Arthrometry

Arthrometry

Arthrometry has long been used as a measure of success or failure for ACL and PCL reconstructions. Arthrometers provide the clinician with a quantifiable method of testing ligamentous laxity of the ACL and PCL. Introduced in 1983 by Daniel et al,¹⁷⁵ the KT1000 was one of the first commercially produced arthrometers used clinically for measuring anterior-posterior laxity (Fig. 19-30).¹⁷⁶ A number of other devices have been developed, but the KT1000 (and KT2000, a KT1000 with an X-Y plotter) remains one of the most commonly cited and clinically used arthrometers.

A number of investigations have been undertaken to evaluate the validity and reliability of arthrometry.^176-180 When one interprets, compares, and applies these results, several factors that must be considered are the displacement force used, the displacement difference considered diagnostic, and study design issues (single testers, multiple testers, etc.).

The reliability of arthrometry can be estimated by a statistical technique called intraclass correlation coefficients (ICCs). Depending on the formula used, this test provides either a measure of reproducibility of individual evaluators or the reproducibility of a measure between evaluators. ICCs can range from 0 (totally unreliable) to 1 (perfect reliability). Reported ICCs for ACL testing with a KT1000 or a KT2000 range from 0.65 to 0.99.^62,181-184 In general, results from testers experienced in arthrometry are more reliable than those from novices, and serial measures between testers are less reliable than serial measures from the same tester. The active quadriceps tests are generally less reliable than any of the passive displacement force tests. The reliability results of the common passive displacement forces (67 N, 89 N, 134 N, maximum manual) are mixed, with no particular displacement force being consistently more reliable than the others.

PCL arthrometry appears to be somewhat less reliable than ACL arthrometry. Huber et al¹⁸⁵ reported ICCs ranging from 0.59 to 0.84 for PCL testing with a KT1000. Similar to ACL testing, results from novice testers were generally less reliable than those from experienced testers, and the reliability of results between different testers was less than that for serial measures from the same tester.

Regardless of the accuracy and reliability reported in the literature, clinicians performing arthrometry must be meticulous in their measurement technique. Daniel¹⁸⁶ suggests that the two greatest sources of error in KT1000 measurements are inappropriate patellar pad stabilization and lack of muscle relaxation. Other key points suggested by Daniel to reduce measurement error include proper lower extremity alignment, accurate placement of the arthrometer, and consistent speed and direction of the application of force.¹⁸⁶

Even though arthrometry provides a measure of an important facet in reconstruction outcome, studies have failed to demonstrate consistent associations between arthrometry and other measures, including functional hop tests and self-report outcome measures.^187–190 The results from these studies suggest that one should not rely solely on arthrometric scores to define reconstruction success or failure.

Arthrometer testing

An arthrometer measures joint motion. The most commonly reported arthrometer in the literature for assessing knee joint motion is the KT-1000 (Med Metrics Corp. Inc., San Diego, California, USA). To perform anteroposterior testing, patients are placed in the Lachman test position (see above) with the knee flexed to 30°. In this position the patella is engaged in the trochlea, so that it does not move during assessment of tibial movement relative to the femur. The arthrometer unit is placed on the anterior tibia and held in place with Velcro straps around the calf. Leg rotation is avoided by supporting the heel in a shallow rubber cup on the couch.

The arthrometer handle applies a force to the tibia usually of 67 N (15 lb) and 89 N (20 lb). The difference in anterior displacement between the two forces is called the ‘compliance index’ and is a frequently quoted measure of knee joint stability. Alternatively, maximal manual force may be used and the injured and non-injured legs compared (side-to-side measurement). Tibial translation (to the nearest 0.5 mm) is measured by the change in relative alignment of pads placed on the tibial tuberosity and patella. However, the translation values seen with arthrometry do not represent actual bony motion specifically. When arthrometer readings are compared with stress radiographs, different values are obtained (Staubli and Jakob, 1991), suggesting that an amount of tissue compression is occurring.

Arthrometer measurement has been found to be consistently accurate. Using maximal manual testing and side-to-side measurement, 90% of conscious and 100% of anaesthetized patients with acute ACL tears had measurements greater than 3 mm (Daniel, Malcom and Losse, 1985). Using 141 uninjured subjects, Bach, Warren and Wickiewicz (1990) showed 99% to have side-to-side measurements less than 3 mm using a force of 89 N.

A number of factors can influence measurement consistency and accuracy. First, muscle relaxation must be obtained. Comparing conscious and anaesthetized patients at force values of 67 N and 136 N, Highgenboten, Jackson and Meske (1989) found side-to-side differences greater than 2 mm in 64% and 81% in conscious patients, but 72% and 83% in anaesthetized patients, respectively. Greater muscle relaxation can be obtained as patients become familiar with the testing procedure, and repeated measurements have certainly been shown to be more effective than isolated tests (Wroble et al., 1990). In addition, arthrometer measurement has been found to be operator dependent (Forster and Warren-Smith, 1989). Consistently accurate results will only be obtained with trained testers who have gained significant expertise. Larger testing forces tend to produce better reproducibility, with maximal manual testing giving the most accurate results with all instruments (Torzilli, 1991; Anderson et al., 1992).

Results

Arthroscopic management of type III tibial eminence fractures allows for anatomic reduction of these fractures with more rapid healing than conventional arthrotomy techniques. We have reported on nine children who were followed for an average of 3.5 years. All patients underwent KT1000 arthrometry with no knee laxity. All patients had excellent function with return to full activities.⁹

Discussion

The main finding of this systematic review is that there is high heterogeneity on the laxity outcomes of arthrometer-assisted anterior drawer and talar tilt test in individuals with CAI. The high heterogeneity may be due to the wide range of arthrometers used, their different mechanism of action and the different assessment procedures including ankle positioning, load applied and methods of measurement.

Diagnosis of ankle laxity through physical examination is known to be dependent of the clinician's sensitivity and experience (18). To overcome this gap, several mechanical testing devices have emerged to objectively measure the ankle ligamentous laxity. These are able to measure the soft tissue structures’ passive stiffness and identify the presence of increased laxity within the talocrural and subtalar joints (90,91). A wide range of arthrometers are presented in the literature to test the ankle laxity. The most commonly used is the Telos device which is used concomitantly with radiographic imaging. The arthrometers that are compatible with imaging devices provide a clearer picture of talus initial and stressed position. When calculating talar position and displacement, the variability in the measurement methods may be a source of heterogenous outcomes. Many measuring methods were reported to measure the anterior drawer test and it remains elusive which measuring method is the best. Using a Telos stress device, Beynnon et al (44) compared 4 common different measuring methods and found that calculating distance between posterior lip of the distal tibia to the talar dome was the most accurate method to ascertain the talus position in relation to the tibial articular surface. The open lateral angle was the most common radiographic method to measure the talar tilt. The open lateral angle measures the angle between 2 lines, one parallel to the distal articular surface of the tibial joint and other parallel to the articular surface of the talar dome. However, nearly half of the studies that radiographically measured the talar tilt did not report the measuring method. Other arthrometers (Blue-Bay device, LigMaster, Quasi Static Anterior Ankle Tester, Dynamic Anterior Ankle Tester or Ankle Flexibility Tester) are used to measure anterior drawer and/or talar tilt, but without the need of concomitant imaging procedures. Although these devices cannot provide an accurate and objective measurement of the talus position relative to the tibia, they allow the assessment of peripheral soft tissue response to external load and measure the ligaments’ stiffness. The measurement of anterior drawer and talar tilt varied according the devices testing characteristics (eg, associated inter rotation, application of internal/external rotation or inversion-eversion torque, ankle position and force applied), the measurement features (radiographic, transducers, electronic or motion sensors) and output (radiographic bony displacement, internal/external or inversion-eversion rotation and/or force-displacement curves) which can explain the heterogenous results found for the anterior drawer and talar tilt across included studies.

Concerning the testing characteristics, the ankle should ideally be tested in neutral flexion and then with 10 to 15° of plantar flexion. Adding some tibial internal rotation (10-25°) may enable some ankle pivoting movement and better test the ankle lateral ligament. When testing the anterior drawer, 150 Newtons of anterior-directed force was the most common method. A radiographic device should be used concomitantly and accurately measure the displacement at the talocrural joints. For the talar tilt test, preferably a 150 Newtons or 4000 Newton-Meters inversion/eversion force should be applied to measure the joint displacement or ligament stiffness.

Measurement of ankle instability was made on sagittal (for anterior drawer test) and coronal (for talar tilt test) planes. Assessing also the axial plane would provide additional insights on how the talus rotates on the calcaneus during anterior drawer and talar tilt tests. During these tests, when the lateral ankle ligaments are incompetent or injured, the talus internally rotates under the tibial articular surface performing a pivot movement which may be related to rotational anterolateral instability (92). Some authors tried to highlight axial plane instability by applying anterior translation with the foot in internal rotation (33,34,46). Other authors choose to describe talar tilts as “inversion” by applying an inversion-eversion rotation torque (61,76,77,83). This points out to the assumption that the pathological movements are multiplanar.

Conservative and surgical interventions were generally effective in restoring ankle stability. However, due to substantial heterogeneity in the measuring methods and outcomes, a comparison between techniques is not possible. Nonetheless, higher postoperative functional outcomes were associated with a greater improvement in objective ankle laxity (20-22). An expert consensus study from the ESSKA-AFAS Ankle Instability Group (93) has recommended at least 3 to 6 months of conservative treatment in CAI patients with mechanical laxity. When conservative treatment fails, the surgical repair (open or endoscopic) is indicated and the Broström-Gould is considered the gold-standard first-line surgical approach (94). When dealing with elite athletes, there is a trend towards earlier surgical treatment (after failure of conservative treatment) in patients presenting with mechanical ligament laxity. In patients with generalized hyperlaxity or poor ligament quality the safest technique is graft reconstruction

Despite the heterogeneity found, there are important findings that can be translated into the clinical practice. An objective definition for mechanical ankle instability was not consensual but there are a few cut-offs that were more frequently found. An arthrometer-assisted ankle anterior drawer ≥10 mm or side-to-side difference of ≥3 mm might be considered as mechanical laxity. A talar tilt ≥10 degrees or side-to-side difference of ≥6 degrees might also be considered as mechanical laxity. These cut-offs may be used to assess the ankle laxity status and guide clinical interventions to restore ankle stability.

Future studies should focus on standardize the testing and measuring methodologies of ankle laxity to normalize the results and achieve more objective and definitive conclusions. This is important to determine a consensual and objective definition of mechanical ankle instability and establish cut-offs that can serve as indication for surgical treatment. Future research should also focus on developing an instrument that is capable of measuring ankle rotation and that is compatible is other imaging devices to correlate the ankle competence (ligamentous laxity) with morphological appearance of ankle ligaments.

We acknowledge several limitations inherent to this systematic review. Included studies showed heterogeneity regarding the devices and measurement methods for objectively measuring ankle laxity which resulted in variable laxity outcomes when comparing injured to uninjured ankles and pre to postoperative assessment of injured ankles. This halts the generalizability of our findings. The variability found on pathologic cut-offs did not allow to reach an objective consensual definition of mechanical ankle instability. Heterogeneity of surgical and conservative interventions and different methods used to measure laxity among those studies precluded the comparison between different interventions and assess which interventions resulted in better outcomes. Sample size was neglected in most of studies and may have led to type I or II errors. Lack of assessment blinding and patient randomization in the included studies indicates that studies have a high risk of detection and selection bias, respectively. The absence of assessing the inter- and intrarater reliability could have led to experimenter bias. When reporting the laxity outcomes of imaging-assisted arthrometric assessment, reporting the rater reliability plays a crucial role to know the homogeneity among repeated measurements and different assessors. Lastly, some of the devices were not appropriate to assess the talus position due to the lack of an imaging-assisted arthrometer or not suitable to assess the talus spacial 3-dimensional movement.

In conclusion, to the best of our knowledge, this is the first systematic review that tries to summarize all available evidence on the measurement of ankle laxity and reports the described measuring methods and pathological cut-offs of ankle laxity in patients with CAI. This systematic review showed high heterogeneity in the scientific literature regarding the arthrometric devices, use of concomitant imaging and measuring methods of arthrometer-assisted anterior drawer and talar tilt tests which led to variable laxity outcomes in individuals with CAI. Future research should focus in standardize the measuring methods, determine a consensual and objective definition of pathological ankle laxity and establish cut-offs that can serve to refine surgical indication.

Patients/Materials and Methods

Physical Examination

Diagnosis of a failed ACL reconstruction is generally straightforward, with the physical examination findings being an abnormal Lachman test with a soft endpoint and a demonstrable pivot shift. The anterior drawer test may be more obvious than in the primary situation, as many patients have lost a meniscus and therefore have lost a component of secondary restraint. The KT-1000 arthrometer plays a role in the evaluation of ACL-reconstructed patients and the assessment of failed ACL surgery. In general, the maximum manual side-to-side difference is greater than 5 mm. This is an accepted criterion for the arthrometric determination of failure.

Results

Study Selection

Database search yielded a total of 950 records and an additional 19 studies were identified through hand reference screening. A total of 68 studies met the eligibility criteria and were included in this systematic review (Fig. 1) comprising a total of 3,235 ankles with CAI.

Fig. 1. PRISMA flow chart.

Stability

3.4 Changes in arthrometry side-to-side difference

There was significant difference in post-operative arthrometry side-to-side difference. Thus, pre- to post-operative change was analysed. Seven studies reported pre-operative and post-operative mean arthrometry side-to-side difference, involving 285 patients in the selective group and 403 patients in the complete group.^{8–10,22,30,31,33,34} The complete group showed significantly greater improvements in mean arthrometry side difference from pre-operative to post-operative phases (MD=−1.53, 95%CI: (−2.36)–(-0.69), p<0.01, I²=92%) (Fig. 3).

Fig. 3. Forest plot showing mean difference of pre-operative to post-operative change in arthrometry side-to-side difference.