Article Text
Abstract
Objective: This study tested the reliability of the ORI laxometer, a newly developed non-invasive device for testing inferior translation of the humeral head in humans.
Design: The instrument was designed to measure displacement between the top of the acromion and the head of the humerus when loaded in a similar fashion to the sulcus sign.
Participants and interventions: Sixteen healthy subjects (32 shoulders) were measured for inferior glenohumeral joint laxity. One observer used the laxometer in 16 subjects on three separate occasions for the intra-observer trial. Three observers measured the inferior shoulder laxity of six subjects on one occasion in the inter-observer trial. Asymptomatic and unstable shoulders were also compared in 12 subjects with shoulder instability.
Main outcome measurements: Translation in the glenohumeral joint
Results: The range of inferior translation of glenohumeral joint in these subjects was between 0.01 mm and 6.5 mm with a mean of 1.5 mm. The intraclass correlation coefficient (ICC) for inter-observer reliability was 0.74. For intra-observer reliability the ICC was 0.76. These results are considered to be good to excellent. There was no advantage gained by using data from all five cycles of testing compared with three cycles. In patients with shoulder instability, laxometer measurements were significantly greater in their unstable shoulders than in their normal shoulders.
Conclusions: The laxometer is easy to use, painless and gives objective measures for inferior glenohumeral laxity, with good intra- and inter-observer reliabilities. The ORI laxometer may be useful for assessing and monitoring global glenohumeral joint laxity which usually involves inferior laxity.
Statistics from Altmetric.com
Shoulder laxity is a description of the looseness of the passive stabilisers (bony geometry, labrum, capsule and glenohumeral ligaments) of the shoulder. The glenohumeral joint requires a certain level of laxity to be able to maintain its range of movement. Instability has been defined as a symptom secondary to increased laxity1; that is, instability results in poor control of humeral movement on the glenoid. The normal range of glenohumeral joint laxity is not known.2 Several studies have evaluated shoulder laxity in asymptomatic athletes and high school and college age students.3 4 The laxity of the glenohumeral joint decreases with age.5 6 There are many clinical tests described for assessing shoulder laxity and instability.7 8 However, most of the tests are subjective, do not provide ordinal data and have poor inter- and intra-observer reliability.9 Neer and Foster10 originally described the sulcus sign test as a test for multidirectional instability.11 A sulcus sign is a palpable and visible dimple created beneath the acromion when applying an inferior force (pulling downward) on the subject’s arm in a sitting position (fig 1B). Tzannes and Murrell studied multidirectional instability, which was defined as the presence of grade 2 or greater laxity in more than one direction on examination under anaesthesia. They found that the sulcus sign was a good indicator of multidirectional instability.7 8 However, when performing this test, the inferior force applied is not standardised and assessing the size of this dimple is subjective. The inter-observer reliability was only fair (ICC 0.60).7
There have been very few attempts to quantify shoulder laxity with instrumentation, such as the KT-1000 in the knee. Jörgensen and Bak12 used a Donjoy knee laxity tester (dj Orthopaedics, Vista, CA, USA) to study anteroposterior translation of the glenohumeral joint. Sauers et al13 developed an instrumented arthrometer to assess glenohumeral joint laxity anteriorly and posteriorly at four force levels.
To our knowledge, however, there is no instrument that measures inferior glenohumeral joint translation or laxity. The purpose of this study was to develop such a device and assess its reliability.
METHODS
Shoulder laxometer
The laxometer was designed to measure the axial displacement between the acromion and the elbow joint when a fixed load was applied to the forearm. When loading the arm, the relative displacement between the acromion and the elbow was assumed to be a reliable measure of the movement between the humeral head and the glenoid.
The ORI laxometer consists of four parts: a C-shaped aluminium frame device and shoulder cuff, a slab and pulley system, dedicated electronics and a computer. A schematic illustration of the C-shaped aluminium frame device, with dimensions, is shown in fig 2. The height of the aluminium frame was 42.5 cm. The upper section of the aluminium frame, which connected with the shoulder cuff, measured 15.5 cm. The bottom section of the laxometer measured 18.5 cm. An electronic sensor (displacement transducer) was placed on the bottom piece 9.5 cm from the rear at the frame. A plastic block 3 cm thick by 7.5 cm in length was fixed to the middle of the rear section such that the elbow was aligned with the displacement transducer. The aluminium frame had three levels of two holes that enabled the height of the plastic block to be adjusted.
A shoulder cuff made of plastic was designed to sit on the top of the shoulder and specifically on the acromion (fig 1A). It measured 13 cm across, 6 cm in depth and 16 cm in length. The shoulder cuff could be adjusted to fit over either the right or left shoulder. The cuff was modified during manufacture to ensure that it firmly fitted any shape of shoulder. This was aided by strapping, which was fixed around the torso to firmly hold the device in location as in fig 1A.
Additional strapping, running from the shoulder cuff to the wrist, passively held the arm at 90° fixed flexion. This enabled the elbow to be used as a second point of reference. The displacement transducer had a flat end that was pressed firmly against the underside of the elbow and supported with a spring. The displacement transducer could be adjusted to accommodate arms of different sizes. The slab and pulley system was designed for loading the glenohumeral joint. It included a form-fitting L-shaped slab, a pulley system and a 15 kg-weight stack. The form-fitting L-shaped slab measured 7 cm in height and 18 cm in length. A distal humeral cuff, measuring 9 cm across and 3 cm in depth, was positioned on the distal part of the arm. A proximal forearm cuff, measuring 8 cm across and 2.5 cm in depth, was positioned on the proximal part of the forearm, fixing the arm at 90° flexion. The 90° forearm slab was used for applying load to the glenohumeral joint.
The slab was connected to a stack of three 5 kg (15 kg total) weights via a cable and pulley system and the free end of the cable was attached to a handle. When the operator pulled the handle in an upward motion, the weights were lifted off the floor, with the weight being equally distributed between the subject’s arm and the operator. As such, the total weight applied to the subject’s arm through the glenohumeral joint was 7.5 kg while the subject was seated. This force is comparable to that which can reveal a sulcus sign if present (fig 1B).
Raw voltage signal and calibration
The displacement transducer was connected to dedicated 1 electronics, which generated a voltage proportional to the displacement of the displacement transducer. The signal was collected via an analog-to-digital data acquisition (DAQ) card – 6024E (National Instruments, Austin, TX, USA) via a laptop computer. The raw voltage signals were collected at a rate of 100 Hz, smoothed using a moving average and dissipated into 100 samples per second. Data was displayed in near real-time on a computer monitor via a displacement-time plot (fig 3). At the end of the test the data was saved in ASCII file format. All programming was compiled in Labview 6.1 (National Instruments Ltd). Calibration involved moving the displacement transducer a 10 mm distance. The voltage reading from the transducer was then collected.
Measurement protocol
Ethics approval was obtained from the Southeast Sydney Area Human Ethics Committee (Southern Division) for inter-observer and intra-observer trials. Sixteen subjects (8 females and 8 males with a mean age of 28.9 (SD 6.3) years) were recruited from staff of the Orthopaedic Research Institute and medical students from the University of New South Wales. Informed consent was obtained from each subject before testing. Exclusion criteria included: previous surgery, fracture or recent pain to either shoulder.
Reliability of the laxometer
Inter-observer reliability trial
Three observers, identified here as A, B (orthopaedic surgeons) and C (a biomechanical engineer), were used for the inter-observer reliability trial. Six subjects were chosen for the inter-observer testing from the total sample size of 16 and tested for inferior laxity assessment of both shoulders. All observers were blinded to the results of each other’s assessments. The testing was conducted within a 1-week period with each observer testing the subjects on a different day.
Intra-observer reliability trial
The intra-observer reliability trial was based on the total sample size of 16 subjects. One of the observers tested both shoulders of the 16 subjects with the laxometer on three different occasions, each separated by a week, over a 2-month period.
Laxometer testing in patients with shoulder instability
For comparison, laxometer readings were obtained from the shoulders of 12 patients who had one unstable shoulder and one asymptomatic shoulder.
Experimental procedure
Prior to testing, all observers were briefed on the study protocol and had one training session to ensure familiarisation with the testing protocol. Each subject was seated on a 44 cm high chair and the device placed on the subject’s shoulder as illustrated in fig 1A.
Once comfortable, the operator loaded the glenohumeral joint by lifting the weights off the ground, holding for 10 s and then lowering the weights back to the ground for 10 s. This was repeated over five cycles, the entire test taking 120 s to complete. Data from each cycle was captured as outlined above for subsequent analysis. The voltage signal of each cycle was noted to have four components: (1) a loading phase (weights lifted off the floor); (2) a sustained phase (weights held steady off the floor for 10 s); (3) an unloading phase (weight placed back to ground); and (4) a recovery phase (weights remained grounded for 10 s) (fig 3). Data from all five cycles were compared to data from the middle three cycles to determine which gave the more reliable testing protocol.
The absolute technical error of measurement (TEM) was calculated by taking the sum of the squared differences of the individual measurements, dividing by twice the number of subjects and then taking the square root of the product, as described by Pederson and Gore.14 The relative technical error of measurement (%TEM) was then calculated by dividing the TEM by the average mean of the two sets of measurements and multiplying by 100.
Statistical analysis
Intraclass correlation coefficients (ICCs) were calculated using a two-way random effects model with absolute agreement (2,1), using SPSS version 11.0 statistical software (SPSS Inc., Chicago, IL, USA). In accordance with the suggestions of Fleiss,15 ICC values of <0.4 were taken to represent poor reliability, values above 0.75 excellent reliability and values between 0.4 and 0.75 fair to good reliability. Normal shoulders and unstable shoulders were compared from the same patients using a Wilcoxon signed rank test with SigmaStat software (Systat Software Inc., Point Richmond, CA, USA).
RESULTS
Laxometer
The device took approximately 5 min/shoulder to apply. The testing protocol lasted 120 s/shoulder. The device was well tolerated by all subjects. The range of inferior shoulder laxity measured by the ORI laxometer on our subjects was 0.01−6.5 mm, with a mean of 1.5 mm.
Inter-observer reliability
The inter-observer reliability test of the laxometer assessment of inferior laxity was based on six subjects (fig 4A−D) (3 females and 3 males aged 27−46 years, mean 34 (SD 7.5) years). The ICCs based on five cycles using four combinations of points on the displacement-time plot ranged from 0.65 to 0.74, indicating good reliability, with the most reliable combination of points being (4-3) (table 1).
The ICCs based on measurements made from the middle three cycles ranged from 0.64–0.74 with the reliability being equally good for each of the combinations of points on the displacement-time plot (table 1).
Intra-observer reliability
The intra-observer reliability estimate for inferior shoulder laxity was based on 16 subjects (fig 5A−D) (8 females and 8 males, mean age 29 (SD 6.3) years. The ICCs obtained during laxometer assessment of inferior laxity in millimetres based on five cycles were between 0.75 and 0.76 (table 2), or good to excellent. Those based on three cycles were also good to excellent (0.73–0.76) (table 2). The results showed that all calculation points were equally reliable and there was no advantage to using data from all five cycles of testing compared with three cycles.
Shoulder instability study
The median amount of translation for the 12 unstable shoulders was 3.05 mm and for the asymptomatic shoulders was 1.59 mm. This difference was statistically significant at the p<0.05 level.
Technical error measurement
TEM of the device was 0.76 mm and %TEM was 4.8% for the reading (4-3) point based on five cycles and measurements from the 16 subjects, both right and left shoulder.
DISCUSSION
A method has been devised to measure objectively inferior glenohumeral joint laxity in human subjects. The laxometer is non-invasive and adjustable for both shoulders and for different arm lengths. Inferior glenohumeral joint displacement following a defined load is calculated to the nearest one-tenth of a millimetre. To date, we have exclusively used the laxometer for research purposes, although its straightforward design is potentially modifiable for clinical use. For example, since inferior translation is measurably greater in unstable shoulders than in stable shoulders, this device could be used in specialised musculoskeletal clinics to assess shoulder instability pre-op and post-op.
The reliability of our device was good (0.74) between observers and excellent (0.76) when used by one observer on repeated occasions. The device was very accurate and displayed good to excellent inter-observer and intra-observer reliability.
There are a number of sources of measurement error when estimating glenohumeral joint laxity.16 The shoulder laxometer described in this paper attempted to minimise errors by using a uniform test position, by placing the device on the tested shoulder in a fixed position (ie, on top of the acromion process), and by using a standardised force.
The amount of inferior laxity noted by the laxometer (0.01–6.5 mm) was similar to that found in other studies. Tillander and Norlin17 used pins and sliding ruler intraoperatively in clinically stable shoulders (4 (SD 2) mm), in similar unstable shoulders (9 (SD 4) mm) and in cadavers (3 (SD 4) mm).18 von Eisenhart-Roth et al19 evaluated atraumatic instability using open magnetic resonance imaging and a three-dimensional post procession (4.7 mm), and Jerosch et al20 used ultrasound in healthy individuals (2.4 mm) and in anaesthetised patients with multidirectional instability (6.1 mm).
There are some limitations to the study device, which should be considered. Our device measures only inferior glenohumeral joint laxity and does not measure anterior or posterior joint translation. The reliability of MRI, x ray and ultrasound methods for assessing the glenohumeral joint have not been determined. We have identified two studies that measured anterior and posterior laxity of the glenohumeral joint in human subjects. One was by Jörgensen and Bak12 who used a Stryker knee laxity tester to study anteroposterior translation of the glenohumeral joint in patients with traumatic or atraumatic instability. Sauers et al16 developed an instrumented arthrometer to assess glenohumeral joint laxity in two directions (anteriorly and posteriorly) at four force levels.
Global glenohumeral joint laxity and multidirectional instability are common and often problematic. The laxometer used in our study may be useful in diagnosis and assessing management strategies including surgery of multidirectional instability of the shoulder. The effect of pain and the effects of relative tone and strength of the deltoid were not tested in our study. The training effect of our device was tested in the intra-observer reliability studies and no effect was found.
The ORI laxometer in its current format is relatively bulky and requires a computer, and is more expensive and time-consuming than clinical examination. However, it is cheaper and easier to perform than MRI, x ray and ultrasound methods of estimating glenohumeral joint laxity.
There are many clinical tests described for assessing shoulder laxity and instability; however, these tests are subjective, do not provide ordinal data and often have poor inter-observer and intra-observer reliability. The device described in this paper objectively determined inferior shoulder laxity with a low measurement error and good inter-observer and intra-observer reliability.
What is already known on this topic?
Shoulder laxity is hard to quantify.
What this study adds
A new, inexpensive, reliable method to objectively measure inferior glenohumeral joint laxity.
REFERENCES
Footnotes
Competing interests: None.