Measurement of passive joint range of motion (ROM) is an important part of physical assessment to determine appropriate treatment. Understanding joint ROM limitations enables the clinicians to recognize the kinematics challenges presented during motion. The universal goniometer (UG) is the most common and inexpensive tool used in clinical settings to record joint ROM during physical assessment. The UG is a 180° or 360° protractor with a single axis joining two arms. One arm is movable around the axis while the other arm is stationary.1 Universal goniometers are available in different sizes to suit the joint being measured (Figure 1). However, a review of the literature examining the reliability of the UG demonstrated considerable variation in results for joint ROM measurements.2 Ability to make direct comparison between studies was restricted due to a lack of similarity in the methodologies implemented.2 The number of testers, experience level, number of sessions, time between the sessions, and subject position varied across the studies found. The review also highlighted the gap in current research about the reliability of the UG and the requirement for a more reliable measuring tool.
Currently, three-dimensional (3D) analysis systems such as the Vicon system (Vicon Motion Systems Ltd) have been shown to produce accurate and reliable measurement.3 In addition, these systems have been used in several studies as benchmark to assess the feasibility of other potential systems/devices.3–7 Three-dimensional gait analysis systems are often financially and technically inaccessible in clinical environments, especially where space is a challenge. A specialist training is required to operate these systems. In addition, the outcomes produced by these systems are difficult to understand for some clinicians.3 Hence, the use of alternative systems such as two-dimensional (2D) video analysis systems might be clinically useful. In addition, these 2D video analysis systems provide immediate interpretation that may be useful in the field of prosthetics and orthotics for tuning or alignment of devices.
In this study, the PnO Clinical Movement Data (PnO CMD; previously known as Siliconcoach) was considered as an example of a 2D video analysis system capable of motion capture and gait analysis (Figure 2). The use of a 2D system in clinical practice may aid in documentation of the fitting process and promote improved communication between clinicians. The PnO CMD system is affordable, practical, and easy to use in clinical environments. A single video camera is required to capture sagittal plane ROM. The video may be uploaded into the PnO CMD software for joint ROM measurement. The option to play back the video frame by frame may be useful in identification of passive ROM.
The reliability of the measurement of ankle, knee, and hip angles at initial contact, midstance, and terminal stance phases of gait cycle using PnO CMD has previously been investigated.2 This study also investigated the use of predefined anatomical markers, with the authors concluding that using predefined anatomical markers increased reliability. The intratester reliability varied across the gait cycle and was found to be best in terminal stance (ICC = 1.00).8 In addition, this study concluded high intertester ICC value (ICC = 0.86) for measurement of knee angle at initial contact. A further study9 aimed to determine the reliability of PnO CMD in assessing dynamic and static ROM of the knee joint. Acceptable ICC values (ICC > 0.6) for dynamic and static motion were found.9 A further study10 investigated the intratester and intertester reliability of measuring resisted isometric knee flexion during single leg squat using PnO CMD. Intertester reliability was found to be acceptable for all three testers (ICC > 0.6).10 It was concluded that intratester reliability was found to be better than intertester reliability.
This pilot study aims to investigate the intratester reliability and intertester reliability of 2D video analysis system PnO CMD compared with the UG in measuring passive ROM of the lower-limb joints during physical assessment among healthy candidates. In addition, this study aims to establish the effect of markers on measurement reliability. This study was carried out among healthy subjects because no evidence was found in the literature examining the reliability of this system for measurement of passive joint ROM.
PARTICIPANTS AND TESTERS
Appropriate ethical approval was obtained from the Biomedical Engineering Departmental Ethics Committee, University of Strathclyde. Recruitment posters for testers and participants were displayed within the Department of Biomedical Engineering of the University of Strathclyde. Individuals who showed interest were asked to contact the research team and were provided with additional information and the participant information sheet. A period of 3 days was given to each individual to make a decision on his or her participation. After that, informed written consent was obtained from the participants and testers at the introductory session. Inclusion criteria were as follows: Participants were adult (age >18 years) and did not experience any musculoskeletal or neurological conditions or from any condition resulting in any lower-limb sensory deficit. Participants were excluded if they were unable to attend the scheduled measuring and recording sessions or if there was a change in physical status or injury during the trial period.
Testers were required to be qualified allied health professionals or fourth-year prosthetics and orthotics students with current experience in measurement of joint ROM. Fourth-year prosthetics and orthotics students use the UG throughout the course and have sufficient training in using it. In addition, only students who have completed their clinical placement were included. Testers were excluded if they were unable to attend the scheduled measuring sessions or were unable to complete the video analysis within a given time frame or had no experience in measuring joint ROM.
Sagittal plane ROM of the hip, knee, and ankle joint of the dominant leg during physical assessment was measured with both tools, with and without markers. Markers were applied on the following bony landmarks: shoulder, greater trochanter, lateral femoral condyle of the knee, lateral malleolus, and fifth metatarsal head. Markers consisted of bright-colored adhesive Velcro cut into circular shapes (25 mm) and placed by the same researcher to reduce the variability.
An introductory session was arranged where explanation about the trial was given to the participants. For practical reasons, the participants were divided into two groups. Each group attended two half-day sessions; one with markers and one without markers, with approximately 1-week gap between the sessions. Participants were provided with Lycra suits to wear for all the sessions to minimize the movement of the marker, which may occur due to loose clothing. In addition, each participant was given a time slot to attend for video recording within a separate video recording session. In the video recording session, which lasted for approximately 30 minutes, the camera (Sony Handycam HDR-CX150 Camcorder video camera with 3.1 megapixels) was positioned using a tripod perpendicular to the examination table where the participant was lying down at an appropriate distance to capture the image of the participant from the shoulder to toe. The researcher moved each joint individually into maximum flexion and extension while a video of the motion is captured. The same procedure was repeated again using markers.
An introductory session was arranged where a PowerPoint presentation explaining the measuring method with PnO CMD and UG was given by the researcher to standardize the measuring methods. Additional information was provided in a measuring instruction manual. For the purpose of this study, each tester was asked to record the following measurements on the dominant leg using both devices: maximum hip flexion, maximum hip extension, maximum knee flexion, maximum knee extension, maximum ankle plantarflexion, and maximum ankle dorsiflexion. The testers were asked to attend four half-day sessions; two marker sessions and two no-marker sessions with approximately 1-week gap between each session. In each session, each tester measured the hip, knee, and ankle joint ROM of four participants.
INTRATESTER RELIABILITY OF PNO CLINICAL MOVEMENT DATA
Each participant’s joint ROM of the dominant leg was video recorded with and without markers. After that, each tester was instructed to evaluate each video three times in a time frame of 3 weeks with approximately 1-week gap between each evaluation. Each session lasted for approximately 2 hours. The order of evaluation of the joints (with and without markers) was randomized. In addition, video sequences and the files were randomized between each evaluation. The testers were guided to pause the video at the stage when the researcher holds the joint at the end of the range for 3 seconds and to take the measurement at that position. The order of evaluation of the joints and video sequences was randomized. Special assessment sheets were used to record the ROM measurements using the identification codes provided to each tester and participant for blind analysis.
INTRATESTER RELIABILITY OF UNIVERSAL GONIOMETER
Before each session, the inclusion and exclusion criteria were checked for each participant to ensure that the participant’s status has not changed. Each tester measured the ROM of each participant (hip, knee, and ankle) 3 times in each session. Each session lasted for approximately 3 hours. The order of measuring participants was randomized, and the order of each session (marker/no marker) was also be randomized. Each tester had approximately 10 minutes to measure maximum sagittal plane motion of each participant’s hip, knee, and ankle joints. Special assessment sheets were used to record the ROM measurements using the identification codes provided for each tester and participant for blind analysis.
INTERTESTER PNO CLINICAL MOVEMENT DATA AND UNIVERSAL GONIOMETER
The mean of the three repetitions for each device was calculated with and without markers and compared for each joint between the two devices. It should be noted that intertester values were only able to be calculated if intratester values across all the testers were above the satisfactory level.
To achieve power of 80% at the 5% level of significance, 3 testers (final-year prosthetics and orthotics students) and 8 participants (healthy subjects) were included in the study. Intraclass correlation coefficient model (2, 1) was used after initial summary statistics were produced. This ICC reliability tool was used to assess and compare the reliability of the PnO CMD and the UG along with Bland and Altman plots and an appropriate paired test assessing the significance of actual differences. Intraclass correlation coefficient values above 0.60 were considered to be satisfactory for research purposes.11
PARTICIPANTS AND TESTERS
In this study, a total of eight healthy subjects were recruited. Fourth-year prosthetics and orthotics students were included as testers, and no qualified allied health professionals were recruited.
PnO Clinical Movement Data
The lowest ICC values found were for ankle dorsiflexion for all the testers. However, the values were above satisfactory limits (ICC > 0.6). The highest ICC values for all the testers were found for knee flexion measurements (Table 1). All ICC values were above satisfactory limits and significant. In addition, all Bland and Altman plots illustrated small dispersion and equal distribution of the points above and below zero, confirming the high ICC results found.
Some ICC values were found to be below the satisfactory limits (ICC < 0.06). The lowest ICC value was found for hip extension for one tester, whereas the highest ICC value was found for ankle plantarflexion for one tester (Table 1). In addition, the Bland and Altman plots validated the ICC results achieved.
PnO Clinical Movement Data
The ICC values for all the joints measured were found to vary from 0.24 to 0.98. The ICC values for ankle dorsiflexion for all the testers were found to be lower in comparison to the other motions measured and below the satisfactory limits (ICC < 0.6). In addition, ICC value for one tester for hip extension was found to be lower than the satisfactory limits (0.53). The highest ICC values for all the testers were found for hip flexion measurements as indicated in Table 1. All the Bland and Altman plots verified the ICC results found.
The ICC values across all the joints measured were found to vary from 0.39 to 0.93. The lowest ICC value was found for ankle dorsiflexion for one tester, whereas the highest was found for ankle plantarflexion for another tester (Table 1). Furthermore, the Bland and Altman plots illustrated widespread scattering of the points, confirming the low ICC results achieved.
PnO Clinical Movement Data
The ICC values for all the joints measured ranged from 0.94 to 0.99 and were above the satisfactory level (ICC > 0.60) and significant (Table 1). In addition, all Bland and Altman plots showed random scattered points equally distributed above and below zero; hence, validating the high ICC values obtained.
Only ICC value for ankle plantarflexion was able to be calculated, and this was found to be below the satisfactory level (ICC = 0.39; Table 1). The Bland and Altman plot illustrated large dispersion, confirming the low ICC result found.
PnO Clinical Movement Data
The ICC values for the all the joints measured except ankle dorsiflexion and hip extension were found to range from 0.91 to 0.97. The ICC value for hip extension for one of the tester was below the satisfactory level (ICC = 0.53), which prevented the calculation for intertester reliability. All the ICC values across all the testers for ankle dorsiflexion were below the satisfactory levels (ICC < 0.6), hence; intertester reliability was not concluded for this joint (Table 1). All Bland and Altman plots confirmed the high ICC values achieved. The plots illustrated small dispersion and equal distribution of the points above and below zero.
Only intratester ICC values for ankle plantarflexion across all the testers were above the satisfactory level, but the resulting intertester reliability was lower than the satisfactory level (ICC = 0.47; Table 1). This low ICC value was confirmed by the Bland and Altman plot.
All intratester and the intertester ICC values obtained using PnO CMD with markers for all testers were found to be above the satisfactory limit (ICC > 0.60) with small variations in values, which demonstrates the reliability of using this tool with markers (Table 1; Figures 3–6). Furthermore, it was observed in this study that all intratester and intertester ICC values for the UG (with/without markers) across all joints ranged considerably and, in some cases, were below the satisfactory limits (ICC < 0.60; Table 1; Figure 3–5). This demonstrates the unreliability of using this tool in comparison to PnO CMD (Table 1; Figure 6).
Intratester ICC values for ankle plantarflexion were the only values found to be above satisfactory limits (ICC > 0.6) for both tools across all the testers (Figure 3–5). The resulting intertester reliability for ankle plantarflexion for PnO CMD was found to be higher than intratester reliability (Table 1; Figure 6). On the other hand, the intertester reliability for ankle plantarflexion for the UG was found to be lower than intratester reliability (Table 1; Figure 6). PnO Clinical Movement Data (with/without marker) ICC values for ankle dorsiflexion for all testers were found to be lower in comparison to the other motions measured (Table 1).
To the best of our knowledge, only ICC values of the UG can be compared with the results previously found in the literature, as no study was found investigating the reliability of PnO CMD for passive motion of hip, knee, and ankle joint. One study, Kilgour et al.,12 was found that investigated the intratester reliability for the measurement of hip, knee, and ankle motion using the UG with markers among healthy candidates. The ICC values reported for hip extension, knee extension, and ankle dorsiflexion using similar testing positions were all above the satisfactory limits, findings which do not agree with the findings of this study (Table 1). Peters et al.13 used a similar testing position as in the current study to investigate intratester reliability of the UG without markers for measurements of knee joint motion among healthy candidates. Again, reported ICC values were higher than the values reported in this study.
As recommended in the literature, subjects' position and measurement procedures were standardized in this study with both tools. Rothstein et al.14 and Youdas et al.15 reported an increase in ICC values for the UG when subjects' position was standardized. Another source of error stated in the literature is the variance found between the clinicians in the identification of bony landmarks.16 Markers proved to be useful with PnO CMD as it increased the reliability. However, the effect of markers on the UG reliability was not clear. It was noted that the ICC values for the UG (with/without markers) ranged from weak to excellent with no pattern observed, making it hard to draw any conclusions. A possible explanation is the unequal manual force applied while measuring. Testers were instructed in each UG session to move each joint to maximum ROM, which may have resulted in different force applied between the testers and sessions. On the other hand, this variance was not present with PnO CMD as all the joints were moved by the same researcher and the same video captured on a single occasion was used for all the evaluation.
This study represents an initial step in using advanced technology (PnO CMD) in clinical practice to measure passive joint ROM. High reliability has been illustrated for sagittal plane passive ROM of hip, knee, and ankle joint, and this increases with the use of markers. Markers can be created from cheap available materials such as Velcro and used effectively. This study has established a methodology that can now be applied and tested in the prosthetic and orthotic populations (e.g., subject with amputation or subjects with cerebral palsy). In addition, as this study only involved fourth-year prosthetics and orthotics students, this may have affected the ICC values obtained using the UG and PnO CMD. It would be beneficial to investigate if experience or professional background has an effect on reliability of PnO CMD and UG as this has not been reported sufficiently in the literature.2 Test-retest design (intrasessional intratester) was applied in this study where all the repeated measurements were taken in the same session. Kilgour et al.12 and Wakefield et al.17 reported lower intersessional intratester reliability in comparison to intrasessional intratester reliability for the UG of hip, knee, and ankle joint. Thus, investigating the intersessional intratester for PnO CMD is recommended as clinicians typically measure on different occasions. This study only investigated the reliability of both tools for a single testing position used for hip, knee, and ankle motion. Investigation of reliability of PnO CMD is required using different test positions for joint ROM. Further research will be required to examine the accuracy of this system in measurement of passive joint ROM.
In conclusion, PnO CMD was found to be more reliable than the UG in measuring passive sagittal ROM of the lower-limb joints motion among healthy candidates. In addition, it was found that using markers increased the intratester and intertester reliability of PnO CMD. The present work opens up possibilities for using new technology in joint ROM measurements to achieve more reliable measurements.
1. Clarkson H. Joint Motion and Function Assessment: A Research-Based Practical Guide
. Philadelphia: Lippincott Williams & Wilkins; 2005.
2. Mohsin F, McGarry A, Bowers R. Factors influencing the reliability of the universal goniometer in measurement
of lower-limb range of motion: a literature review. J Prosthet Orthot
3. Davis RB. Clinical gait analysis. IEEE Eng Med Biol Mag
4. Ugbolue UC, Papi E, Kaliarntas KT, et al. The evaluation of an inexpensive, 2D, video based gait assessment system for clinical use. Gait Posture
5. Clarke L, Murphy A. Validation of a novel 2D motion analysis system to the gold standard in 3D motion analysis for calculation of sagittal plane kinematics. Gait Posture
6. Otte K, Kayser B, Mansow-Model S, et al. Accuracy and reliability of the kinect version 2 for clinical measurement
of motor function. PLoS One
7. Lee KM, Chung CY, Kwon DG, et al. Reliability of physical examination in the measurement
of hip flexion contracture and correlation with gait parameters in cerebral palsy. J Bone Joint Surg Am
8. Davidson E, Bowers R. A pilot study investigating the intra and inter-rater reliability of siliconcoach within the field of gait analysis. ISPO 2013 World Congress
. Hyderabad, India 2013.
9. Cronin J, Nash M, Whatman C. Assessing dynamic knee joint range of motion
using siliconcoach. Phys Ther Sport
10. Sparkes V, Brophy R, Sheeran L. Reliability of measuring critical knee flexion angle during single squat using Silicon coach™ in recreational females. Bone Joint J
11. Evers A. The revised dutch rating system for test quality. Int J Testing
12. Kilgour G, McNair P, Stott NS. Intrarater reliability of lower limb sagittal range-of-motion measures in children with spastic diplegia. Dev Med Child Neurol
13. Peters PG, Herbenick MA, Anloague PA, et al. Knee range of motion: reliability and agreement of 3 measurement
methods. Am J Orthop (Belle Mead NJ)
14. Rothstein JM, Miller PJ, Roettger RF. Goniometric reliability in a clinical setting. Elbow and knee measurements. Phys Ther
15. Youdas JW, Bogard CL, Suman VJ. Reliability of goniometric measurements and visual estimates of ankle joint active range of motion obtained in a clinical setting. Arch Phys Med Rehabil
16. Peeler JD, Anderson JE. Reliability limits of the modified Thomas test for assessing rectus femoris muscle flexibility about the knee joint. J Athl Train
17. Wakefield CB, Halls A, Difilippo N, Cottrell GT. Reliability of goniometric and trigonometric techniques for measuring hip-extension range of motion using the modified Thomas test. J Athl Train