
      Peak torque extension 60°/s
       Right limb.94.86.970.67–2.844.186.06.216.711.3–15.7317.04
       Left limb.95.88.982.70−0.245.675.45.515.010.2−12.317.70
      Average peak torque extension 60°/s
       Right limb.84.68.931.28−4.026.599.210.225.418.9−24.1226.68
       Left limb.85.69.934.45−0.409.308.99.924.618.2−20.1529.05
      Total work extension 60°/s
       Right limb.85.70.933.96−24.4432.3648.89.9135.318.2−131.34139.26
       Left limb.88.76.959.45−16.5135.40645.19.3124.917.2−115.45144.75
      Peak torque extension 180°/s
       Right limb.98.95.99−0.17−1.481.132.23.56.26.4−6.376.03
       Left limb.97.93.990.78−0.812.372.84.37.78.0−6.928.48
      Average peak torque extension 180°/s
       Right limb.96.91.98−0.84−2.310.452.34.66.48.5−7.245.56
       Left limb.94.88.970.57−0.962.102.75.37.49.8−6.837.97
      Total work extension 180°/s
       Right limb.96.92.98−17.91−53.6517.8362.74.7173.78.8−191.61155.79
       Left limb.95.89.985.88−33.7845.5668.25.3189.115.2−183.22194.98
      Peak torque flexion 60°/s
       Right limb.92.79.97−3.09−5.40−0.784.59.112.517.0−15.599.41
       Left limb.94.87.97−1.59−3.430.273.47.09.313.0−10.897.71
      Average peak torque flexion 60°/s
       Right limb.91.76.96−3.12−5.46−0.774.610.112.718.8−15.129.58
       Left limb.92.82.96−1.70−3.660.263.68.29.915.2−11.68.2
      Total work hamstrings 60°/s
       Right limb.78.54.90−29.34−54.14−4.6447.317.7131.133.4−160.44101.76
       Left limb.78.55.89−21.44−44.141.2741.816.2115.930.4−137.3494.46
      Peak torque hamstrings 180°/s
       Right limb.91.77.96−2.28−4.10−0.463.510.29.819.1−12.087.52
       Left limb.95.89.980.23−1.151.612.47.06.613.0−6.146.83
      Average peak torque hamstrings 180°/s
       Right limb.91.80.96−1.47−3.090.153.010.88.220.2−9.676.73
       Left limb.95.90.98−0.66−1.780.462.07.25.513.5−6.164.84
      Total work hamstrings 180°/s
       Right limb.91.81.96−38.24−82.826.3481.212.2225.022.6−263.24186.76
       Left limb.92.83.97−36.63−75.151.9171.110.6196.919.8−233.53160.27
      The mean ROM values at the 60°/s test condition was for tests 1, 2, and 3 were 89.5°/89.4°, 89.3°/89.8°, and 90.2°/89.4° for the right and left limb, respectively. For the 180°/s test condition, corresponding numbers were 88.9°/88.7°, 88.9°/88.8°, and 88.9°/88.9°. No significant differences (P>.05) were found in ROM of the tests, neither between knee extension and flexion, nor between tests. The CVs between the repetitions within each test for knee extension and flexion at the 60°/s and 180°/s test condition are given in table 6. No significant differences (P>.05) were found between the right and the left limb, or between the same condition at tests 1, 2, or 3. However, CVs were significantly larger (P<.001 to P=.003) for the 180°/s test condition than for 60°/s, both for knee extension and flexion.
      Table 6CV Between Repetitions for All Test Conditions (n=27)
      ExtensionFlexion
      Test ConditionRightLeftRightLeft
      60°/s
       Test 17.4±3.86.9±3.48.4±5.38.7±4.2
       Test 26.8±3.86.3±3.88.2±4.28.4±4.2
       Test 36.3±3.47.2±5.58.0±5.28.1±4.6
      180°/s
       Test 114.8±3.114.1±3.515.4±7.715.2±8.8
       Test 214.9±3.313.6±4.416.8±10.013.4±4.6
       Test 314.1±4.013.8±3.814.6±5.013.3±6.7
      NOTE. Values are mean ± SD.

      Discussion

      To our knowledge, this is the first study to address the inter- and intrarater reliability of isokinetic dynamometry during knee extension and flexion in postmenopausal women with osteopenia. The main finding of this study is that the levels of agreement across both the 2 different raters and the different times are high, with ICC levels comparable to those of previous reliability studies concerning both healthy adults and adults with different health conditions.
      • Hartmann A.
      • Knols R.
      • Murer K.
      • de Bruin E.D.
      Reproducibility of an isokinetic strength-testing protocol of the knee and ankle in older adults.
      • Symons T.B.
      • Vandervoort A.A.
      • Rice C.L.
      • Overend T.J.
      • Marsh G.D.
      Reliability of isokinetic and isometric knee-extensor force in older women.
      • Symons T.B.
      • Vandervoort A.A.
      • Rice C.L.
      • Overend T.J.
      • Marsh G.D.
      Reliability of a single-session isokinetic and isometric strength measurement protocol in older men.
      • Carpenter M.R.
      • Carpenter R.L.
      • Peel J.
      • et al.
      The reliability of isokinetic and isometric leg strength measures among individuals with symptoms of mild osteoarthritis.
      • Kean C.O.
      • Birmingham T.B.
      • Garland S.J.
      • Bryant D.M.
      • Giffin J.R.
      Minimal detectable change in quadriceps strength and voluntary muscle activation in patients with knee osteoarthritis.
      • de Amorim A.M.
      • Leme L.E.
      Isokinetic dynamometry in elderly women undergoing total knee arthroplasty: a comparative study.
      • Pohl P.S.
      • Startzell J.K.
      • Duncan P.W.
      • Wallace D.
      Reliability of lower extremity isokinetic strength testing in adults with stroke.
      • Noorizadeh D.S.
      • Talebian S.
      • Olyaei G.
      • Montazeri A.
      Reliability of isokinetic normalized peak torque assessments for knee muscles in post-stroke hemiparesis.
      • Flansbjer U.B.
      • Lexell J.
      Reliability of knee extensor and flexor muscle strength measurements in persons with late effects of polio.
      • Qittan M.
      • Wiesinger G.F.
      • Crevenna R.
      • et al.
      Isokinetic strength testing in patients with chronic heart failure: a reliability study.
      The 95% CIs of the ICC2,1 were, however, generally wider for knee flexion, indicating better agreement for knee extension. With regard to different test velocities, the agreement was high for both the 60°/s and 180°/s test conditions.
      Even after stating that the main findings of this study establish a high inter- and intrarater reliability of the method under investigation, it is important to emphasize that there is no absolute consensus for defining cutoffs for the ICC level.
      • Lexell J.E.
      • Downham D.Y.
      How to assess the reliability of measurements in rehabilitation.
      • Weir J.P.
      Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM.
      In the hallmark publication from 1979, Shrout and Fleiss
      • Shrout P.E.
      • Fleiss J.L.
      Intraclass correlations: uses in assessing rater reliability.
      suggested that ICCs should be categorized as acceptable, fair, or good. Fleiss
      • Fleiss J.L.
      The design and analysis of clinical experiments.
      later suggested .75 as the cutoff for good reliability, which is in line with the work of Portney and Watkins,
      • Portney L.
      • Watkins M.
      Foundations of clinical research: applications to practice.
      who define ICCs less than .50 as indicating poor reliability, .50 to .75 as moderate reliability, and greater than .75 as good reliability. Our ICC2,1 values were all greater than .75, with the majority greater than .90. The ICC is, however, affected by the heterogeneity of the sample under investigation.
      • Weir J.P.
      Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM.
      If the variation in the scores is high, the ICCs may be higher than if the sample are more homogeneous.
      • Scholtes V.A.
      • Terwee C.B.
      • Poolman R.W.
      What makes a measurement instrument valid and reliable?.
      • Weir J.P.
      Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM.
      The SD of the mean scores of the absolute strength values, and also the 95% CI of the ICC2,1 reveal considerable variability in this material. Consequently, the observed agreement alone is insufficient to conclude that the method has high reliability.
      To provide a comprehensive analysis of the reliability, unaffected by the variation between subjects,
      • Lexell J.E.
      • Downham D.Y.
      How to assess the reliability of measurements in rehabilitation.
      • Scholtes V.A.
      • Terwee C.B.
      • Poolman R.W.
      What makes a measurement instrument valid and reliable?.
      we therefore calculated the SEM and the SRD of the different test conditions. Since the different test conditions have inherent different scores, we reported the SEM and the SRD also as percentages. The SEM% for knee extension was 10.2% or less for all test conditions, whereas somewhat larger values up to 17.7% were established for knee flexion. In total, our SEM% levels can be characterized as low to moderate.
      • Flansbjer U.B.
      • Lexell J.
      Reliability of knee extensor and flexor muscle strength measurements in persons with late effects of polio.
      • Pincivero D.M.
      • Lephart S.M.
      • Karunakara R.A.
      Reliability and precision of isokinetic strength and muscular endurance for the quadriceps and hamstrings.
      Larger measurement errors for flexion than extension have previously been reported in reliability studies concerning isokinetic dynamometry of the knee in elderly subjects.
      • Pohl P.S.
      • Startzell J.K.
      • Duncan P.W.
      • Wallace D.
      Reliability of lower extremity isokinetic strength testing in adults with stroke.
      • Noorizadeh D.S.
      • Talebian S.
      • Olyaei G.
      • Montazeri A.
      Reliability of isokinetic normalized peak torque assessments for knee muscles in post-stroke hemiparesis.
      From our clinical experience with isokinetic testing, we believe that subjects more intuitively capture the necessity to perform with substantial effort during the extension movement, because in contrast to flexion, extension is not assisted by gravity. If the subjects relax after reaching full extension, the limb will passively go into flexion because of the nature of the isokinetic test. Thus, it is essential to emphasize the need to also perform with maximum effort during flexion, in particular for patients who are not used to heavy resistance strength training. The resultant SRD% values were also higher for knee flexion, and higher for the intrarater condition, suggesting that improvements should exceed 15% to 20% for knee extension and 25% to 30% for knee flexion, to reflect a true change. However, even though the SRD% provides a threshold for true change, it is important not to regard the SRD% as equivalent to the minimally important change (MIC). Since the SEM% and the SRD% are distribution based measures, dependent on the method and the subjects included, they provide no information on the meaning of the change.
      • Turner D.
      • Schunemann H.J.
      • Griffith L.E.
      • et al.
      The minimal detectable change cannot reliably replace the minimal important difference.
      Rather, the MIC should be defined from an external anchor-based criterion, like a global rating of change.
      • Turner D.
      • Schunemann H.J.
      • Griffith L.E.
      • et al.
      The minimal detectable change cannot reliably replace the minimal important difference.
      First when both the SRD and the MIC are established, the correspondence between the SRD and the MIC can be used to decide the clinical importance of the change.
      • de Vet H.C.
      • Terwee C.B.
      • Ostelo R.W.
      • Beckerman H.
      • Knol D.L.
      • Bouter L.M.
      Minimal changes in health status questionnaires: distinction between minimally detectable change and minimally important change.
      A prerequisite for valid isokinetic assessments is that the subjects are able to maintain steady movement throughout the given ROM, including an acceptable CV within each trial. Our findings showed that the ROM was near complete (90°–0° knee flexion) both for the 60°/s and 180°/s test conditions. However, a significant difference was found in that CVs for both extension and flexion were larger when testing at 180°/s. As larger CVs have been suggested to reflect submaximal rather than maximal efforts,
      • Lin P.C.
      • Robinson M.E.
      • Carlos Jr, J.
      • O'Connor P.
      Detection of submaximal effort in isometric and isokinetic knee extension tests.
      this could suggest inadequate consistency between the repetitions of each test. However, the 180°/s test condition, in contrast to the 60°/s test condition, is intended to engender some degree of fatigue, revealed in the test reports as lower work during the last third of the repetitions. The ICCs and SEM% were overall not higher for the 180°/s test condition, which indicates that the development of fatigue is consistent between tests. Thus, the larger CVs reflect the inherent characteristics of the test condition, rather than a threat to the reliability.
      We evaluated peak torque, average peak torque, and total work for 2 different test conditions. In clinical practice, it can be inconvenient to perform such an extended test protocol. With regard to reliability, we found no benefit of reporting average peak torque or total work instead of peak torque. Because 5 repetitions at 60°/s with peak torque as the main outcome is the overall most established protocol for isokinetic assessments of knee extension and flexion, we also suggest this protocol when testing postmenopausal women with osteopenia. However, if muscle performance during higher velocities is the primary interest, the 180°/s protocol is adequate. Limb dominance and starting order seem to be of no great concern for this patient group. However, some significant strength increases were found for knee flexion in the 60°/s test condition. This might suggest a potential learning effect, which could be taken into account by providing additional trial sessions before conducting the tests.

      Study Limitations

      Only patients with osteopenia were included in the study, and the results cannot be generalized to patients with more severe established osteoporosis. The test order in this study implied that all subjects were tested first by rater 1 and thereafter by rater 2 at test day 1. A random test order could be suggested to distribute potential differences within subjects evenly on the 2 raters. However, no differences that suggest systematic subject-related differences from test 1 to 2 were detected in this study.

      Conclusions

      Isokinetic assessments of thigh muscle strength in postmenopausal women with osteopenia are of high reliability, with a level of agreement comparable to the levels found in previous reliability studies concerning the elderly population.
      • a
        Biodex Medical Systems, Inc, 20 Ramsay Rd, Shirley, NY 11967-4704.
      • b
        SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

      References

        • Doherty T.J.
        The influence of aging and sex on skeletal muscle mass and strength.
        Curr Opin Clin Nutr Metab Care. 2001; 4: 503-508
        • Lindle R.S.
        • Metter E.J.
        • Lynch N.A.
        • et al.
        Age and gender comparisons of muscle strength in 654 women and men aged 20-93 yr.
        J Appl Physiol. 1997; 83: 1581-1587
        • Dvir Z.
        Isokinetics muscle testing, interpretation and clinical applications.
        2nd ed. Churchill Livingstone, Edinburgh2004
        • Gaines J.M.
        • Talbot L.A.
        Isokinetic strength testing in research and practice.
        Biol Res Nurs. 1999; 1: 57-64
        • Hartmann A.
        • Knols R.
        • Murer K.
        • de Bruin E.D.
        Reproducibility of an isokinetic strength-testing protocol of the knee and ankle in older adults.
        Gerontology. 2009; 55: 259-268
        • Schroeder E.T.
        • Wang Y.
        • Castaneda-Sceppa C.
        • et al.
        Reliability of maximal voluntary muscle strength and power testing in older men.
        J Gerontol A Biol Sci Med Sci. 2007; 62: 543-549
        • Symons T.B.
        • Vandervoort A.A.
        • Rice C.L.
        • Overend T.J.
        • Marsh G.D.
        Reliability of isokinetic and isometric knee-extensor force in older women.
        J Aging Phys Act. 2004; 12: 525-537
        • Symons T.B.
        • Vandervoort A.A.
        • Rice C.L.
        • Overend T.J.
        • Marsh G.D.
        Reliability of a single-session isokinetic and isometric strength measurement protocol in older men.
        J Gerontol A Biol Sci Med Sci. 2005; 60: 114-119
        • Carpenter M.R.
        • Carpenter R.L.
        • Peel J.
        • et al.
        The reliability of isokinetic and isometric leg strength measures among individuals with symptoms of mild osteoarthritis.
        J Sports Med Phys Fitness. 2006; 46: 585-589
        • Kean C.O.
        • Birmingham T.B.
        • Garland S.J.
        • Bryant D.M.
        • Giffin J.R.
        Minimal detectable change in quadriceps strength and voluntary muscle activation in patients with knee osteoarthritis.
        Arch Phys Med Rehabil. 2010; 91: 1447-1451
        • de Amorim A.M.
        • Leme L.E.
        Isokinetic dynamometry in elderly women undergoing total knee arthroplasty: a comparative study.
        Clinics (Sao Paulo). 2006; 61: 215-222
        • Cetin N.
        • Dilek A.
        • Aytar A.
        • Akman M.N.
        Reproducibility of isokinetic test findings for assessment of wrist spasticity in stroke patients.
        Isokinet Exerc Sci. 2008; 16: 61-67
        • Hsu A.L.
        • Tang P.F.
        • Jan M.H.
        Test-retest reliability of isokinetic muscle strength of the lower extremities in patients with stroke.
        Arch Phys Med Rehabil. 2002; 83: 1130-1137
        • Pohl P.S.
        • Startzell J.K.
        • Duncan P.W.
        • Wallace D.
        Reliability of lower extremity isokinetic strength testing in adults with stroke.
        Clin Rehabil. 2000; 14: 601-607
        • Tripp E.J.
        • Harris S.R.
        Test-retest reliability of isokinetic knee extension and flexion torque measurements in persons with spastic hemiparesis.
        Phys Ther. 1991; 71: 390-396
        • Noorizadeh D.S.
        • Talebian S.
        • Olyaei G.
        • Montazeri A.
        Reliability of isokinetic normalized peak torque assessments for knee muscles in post-stroke hemiparesis.
        Gait Posture. 2008; 27: 715-718
        • Flansbjer U.B.
        • Lexell J.
        Reliability of knee extensor and flexor muscle strength measurements in persons with late effects of polio.
        J Rehabil Med. 2010; 42: 588-592
        • Tiffreau V.
        • Ledoux I.
        • Eymard B.
        • Thevenon A.
        • Hogrel J.Y.
        Isokinetic muscle testing for weak patients suffering from neuromuscular disorders: a reliability study.
        Neuromuscul Disord. 2007; 17: 524-531
        • Qittan M.
        • Wiesinger G.F.
        • Crevenna R.
        • et al.
        Isokinetic strength testing in patients with chronic heart failure: a reliability study.
        Int J Sports Med. 2001; 22: 40-44
        • Lexell J.E.
        • Downham D.Y.
        How to assess the reliability of measurements in rehabilitation.
        Am J Phys Med Rehabil. 2005; 84: 719-723
        • Scholtes V.A.
        • Terwee C.B.
        • Poolman R.W.
        What makes a measurement instrument valid and reliable?.
        Injury. 2011; 42: 236-240
        • Weir J.P.
        Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM.
        J Strength Cond Res. 2005; 19: 231-240
        • Turner D.
        • Schunemann H.J.
        • Griffith L.E.
        • et al.
        The minimal detectable change cannot reliably replace the minimal important difference.
        J Clin Epidemiol. 2010; 63: 28-36
        • Kanis J.A.
        • WHO Study Group
        Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report.
        Osteoporos Int. 1994; 4: 368-381
        • Kahn K.
        • McKay H.
        • Kannus R.
        • Bailey D.
        • Wark J.
        • Bennel K.
        Physical activity and bone health.
        Human Kinetics, Champaign, IL2001
        • Blake G.M.
        • Fogelman I.
        The role of DXA bone density scans in the diagnosis and treatment of osteoporosis.
        Postgrad Med J. 2007; 83: 509-517
        • Dinger M.K.
        • Oman R.F.
        • Taylor E.L.
        • Vesely S.K.
        • Able J.
        Stability and convergent validity of the Physical Activity Scale for the Elderly (PASE).
        J Sports Med Phys Fitness. 2004; 44: 186-192
        • Washburn R.A.
        • Smith K.W.
        • Jette A.M.
        • Janney C.A.
        The Physical Activity Scale for the Elderly (PASE): development and evaluation.
        J Clin Epidemiol. 1993; 46: 153-162
        • Reichard L.B.
        • Croisier J.L.
        • Malnati M.
        • Katz-Leurer M.
        • Dvir Z.
        Testing knee extension and flexion strength at different ranges of motion: an isokinetic and electromyographic study.
        Eur J Appl Physiol. 2005; 95: 371-376
        • Eitzen I.
        • Eitzen T.J.
        • Holm I.
        • Snyder-Mackler L.
        • Risberg M.A.
        Anterior cruciate ligament-deficient potential copers and noncopers reveal different isokinetic quadriceps strength profiles in the early stage after injury.
        Am J Sports Med. 2010; 38: 586-593
        • Tsepis E.
        • Giakas G.
        • Vagenas G.F.
        • Georgoulis A.
        Frequency content asymmetry of the isokinetic curve between ACL deficient and healthy knee.
        J Biomech. 2004; 37: 857-864
        • Holm I.
        Quantification of muscle strength by isokinetic performance.
        Biomechanics Laboratory/Physiotherapy Dept, National Hospital, Orthopaedic Centre, University of Oslo, Oslo1996
        • Pincivero D.M.
        • Heller B.M.
        • Hou S.I.
        The effects of ACL injury on quadriceps and hamstring torque, work and power.
        J Sports Sci. 2002; 20: 689-696
        • Shrout P.E.
        • Fleiss J.L.
        Intraclass correlations: uses in assessing rater reliability.
        Psychol Bull. 1979; 86: 420-428
        • Fleiss J.L.
        The design and analysis of clinical experiments.
        Wiley, New York1986
        • Portney L.
        • Watkins M.
        Foundations of clinical research: applications to practice.
        2nd ed. Prentice Hall, Upper Saddle River2000
        • Pincivero D.M.
        • Lephart S.M.
        • Karunakara R.A.
        Reliability and precision of isokinetic strength and muscular endurance for the quadriceps and hamstrings.
        Int J Sports Med. 1997; 18: 113-117
        • de Vet H.C.
        • Terwee C.B.
        • Ostelo R.W.
        • Beckerman H.
        • Knol D.L.
        • Bouter L.M.
        Minimal changes in health status questionnaires: distinction between minimally detectable change and minimally important change.
        Health Qual Life Outcomes. 2006; 4: 54
        • Lin P.C.
        • Robinson M.E.
        • Carlos Jr, J.
        • O'Connor P.
        Detection of submaximal effort in isometric and isokinetic knee extension tests.
        J Orthop Sports Phys Ther. 1996; 24: 19-24