Mark W. Lenhoff

Bootstrap prediction and confidence bands: a superior statistical method for analysis of gait data

Gait analysis studies typically utilize continuous curves of data measured over the gait cycle, or a portion of the gait cycle. Statistical methods which are appropriate for use in studies involving a single point of data are not adequate for analysis of continuous curves of data. This paper determines the operating characteristics for two methods of constructing statistical prediction and confidence bands. The methods are compared, and their performance is evaluated using cross-validation methodology with a data set of the sort commonly evaluated in gait analysis. The methods evaluated are the often-used point-by-point Gaussian theory intervals, and the simultaneous bootstrap intervals of Sutherland et al. The Development of Mature Walking, MacKeith Press, London, 1988 and Olshen et al. Ann. Statist. 17 (1989) 1419-40. The bootstrap bands are shown to provide appropriate coverage for continuous curve gait data (86% coverage for a targeted coverage of 90%). The Gaussian bands are shown to provide inadequate coverage (54% for a targeted coverage of 90%). The deficiency in the Gaussian method can lead to inaccurate conclusions in gait studies. Bootstrap prediction and confidence bands are advocated for use as a standard method for evaluating gait data curves because the method is non-parametric and maintains nominal coverage levels for entire curves of gait data.

Accuracy and Reliability of Three Different Techniques for Manual Goniometry for Wrist Motion: A Cadaveric Study

PURPOSE Despite the ubiquitous use of manual goniometry in measuring objective outcomes of hand surgery and therapy, there are limited data concerning its accuracy or repeatability for wrist motion. The purpose of this study was to evaluate the accuracy and reliability (both inter- and intra-rater) in measuring wrist flexion and extension using 3 manual goniometric alignment techniques (ulnar, radial, and dorsal-volar) in cadaveric upper extremities, using fluoroscopic verification of posture as a gold standard. In addition, we sought to assess the accuracy and reliability of the dorsal-volar technique for measurement of radioulnar deviation. METHODS External fixators were applied to 10 cadaveric wrists with intramedullary cannulated rods in the radius and third metacarpal for gold-standard fluoroscopic verification of posture. Manual goniometric measurements with each technique were captured by 2 raters (a hand surgeon and a hand therapist) for reliability measurements and by a single rater for accuracy. Wrists were positioned at angles of maximum flexion, extension, and radial and ulnar deviation for reliability testing and at preselected angles across the range of motion for accuracy testing. At each position, wrist angle was measured with a 1 degrees increment goniometer, and fluoroscopic angles were measured digitally. Intraclass correlation coefficients and root mean square values were calculated for all combinations, and analysis of variance was used to test differences between techniques. RESULTS No technique was statistically less accurate than any other (6 degrees to 7 degrees ). Each method was found to have high intra-rater reliability. For measurement of wrist flexion and extension, the dorsal-volar technique demonstrated the greatest inter-rater reliability, as compared to ulnar and radial, respectively. CONCLUSIONS Although each measurement technique demonstrated a similar degree of accuracy and intra-rater reliability, the dorsal-volar technique demonstrates the greatest level of inter-rater reliability for measurement of wrist flexion and extension. This information is important clinically, as measurements are regularly exchanged between hand surgeons and therapists as a basis for decisions regarding patient care.

Medial Collateral Ligament Injuries and Subsequent Load on the Anterior Cruciate Ligament A Biomechanical Evaluation in a Cadaveric Model

Background Numerous studies have documented the effect of complete medial collateral ligament injury on anterior cruciate ligament loads; few have addressed how partial medial collateral ligament disruption affects knee kinematics. Purpose To determine knee kinematics and subsequent change in anterior cruciate ligament load in a partial and complete medial collateral ligament injury model. Study Design Controlled laboratory study. Methods Ten human cadaveric knees were sequentially tested by a robot with the medial collateral ligament intact, in a partial injury model, and in a complete injury model with a universal force-moment sensor measuring system. Tibial translation, rotation, and anterior cruciate ligament load were measured under 3 conditions: anterior load (125 N), valgus load (10 N·m), and internal-external rotation torque (4 N·m; all at 0° and 30° of flexion). Results Anterior and posterior translation did not statistically increase with a partial or complete medial collateral ligament injury at 0° and 30° of flexion. In response to a 125 N anterior load, at 0°, the anterior cruciate ligament load increased 8.7% (from 99.5 to 108.2 N; P =. 006) in the partial injury and 18.3% (117.7 N; P <. 001) in the complete injury; at 30°, anterior cruciate ligament load was increased 12.3% (from 101.7 to 114.2 N; P =. 001) in the partial injury and 20.6% (122.7 N; P <. 001) in the complete injury. In response to valgus torque (10 N·m) at 30°, anterior cruciate ligament load was increased 55.3% (30.4 to 47.2 N; P =. 044) in the partial injury model and 185% (86.8 N; P =. 001) in the complete injury model. In response to internal rotation torque (4 N·m) at 30°, anterior cruciate ligament load was increased 29.3% (27.6 to 35.7 N; P =. 001) in the partial injury model and 65.2% (45.6 N; P <. 001) in the complete injury model. The amount of internal rotation at 30° of flexion was significantly increased in the complete injury model (22.8°) versus the intact state (19.5°; P <. 001). Conclusion Partial and complete medial collateral ligament tears significantly increased the load on the anterior cruciate ligament. In a partial tear, the resultant load on the anterior cruciate ligament was increased at 30° of flexion and with valgus load and internal rotation torque. Clinical Relevance Patients may need to be protected from valgus and internal rotation forces after anterior cruciate ligament reconstruction in the setting of a concomitant partial medial collateral ligament tear. This information may help clinicians understand the importance of partial injuries of the medial collateral ligament with a combined anterior cruciate ligament injury complex.

Reliability of plantar pressure platforms

Plantar pressure measurement is common practice in many research and clinical protocols. While the accuracy of some plantar pressure measuring devices and methods for ensuring consistency in data collection on plantar pressure measuring devices have been reported, the reliability of different devices when testing the same individuals is not known. This study calculated intra-mat, intra-manufacturer, and inter-manufacturer reliability of plantar pressure parameters as well as the number of plantar pressure trials needed to reach a stable estimate of the mean for an individual. Twenty-two healthy adults completed ten walking trials across each of two Novel emed-x(®) and two Tekscan MatScan(®) plantar pressure measuring devices in a single visit. Intraclass correlation (ICC) was used to describe the agreement between values measured by different devices. All intra-platform reliability correlations were greater than 0.70. All inter-emed-x(®) reliability correlations were greater than 0.70. Inter-MatScan(®) reliability correlations were greater than 0.70 in 31 and 52 of 56 parameters when looking at a 10-trial average and a 5-trial average, respectively. Inter-manufacturer reliability including all four devices was greater than 0.70 for 52 and 56 of 56 parameters when looking at a 10-trial average and a 5-trial average, respectively. All parameters reached a value within 90% of an unbiased estimate of the mean within five trials. Overall, reliability results are encouraging for investigators and clinicians who may have plantar pressure data sets that include data collected on different devices.

Development and validation of a computational model of the knee joint for the evaluation of surgical treatments for osteoarthritis

A three-dimensional (3D) knee joint computational model was developed and validated to predict knee joint contact forces and pressures for different degrees of malalignment. A 3D computational knee model was created from high-resolution radiological images to emulate passive sagittal rotation (full-extension to 65°-flexion) and weight acceptance. A cadaveric knee mounted on a six-degree-of-freedom robot was subjected to matching boundary and loading conditions. A ligament-tuning process minimised kinematic differences between the robotically loaded cadaver specimen and the finite element (FE) model. The model was validated by measured intra-articular force and pressure measurements. Percent full scale error between FE-predicted and in vitro-measured values in the medial and lateral compartments were 6.67% and 5.94%, respectively, for normalised peak pressure values, and 7.56% and 4.48%, respectively, for normalised force values. The knee model can accurately predict normalised intra-articular pressure and forces for different loading conditions and could be further developed for subject-specific surgical planning.

Electromyographic activity in stiff and normal elbows during elbow flexion and extension.

This prospective, controlled pilot study investigated electromyographic activity in elbow muscles during active elbow flexion and extension and during prolonged elbow extension with and without resistance. Biceps brachii, brachialis, brachioradialis, and triceps activity was measured in 10 subjects with elbow motion deficits after injury and in 10 controls. Surface electromyography was recorded during active elbow flexion, extension, and passively positioned elbow extension with and without 3 lb on the distal forearm. All muscles of the stiff elbow group had greater activity compared with controls during active elbow flexion and extension. Biceps brachii of the stiff group showed antagonist activity equivalent to agonist. For all other flexors in both groups, agonist activity was greater than antagonist. During prolonged extension, biceps brachii, unweighted, showed greater activity in the stiff group than in controls. Both groups showed greater flexor activity when weighted. Across time, activity was sustained or increased in all muscles in both groups.

Proximal femoral varus rotation osteotomy in cerebral palsy: a prospective gait study.

This prospective study examined the kinematic and temporal/spatial effects of proximal femoral varus rotation osteotomy (VRO) on the gait of individuals with cerebral palsy from preoperative to 1-year postoperative status. Participants were a consecutive sample of 37 individuals (14 males, 23 females). The analysis consisted of three-dimensional kinematics, temporal/spatial measures, and functional status. A curve representing the difference between the preoperative and 12-month postoperative conditions was calculated for each joint motion. Ninety-five percent confidence intervals were calculated about the mean difference curves using a bootstrapping technique. VRO resulted in improved cosmesis and objective improvement in gait, including a statistically significant increase in hip external rotation and hip extension, a decrease in anterior pelvic tilt, and an increase in knee extension strength. The use of confidence bands to identify surgical outcomes with respect to kinematic variables has enormous value for patients and professionals. Long-term follow-up is needed to see if the aforementioned gains improve.

Development of an anatomical wrist joint coordinate system to quantify motion during functional tasks

The purpose of this study was to develop a three-dimensional (3D) motion analysis based anatomical wrist joint coordinate system for measurement of in-vivo wrist kinematics. The convergent validity and reliability of the 3D motion analysis implementation was quantified and compared with manual and electrogoniometry techniques on 10 cadaveric specimens. Fluoroscopic measurements were used as the reference. The 3D motion analysis measurements (mean absolute difference [MAD] = 3.6°) were significantly less different (P < .005) than manual goniometry (MAD = 5.7°) but not (P = .066, power = 0.45) electrogoniometry (MAD = 5.0°) compared with fluoroscopy. The intraclass correlation coefficient (ICC[2,1]) was highest for 3D motion analysis compared with manual and electrogoniometry, suggesting better reliability for this technique. To demonstrate the utility of this new wrist joint coordinate system, normative data from 10 healthy subjects was obtained while throwing a dart.

The effect of wrist surgery on the kinematic consistency of joint axis reconstruction in a static posture.

Three‐dimensional analysis of wrist motion is a growing focus in orthopedic research, however, our understanding of its validity (accuracy and reliability) remains limited. Nine human cadavers were tested to estimate wrist joint axes alignment in a postural static pose. The objective was to investigate a raters ability to reliably align three skin‐ tracked wrist joint coordinate system (WJCS) definitions across baseline and reconstructive wrist states (intact, mid‐carpal arthrodesis, and proximal‐row carpectomy). Two WJCSs (legacy, anatomic) were based on palpated bony landmarks and the third (functional) was based on both landmarks and passive flexion‐extension motion. A coordinate frame based on the anatomic definition was tracked with bone pins and served as a reference. Each WJCS was tested in each wrist state and in three forearm position (45° pronation, neutral, 45° supination). The angular offset about each axis of the WJCS frames were calculated with respect to the reference in flexion‐extension, radial‐ulnar deviation, and pronation‐supination for every iteration. Reliability and root mean square deviation values were analyzed across wrist states. Our data suggest that no WJCS is uniformly more reliable than another. The functional WJCS definition was most consistent across intact and post‐operative states for pronation‐supination offset, but this was dependent on rater interpretation. It still however offers the practical benefit of requiring fewer landmarks.

AN EXTENSION TO THE ISB WRIST JOINT COORDINATE SYSTEM: AN IN VITRO ACCURACY STUDY

INTRODUCTION Investigators have used a variety of marker sets and methods to describe upper extremity kinematics. However, without a standard set of anatomical and joint coordinate system definitions for 3-D motion analysis (3DMA), data sets cannot be compared between investigators. In an effort to summarize the descriptions of upper extremity motion, Wu and colleagues [1] published a set of joint coordinate system recommendations. This proposed kinematic solution is an extension from that standard. The purpose of this investigation is to develop an anatomically based wrist joint coordinate system (WJCS) and appropriate tracking frames. To test this new WJCS a comparison between the 3-D motion analysis (3DMA) implementation, manual goniometry (MG), and electro-goniometry (EG) was performed. We hypothesize that 3DMA would provide increased accuracy over the other methods.