Article Text

PDF

Static anteroposterior knee laxity tests are poorly correlated to quantitative pivot shift in the ACL-deficient knee: a prospective multicentre study
  1. Eleonor Svantesson1,
  2. Eric Hamrin Senorski2,
  3. Julia Mårtensson1,
  4. Stefano Zaffagnini3,
  5. Ryosuke Kuroda4,
  6. Volker Musahl5,
  7. Jón Karlsson1,6,
  8. Kristian Samuelsson1,6
  9. PIVOT study group
    1. 1 Department of Orthopaedics, Institute of Clinical Sciences, The Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
    2. 2 Department of Health and Rehabilitation, Institute of Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
    3. 3 Laboratorio di Biomeccanica e Innovazione Tecnologica, Istituto Ortopedico Rizzoli, Bologna, Italy
    4. 4 Department of Orthopaedic Surgery, Kobe University, Kobe, Japan
    5. 5 Department of Orthopaedic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
    6. 6 Department of Orthopaedics, Sahlgrenska University Hospital, Mölndal, Sweden
    1. Correspondence to Dr Eleonor Svantesson, Department of Orthopedics, Institute of Clinical Sciences, The Sahlgrenska Academy, University of Gothenburg, Mölndal SE-431 80, Sweden; eleonor.svantesson{at}outlook.com

    Abstract

    Objective To determine the relationship between preoperative static knee joint laxity and non-invasive quantitative pivot shift (QPS) in patients with anterior cruciate ligament (ACL) rupture.

    Methods Patients with an ACL injury participating in a multicentre trial were analysed if they had complete preoperative data on the following laxity tests: the rolimeter, the KT-1000 (134 N and manual maximum force), the Lachman, the anterior drawer and QPS. The QPS was assessed via a non-invasive inertial sensor system and an image analysis system for tibial acceleration and lateral tibial translation, respectively. Awake examination and examination under anaesthesia (EUA) were performed. Correlation between absolute values of static laxity and the QPS for each leg was assessed by Spearman’s rho. The Lachman and the anterior drawer were dichotomised into low- and high-grade, and differences between the groups in terms of continuous values of QPS were assessed.

    Results A total of 58 patients were included (41.4% women, mean age 27.1±9.8 years). Awake static laxity and QPS acceleration were negatively correlated in the ACL-deficient knee, meaning that a greater acceleration correlated to a lesser static tibial translation, and vice versa. The mean QPS acceleration correlated with the static tests as follows: the rolimeter r=−0.30 (P=0.024), the KT-1000 134 N r=−0.25 (P=0.06) and the KT-1000 manual maximum r=−0.37 (P=0.004). A negative correlation between awake QPS acceleration and the static tests was also shown for the non-involved knee. Patients with a high-grade Lachman’s test in the EUA had significantly greater QPS acceleration (P=0.0002) and QPS translation (P<0.001) compared with patients with a low-grade. The corresponding analysis for the anterior drawer showed a significantly greater QPS translation in the high-grade group (P=0.01), while no differences were found in the QPS acceleration.

    Conclusion Static anteroposterior and dynamic knee laxities, as presented by QPS, are poorly correlated in the ACL-deficient knee and should therefore be considered as separate entities of the knee examination. These findings strengthen the implementation of non-invasive technology for quantification of the pivot shift when establishing treatment algorithms for ACL reconstruction.

    Level of evidence Level III, prospective cohort.

    • knee
    • ACL/PCL
    • ligament
    • repair/reconstruction
    • arthroscopy

    Statistics from Altmetric.com

    What are the new findings

    • Static anteroposterior knee laxity tests are poorly correlated to quantitative pivot shift in the anterior cruciate ligament (ACL)-deficient knee.

    • Patients with greater anteroposterior knee laxity could falsely present a lower tibial acceleration during the pivot shift, probably due to guarding. Therefore, examination under anaesthesia is important to accurately assess knee joint status.

    • Static and dynamic knee joint laxity should be considered as separate entities of knee kinematics, and need to be assessed separately when treatment algorithms for ACL reconstruction are established.

    Introduction

    Assessment of knee joint laxity is a one of the key points in the management and diagnosis of anterior cruciate ligament (ACL) injury. Static knee laxity tests consider a single degree of freedom of motion and the Lachman test, which determines the anterior-posterior (A-P) laxity, is considered the most sensitive test for detection of an ACL injury.1 The most common devices for quantification of A-P laxity are the KT-1000 and the rolimeter.2–4 The unidirectional force applied during a static laxity test manoeuvre is, however, subphysiological and is a simplistic evaluation of laxity. This may explain the poor correlation between static laxity and functional status.5–7

    On the other hand, the pivot shift test is a dynamic laxity test that simulates a more physiological multiaxial loading for assessment of combined rotatory and translational knee laxity. Therefore, the test is referred to as the most specific test for the detection and quantification of ACL insufficiency.1 By applying valgus stress and an internal rotation torque on the tibia while the knee is moved from extended towards flexed position, the sudden reduction of the anteriorly subluxated tibia at about 30 degrees of flexion can be felt as an episode of ‘give way’ for the patient, or a ‘clunk’ in the examiner’s hands.8 Although the test is the best indicator of a patient’s subjective instability,6 the validity of the test has been limited by a wide variability of execution techniques,9–11 subjective grading12 and great intravariability and intervariability.13 14 To overcome this, a standardised pivot shift test has been introduced10 and technology that allows non-invasive quantitative measurement of the kinematics during the test has been developed.15–17

    It is of importance to accurately assess preoperative knee joint laxity in order to individualise the surgical approach for a successful restoration of stability. Lately, emphasis has been given to the evaluation of rotatory knee joint laxity, and it has been proposed that a lateral extra-articular tenodesis procedure might be necessary in order to restore functional stability in patients with severe rotatory laxity.18 19 Which preoperative factors to be incorporated in the algorithm of surgical approach are yet to be established. Because of this, it is of importance to understand how static and dynamic tests of knee laxity correlate preoperatively. The link between non-invasive methods for quantitative pivot shift and static A-P knee laxity tests is unknown. Therefore, the purpose of this study was to determine the preoperative relationship between static and dynamic laxity in a multicentre setting. Because of the different properties of the forces applied to static and dynamic knee laxity tests, it was hypothesised that static and dynamic knee joint laxity would be poorly correlated in ACL-deficient knees preoperatively.

    Methods

    This study was based on a multicentre prospective cohort study, The Prospective International Validation of Outcome Technology (PIVOT) trial, and involved four clinical centres, which all applied the same study protocol. Patients were enrolled between December 2012 and February 2015 if fulfilling the inclusion criteria, being 14–50 years of age, having injury to at least one bundle of the ACL and undergoing hamstring autograft ACL reconstruction within 1 year from injury. Patients who fulfilled any of the following criteria were excluded: had prior ligament surgery of the involved knee, cartilage lesions of grade 3 or 4, inflammatory arthritis, concomitant posterior cruciate ligament or collateral ligament grade III injury, any condition hindering the ability to walk or participate in level I and II activities, or had current or previous injury or surgery to the contralateral knee. Additionally, for the specific analysis performed in the present study, only patients with complete data on all preoperative laxity test aimed for the investigation were included. Thus, only patients with complete preoperative data on quantitative pivot shift (QPS), rolimeter, KT-1000 laxity measurements, Lachman’s test and anterior drawer test were included. The PIVOT trial was approved across all centres by the institutional review boards. Information about the study was given to all participating patients before written consent was collected.

    Static laxity examination

    The preoperative static knee ligament testing was performed in office, with the patient in awake state, as well as immediately prior to surgery start, when the examination under anaesthesia (EUA) was performed. The assessment procedure was standardised across all involved centres, and both manual and instrumented measurements were included. First, the Lachman test was performed in 25° of knee flexion and graded relative to the contralateral knee according to the International Knee Documentation Committee (IKDC) guidelines.12 Subsequently, the KT-1000 knee arthrometer was used to assess anterior translation with 134 Newton (N) force at 25° of knee flexion. This was followed by applying a manual maximum force, which was enabled by manually pushing the tibia forward as far as possible until three consistent measurements of A-P translation had been observed on the KT-1000 display. Finally, the rolimeter was used to quantify anterior translation of the tibia during the Lachman test in 25° of knee flexion. All knee examinations were first performed on the non-involved leg, followed by the involved knee. The examination procedures were identically performed in the EUA setting with the exception of not including the KT-1000 examinations due to practical reasons in the operating room.

    Quantitative pivot shift examination

    The preoperative examination of the pivot shift was performed in the same settings as the static laxity examination. To standardise the examination, all four centres were provided with video and written instructions of how to perform the test.10 20 The QPS was assessed by inertial sensor and image analysis, enabling quantification of the acceleration and the lateral translation of the tibia, respectively. The inertial sensor system (KiRa; Orthokey, USA) records the tibial acceleration during the pivot shift.21 22 The sensor, which includes a triaxial accelerometer and gyroscope, is positioned over Gerdy’s tubercle and held in place by a hypoallergenic elastic strap. The validity of this device has been described previously.23 The lateral tibial translation was calculated via image analysis system.24 Three markers were placed on the skin of the lateral aspect of the knee, each at the following specific landmarks: the lateral femoral epicondyle, Gerdy’s tubercle and the fibular head. The position of the markers is tracked by using commercial tablet (iPad; Apple, Cupertino, California, USA) software for video recording of the pivot shift test. A software application on the tablets automatically calculates the tibial translation, which is also displayed as a graph.25 Previous research has concluded that the method is valid for evaluation of the pivot shift.16 The intraobserver and interobserver intraclass correlation coefficient has previously been reported as >0.90 for the image analysis system and, for the inertial sensor system, the corresponding number is 0.90, which indicates excellent repeatability.25–27

    Statistical analysis

    The SAS statistical analysis system (SAS, V.9.4; SAS Institute, Cary, North Carolina, USA) was used for all calculations. Descriptive statistics were presented as mean values ± standard deviation (SD) for continuous variables and counts and percentages for categorical variables. The results of the QPS examinations were analysed as continuous variables for each leg separately (the involved and the non-involved knee) and Spearman’s rho was used to assess the correlation with continuous variables of static laxity test in terms of KT-1000 (134 N), KT-1000 with manual maximum force and the rolimeter. The analyses between each QPS examination and static test were performed for both the examination in the awake state and EUA. The results of the Lachman test and the anterior drawer test were dichotomised into low-grade (IKDC grades A and B) and high-grade (IKDC grades C and D). Differences between the low-grade and high-grade groups in terms of QPS were assessed by applying a two sample t-test with unequal variance where the QPS was analysed as a continuous variable. All significance tests were two-sided. Significance was set at P <0.05.

    Results

    A total of 107 patients were enrolled in this multicentre trial. Of these, 58 patients had available data on all knee laxity tests of interest in the present study and were included. The cohort consisted of 41.4% women and the average age was 27.1±9.8 years. A majority of the patients (67.2%) had a concomitant meniscal lesion. The demographic characteristics of the study group are shown in table 1.

    Table 1

    Demographic variables of the study group (n=58)

    Static A-P laxity and QPS in terms of tibial acceleration was negatively correlated in the ACL-deficient knee when examined in the awake state, meaning that a higher absolute value of tibial acceleration corresponded to a lower absolute value of static A-P translation and vice versa. The mean QPS acceleration in awake state was 3.71±1.66 m/s2. The correlations with the static tests were the rolimeter r=−0.30 (P=0.024), the KT-1000 at 134 N test force r=−0.25 (P=0.06) and KT-1000 manual maximum r=−0.37 (P=0.004) (table 2). In the EUA setting, however, absolute values of static A-P laxity, as measured with the rolimeter, and QPS acceleration was not significantly correlated in any direction (r=0.049, P=0.71) (table 3). The similar correlation analysis for QPS in terms of lateral tibial translation and static A-P laxity showed no significant correlation between any static test and QPS translation in neither the awake nor the EUA setting (tables 2 and 3).

    Table 2

    Correlation between the awake assessment of static laxity and quantitative pivot shift for the involved and the non-involved knee, respectively

    Table 3

    Correlation between the examination under anaesthesia assessment of static laxity and quantitative pivot shift for the involved and the non-involved knee, respectively

    Additionally, the same analysis protocol was applied for the non-involved knee. Likewise, a significant correlation in negative direction was found between static A-P laxity and QPS acceleration when examined in awake state. The mean QPS acceleration in the awake state was 2.88±1.11 m/s2 and correlated as follows to the A-P laxity tests: the rolimeter r=−0.31 (P=0.04), the KT-1000 at 134 N test force r=−0.27 (P=0.038) and KT-1000 manual maximum r=−0.37 (P=0.0038) (table 2). No significant correlation was found between QPS acceleration and the rolimeter in EUA (r=−0.24, P=0.069) (table 3). The corresponding analysis for the correlation between QPS lateral tibial translation and static A-P laxity showed that these were not significantly correlated in neither the awake state nor in the EUA. (tables 2 and 3).

    Thirty-nine patients were evaluated as high-grade while 19 patients as low-grade in the Lachman assessment in the awake setting. The groups did not differ significantly in terms of neither the QPS acceleration (P=0.78) nor the QPS lateral tibial translation (P=0.096). During the EUA, the number of patients categorised as high-grade Lachman’s test was 45 patients and both the QPS acceleration and the QPS lateral tibial translation differed significantly between the groups. The mean QPS acceleration in the low-grade group was 3.15±1.07 m/s2 compared with 6.14±4.66 m/s2 in the high-grade group (P=0.0002). The corresponding values for the QPS tibial translation were 1.53±0.62 mm compared with 3.75±2.44 mm (P<0.001) (table 4).

    Table 4

    Comparison of dynamic and static measures of knee laxity

    The awake assessment of the anterior drawer categorised 34 patients as low-grade and 24 patients as high-grade anterior drawer. The groups differed significantly in terms of QPS acceleration since the mean QPS acceleration for the low-grade group was 3.11±1.65 m/s2 compared with 4.56±1.29 m/s2 in the high-grade group (P=0.0007), and the QPS tibial translation for the low-grade group was 1.86±1.33 mm compared with 3.07±2.13 mm in the high-grade group (P=0.02). In the EUA, the number of patients with a high-grade anterior drawer increased (n=30). There was no significant difference between patients in the low-grade and high-grade group regarding QPS acceleration (P=0.71). However, the groups differed significantly in terms of QPS tibial translation in the EUA setting (P=0.01), where patients in the high-grade group had a greater QPS tibial translation (table 4).

    Discussion

    The most important finding of this study was that dynamic rotatory knee laxity, assessed by QPS, and static A-P knee laxity were poorly correlated in the ACL-deficient knee. In fact, the tibial acceleration during the pivot shift was negatively correlated to static A-P laxity during the awake assessment, although the correlation was not strong. This indicates that a large A-P translation may not necessarily lead to a pronounced rotatory laxity and suggests that static A-P and dynamic laxities should be considered as separate entities for evaluation of knee kinematics in the ACL-deficient knee. Assessment of static A-P knee laxity is valuable in the clinical setting for diagnosis of an ACL rupture; however, considering its poor correlation to rotatory laxity, and the fact that the latter stronger correlates to functional instability,28 29 the degree of rotatory laxity should primarily be incorporated into treatment algorithms of ACL reconstruction. Another important finding was that the EUA influenced the analyses considerably, indicating that laxity testing under anaesthesia is valuable for a correct assessment of preoperative knee status.

    This is the first study investigating how the recent non-invasive technology for QPS correlates with static A-P knee laxity in the ACL-deficient and the healthy knee. The correlation analysis between the lateral tibial translation during pivot shift and static anterior translation in absolute values of each leg was non-significant throughout all applied tests and settings. However, a weak, yet significant, negative correlation was found between the tibial acceleration during pivot shift and the KT-1000 as well as the rolimeter for both the involved and the uninvolved leg when examined in the awake state. Nevertheless, this correlation was non-significant when examining the patients under anaesthesia. It is well established that patients’ guarding may aggravate the execution of the pivot shift in the clinical setting.30 It is possible that patients with more severe A-P laxity feel especially displeased during the pivot shift examination and that the guarding is more pronounced in these patients. This could explain the negative correlation found between the tibial acceleration and A-P translation in this study, particularly since the significant correlation extinguished in the EUA setting, where voluntary muscle defence is constrained. Previous studies have stressed the importance of EUA to improve the clinical reliability and evaluation of the pivot shift test, and the results of this study is in harmony with those previous results.31 32

    In addition to the correlation analysis performed for absolute values of static and dynamic laxity of each leg, this study was complemented with a correlation analysis of QPS and dichotomised groups of high-grade respectively low-grade Lachman and anterior drawer tests. Interestingly, there was no difference in QPS between patients with a low-grade and high-grade Lachman when the tests were performed in the awake setting. However, the number of patients with a high-grade Lachman’s test increased in the EUA and a significant difference was found for both tibial acceleration and translation of the pivot shift between the groups. This finding further underlines the value of EUA and suggests that, although the absolute values of static and rotatory laxity (of each leg separately) did not correlate in the present study, a discrepancy of QPS likely exists between patients in the lower and upper span of static laxity when this is evaluated according to the IKDC as the side-to-side difference.

    An ACL injury appears to be in the nearest obligate for a pivot shift to occur.13 33 34 There are, however, several complex interactions involved in the phenomenon of the pivot shift such as concomitant injuries, bony morphology and general ligament laxity.19 30 35–37 When the ACL is injured, it is likely that these factors gain a stronger influence on the knee kinematics. Since static tests are unable to detect restraints acting at different levels during range of motion and under different stress conditions to the joint, the contribution of the aforementioned factors on laxity are at risk of being indiscernible when static tests are performed. Although it has been shown that the acceleration tends to vary more among examiners than the quantification of the lateral tibial translation during the pivot shift,20 it is not yet fully understood how the non-invasively quantified acceleration and tibial translation of the pivot shift correlate with each other. However, an interesting finding in the present study was that the tibial acceleration and translation were differently correlated to the static tests. For example, the tibial acceleration showed a consistent negative correlation with the static laxity tests in the awake setting, while no such correlation was found for the tibial translation in relation to the static tests. This finding encourages deepened investigation in the matter in order to understand the characteristics and the differences of the two entities of the pivot shift. However, it is known that the tibial acceleration is more strongly correlated to the clinical grading of the pivot shift than the lateral translation,38 39 and it has also been shown that the magnitude of the translation is of lesser importance in distinguishing grades of the pivot shift compared with the acceleration and velocity.40 Thus, it is possible that the acceleration and the velocity of the reduction phenomenon should be emphasised over translation when rotatory knee laxity is evaluated.

    The main limitation to the present study was that the same examiner performed both the static and the dynamic tests and was not blinded to the results of each test. The assessment was, however, performed according to a standardised protocol,8 10 and all the tests that were subjectively graded by the surgeon were performed prior to assessing the QPS, which should minimise the influence of QPS on the examiner’s subjective evaluation. Moreover, although each test was performed according to a standardised technique, the execution of the tests would most likely vary to some extent among the examiners. The choice of excluding all patients who did not have complete data on all the investigated variables was made in order to analyse the exact same population for all the correlation analyses, thus, the variability of injury pattern and severity of concomitant injuries should have influenced all the analyses equally. It should also be noted that none of the analyses were adjusted for the possible influence of concomitant injuries.

    Conclusion

    Static A-P and dynamic knee laxities, as presented by QPS, are poorly correlated in the ACL-deficient knee and should therefore be considered as separate entities of the knee examination. These findings strengthen the implementation of non-invasive technology for quantification of the pivot shift when establishing treatment algorithms for ACL reconstruction.

    References

    View Abstract

    Footnotes

    • Contributors ES and EHS have substantially contributed to the acquisition of data, analysis of data and are responsible for drafting the work and revising it critically for important intellectual content. JM, VM and KS have substantially contributed to the acquisition of data, analysis of data and critically revising the work for important intellectual content. RK, SZ and JK have done large contributions to revise and design the manuscript. All authors have given their final approval of the manuscript to be published. In addition, all authors are in agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

    • Funding The study was funded through a grant from the International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine (ISAKOS) and the Orthopaedic Research and Education Foundation (OREF) (research grant no. 708661).

    • Competing interests RK reports personal fees from Medacta International, personal fees from Arthrex Inc., personal fees from B. Braun Aesculap Japan, personal fees from Zimmer Biomet, personal fees from Arthrex Japan GK, personal fees from Smith & Nephew KK, personal fees from Johnson & Johnson KK, grants from Stryker Japan KK, grants from Zimmer Biomet, grants from Smith & Nephew Orthopedics KK, grants from Astellas Pharm Inc., grants from Chugai Pharmaceutical Co., Ltd, grants from Taisyo Toyama Pharmaceutical Co., Ltd, grants from Pfizer Japan Inc., outside the submitted work. VM reports other (consulting) from Smith & Nephew, outside the submitted work.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Collaborators The Pivot Study Group: James J Irrgang, Freddie H Fu, Adam Popchak, Paulo Araujo, Darren De SA, Neel Patel, Jayson Lian (Department of Orthopaedic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA); Yuichi Hoshino, Masahiro Kurosaka, Kouki Nagamune (Kobe University, Kobe, Japan); Giulio Maria Marcheggiani Muccioli, Cecilia Signorelli, Nicola Lopomo, Alberto Grassi, Federico Raggi (Istituto Ortopedico Rizzoli, Bologna, Italy); David Sundemo (Sahlgrenska University Hospital, Gothenburg, Sweden).

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.