The Knee 14 (2007) 2 – 8

Review

Psychometric properties of measurement tools for quantifying knee joint position and movement: A systematic review Pagamas Piriyaprasarth ⁎, Meg E. Morris School of Physiotherapy, The University of Melbourne, Victoria, 3010, Australia Received 26 June 2006; received in revised form 3 October 2006; accepted 15 October 2006

Abstract This systematic review critically evaluates literature on the reliability and validity of measurement tools for quantifying knee joint angles and knee movement. A search was conducted of seven medical databases and one biomedical engineering database, yielding 43 articles that reported reliability or validity. Tools for quantifying knee joint angles included standard handheld goniometers, fluid-based goniometers, gravity-based goniometers, photographs and two dimensional (2-D) motion analysis. Knee movement was measured with electrogoniometers, 2-D and three dimensional (3-D) motion analysis. Intraclass correlation coefficients for testing knee angles ranged from 0.51–1.00 for intratester reliability and 0.43–0.99 for intertester reliability. For quantifying knee position, sequential MRI and 2-D had the least error of measurement, followed by hand held goniometers and photographs. For dynamic measurements, electrogoniometers and 3D motion analysis were most reliable and had low error of measurement. Strong concurrent validity was found between hand held goniometers and radiographs, as well as between hand held goniometers and 3-D motion analysis. © 2006 Elsevier B.V. All rights reserved. Keywords: Knee; Systematic review; Review; Reliability; Measurement

Contents 1. 2. 3.

Introduction . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . 3.1. Search strategy yields . . . . . . . . . . 3.2. Quality of review articles . . . . . . . . 3.3. Participants . . . . . . . . . . . . . . . 3.4. Testers. . . . . . . . . . . . . . . . . . 3.5. Static knee joint angle measurements . . 3.6. Dynamic knee joint angle measurements 3.7. Statistical analysis . . . . . . . . . . . . 3.8. Validity of measurement tools. . . . . . 3.9. Reliability of measurement tools . . . . 4. Discussion . . . . . . . . . . . . . . . . . . . 5. Conclusion . . . . . . . . . . . . . . . . . . . 6. Potential conflict of interest . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

⁎ Corresponding author. Tel.: +61 3 9265 1547. E-mail address: [email protected] (P. Piriyaprasarth). 0968-0160/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.knee.2006.10.006

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

3 3 3 3 3 3 4 4 5 5 5 5 6 6 6 6 6

P. Piriyaprasarth, M.E. Morris / The Knee 14 (2007) 2–8

1. Introduction This systematic review critically evaluates the methods used to quantify knee joint flexion and extension in both static and dynamic conditions. The Cochrane Collaboration guidelines for conducting a systematic review [1] were used to generate protocols to address the research question: “How reliable and valid are measurement tools for quantifying sagittal plane knee joint angles and movements in humans?” Several measurement tools have been used to quantify knee joint angles and knee movements. The selection of a measurement tool depends on the purpose of testing [2] and psychometric properties such as reliability and validity [3]. To be valid, tools for quantifying knee position and movement need to produce minimal measurement error. Measurement error can arise from the tool, the tester or from variability in the performance of the individual [4,5]. Static knee joint measurements include clinical tests of passive and active range of motion [6]. Passive range of motion tests are typically performed by an external force, delivered by a human [3,7–9] or custom-built instrument to position the knee [10,11]. Active range of motion tests are performed by individuals who are instructed to move their knee joint to a desired position. Both active and passive range of motion are usually measured at the end range of knee flexion [3,8,12,13] or extension [3,14–16]. This information assists clinicians and coaches to detect pathological knee joint conditions according to the degree of extensibility and stiffness [2]. Active range of motion tests can provide information about muscle function [2], whilst passive range of motion tests are typically used to quantify the integrity and function of non-contractile structures such as the knee joint capsule, cartilages and ligaments [2]. This review included all measurement tools used for quantifying either active or passive knee joint motion. Dynamic knee joint measurements are seldom performed in the clinical setting due to the limitations of clinical measurement tools for quantifying motion [17]. Assessment of knee joint movement is mostly performed in the laboratory where accurate dynamic knee joint measurements are possible [18– 21]. Tools used include two dimensional (2-D) motion analysis, three dimensional (3-D) motion analysis, electrogoniometers and accelerometers for measuring walking [20,22], stair climbing [23] and moving from sitting to standing [24]. 2. Methods One of the authors (PP) conducted an electronic search using seven medical databases for articles published between 1966 and March 2006 (Cochrane Library, Medline@OVID (1966 to March Week 2 2006), CINAHL (1982 to March Week 2 2006), Embase (1988 to 2006), Sport Discus (1830 to March 2006), PsyInfo and Pubmed) and an engineering database (IEEE Xplore). The keywords used in the search pertained to knee joint measurement, validity and reliability. Reference lists from all papers were hand searched to identify additional studies. After completing the search, the total yields were retrieved. Two reviewers (PP and MM) independently assessed the relevant articles from titles and abstracts, applying pre-determined inclusion and exclusion

3

criteria. When a decision could not be made from titles and abstracts, the full text was examined. To be included, studies needed to be published in English and to measure knee position or movement in the sagittal plane. Studies also had to report the psychometric properties of the measurement tool or procedure. Articles that reported research conducted in animals, cadavers or using mechanical models were excluded. Two reviewers compared the studies selected in a meeting and consensus was used to reach an agreement when disagreement existed. For the quality assessment, two reviewers (PP and EG) independently appraised the methodologies used for each full text article. The key quality checklist items were participant characteristics, testers, tool descriptions and the selection of statistical methods used to analyze psychometric properties. Participant characteristics documented included age and sex, methods of participant recruitment and inclusion and exclusion criteria. The tester checklist included descriptions of the testers, their experience and their training. The authors (PP and MM) extracted and tabulated information on the participant and tester numbers, the type of reliability and validity studies, measurement protocols, statistical analyses, reliability coefficients and validity indices.

3. Results 3.1. Search strategy yields Of the 921 articles yielded, 837 unsuitable articles were discarded. Eighty-four articles were identified as potentially relevant and 42 of these met the inclusion criteria. Four further articles [13,25–27] were obtained from a hand search through references of the articles obtained. The full texts of the 46 articles were then obtained and three more articles were discarded as they did not meet the criteria. In total, 43 studies met the selection criteria and were included for further analysis. There were 18 reliability studies and nine which reported the validity of measurement tools. Sixteen studies reported both reliability and validity. A summary of the key studies is presented in Table 1. 3.2. Quality of review articles The quality evaluation for the selected articles is summarized in Table 2. Convenience sampling methods were generally used. The number of testers and their experience in using the measurement tools were usually reported [7,13,28,29] and standardized measurement procedures were used in most studies [8,10,13,16,25,30]. In some investigations, the testers used their own non-validated techniques to quantify knee joint angles [3,7]. Statistical analysis methods were reported for all studies. 3.3. Participants The studies included were limited to healthy populations [12,16,25,26,31] and some pathological knee conditions [11,29]. Most of the reliability and validity studies were conducted in healthy adults and six studies used paediatric samples. Of these, two included healthy children [9,32], two studies examined those with cerebral palsy [33,34] and another two included both healthy children and those with cerebral palsy [28,35]. Four investigations were on adults with knee joint pathologies. These included studies of surgery to correct lower limb alignment [29] or knee joint restrictions [11]. Participants who had knee joint range of motion measurements as a part of their assessment were also included [3,7]. The number of participants ranged from 15–60. Two studies tested a single participant to examine the variation of a measurement technique rather than variations between participants [10,30]. Detailed justifications for sample sizes were not provided in any study.

4

P. Piriyaprasarth, M.E. Morris / The Knee 14 (2007) 2–8

Table 1 Identified studies of psychometric properties of knee joint angle and movement measurement techniques Measurement tools/techniques

Psychometric properties Intertester reliability

Static knee joint angle measurements Visual estimation – Standard hand held goniometer

Studies

Intratester reliability Validity

SEM

ICC = 0.82–0.98

–

[7,11,38]

2.37°

[ 3 , 7 – 9 , 11 – 1 3 , 1 6 , 2 5 , 2 6 , 2 9 – 31,34,39,45]

–

–

[3,28]

–

–

[3] [11,12] [16,32,37,42]

r = 0.25 to 0.94 ICC = 0.82–0.98 r = 0.33–0.99 ICC = 0.98–0.99

Parallelogram goniometer Gravity-based goniometer

CV = 13%–21% r = 0.50–0.99 ICC = 0.62–0.99 r = 0.60–0.92 ICC = 0.61–0.92 r = 0.80–0.92 ICC = 0.71–0.99 ICC = 0.43–0.98 –

Fluid-based goniometer Internet-based goniometer

r = 0.83 ICC = 0.96–1.00

CV = 10%–20% r = 0.86–0.97 ICC = 0.85–0.99 r = 0.34–0.99 ICC = 0.97–0.99 r = 0.96–0.97 ICC = 0.96–0.97 ICC = 0.85–0.99 r = 0.99 CV = 0.5%–3.2% ICC = 0.80–0.91 – ICC = 1.00

Goniometer and photograph Goniometer and videography Sequential MRI 2-D motion analysis

– ICC = 1.00 r = 0.96 –

ICC = 0.51–0.87 ICC = 1.00 r = 0.95–0.97 ICC = 0.69–1.00

– – r = 0.99 ICC = 1.00

– [8] LOA = − 1.9° to [10] 1.8° 1.65°–3.95° [14,15] – [24] 1.05°–1.29° [18] 0.84°–1.19° [36]

–

r N 0.90

0.91°–2.3°

Small plastic goniometer Metal goniometer

Dynamic knee joint angle measurements Accelerometers, gyroscope and – markers Electrogoniometer –

r = 0.33–0.99 r = − 0.33–0.79

– –

r = 0.82–0.83 ICC = 0.99–1.00

Mean difference = 1°– 2.5°–3.0° 1.2° – ICC = 0.82–0.89 – 1.57°–1.70° r = 0.99 CMD (r2 ) = 0.88– r = 0.99 3.5° 0.988 CMD (r2) = 0.91– CMC = 0.94–1.00 rms = 0.6°–1.6° 0.99

2-D motion analysis 3-D motion analysis

r = 0.93–0.95

[47] [19,21,27,43,54] [23,36] [20,26,33,35,40,41,55,56]

r = reliability coefficient, ICC = intraclass correlation coefficient, CV = coefficient of variation, SEM = standard error of measurement, LOA = Limits of agreement, CMD = coefficients of multiple determination, CMC = coefficient of multiple correlation, rms = root mean square.

3.4. Testers Intratester reliability was usually evaluated by a single tester [14,16,20,21,23,27,33,35–37]. There were 2–14 testers in studies of intertester reliability. [3,7–13,16,24,26,28–31,34,38–42]. These included physiotherapists [3,7,10,11,13,16,24,28,30,31,34,42], physiotherapy students [8,9,12], physicians [42] and students of biomedical engineering [18]. The experience of the testers was not

uniformly reported in the 43 studies. The use of paired testers in some investigations [3,7,29] aimed to minimize bias in selecting testers in intertester reliability studies. 3.5. Static knee joint angle measurements Visual estimation techniques were used to quantify static knee joint angles in three investigations [7,11,38]. Measurement tools for

Table 2 Methodological quality of the reviewed studies Subject characteristics (n = 43) Definition

Sampling method

Inclusion criteria/ exclusion criteria

Adequate (n = 31) Partial (n = 6) Inadequate (n = 6)

Not stated (n = 23) Convenience (n = 19) Populationbased (n = 1)

Stated (n = 26)

n = number of studies.

Limited (n = 12) Not stated (n = 5)

Tester

Adequate (n = 4) Partial (n = 17) Inadequate (n = 21) Not stated (n = 1)

Tool description

Described (n = 42) Limited (n = 1)

Psychometric properties Reliability design (n = 33)

Protocol description

Statistical analysis

Validity design (n = 26)

Statistical analysis

Intratester (n = 6)

Stated (n = 32) Limited (n = 1)

Adequate (n = 33)

Construct (n = 2) Concurrent (n = 24)

Adequate (n = 24) Inadequate (n = 2)

Intertester (n = 5) Intra- and intertester (n = 15) Test–retest (n = 7)

P. Piriyaprasarth, M.E. Morris / The Knee 14 (2007) 2–8

the static measures included standard hand held goniometers [3,7,8,11–13,16,25,26,29–31,34,37,39], gravity based tools such as fluid based goniometers [8] or flexometers [42]. Reflective markers coupled with photographs and parallelogram goniometers [11,12] were also reported. Parallelogram goniometers are modified goniometers with a movable arm elevated above a stationary arm. The movable arm minimizes the difficulty in attaching the goniometer to knee joint segments. Two dimensional motion analysis [24] and MRI [18] were also used for static knee joint angle measurements. 3.6. Dynamic knee joint angle measurements Measurement tools used for dynamic movements such as walking [21] or moving from sitting to standing [24] included electrogoniometers [19,21,27,43,44], 2-D [23,36] and 3-D motion analysis [19,23,26,35,40]. 3.7. Statistical analysis Common methods used for statistical analysis were Pearson's product correlation coefficients (r) and intraclass correlation coefficients (ICC). The r is an index of the degree of association between two sets of continuous data whilst ICCs provide an index of the mean change between two or more variables [5]. Different ICC models were reported for the 43 studies [7,11,16,24,31], with some reporting the use of ICCs without identifying the model [3,10,12,14,15,28,36,38]. Using solely r or ICC values might not be able to fully explain differences between measurement tools because these correlation coefficients do not indicate the magnitude of measurement error [4]. The standard error of measurement (SEM) was used to provide information on measurement error in the same unit of measurement for only a small number of investigations [14,18,36,45]. It is recommended that future studies use the SEM in addition to ICCs or r to enable the error of measurement to be estimated. 3.8. Validity of measurement tools Twenty-six studies documented the validity of measurement tools for quantifying sagittal knee joint angles or motion. Criterion related validity in the form of concurrent validity was most often reported [8,11,12,23,25,31,36,46]. Radiographs were used as the criterion measurements in four concurrent validity studies, two of which used standard plastic hand held goniometers [25,31] and two which used parallelogram goniometers [11,12]. Standard hand held goniometers were sometimes used as the criterion measure against fluid-based goniometers [8], 3-D [26] and internet-based goniometers [10]. For the criterion related validity studies, there was a large range of r values and variation in ICCs which may have been related to differences in measurement tools and procedures. For the included studies, r values ranged from 0.25 [11] to 0.99 [25,47], indicating poor to excellent concurrent validity [5]. For the studies where ICCs ranged from 0.82–1.00 [7,10], it could be assumed that measurement tools could be used interchangeably. The correlation coefficients for goniometers and visual estimation against radiographs suggested that there was poor criterion related validity for measuring knee extension [7]. Visual estimation was tested against standard plastic goniometers and photo images in three studies. Visual estimation of

5

knee extension was not as accurate as goniometric measurement [11,12] or data taken from a photo image [38]. There was a strong positive relationship between visual estimation techniques and goniometric measurements (ICC or r N 0.80) [7,11] except for the visual estimation of knee extension [11]. Measurements of knee extension angles using visual estimation showed lower correlations with goniometric measurements than for knee flexion. Gravity-based goniometers and linear measurements of heel distance in different knee positions were used in healthy children, with a correlation r of 0.79 between these measures [32]. A high Pearson's correlation coefficient does not always represent good agreement between measurements taken from two different tools [5]. Without reporting the SEM or the mean differences between measurement tools, the actual agreement cannot be fully ascertained. The magnitude of knee movement and the direction of movement were found to influence concurrent validity. Brosseau [12] tested the validity of a parallelogram goniometer against radiographs in healthy participants and reported that measurements of small range knee movements were less accurate than larger movements. The criterion related validity of the range of motion measurements in knee flexion was better than for knee extension [11,12,25,31]. Brosseau [11] and Enwemeka [25] found that measurements of knee extension angles was less accurate than knee flexion angles using goniometers, even though the participant samples were different. Brosseau [11] tested participants with limited knee joint range of motion whilst Enwemeka [25] used healthy participants. 3.9. Reliability of measurement tools Intertester reliability, intratester reliability, inter-device reliability and test-retest reliability were reported in 33 studies on sagittal plane knee joint measurements. Several measurement tools were found to be reliable for measuring knee angle in flexion and extension [3,10,11,15,16]. These were standard hand held goniometers, fluid-based goniometers, gravity-based goniometers, 2-D motion analysis and MRI. Most of the 33 studies used standardized measurement protocols for quantifying knee joint angles or movements such as specified by the American Academy of Orthopaedic Surgeons [48] and Norkin and White [49]. A standardized protocol optimizes reliability because repeated measurements are similar and the error of measurement arising from variation in measurement protocols is minimized [3,7]. Although a standardized measurement protocol was followed, variations between testers were greater than measurements taken by the same person. This was shown by the finding that intratester reliability of knee joint angle measurement was greater than intertester reliability [3,7,11–13,29,30,39]. This was regardless of the direction of movement and measurement techniques such as passive or active tests. The time between testing also affected reliability even though the same person performed the tests in most studies [14,15,42]. The period varied from separate tests on the same day [3,7] to 30 days between tests [14,15]. Measurements performed on the same day were more likely to have high reliability (ICC = 0.78–0.87) than the measurements performed on different days (ICC = 0.53–0.75) [14,15,28,42]. Presumably testers are more reliable at repeating measurement procedures over short periods. The number of measurements ranged from 1–20. If more than one measurement was performed, the means of several trials were usually used to estimate reliability. Averaging data can increase the

6

P. Piriyaprasarth, M.E. Morris / The Knee 14 (2007) 2–8

reliability coefficient by minimizing the magnitude of differences between measurements, as shown in three studies [10,20,34]. Goniometers were used to measure knee joint angles in participants with conditions such as diplegia [28] and knee joint restriction [11]. The reliability of measurements in diplegic children was greater than for control participants for reasons that were not easily explained. Likewise, Brosseau [11] found that measurements of pathological knees were more reliable than for normal knees although no explanation was given to account for this finding. The direction of knee movement also influenced reliability. Measurements of knee joint flexion were more reliable than for knee extension [3,12,16]. This applied to standard goniometers [3,7,16] and parallelogram goniometers [11]. Differences in measurement error for knee joint flexion and extension were not seen for pendulum goniometers [37], photographs [15] or internet-based goniometers [10]. From this systematic review, it is apparent that goniometric measurements are more reliable than visual estimation for quantifying knee position and movement [7,11]. Standard goniometers appear to be able to be used interchangeably with parallelogram goniometers [11,12], fluid-based goniometers [8] and internet-based goniometers [10]. Sequential MRI [18] and 2-D motion analysis [36] had the least error of measurement, being less than 1.3°. For measurements of knee movement, 2-D analyses were used in dynamic tasks in standing and during sit-to-stand with relatively good intertester reliability (ICC = 0.69–0.92) [24,36]. There were few reports of the error associated with measurements of knee joint movement during gait [20,21,35,40]. High reliability (r N 0.90) was found for both intratester reliability of 3-D analysis [20,26,33,35,41] and the intratester reliability of electrogoniometers [27]. SEMs of less than 3.5° were reported for repeated measurements using electrogoniometers or 3-D analysis [19,21,41]. Thus for quantifying knee movement, either electrogoniometers or motion analysis systems appear reliable. Error of measurement from 3-D analysis arises primarily from marker placement discrepancies [20,50]. For some electrogoniometers, the propensity for the thigh attachment to slip during movement can increase error [27].

4. Discussion This systematic review found several tools to be reliable and valid for quantifying knee position and movement. Hand held goniometers, gravity-based goniometers, fluid-based goniometers, internet-based goniometers, 2-D motion analysis and measurements taken from sequential MRI can be used with confidence for quantifying sagittal knee joint position. For static knee joint angle measurements, defining the coordinate axes using MRI and 2-D motion analysis generated low error of measurement, with high reliability coefficients. 2D motion analysis had less error than goniometric measurement and photographs. Static knee extension measures were not as reliable as flexion measures. For dynamic measures of knee motion, electrogoniometers and 3-D analysis had similar levels of measurement error. Due to parallax error, 2-D analysis appears to be less suitable for measuring knee motion in all planes compared with 3-D analysis [51,52]. Parallel findings to previous literature reviews were found for the extent of variability among testers and differences in measurement protocols [2,37]. The reliability of measure-

ments was greatest when the measurements were performed by the same tester and in a controlled environment using standardized testing protocols. Within-session repeated measurements were more consistent than between-session measurements. Taking the mean of several measurements also increased reliability, compared to taking a single measure. Lower reliability coefficients were reported when the repeated measurements were performed by different testers than by the same tester. The lowest ICCs were reported in measurements using different testers with parallelogram goniometers to measure knee extension angles [12]. Low ICCs were also seen for measurements of both knee flexion and extension angles using photographs by the same tester within 30 day intervals [14]. Differing statistical analysis methods were a contributing factor affecting the interpretation of reliability coefficients and validity indices [5,53]. The factors that affected the reliability of dynamic knee joint angle measurements were different than for static measures. The consistency of participant performance and the ability of the tester to accurately position the measurement tools had a major influence on dynamic measures. Errors from the measurement tool were minimized by careful calibration and application [19–21]. 5. Conclusion To conclude, this systematic review found that several reliable and valid methods are available to quantify knee joint position and movement. Reliability and validity of measurement were enhanced by using validated tools, standardized measurement procedures and by minimizing the number of testers. Knee joint flexion measurements were found to be more reliable than knee joint extension measurements. Using the mean of several measurements enhanced reliability. Adopting standardized measurement protocols and trained testers were also effective strategies for reducing measurement error. 6. Potential conflict of interest There is no potential conflict of interest involved in this review. Acknowledgements The authors would like to thank Erika Gosney for being one of the independent reviewers in appraising the quality of selected articles. Special thanks for Dr. Frances Huxham, Dr. Jenny McGinley, Pamela Fok and Brook Galna for their critical comments on the article. References [1] Green S, Higgins JPT. Cochrane Handbook for Systematic Reviews of Interventions 4.2.5 [updated May 2005]. The Cochrane Library, Issue 3. Chichester: Wiley; 2005.

P. Piriyaprasarth, M.E. Morris / The Knee 14 (2007) 2–8 [2] Miller PJ. Assessment of joint motion. In: Rothstein JM, editor. Measurement in physical therapy. London: Churchill Livingstone; 1985. [3] Rothstein JM, Miller PJ, Roettger RF. Goniometric reliability in a clinical setting: elbow and knee measurements. Phys Ther 1983;63:1611–5. [4] Stratford P, Agostino V, Brazeau C, Gowitzke BA. Realibility of joint angle measurement: discussion of methodology issues. Physiother Can 1984;36:5–9. [5] Russek L. Factors affecting interpretation of reliability coefficients. J Orthop Sport Phys Ther 2004;34:341–9. [6] Gajdosik RL, Bohannon RW. Clinical measurement of range of motion: review of goniometry emphasizing reliability and reliability. Phys Ther 1987;67:1867–72. [7] Watkins MA, Riddle DL, Lamb RL, Personius WJ. Reliability of goniometric measurements and visual estimates of knee range of motion obtained in a clinical setting. Phys Ther 1991;71:90–6. [8] Rheault W, Miller M, Nothnagel P, Straessle J, Urban D. Intertester reliability and concurrent validity of fluid-based and universal goniometers for active knee flexion. Phys Ther 1988;68:1676–8. [9] Rakos DM, Shaw KA, Fedor RL, LaManna M, Yocum CC, Lawrence KJ. Interrater reliability of the active-knee-extension test for hamstring length in school-aged children. Pediatr Phys Ther 2001;13: 37–41. [10] Russell TG, Jull GA, Wootton R. Can the Internet be used as a medium to evaluate knee angle? Man Ther 2003;8:242–6. [11] Brosseau L, Balmer S, Tousignant M, O'Sullivan JP, Goudreault C, Goudreault M, et al. Intra- and intertester reliability and criterion validity of the parallelogram and universal goniometers for measuring maximum active knee flexion and extension of patients with knee restrictions. Arch Phys Med Rehabil 2001;82:396–402. [12] Brosseau L, Tousignant M, Budd J, Chartier N, Duciaume L, Plamondon S, et al. Intratester and intertester reliability and criterion validity of the parallelogram and universal goniometers for active knee flexion in healthy subjects. Physiother Res Int 1997;2:150–66. [13] Boone DC, Azen SP, Lin C-M, Spence C, Baron C, Lee L. Reliability of goniometric measurements. Phys Ther 1978;58:1355–60. [14] Karkouti E, Marks R. Reliability of photographic range of motion measurements in a healthy sample: knee and ankle joint measurement. Physiother Can 1997;49:24–31. [15] Marks R, Karkouti E. Evaluation of the reliability of reflective marker placements. Physiother Res Int 1996;1:50–61. [16] Clapper MP, Wolf SL. Comparison of the reliability of the Orthoranger and the standard goniometer for assessing active lower extremity range of motion. Phys Ther 1988;68:214–8. [17] Karpovich PV, Herden EL. Electrogoniometric study of joints. US Armed Forces Med J 1960;11:424–50. [18] Lerner AL, Tamez-Pena JG, Houck JR, Yao J, Harmon HL, Salo AD, et al. The use of sequential MR image sets for determining tibiofemoral motion: reliability of coordinate systems and accuracy of motion tracking algorithm. J Biomech Eng 2003;125:246–53. [19] Rowe PJ, Myles CM, Hillmann SJ, Hazlewood ME. Validation of flexible electrogoniometry as a measure of joint kinematics. Physiotherapy 2001;87:479–88. [20] Growney E, Meglan D, Johnson M, Cahalan T, An KN. Repeated measures of adult normal walking using a video tracking system. Gait Posture 1997;6:147–62. [21] Isacson J, Gransberg L, Knutsson E. Three-dimensional electrogoniometric gait recording. J Biomech 1986;19:627–35. [22] Chao EY, Laughman RK, Schneider SS, Stauffer RN. Normative data of knee joint motion and ground reaction force in adult level walking. J Biomech 1983;16:219–33. [23] Selfe J. Validity and reliability of measurements taken by the Peak 5 motion analysis system. J Med Eng Technol 1998;22:220–5. [24] Jeng SF, Schenkman M, Riley PO, Lin SJ. Reliability of a clinical kinematic assessment of the sit-to-stand movement. Phys Ther 1990;70:511–20.

7

[25] Enwemeka CS. Radiographic verification of knee goniometry. Scand J Rehabil Med 1986;18:47–9. [26] Bratton CV, Ross EC. Validity and intertester reliability of static joint angle measurements made with the PEAK5 motion analysis system. Phys Ther 1994;74:S113. [27] Kettlekamp DB, Johnson DL, Smidt GL, Chao EY, Walker M. An electrogoniometric study of knee motion in normal gait. J Bone Jt Surg 1970;52:775–90. [28] Kilgour G, McNair P, Stott SN. Intrarater reliability of lower limb sagittal range-of-motion measures in children with spastic diplegia. Dev Med Child Neurol 2003;45:391–9. [29] Barker KL, Lamb SE, Burns M, Simpson AHR. Repeatability of goniometer measurements of the knee in patients wearing an Ilizarov external fixator: a clinic-based study. Clin Rehabil 1999;13: 156–63. [30] Mayerson NH, Milano RA. Goniometric measurement reliability in physical medicine. Arch Phys Med Rehabil 1984;65:92–4. [31] Gogia PP, Braatz JH, Rose SJ, Norton BJ. Reliability and validity of goniometric measurements at the knee. Phys Ther 1987;67:192–5. [32] Hein V. Comparison of a new linear instrument and the gravity goniometer for assessing knee extension ROM among children. J Sport Rehabil 1998;7:61–8. [33] Mackey AH, Walt SE, Lobb GA, Stott NS. Reliability of upper and lower limb three-dimensional kinematics in children with hemiplegia. Gait Posture 2005;22:1–9. [34] Stuberg WA, Fuchs RH, Miedaner JA. Reliability of goniometric measurements of children with cerebral palsy. Dev Med Child Neurol 1988;30:657–66. [35] Steinwender G, Saraph V, Scheiber S, Zwick EB, Uitz C, Hackl K. Intrasubject repeatability of gait analysis data in normal and spastic children. Clin Biomech 2000;15:134–9. [36] Gribble P, Hertel J, Denegar C, Buckley W. Reliability and validity of a 2-D video digitizing system during a static and a dynamic task. J Sport Rehabil 2005;14:137–49. [37] Gajdosik R, Lusin G. Hamstring muscle tightness. Reliability of an active-knee-extension test. Phys Ther 1983;63:1085–90. [38] Kowalsky DS, Audette DJ, Florek TK, Kaplan JS. Reliability of visual estimation of knee ROM. Physiother Can 1994;46:33. [39] Dijkstra PU, de Bont LG, van der Weele LT, Boering G. Joint mobility measurements: reliability of a standardized method. Cranio 1994;12:52–7. [40] Hagemeister N, Parent G, Van De Putte M, St-Onge N, Duval N, De Guise J. A reproducible method for studying three-dimensional knee kinematics. J Biomech 2005;38:1926–31. [41] Besier TF, Sturnieks DL, Alderson JA, Lloyd DG. Repeatability of gait data using a functional hip joint centre and a mean helical knee axis. J Biomech 2003;36. [42] Ekstrand J, Wiktorsson M, Oberg B, Gillquist J. Lower extremity goniometric measurements: a study to determine their reliability. Arch Phys Med Rehabil 1982;63:171–5. [43] Tesio L, Monzani M, Gatti R, Franchignoni F. Flexible electrogoniometers: kinesiological advantages with respect to potentiometric goniometers. Clin Biomech 1995;10:275–7. [44] Willemsen AT, van Alste JA, Boom HB. Real-time gait assessment utilizing a new way of accelerometry. J Biomech 1990;23:859–63. [45] Gribble PA, Guskiewicz KM, Prentice WE, Shields EW. Effects of static and hold-relax stretching on hamstring range of motion using the FlexAbility LE1000. J Sport Rehabil 1999;8:195–208. [46] Houck J, Yack HJ, Cuddeford T. Validity and comparisons of tibiofemoral orientations and displacement using a femoral tracking device during early to mid stance of walking. Gait Posture 2004;19: 76–84. [47] Dejnabadi H, Jolles BM, Aminian K. A new approach to accurate measurement of uniaxial joint angles based on a combination of accelerometers and gyroscopes. IEEE Trans Biomed Eng 2005;52:1478–84. [48] American Academy of Orthopedic Surgeons. Joint motion: method of measuring and recording. Edinburgh: Churchill Livingstone; 1986.

8

P. Piriyaprasarth, M.E. Morris / The Knee 14 (2007) 2–8

[49] Norkin CC, White DJ. Measurement of joint motion: a guide to goniometry. Philadelphia: F. A. Davis Co; 1985. [50] Lafortune MA, Lambert C, Lake M. Skin marker displacement at the knee joint. Proceedings 2nd North American congress on biomechanics, Chicago; 1992. p. 101–2. [51] Sih BL, Hubbard M, Williams KR. Correcting out-of-plane errors in twodimensional imaging using nonimage-related information. J Biomech 2001;34:257–60. [52] Nigg BM, Cole GK. Optical methods: errors in two-dimensional motion analysis. In: Nigg BM, Herzog W, editors. Biomechanics of the musculo-skeletal system. New York: Wiley; 1994. p. 283–4. [53] Eliasziw M, Young SL, Woodbury MG, Fryday-Field K. Statistical methodology for the concurrent assessment of interrater and intrarater

reliability: using goniometric measurements as an example. Phys Ther 1994;74:777–88. [54] Kuiken TA, Amir H, Scheidt RA. Computerized biofeedback knee goniometer: acceptance and effect on exercise behavior in post-total knee arthroplasty rehabilitation. Arch Phys Med Rehabil 2004;85: 1026–30. [55] Andriacchi TP, Alexander EJ, Toney MK, Dyrby C, Sum J. A point cluster method for in vivo motion analysis: applied to a study of knee kinematics. J Biomech Eng 1998;120:743–9. [56] Hart RA, Mote CD, Skinner HB. A finite helical axis as a landmark for kinematic reference of the knee. J Biomech Eng 1991;113:215–22.