Knee Surg Sports Traumatol Arthrosc DOI 10.1007/s00167-015-3563-2
KNEE
Diagnostic accuracy of physical examination for anterior knee instability: a systematic review Marie‑Claude Leblanc1 · Marcin Kowalczuk1 · Nicole Andruszkiewicz2 · Nicole Simunovic3 · Forough Farrokhyar3,4 · Travis Lee Turnbull5 · Richard E. Debski6 · Olufemi R. Ayeni1,3
Received: 20 January 2015 / Accepted: 2 March 2015 © European Society of Sports Traumatology, Knee Surgery, Arthroscopy (ESSKA) 2015
Abstract Purpose Determining diagnostic accuracy of Lachman, pivot shift and anterior drawer tests versus gold standard diagnosis (magnetic resonance imaging or arthroscopy) for anterior cruciate ligament (ACL) insufficiency cases. Secondarily, evaluating effects of: chronicity, partial rupture, awake versus anaesthetized evaluation. Methods Searching MEDLINE, EMBASE and PubMed identified studies on diagnostic accuracy for ACL insufficiency. Studies identification and data extraction were performed in duplicate. Quality assessment used QUADAS tool, and statistical analyses were completed for pooled sensitivity and specificity. Results Eight studies were included. Given insufficient data, pooled analysis was only possible for sensitivity on
Lachman and pivot shift test. During awake evaluation, sensitivity for the Lachman test was 89 % (95 % CI 0.76, 0.98) for all rupture types, 96 % (95 % CI 0.90, 1.00) for complete ruptures and 68 % (95 % CI 0.25, 0.98) for partial ruptures. For pivot shift in awake evaluation, results were 79 % (95 % CI 0.63, 0.91) for all rupture types, 86 % (95 % CI 0.68, 0.99) for complete ruptures and 67 % (95 % CI 0.47, 0.83) for partial ruptures. Conclusion Decreased sensitivity of Lachman and pivot shift tests for partial rupture cases and for awake patients raised suspicions regarding the accuracy of these tests for diagnosis of ACL insufficiency. This may lead to further research aiming to improve the understanding of the true accuracy of these physical diagnostic tests and increase the reliability of clinical investigation for this pathology. Level of evidence IV.
Electronic supplementary material The online version of this article (doi:10.1007/s00167-015-3563-2) contains supplementary material, which is available to authorized users.
Keywords Anterior cruciate ligament · Rupture · Diagnostic accuracy · Physical examination · Knee injuries
* Olufemi R. Ayeni [email protected]
Introduction
1
Division of Orthopedic Surgery, Department of Surgery, McMaster University, 4E15‑ 1200 Main St W, Hamilton, ON L8N 3Z5, Canada
2
Department of Public Health Sciences, Queen’s University, Kingston, ON, Canada
3
Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada
4
Department of Surgery, McMaster University, Hamilton, ON, Canada
5
Department of BioMedical Engineering, Steadman Philippon Research Institute, Vail, CO, USA
6
Orthopaedic Robotics Laboratory, University of Pittsburgh, Pittsburgh, PA, USA
Rupture of the anterior cruciate ligament (ACL) is one of the most frequent knee injuries [21]. The highest incidence has been reported among young and active individuals aged 15–25 years, making early diagnosis essential [14, 35]. The natural history of ACL-deficient knees with recurrent instability includes acceleration of chondral wear and increased incidence of meniscal tears [19, 26]. Despite the presence of an increased incidence of osteoarthritis in ACL stabilized knees, it is clear from the literature that the incidence of meniscal tears and chondral injuries can be reduced with ACL reconstruction [19, 25]. Due to associated recurrent instability, conservative treatment of ACL ruptures can
13
produce unsatisfactory results leading to losses of productivity, competition and scholarship opportunities [19, 26, 35]. Diagnosis is usually based on a combination of history, physical examination and diagnostic imaging. Three of the most common physical tests used for diagnosis are the Lachman, the pivot shift and the anterior drawer test [29]. The sensitivity and specificity of these tests has been largely reported in the literature, with great variability. Reported sensitivity and specificity values for the Lachman test ranged from 48 to 100 and 42 to 100 %, respectively. The pivot shift test has a reported sensitivity ranging from 0 to 98.4 % and a specificity ranging from 82 to 100 % [6, 17, 24, 32, 34]. For the anterior drawer test, the reported sensitivity ranges from 9 to 95.2 % and the specificity from 23 to 100 %. The best moment to obtain an unbiased examination is immediately after injury, when there is no significant hemarthrosis, painful spasm or muscle guarding [26]. Unfortunately, this scenario is more the exception than the rule in the recreational athlete and the physical examination can often be equivocal. In an effort to summarize the value of physical examination, seven systematic reviews have been produced over the last 15 years [6, 17, 22, 24, 32, 34, 36]. Unfortunately, most of these systematic reviews did not include studies published since the year 2000. Moreover, the quality of the studies published prior to the year 2000 was less controlled and more heterogeneous, making pooled analysis difficult. Of the previously conducted systematic reviews, the effect of anaesthesia and lesion chronicity on physical exam accuracy was reported by only a single study and did not distinguish between partial and complete ruptures [6]. The current systematic review aimed to determine the diagnostic accuracy of physical examination techniques (Lachman, pivot shift or anterior drawer tests) versus a gold standard diagnosis [magnetic resonance imaging (MRI) or arthroscopy] in cases of anterior knee instability secondary to ACL insufficiency. Furthermore, the impact of various clinical factors on physical examination accuracy was assessed. This included: acute versus chronic injury, complete versus partial ACL rupture and examination under anaesthesia (EUA) versus awake clinical setting. The hypothesis was that advanced MRI and arthroscopic techniques combined with improved study design in articles presented since the year 2000 would yield quality data to support evaluation of the diagnostic accuracy of physical examinations. In addition, the hypothesis was made that the diagnostic accuracy would be lower for acute injuries, partial ruptures and examinations conducted in the awake clinical setting. This is the first systematic review on the subject focusing on contemporary literature presented since the year 2000.
13
Knee Surg Sports Traumatol Arthrosc
Materials and methods This systematic review was performed according to PRISMA guidelines [23]. Study eligibility Studies were eligible for inclusion if they primarily assessed the diagnostic accuracy of physical examination (Lachman, pivot shift or anterior drawer tests) relative to MRI or arthroscopy as a gold standard for diagnosis. The study population included all patients with anterior knee instability secondary to ACL insufficiency. The inclusion criteria were as follows: (1) patients with a knee injury, (2) physical diagnosis with at least one physical test (clinic or EUA), (3) correlation with a gold standard (MRI, arthroscopy, arthrotomy), (4) in vivo human studies, (5) adults and (6) studies published in English or French. Exclusion criteria included: (1) review articles, (2) knee dislocation with multiple ligamentous injuries, (3) studies on specific injuries other than primary ACL, (4) no specification of the physical diagnostic test used, (5) studies which only discussed surgical techniques and (6) publications published prior to the year 2000. Systematic reviews and biomechanical (non-human) studies were excluded. Information sources and research strategy Electronic databases [MEDLINE (1946–21 July 2014), EMBASE (1980–22 July 2014) and PubMed (1950s–21 July 2014)] were searched for knee ligamentous injury studies reporting data on the diagnostic accuracy comparison between physical examination and MRI, arthroscopy or arthrotomy. The complete search strategy is presented in Appendix 1 of Supplementary Material. Articles from the different databases were compiled and duplicates removed. The results were uploaded to a bibliographic management database (RefWorks, version 2.0; Bethesda, MD). Study selection Two reviewers (MC.L., M.K.) initially completed a title and abstract review screening for eligible studies. Following this, a full-text review was conducted and the references for each article were also hand-searched for other eligible studies. Both reviewers screened all studies independently. Disagreements were resolved by a consensus discussion involving the senior author. Exclusion of articles published before the year 2000 was completed after reviewing all full texts. Those publications (pre year 2000) were excluded due to lack of consistent methodology, limited imaging options compared to contemporary practice and vague
Knee Surg Sports Traumatol Arthrosc
definitions of ACL injuries. Any article not available online was obtained in print through our institutional library. Data collection Data were collected independently and in duplicate from the included articles by two reviewers (MC.L., and M.K. or LP.B. or N.F.) using a standardized data collection form. Extracted data included: title, author, year of publication, location of study, study design, level of evidence, number of patients, mean age, male to female ratio, number of knees examined, number of ACL ruptures, number of normal ACLs, type of injury (acute, chronic, complete, or partial), physical test, setting of test (EUA or clinic), gold standard test, description of physical test, physical exam result and gold standard exam result. Sensitivity and specificity data were collected as counts whenever reported. Any discrepancies regarding collected data were resolved through discussion and consensus between reviewers (MC.L., M.K.) and the senior author. Quality assessment The methodological quality of each study was assessed independently by two reviewers (MC.L. and M.K.) using the QUADAS tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews [40]. QUADAS is a validated clinometric tool used to assess the overall quality of diagnostic accuracy studies through individual quality component questions. Statistical analysis and data measures Sensitivity and specificity data were extracted as counts from each of the studies where applicable. The figures were changed into counts if reported in percentages. Sensitivity was defined as the percentage of individuals who were correctly identified as having a ruptured ACL and specificity was defined as the percentage of individuals who were correctly identified as not having a ruptured ACL. In order to maintain consistency, studies that reported grades for each of the three tests were evaluated using a defined method. With regard to the Lachman and anterior drawer tests, a grade 1 or higher was consistent with an ACL rupture, while grade 0 examinations were deemed normal. For the pivot shift test, a glide, clunk, gross or severe grade was considered to be a ruptured ACL. This grading classification assisted with calculating true positives and false negatives for each of the tests in each article. This means according to the aforementioned classification that patients with a ruptured ACL and a grade 1 or higher were defined as true positives, while patients with a ruptured ACL and a grade 0 were defined as false negatives. Only a single
study included healthy participants without an ACL rupture, which allowed for the calculation of true negatives and false positives to determine the specificity of each applicable test within the study [5]. Therefore, without both sensitivity and specificity data for more than one study, a full meta-analysis of diagnostic accuracy and calculation of likelihood ratios of positive and negative tests could not be performed. A random effects model (DerSimonian–Laird) was used to calculate the pooled weighted proportions of sensitivity due to the inherent heterogeneity applied to the observational studies upon collecting the data for true positives and false negatives from the studies. Pooled sensitivities with the corresponding 95 % confidence intervals (CI) were reported, and forest plots of the pooled proportions were presented. StatsDirect 2.7 (StatsDirect Ltd, UK) was used for data analysis.
Results Study identification, descriptions and quality assessment The literature search initially yielded 2529 studies of which 103 underwent full-text review. Ultimately, eight studies were included for final analysis (Fig. 1) [5, 9–11, 28, 29, 38, 39]. The majority of the studies were level IV evidence. The total pooled number of ACL-deficient knees was 1196. Two studies, Beldame et al. [5] and Wagemakers et al. [39], reported on both ACL-deficient and non-ACL-deficient knees and six used only an ACL-deficient population. As for the gold standard correlation, Wagemakers et al. [39] used MRI, while the others studies used arthroscopy. Two studies, Dhillon et al. [10] and Espregueira-Mendes et al. [11], completed their physical examination under anaesthesia, and six studies evaluated the patient in an awake clinical setting (Table 1). The result for initial inter-reviewer agreement was 0.980 (95 % CI 0.978–0.982) for title screen, 0.967 (95 % CI 0.959–0.974) for abstract screen and 0.796 (95 % CI 0.699–0.862) for full text screen. The quality assessment of the included studies is reported in Table 2. Pooled results A pooled sensitivity analysis was only possible for Lachman and pivot shift tests. Additionally, the effect of the type of rupture (partial vs. complete) and the result in an awake clinical setting were reported. The detailed results are described in Figs. 2 and 3, in Table 3 as well as in Appendix 2 of Supplementary Material through 11. For combined (partial and complete) ruptures, the pooled sensitivity was 89 % (95 % CI 76–98 %) for the Lachman test
13
Knee Surg Sports Traumatol Arthrosc
2529 Studies Identified MEDLINE: 1088 Studies PUBMED: 805 Studies EMBASE: 636 Studies
825 duplicates
1704 studies for title review
Excluded studies: 1417
287 studies for abstract review
Excluded studies: 209
103 studies for full text review
Excluded studies: 95
Handpick studies: 25
8 studies included in the systematic review
-
Before year 2000: 61 No specification on physical test: 6 Quantifying physical test: 5 Cadaver: 1 Systematic review: 6 No physical test results: 12 Editorial: 3 Prone Lachman: 1
Fig. 1 Flow diagram of literature search process
and 79 % (95 % CI 63–91 %) for the pivot shift test. For complete ruptures, the pooled sensitivity was 96 % (95 % CI 90–100 %) for the Lachman test and 86 % (95 % CI 68–99 %) for the pivot shift test. For partial ruptures, the pooled sensitivity was lower and more variable: 68 % (95 % CI 25–98 %) for the Lachman test and 67 % (95 % CI 47–83 %) for the pivot shift test. Most of the included studies did not provide data for calculation of specificity. This is because patients with positive MRI or arthroscopy findings underwent physical examinations, and as a result, data on the true negatives and false positives were not available. Only two studies provided complete data on both ACL-deficient knees and non-ACL-deficient knees [5, 39]. They reported specificity for the anterior drawer test of 83.9 % (95 % CI not available) and 57.0 % (95 % CI 0.48, 0.67). Only Beldame et al. [5] reported on specificity for the Lachman and pivot shift tests, at 78.1 % (95 % CI 0.61, 0.89) and 86.4 % (95 % CI not available), respectively. Due to insufficient data, pooled sensitivity results were not calculated for the anterior drawer test, EUA and chronicity. Only two studies reported sensitivity results for combined or complete ruptures using the anterior drawer test, and only Beldame et al. reported results for partial ruptures for the anterior drawer test [5, 29]. Also, only two studies used EUA [10, 11]. Dhillon et al. [10] reported a
13
sensitivity of 100 % (largest 95 % CI 0.87, 1.00) for Lachman and pivot shift physical examinations in all settings including combined (partial and complete) and individual rupture types (Table 3). Espregueira-Mendes et al. [11] only reported on complete ruptures, with a sensitivity of 100 % (95 % CI 0.88, 1.00) for the Lachman and 82 % (95 % CI 0.63, 0.94) for the pivot shift (Table 3). Data regarding the effect of the chronicity of the lesion were also too scarce to analyse.
Discussion The most important finding of this systematic review was that although both Lachman and pivot shift tests are sensitive in diagnosing ACL ruptures, the clinical setting (awake vs. non-awake) and extent of injury (partial vs. complete rupture) have an impact on diagnostic accuracy. Biomechanical data from Araki et al. [3] support this observation by demonstrating that partial tears in ACL knees had less laxity in all patients and a detectable end point on Lachman examination in only 45 % of the patients. Concerning the effect of anaesthesia, the decrease in sensitivity with an awake patient in a clinical setting reinforces the effect of muscular guarding on resisting tibial anterior translation during Lachman and pivot shift examination.
Ipswich, UK
Porto, Portugual
Dhillon [10]
EspregueiraMendes [11] Panisset al. [28]
Netherlands (five centres)
Wagemakers [39]
48
112
418
28
182
300
112
Knee injuries within 5 wks
ACL deficient knee ACL deficient knee 28
48
112
17
32
N/A
208
28
103
177
70
Total ACL Complete ACL rupture
80 Knee injury with indication for knee arthroscopy ACL deficient 300 knee ACL deficient 182 knee ACL deficient 28 knee ACL deficient 418 knee
Nb patients Context
134 Cross-sectional diagnostic study/ IV
Retrospective care series/IV Retrospective care series/IV
Prospective case serie/IV
Prospective case– control/III Retrospective diagnostic/IV Retrospective/IV
Prospective case– control/III
Design/evidence
11
16
N/A
67
0
27
123
10
Partial ACL rupture
Arthroscopy
Arthroscopy
Arthroscopy
AD
Lachman
MRI
Arthroscopy
AD Lachman PS Arthroscopy
Lachman PS
AD Lachman PS Arthroscopy
Lachman PS
Lachman PS
Awake
Awake
Awake
Awake
EUA
EUA
Awake
Awake
Gold standard Setting
AD Lachman PS Arthroscopy
Clinical test
ACL anterior cruciate ligament, AD anterior drawer, PS pivot shift, MRI magnetic resonance imaging, wks weeks, N/A none applicable, Nb number, EUA examination under anaesthesia
Taipei, Taiwan
Tsai [38]
Peeler [29]
France (Bordeaux, Grenoble, Lyon, Pau, Toulouse) Manitoba, Canada
Rouen Cedex, France/University Lyon, France
Beldame [5]
Dejour [9]
Location/type of hospital
References
Table 1 Characteristics of included studies
Knee Surg Sports Traumatol Arthrosc
13
Knee Surg Sports Traumatol Arthrosc
Table 2 Validity scoring QUADAS References
Items number 1
2
3
4
5
6
7
8
9
10
11
12
13
14
Beldame [5] Dejour [9] Dhillon [10] Espregueira-Mendes [11] Panisset [28] Peeler [29] Tsai [38]
No No No No No No No
Yes Yes Yes Yes Unclear Yes No
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Unclear Yes Yes Yes Yes Unclear Unclear
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Unclear No No No Unclear Unclear No
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Unclear No
Yes Yes Yes Yes Yes Yes Yes
Wagemakers [39]
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
1. Was the spectrum of patients representative of the patients who will receive the test in practice? 2. Were selection criteria clearly described? 3. Is the reference standard likely to correctly classify the target condition? 4. Is the time period between reference standard and index test short enough to be reasonably sure that the target condition did not change between the two tests? 5. Did the whole sample or a random selection of the sample, receive verification using a reference standard? 6. Did patients receive the same reference standard regardless of the index test result? 7. Was the reference standard independent of the index test (i.e. the index test did not form part of the reference standard)? 8. Was the execution of the index test described in sufficient detail to permit replication of the test? 9. Was the execution of the reference standard described in sufficient detail to permit its replication? 10. Were the index test results interpreted without knowledge of the results of the reference standard? 11. Was the reference standard results interpreted without knowledge of the results of the index test? 12. Were the same clinical data available when test results were interpreted as would be available when the test is used in practice? 13. Were uninterpretable/intermediate test results reported? 14. Were withdrawals from the study explained?
Proportion meta-analysis plot [random effects]
Proportion meta-analysis plot [random effects] Beldame 2011
0.81 (0.71, 0.89)
Dejour 2013
1.00 (0.99, 1.00)
Peeler 2010
0.86 (0.77, 0.92)
Tsai 2004
0.81 (0.67, 0.91)
Panisset 2008 combined 0.6
0.7
0.8
0.9
Beldame 2011
0.60 (0.45, 0.73)
Dejour 2013
0.89 (0.85, 0.92)
Peeler 2010
0.63 (0.51, 0.74)
0.89 (0.84, 0.92)
Panisset 2008
0.93 (0.89, 0.96)
0.89 (0.76, 0.98)
combined
0.79 (0.63, 0.91)
1.0
proportion (95% confidence interval)
Fig. 2 Lachman, reported and pooled sensitivity, combined ruptures, awake setting
Clinical diagnosis of an ACL insufficient knee is not an exact science. Many clinicians have attempted to develop an accurate and reproducible physical examination in order to facilitate early diagnosis. The number of different tests described in the literature demonstrates that there is no ‘one size fits all’ physical exam to detect ACL rupture [1, 2, 4, 7,
13
0.4
0.6
0.8
1.0
1.2
proportion (95% confidence interval)
Fig. 3 Pivot shift, reported and pooled sensitivity, combined ruptures, awake setting
9, 11–13, 15, 16, 18–21, 25, 26]. The anterior drawer, Lachman and pivot shift tests have been the major physical tests used for diagnosis of ACL insufficiency. Many factors can affect the reproducibility of these physicals tests [6, 24, 29, 32]. Cooperman et al. [8] compared the clinical accuracy
Knee Surg Sports Traumatol Arthrosc Table 3 Summary of pooled sensitivity (95 % CI) for Lachman and pivot shift Type of rupture
Tests
Combine (awake + EUA)
Awake
EUAa
Combine (complete and partial)
Lachman PS Lachman PS Lachman
92 % (0.81, 0.99) 85 % (0.70, 0.96) 98 % (0.94, 1.00) 89 % (0.77, 0.97) 77 % (0.43, 0.98)
89 % (0.76, 0.98) 79 % (0.63, 0.91) 96 % (0.90, 1.00) 86 % (0.68, 0.99) 68 % (0.25, 0.98)
100 % (0.97, 1.00) [10] 100 % (0.97, 1.00) [10] 100 % (0.96, 1.00) [10] (0.88, 1.00) [11] 100 % (0.96, 1.00) [10] 82 % (0.63, 0.94) [11] 100 % (0.87, 1.00) [10]
PS
77 % (0.54, 0.93)
67 % (0.47, 0.83)
100 % (0.87, 1.00) [10]
Complete Partial
EUA examination under anaesthesia, PS pivot shift, CI confidence interval a
Only two studies reported on EUA. This column is not a pooled analysis, but a report of the data available
of the Lachman test executed by an orthopaedic surgeon to that by a physical therapist. They reported a sensitivity of 77 % and specificity of 50 % for the orthopaedic surgeon. For the physical therapist, these numbers decreased to a sensitivity of 65 % and specificity of 42 %. In addition to the experience level of the examiner, many other factors have been reported to affect the reproducibility of the tests such as knee effusion, muscle spasms, thigh circumference, size of examiner hands, timing of examination, chronicity of lesion, type of rupture, associated injuries and EUA [6, 24, 29, 32]. These multiple factors are likely to contribute to the large variability in reported sensitivity and specificity in the literature. Two systematic reviews have reported pooled analyses. Jackson et al. [17] showed a sensitivity of 48 % for anterior drawer, 87 % for Lachman and 61 % for pivot shift. For the specificity, the results were 87 % for anterior drawer, 93 % for Lachman and 97 % for pivot shift [17]. The current review found a comparable sensitivity for the Lachman examination at 92 % but an improved sensitivity of 85 % for the pivot shift. Benjaminse et al. [6] also conducted a detailed pooled analysis, being the first to divide results according to presence or absence of anaesthesia and acute or chronic rupture. The authors reported a sensitivity for combined ruptures in patient without anaesthesia of 55 % for anterior drawer, 85 % for Lachman and 24 % for pivot shift [6]. For the specificity, the results were 92 % for the anterior drawer, 94 % for the Lachman and 98 % for the pivot shift [6]. In contrast, the current review noted an improved sensitivity for the pivot shift in awake patients at 79 %. Lachman testing in awake patients was comparable at 89 %. Unfortunately, as discussed in the pooled analysis results above, insufficient data were available from the selected articles in this review to allow for calculation of specificity for Lachman and pivot shift test as well as statistical analysis for anterior drawer test. Also, the reviews described above included only articles published prior to 2000, yielding a completely different data set from this review. The first evaluation in an awake clinical setting following a traumatic knee injury requires a highly sensitive
physical examination with good reliability. One systematic review looked at the intra- and inter-rater reliability of the physical examination [22]. Given the modest quality of the studies included, they could only show a moderate intra-rater reliability for the Lachman test. The intra-rater reliability for anterior drawer and pivot shift as well as inter-rater reliability for the three tests was inconclusive. This systematic review demonstrates very large variability in the literature regarding the reported diagnostic accuracy of physical examinations. Furthermore, the type of rupture can greatly affect the sensitivity of both Lachman and pivot shift tests. Additionally, a clear trend implicating decreased sensitivity with physical examination in an awake clinical setting was observed. Given the above challenges, a more reproducible method of diagnosis is commonly preferred. Prior to the availability of high-field strength technology, MRI diagnosis had varying accuracies, with sensitivity ranging from 64 to 94.4 % and a specificity ranging from 94.3 to 100 % [27, 31, 37]. With the advent of 3 T MRI systems, diagnostic accuracy has increased to 97.9–100 % for sensitivity and 98.6–100 % for specificity [30, 33]. As such, MRI is likely to be used ever increasingly for diagnosis of ACL insufficiency for the foreseeable future. There were multiple strengths to this systematic review. A comprehensive search strategy of multiple databases using various keywords and subheading was performed to reduce publication bias. Study selection and data extraction were conducted in duplicate to ensure accuracy and further minimize bias. Included articles were limited to studies published after the year 2000 in an attempt to evaluate only high-quality studies relevant to current practice. Despite the methodological strengths of this study, there are certain limitations due to the inherent biases of the included studies, and thus, findings should be interpreted with caution. The included studies were of different observational designs with diverse patient populations, resulting in a large inter-study heterogeneity. Most studies did not report sufficient data to allow a complete diagnostic accuracy analysis. The majority of studies only reported physical examination findings on patients with known ACL
13
injury based on MRI or arthroscopic examination. This allowed sensitivity analysis but precluded calculation of exam specificity in this review. Also, the lack of ACL-deficient patients for comparison in six of the studies limited subsequent analysis. It is therefore difficult to state a definitive recommendation regarding the diagnostic accuracy of the Lachman test, pivot shift test or anterior drawer test. Furthermore, a description of the exact physical exam procedure was not always available, and often only the name of the physical test was stated. This review clearly demonstrates insufficient and often incomplete documentation of the diagnostic accuracy of the most commonly used physical tests to diagnose ACL rupture. From a clinical perspective, this study broadens our understanding of anterior knee instability clinical assessment and the effect of various factors affecting their results, such as awake patients and partial tears. The apparent lack of diagnostic accuracy offered by existing common physical tests may lead to further research aiming to improve the understanding of the true accuracy of these physical diagnostic tests and increase the reliability of clinical investigation for this pathology.
Conclusion The key finding of this systematic review was that although both Lachman and pivot shift tests are sensitive in diagnosing ACL ruptures, the clinical setting (awake vs. nonawake) and extent of injury (partial vs. complete rupture) have an impact on diagnostic accuracy. The current literature did not contain sufficient data to calculate pooled specificity; therefore, no clear recommendation regarding diagnostic accuracy of the physical examination for ACL insufficient knees could be made. Given the advances in the resolution of MRI and concomitant capability for diagnosing ACL ruptures, the possibility of conducting a diagnostic accuracy study for physical examination of ACL ruptures is now available and could greatly improve the understanding of the true accuracy of these physical diagnostic tests. Acknowledgments A special thanks to four collaborators to this systematic review: Louis-Philippe Baisi, Nicole Friel, Hiba Abdul Mannan and Zakia Islam.
References 1. Anderson AF, Lipscomb AB (1989) Preoperative instrumented testing of anterior and posterior knee laxity. Am J Sports Med 17(3):387–392 2. Araki D, Kuroda R, Kubo S, Nagamune K, Hoshino Y, Nishimoto K, Takayama K, Matsushita T, Tei K, Yamaguchi M, Kurosaka M (2011) The use of an electromagnetic measurement system for
13
Knee Surg Sports Traumatol Arthrosc anterior tibial displacement during the Lachman test. Arthroscopy 27(6):792–802 3. Araki D, Kuroda R, Matsushita T, Matsumoto T, Kubo S, Nagamune K, Kurosaka M (2013) Biomechanical analysis of the knee with partial anterior cruciate ligament disruption: quantitative evaluation using an electromagnetic measurement system. Arthroscopy 29(6):1053–1062 4. Araujo PH, Ahlden M, Hoshino Y, Muller B, Moloney G, Fu FH, Musahl V (2012) Comparison of three non-invasive quantitative measurement systems for the pivot shift test. Knee Surg Sports Traumatol Arthrosc 20(4):692–697 5. Beldame J, Bertiaux S, Roussignol X, Lefebvre B, Adam J-M, Mouilhade F, Dujardin F (2011) Laxity measurements using stress radiography to assess anterior cruciate ligament tears. Orthop Traumatol Surg Res 97(1):34–43 6. Benjaminse A, Gokeler A, van der Schans CP (2006) Clinical diagnosis of an anterior cruciate ligament rupture: a meta-analysis. J Orthop Sports Phys Ther 36(5):267–288 7. Braunstein EM (1982) Anterior cruciate ligament injuries: a comparison of arthrographic and physical diagnosis. Am J Roentgenol 138(3):423–425 8. Cooperman JM, Riddle DL, Rothstein JM (1990) Reliability and validity of judgments of the integrity of the anterior cruciate ligament of the knee using the Lachman’s test. Phys Ther 70(4):225–233 9. Dejour D, Ntagiopoulos PG, Saggin PR, Panisset J-C (2013) The diagnostic value of clinical tests, magnetic resonance imaging, and instrumented laxity in the differentiation of complete versus partial anterior cruciate ligament tears. Arthroscopy 29(3):491–499 10. Dhillon AK, Oday A-D, Servant CT (2013) Diagnostic accuracy of ACL tears according to tear morphology. Acta Orthop Belg 79(1):76–82 11. Espregueira-Mendes J, Pereira H, Sevivas N, Passos C, Vasconcelos JC, Monteiro A, Oliveira JM, Reis RL (2012) Assessment of rotatory laxity in anterior cruciate ligament-deficient knees using magnetic resonance imaging with Porto-knee testing device. Knee Surg Sports Traumatol Arthrosc 20(4):671–678 12. Galway HR, MacIntosh DL (1980) The lateral pivot shift: a symptom and sign of anterior cruciate ligament insufficiency. Clin Orthop Relat Res 147:45–50 13. Graham GP, Johnson S, Dent CM, Fairclough JA (1991) Comparison of clinical tests and the KT1000 in the diagnosis of anterior cruciate ligament rupture. Br J Sports Med 25(2):96–97 14. Griffin LY, Agel J, Albohm MJ, Arendt EA, Dick RW, Garrett WE, Garrick JG, Hewett TE, Huston L, Ireland ML et al (2000) Noncontact anterior cruciate ligament injuries: risk factors and prevention strategies. J Am Acad Orthop Surg 8(3):141–150 15. Hardaker WT, Garret WE Jr, Bassett FH (1990) Evaluation of acute traumatic hemarthrosis of the knee joint. South Med J 83(6):640–644 16. Hughston JC, Andrews JR, Cross MJ, Moschi A (1976) Classification of knee ligament instabilities Part 1: the medial compartment and cruciate ligaments. J Bone Joint Surg Am 58(A2):159–172 17. Jackson JL, O’Malley PG, Kroenke K (2003) Evaluation of acute knee pain in primary care. Ann Intern Med 139(7):575–588 18. Jonsson T, Althoff BO, Peterson L, Renström P (1982) Clinical diagnosis of ruptures of the anterior cruciate ligament: a comparative study of the Lachman test and the anterior drawer sign. Am J Sports Med 10(2):100–102 19. Junkin DM Jr, Johnson DL, Fu FH, Miller MD, Willenborg M, Fanelli GC, Wascher DC (2009) Knee ligaments injuries. In: Kibler WB (ed) Orthopaedic knowledge update: sports medicine 4, 4th edn. Rosemont, Illinois, pp 135–154
Knee Surg Sports Traumatol Arthrosc 20. Katz JW, Fingeroth RJ (1986) The diagnostic accuracy of ruptures of the anterior cruciate ligament comparing the Lachman test, the anterior drawer sign, and the pivot shift test in acute and chronic knee injuries. Am J Sports Med 14(1):88–91 21. Kirchhoff C, Brucker PU, Imhoff AB (2013) Evolving concepts in tunnel placement for ACL reconstruction. In: Johnson DH (ed) Operative arthroscopy, 4th edn. Pennsylvania, Philadelphia, pp 772–777 22. Lange T, Freiberg A, Dröge P, Lützner J, Schmitt J, Kopkow C (2014) The reliability of physical examination tests for the diagnosis of anterior cruciate ligament rupture: a systematic review. Man Ther. doi:10.1016/j.math.2014.11.003 23. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, Clarke M, Devereaux PJ, Kleijnen J, Moher D (2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 62(10):e1–e34. doi:10.1371/journal.pmed.1000100 24. Malanga GA, Andrus S, Nadler SF, McLean J (2003) Physical examination of the knee: a review of the original test description and scientific validity of common orthopedic tests. Arch Phys Med Rehabil 84(4):592–603 25. Miller MD (2011) Single-bundle anterior cruciate ligament repair. In: Wiesel SW (ed) Operative techniques in orthopaedic surgery, 1st edn. Pennsylvania, Philadelphia, pp 342–349 26. Miller RH, Azar FM (2013) Knee injuries. In: Canale ST, Beaty JH (eds) Campbell’s operative orthopaedics, 12th edn. Pennsylvania, Philadelphia, pp 2052–2211 27. Oei EHG, Nikken JJ, Verstijnen ACM, Ginai AZ, Myriam Hunink MG (2003) MR imaging of the menisci and cruciate ligaments: a systematic review. Radiology 226(3):837–848 28. Panisset J-C, Duraffour H, Vasconcelos W, Colombet P, Javois C, Potel J-F, Dejour D (2008) Analyses clinique, radiologique et arthroscopique de la rupture du LCA: étude prospective de 418 cas. Rev Chir Orthop Reparatrice Appar Mot 94(8 supp):362–368 29. Peeler J, Leiter J, MacDonald P (2010) Accuracy and reliability of anterior cruciate ligament clinical examination in a multidisciplinary sports medicine setting. Clin J Sport Med 20(2):80–85 30. Sampson MJ, Jackson MP, Moran CJ, Moran R, Eustace SJ, Shine S (2008) Three Tesla MRI for the diagnosis of meniscal
and anterior cruciate ligament pathology: a comparison to arthroscopic findings. Clin Radiol 63(10):1106–1111 31. Schaefer FKW, Schaefer PJ, Brossmann J, Frahm C, Muhle C, Hilgert RE, Heller M, Jahnke T (2006) Value of fat-suppressed PD-weighted TSE-sequences for detection of anterior and posterior cruciate ligament lesions: comparison to arthroscopy. Eur J Radiol 58(3):411–415 32. Scholten R, Opstelten W, van der Plas CG, Bijl D, Deville W, Bouter LM et al (2003) Accuracy of physical diagnostic tests for assessing ruptures of the anterior cruciate ligament: a meta-analysis. J Fam Pract 52(9):689–694 33. Schub DL, Altahawi F, Meisel AF, Winalski C, Parker RD, Saluan PM (2012) Accuracy of 3-Tesla magnetic resonance imaging for the diagnosis of intra-articular knee injuries in children and teenagers. J Pediatr Orthop 32(8):765–769 34. Solomon DH, Simel DL, Bates DW, Katz JN, Schaffer JL (2001) Does this patient have a torn meniscus or ligament of the knee? Value of the physical examination. JAMA 286(13):1610–1620 35. Sutton KM, Bullock JM (2013) Anterior cruciate ligament rupture: differences between males and females. J Am Acad Orthop Surg 21(1):41–50 36. Swain MS, Henschke N, Kamper SJ, Downie AS, Koes BW, Maher CG (2014) Accuracy of clinical tests in the diagnosis of anterior cruciate ligament injury: a systematic review. Chiropr Man Ther 22:25. doi:10.1186/s12998-014-0025-8 37. Thomas S, Pullagura M, Robinson E, Cohen A, Banaszkiewicz P (2007) The value of magnetic resonance imaging in our current management of ACL and meniscal injuries. Knee Surg Sports Traumatol Arthrosc 15(5):533–536 38. Tsai K-J, Chiang H, Jiang C-C (2004) Magnetic resonance imaging of anterior cruciate ligament rupture. BMC Musculoskelet Disord. 5:21. doi:10.1186/1471-2474-5-21 39. Wagemakers HP, Luijsterburg PA, Boks SS, Heintjes EM, Berger MY, Verhaar JA, Koes BW, Bierma-Zeinstra SM (2010) Diagnostic accuracy of history taking and physical examination for assessing anterior cruciate ligament lesions of the knee in primary care. Arch Phys Med Rehabil 91(9):1452–1459 40. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J (2003) The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 3:25
13
KNEE
Diagnostic accuracy of physical examination for anterior knee instability: a systematic review Marie‑Claude Leblanc1 · Marcin Kowalczuk1 · Nicole Andruszkiewicz2 · Nicole Simunovic3 · Forough Farrokhyar3,4 · Travis Lee Turnbull5 · Richard E. Debski6 · Olufemi R. Ayeni1,3
Received: 20 January 2015 / Accepted: 2 March 2015 © European Society of Sports Traumatology, Knee Surgery, Arthroscopy (ESSKA) 2015
Abstract Purpose Determining diagnostic accuracy of Lachman, pivot shift and anterior drawer tests versus gold standard diagnosis (magnetic resonance imaging or arthroscopy) for anterior cruciate ligament (ACL) insufficiency cases. Secondarily, evaluating effects of: chronicity, partial rupture, awake versus anaesthetized evaluation. Methods Searching MEDLINE, EMBASE and PubMed identified studies on diagnostic accuracy for ACL insufficiency. Studies identification and data extraction were performed in duplicate. Quality assessment used QUADAS tool, and statistical analyses were completed for pooled sensitivity and specificity. Results Eight studies were included. Given insufficient data, pooled analysis was only possible for sensitivity on
Lachman and pivot shift test. During awake evaluation, sensitivity for the Lachman test was 89 % (95 % CI 0.76, 0.98) for all rupture types, 96 % (95 % CI 0.90, 1.00) for complete ruptures and 68 % (95 % CI 0.25, 0.98) for partial ruptures. For pivot shift in awake evaluation, results were 79 % (95 % CI 0.63, 0.91) for all rupture types, 86 % (95 % CI 0.68, 0.99) for complete ruptures and 67 % (95 % CI 0.47, 0.83) for partial ruptures. Conclusion Decreased sensitivity of Lachman and pivot shift tests for partial rupture cases and for awake patients raised suspicions regarding the accuracy of these tests for diagnosis of ACL insufficiency. This may lead to further research aiming to improve the understanding of the true accuracy of these physical diagnostic tests and increase the reliability of clinical investigation for this pathology. Level of evidence IV.
Electronic supplementary material The online version of this article (doi:10.1007/s00167-015-3563-2) contains supplementary material, which is available to authorized users.
Keywords Anterior cruciate ligament · Rupture · Diagnostic accuracy · Physical examination · Knee injuries
* Olufemi R. Ayeni [email protected]
Introduction
1
Division of Orthopedic Surgery, Department of Surgery, McMaster University, 4E15‑ 1200 Main St W, Hamilton, ON L8N 3Z5, Canada
2
Department of Public Health Sciences, Queen’s University, Kingston, ON, Canada
3
Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada
4
Department of Surgery, McMaster University, Hamilton, ON, Canada
5
Department of BioMedical Engineering, Steadman Philippon Research Institute, Vail, CO, USA
6
Orthopaedic Robotics Laboratory, University of Pittsburgh, Pittsburgh, PA, USA
Rupture of the anterior cruciate ligament (ACL) is one of the most frequent knee injuries [21]. The highest incidence has been reported among young and active individuals aged 15–25 years, making early diagnosis essential [14, 35]. The natural history of ACL-deficient knees with recurrent instability includes acceleration of chondral wear and increased incidence of meniscal tears [19, 26]. Despite the presence of an increased incidence of osteoarthritis in ACL stabilized knees, it is clear from the literature that the incidence of meniscal tears and chondral injuries can be reduced with ACL reconstruction [19, 25]. Due to associated recurrent instability, conservative treatment of ACL ruptures can
13
produce unsatisfactory results leading to losses of productivity, competition and scholarship opportunities [19, 26, 35]. Diagnosis is usually based on a combination of history, physical examination and diagnostic imaging. Three of the most common physical tests used for diagnosis are the Lachman, the pivot shift and the anterior drawer test [29]. The sensitivity and specificity of these tests has been largely reported in the literature, with great variability. Reported sensitivity and specificity values for the Lachman test ranged from 48 to 100 and 42 to 100 %, respectively. The pivot shift test has a reported sensitivity ranging from 0 to 98.4 % and a specificity ranging from 82 to 100 % [6, 17, 24, 32, 34]. For the anterior drawer test, the reported sensitivity ranges from 9 to 95.2 % and the specificity from 23 to 100 %. The best moment to obtain an unbiased examination is immediately after injury, when there is no significant hemarthrosis, painful spasm or muscle guarding [26]. Unfortunately, this scenario is more the exception than the rule in the recreational athlete and the physical examination can often be equivocal. In an effort to summarize the value of physical examination, seven systematic reviews have been produced over the last 15 years [6, 17, 22, 24, 32, 34, 36]. Unfortunately, most of these systematic reviews did not include studies published since the year 2000. Moreover, the quality of the studies published prior to the year 2000 was less controlled and more heterogeneous, making pooled analysis difficult. Of the previously conducted systematic reviews, the effect of anaesthesia and lesion chronicity on physical exam accuracy was reported by only a single study and did not distinguish between partial and complete ruptures [6]. The current systematic review aimed to determine the diagnostic accuracy of physical examination techniques (Lachman, pivot shift or anterior drawer tests) versus a gold standard diagnosis [magnetic resonance imaging (MRI) or arthroscopy] in cases of anterior knee instability secondary to ACL insufficiency. Furthermore, the impact of various clinical factors on physical examination accuracy was assessed. This included: acute versus chronic injury, complete versus partial ACL rupture and examination under anaesthesia (EUA) versus awake clinical setting. The hypothesis was that advanced MRI and arthroscopic techniques combined with improved study design in articles presented since the year 2000 would yield quality data to support evaluation of the diagnostic accuracy of physical examinations. In addition, the hypothesis was made that the diagnostic accuracy would be lower for acute injuries, partial ruptures and examinations conducted in the awake clinical setting. This is the first systematic review on the subject focusing on contemporary literature presented since the year 2000.
13
Knee Surg Sports Traumatol Arthrosc
Materials and methods This systematic review was performed according to PRISMA guidelines [23]. Study eligibility Studies were eligible for inclusion if they primarily assessed the diagnostic accuracy of physical examination (Lachman, pivot shift or anterior drawer tests) relative to MRI or arthroscopy as a gold standard for diagnosis. The study population included all patients with anterior knee instability secondary to ACL insufficiency. The inclusion criteria were as follows: (1) patients with a knee injury, (2) physical diagnosis with at least one physical test (clinic or EUA), (3) correlation with a gold standard (MRI, arthroscopy, arthrotomy), (4) in vivo human studies, (5) adults and (6) studies published in English or French. Exclusion criteria included: (1) review articles, (2) knee dislocation with multiple ligamentous injuries, (3) studies on specific injuries other than primary ACL, (4) no specification of the physical diagnostic test used, (5) studies which only discussed surgical techniques and (6) publications published prior to the year 2000. Systematic reviews and biomechanical (non-human) studies were excluded. Information sources and research strategy Electronic databases [MEDLINE (1946–21 July 2014), EMBASE (1980–22 July 2014) and PubMed (1950s–21 July 2014)] were searched for knee ligamentous injury studies reporting data on the diagnostic accuracy comparison between physical examination and MRI, arthroscopy or arthrotomy. The complete search strategy is presented in Appendix 1 of Supplementary Material. Articles from the different databases were compiled and duplicates removed. The results were uploaded to a bibliographic management database (RefWorks, version 2.0; Bethesda, MD). Study selection Two reviewers (MC.L., M.K.) initially completed a title and abstract review screening for eligible studies. Following this, a full-text review was conducted and the references for each article were also hand-searched for other eligible studies. Both reviewers screened all studies independently. Disagreements were resolved by a consensus discussion involving the senior author. Exclusion of articles published before the year 2000 was completed after reviewing all full texts. Those publications (pre year 2000) were excluded due to lack of consistent methodology, limited imaging options compared to contemporary practice and vague
Knee Surg Sports Traumatol Arthrosc
definitions of ACL injuries. Any article not available online was obtained in print through our institutional library. Data collection Data were collected independently and in duplicate from the included articles by two reviewers (MC.L., and M.K. or LP.B. or N.F.) using a standardized data collection form. Extracted data included: title, author, year of publication, location of study, study design, level of evidence, number of patients, mean age, male to female ratio, number of knees examined, number of ACL ruptures, number of normal ACLs, type of injury (acute, chronic, complete, or partial), physical test, setting of test (EUA or clinic), gold standard test, description of physical test, physical exam result and gold standard exam result. Sensitivity and specificity data were collected as counts whenever reported. Any discrepancies regarding collected data were resolved through discussion and consensus between reviewers (MC.L., M.K.) and the senior author. Quality assessment The methodological quality of each study was assessed independently by two reviewers (MC.L. and M.K.) using the QUADAS tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews [40]. QUADAS is a validated clinometric tool used to assess the overall quality of diagnostic accuracy studies through individual quality component questions. Statistical analysis and data measures Sensitivity and specificity data were extracted as counts from each of the studies where applicable. The figures were changed into counts if reported in percentages. Sensitivity was defined as the percentage of individuals who were correctly identified as having a ruptured ACL and specificity was defined as the percentage of individuals who were correctly identified as not having a ruptured ACL. In order to maintain consistency, studies that reported grades for each of the three tests were evaluated using a defined method. With regard to the Lachman and anterior drawer tests, a grade 1 or higher was consistent with an ACL rupture, while grade 0 examinations were deemed normal. For the pivot shift test, a glide, clunk, gross or severe grade was considered to be a ruptured ACL. This grading classification assisted with calculating true positives and false negatives for each of the tests in each article. This means according to the aforementioned classification that patients with a ruptured ACL and a grade 1 or higher were defined as true positives, while patients with a ruptured ACL and a grade 0 were defined as false negatives. Only a single
study included healthy participants without an ACL rupture, which allowed for the calculation of true negatives and false positives to determine the specificity of each applicable test within the study [5]. Therefore, without both sensitivity and specificity data for more than one study, a full meta-analysis of diagnostic accuracy and calculation of likelihood ratios of positive and negative tests could not be performed. A random effects model (DerSimonian–Laird) was used to calculate the pooled weighted proportions of sensitivity due to the inherent heterogeneity applied to the observational studies upon collecting the data for true positives and false negatives from the studies. Pooled sensitivities with the corresponding 95 % confidence intervals (CI) were reported, and forest plots of the pooled proportions were presented. StatsDirect 2.7 (StatsDirect Ltd, UK) was used for data analysis.
Results Study identification, descriptions and quality assessment The literature search initially yielded 2529 studies of which 103 underwent full-text review. Ultimately, eight studies were included for final analysis (Fig. 1) [5, 9–11, 28, 29, 38, 39]. The majority of the studies were level IV evidence. The total pooled number of ACL-deficient knees was 1196. Two studies, Beldame et al. [5] and Wagemakers et al. [39], reported on both ACL-deficient and non-ACL-deficient knees and six used only an ACL-deficient population. As for the gold standard correlation, Wagemakers et al. [39] used MRI, while the others studies used arthroscopy. Two studies, Dhillon et al. [10] and Espregueira-Mendes et al. [11], completed their physical examination under anaesthesia, and six studies evaluated the patient in an awake clinical setting (Table 1). The result for initial inter-reviewer agreement was 0.980 (95 % CI 0.978–0.982) for title screen, 0.967 (95 % CI 0.959–0.974) for abstract screen and 0.796 (95 % CI 0.699–0.862) for full text screen. The quality assessment of the included studies is reported in Table 2. Pooled results A pooled sensitivity analysis was only possible for Lachman and pivot shift tests. Additionally, the effect of the type of rupture (partial vs. complete) and the result in an awake clinical setting were reported. The detailed results are described in Figs. 2 and 3, in Table 3 as well as in Appendix 2 of Supplementary Material through 11. For combined (partial and complete) ruptures, the pooled sensitivity was 89 % (95 % CI 76–98 %) for the Lachman test
13
Knee Surg Sports Traumatol Arthrosc
2529 Studies Identified MEDLINE: 1088 Studies PUBMED: 805 Studies EMBASE: 636 Studies
825 duplicates
1704 studies for title review
Excluded studies: 1417
287 studies for abstract review
Excluded studies: 209
103 studies for full text review
Excluded studies: 95
Handpick studies: 25
8 studies included in the systematic review
-
Before year 2000: 61 No specification on physical test: 6 Quantifying physical test: 5 Cadaver: 1 Systematic review: 6 No physical test results: 12 Editorial: 3 Prone Lachman: 1
Fig. 1 Flow diagram of literature search process
and 79 % (95 % CI 63–91 %) for the pivot shift test. For complete ruptures, the pooled sensitivity was 96 % (95 % CI 90–100 %) for the Lachman test and 86 % (95 % CI 68–99 %) for the pivot shift test. For partial ruptures, the pooled sensitivity was lower and more variable: 68 % (95 % CI 25–98 %) for the Lachman test and 67 % (95 % CI 47–83 %) for the pivot shift test. Most of the included studies did not provide data for calculation of specificity. This is because patients with positive MRI or arthroscopy findings underwent physical examinations, and as a result, data on the true negatives and false positives were not available. Only two studies provided complete data on both ACL-deficient knees and non-ACL-deficient knees [5, 39]. They reported specificity for the anterior drawer test of 83.9 % (95 % CI not available) and 57.0 % (95 % CI 0.48, 0.67). Only Beldame et al. [5] reported on specificity for the Lachman and pivot shift tests, at 78.1 % (95 % CI 0.61, 0.89) and 86.4 % (95 % CI not available), respectively. Due to insufficient data, pooled sensitivity results were not calculated for the anterior drawer test, EUA and chronicity. Only two studies reported sensitivity results for combined or complete ruptures using the anterior drawer test, and only Beldame et al. reported results for partial ruptures for the anterior drawer test [5, 29]. Also, only two studies used EUA [10, 11]. Dhillon et al. [10] reported a
13
sensitivity of 100 % (largest 95 % CI 0.87, 1.00) for Lachman and pivot shift physical examinations in all settings including combined (partial and complete) and individual rupture types (Table 3). Espregueira-Mendes et al. [11] only reported on complete ruptures, with a sensitivity of 100 % (95 % CI 0.88, 1.00) for the Lachman and 82 % (95 % CI 0.63, 0.94) for the pivot shift (Table 3). Data regarding the effect of the chronicity of the lesion were also too scarce to analyse.
Discussion The most important finding of this systematic review was that although both Lachman and pivot shift tests are sensitive in diagnosing ACL ruptures, the clinical setting (awake vs. non-awake) and extent of injury (partial vs. complete rupture) have an impact on diagnostic accuracy. Biomechanical data from Araki et al. [3] support this observation by demonstrating that partial tears in ACL knees had less laxity in all patients and a detectable end point on Lachman examination in only 45 % of the patients. Concerning the effect of anaesthesia, the decrease in sensitivity with an awake patient in a clinical setting reinforces the effect of muscular guarding on resisting tibial anterior translation during Lachman and pivot shift examination.
Ipswich, UK
Porto, Portugual
Dhillon [10]
EspregueiraMendes [11] Panisset al. [28]
Netherlands (five centres)
Wagemakers [39]
48
112
418
28
182
300
112
Knee injuries within 5 wks
ACL deficient knee ACL deficient knee 28
48
112
17
32
N/A
208
28
103
177
70
Total ACL Complete ACL rupture
80 Knee injury with indication for knee arthroscopy ACL deficient 300 knee ACL deficient 182 knee ACL deficient 28 knee ACL deficient 418 knee
Nb patients Context
134 Cross-sectional diagnostic study/ IV
Retrospective care series/IV Retrospective care series/IV
Prospective case serie/IV
Prospective case– control/III Retrospective diagnostic/IV Retrospective/IV
Prospective case– control/III
Design/evidence
11
16
N/A
67
0
27
123
10
Partial ACL rupture
Arthroscopy
Arthroscopy
Arthroscopy
AD
Lachman
MRI
Arthroscopy
AD Lachman PS Arthroscopy
Lachman PS
AD Lachman PS Arthroscopy
Lachman PS
Lachman PS
Awake
Awake
Awake
Awake
EUA
EUA
Awake
Awake
Gold standard Setting
AD Lachman PS Arthroscopy
Clinical test
ACL anterior cruciate ligament, AD anterior drawer, PS pivot shift, MRI magnetic resonance imaging, wks weeks, N/A none applicable, Nb number, EUA examination under anaesthesia
Taipei, Taiwan
Tsai [38]
Peeler [29]
France (Bordeaux, Grenoble, Lyon, Pau, Toulouse) Manitoba, Canada
Rouen Cedex, France/University Lyon, France
Beldame [5]
Dejour [9]
Location/type of hospital
References
Table 1 Characteristics of included studies
Knee Surg Sports Traumatol Arthrosc
13
Knee Surg Sports Traumatol Arthrosc
Table 2 Validity scoring QUADAS References
Items number 1
2
3
4
5
6
7
8
9
10
11
12
13
14
Beldame [5] Dejour [9] Dhillon [10] Espregueira-Mendes [11] Panisset [28] Peeler [29] Tsai [38]
No No No No No No No
Yes Yes Yes Yes Unclear Yes No
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Unclear Yes Yes Yes Yes Unclear Unclear
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Unclear No No No Unclear Unclear No
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Unclear No
Yes Yes Yes Yes Yes Yes Yes
Wagemakers [39]
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
1. Was the spectrum of patients representative of the patients who will receive the test in practice? 2. Were selection criteria clearly described? 3. Is the reference standard likely to correctly classify the target condition? 4. Is the time period between reference standard and index test short enough to be reasonably sure that the target condition did not change between the two tests? 5. Did the whole sample or a random selection of the sample, receive verification using a reference standard? 6. Did patients receive the same reference standard regardless of the index test result? 7. Was the reference standard independent of the index test (i.e. the index test did not form part of the reference standard)? 8. Was the execution of the index test described in sufficient detail to permit replication of the test? 9. Was the execution of the reference standard described in sufficient detail to permit its replication? 10. Were the index test results interpreted without knowledge of the results of the reference standard? 11. Was the reference standard results interpreted without knowledge of the results of the index test? 12. Were the same clinical data available when test results were interpreted as would be available when the test is used in practice? 13. Were uninterpretable/intermediate test results reported? 14. Were withdrawals from the study explained?
Proportion meta-analysis plot [random effects]
Proportion meta-analysis plot [random effects] Beldame 2011
0.81 (0.71, 0.89)
Dejour 2013
1.00 (0.99, 1.00)
Peeler 2010
0.86 (0.77, 0.92)
Tsai 2004
0.81 (0.67, 0.91)
Panisset 2008 combined 0.6
0.7
0.8
0.9
Beldame 2011
0.60 (0.45, 0.73)
Dejour 2013
0.89 (0.85, 0.92)
Peeler 2010
0.63 (0.51, 0.74)
0.89 (0.84, 0.92)
Panisset 2008
0.93 (0.89, 0.96)
0.89 (0.76, 0.98)
combined
0.79 (0.63, 0.91)
1.0
proportion (95% confidence interval)
Fig. 2 Lachman, reported and pooled sensitivity, combined ruptures, awake setting
Clinical diagnosis of an ACL insufficient knee is not an exact science. Many clinicians have attempted to develop an accurate and reproducible physical examination in order to facilitate early diagnosis. The number of different tests described in the literature demonstrates that there is no ‘one size fits all’ physical exam to detect ACL rupture [1, 2, 4, 7,
13
0.4
0.6
0.8
1.0
1.2
proportion (95% confidence interval)
Fig. 3 Pivot shift, reported and pooled sensitivity, combined ruptures, awake setting
9, 11–13, 15, 16, 18–21, 25, 26]. The anterior drawer, Lachman and pivot shift tests have been the major physical tests used for diagnosis of ACL insufficiency. Many factors can affect the reproducibility of these physicals tests [6, 24, 29, 32]. Cooperman et al. [8] compared the clinical accuracy
Knee Surg Sports Traumatol Arthrosc Table 3 Summary of pooled sensitivity (95 % CI) for Lachman and pivot shift Type of rupture
Tests
Combine (awake + EUA)
Awake
EUAa
Combine (complete and partial)
Lachman PS Lachman PS Lachman
92 % (0.81, 0.99) 85 % (0.70, 0.96) 98 % (0.94, 1.00) 89 % (0.77, 0.97) 77 % (0.43, 0.98)
89 % (0.76, 0.98) 79 % (0.63, 0.91) 96 % (0.90, 1.00) 86 % (0.68, 0.99) 68 % (0.25, 0.98)
100 % (0.97, 1.00) [10] 100 % (0.97, 1.00) [10] 100 % (0.96, 1.00) [10] (0.88, 1.00) [11] 100 % (0.96, 1.00) [10] 82 % (0.63, 0.94) [11] 100 % (0.87, 1.00) [10]
PS
77 % (0.54, 0.93)
67 % (0.47, 0.83)
100 % (0.87, 1.00) [10]
Complete Partial
EUA examination under anaesthesia, PS pivot shift, CI confidence interval a
Only two studies reported on EUA. This column is not a pooled analysis, but a report of the data available
of the Lachman test executed by an orthopaedic surgeon to that by a physical therapist. They reported a sensitivity of 77 % and specificity of 50 % for the orthopaedic surgeon. For the physical therapist, these numbers decreased to a sensitivity of 65 % and specificity of 42 %. In addition to the experience level of the examiner, many other factors have been reported to affect the reproducibility of the tests such as knee effusion, muscle spasms, thigh circumference, size of examiner hands, timing of examination, chronicity of lesion, type of rupture, associated injuries and EUA [6, 24, 29, 32]. These multiple factors are likely to contribute to the large variability in reported sensitivity and specificity in the literature. Two systematic reviews have reported pooled analyses. Jackson et al. [17] showed a sensitivity of 48 % for anterior drawer, 87 % for Lachman and 61 % for pivot shift. For the specificity, the results were 87 % for anterior drawer, 93 % for Lachman and 97 % for pivot shift [17]. The current review found a comparable sensitivity for the Lachman examination at 92 % but an improved sensitivity of 85 % for the pivot shift. Benjaminse et al. [6] also conducted a detailed pooled analysis, being the first to divide results according to presence or absence of anaesthesia and acute or chronic rupture. The authors reported a sensitivity for combined ruptures in patient without anaesthesia of 55 % for anterior drawer, 85 % for Lachman and 24 % for pivot shift [6]. For the specificity, the results were 92 % for the anterior drawer, 94 % for the Lachman and 98 % for the pivot shift [6]. In contrast, the current review noted an improved sensitivity for the pivot shift in awake patients at 79 %. Lachman testing in awake patients was comparable at 89 %. Unfortunately, as discussed in the pooled analysis results above, insufficient data were available from the selected articles in this review to allow for calculation of specificity for Lachman and pivot shift test as well as statistical analysis for anterior drawer test. Also, the reviews described above included only articles published prior to 2000, yielding a completely different data set from this review. The first evaluation in an awake clinical setting following a traumatic knee injury requires a highly sensitive
physical examination with good reliability. One systematic review looked at the intra- and inter-rater reliability of the physical examination [22]. Given the modest quality of the studies included, they could only show a moderate intra-rater reliability for the Lachman test. The intra-rater reliability for anterior drawer and pivot shift as well as inter-rater reliability for the three tests was inconclusive. This systematic review demonstrates very large variability in the literature regarding the reported diagnostic accuracy of physical examinations. Furthermore, the type of rupture can greatly affect the sensitivity of both Lachman and pivot shift tests. Additionally, a clear trend implicating decreased sensitivity with physical examination in an awake clinical setting was observed. Given the above challenges, a more reproducible method of diagnosis is commonly preferred. Prior to the availability of high-field strength technology, MRI diagnosis had varying accuracies, with sensitivity ranging from 64 to 94.4 % and a specificity ranging from 94.3 to 100 % [27, 31, 37]. With the advent of 3 T MRI systems, diagnostic accuracy has increased to 97.9–100 % for sensitivity and 98.6–100 % for specificity [30, 33]. As such, MRI is likely to be used ever increasingly for diagnosis of ACL insufficiency for the foreseeable future. There were multiple strengths to this systematic review. A comprehensive search strategy of multiple databases using various keywords and subheading was performed to reduce publication bias. Study selection and data extraction were conducted in duplicate to ensure accuracy and further minimize bias. Included articles were limited to studies published after the year 2000 in an attempt to evaluate only high-quality studies relevant to current practice. Despite the methodological strengths of this study, there are certain limitations due to the inherent biases of the included studies, and thus, findings should be interpreted with caution. The included studies were of different observational designs with diverse patient populations, resulting in a large inter-study heterogeneity. Most studies did not report sufficient data to allow a complete diagnostic accuracy analysis. The majority of studies only reported physical examination findings on patients with known ACL
13
injury based on MRI or arthroscopic examination. This allowed sensitivity analysis but precluded calculation of exam specificity in this review. Also, the lack of ACL-deficient patients for comparison in six of the studies limited subsequent analysis. It is therefore difficult to state a definitive recommendation regarding the diagnostic accuracy of the Lachman test, pivot shift test or anterior drawer test. Furthermore, a description of the exact physical exam procedure was not always available, and often only the name of the physical test was stated. This review clearly demonstrates insufficient and often incomplete documentation of the diagnostic accuracy of the most commonly used physical tests to diagnose ACL rupture. From a clinical perspective, this study broadens our understanding of anterior knee instability clinical assessment and the effect of various factors affecting their results, such as awake patients and partial tears. The apparent lack of diagnostic accuracy offered by existing common physical tests may lead to further research aiming to improve the understanding of the true accuracy of these physical diagnostic tests and increase the reliability of clinical investigation for this pathology.
Conclusion The key finding of this systematic review was that although both Lachman and pivot shift tests are sensitive in diagnosing ACL ruptures, the clinical setting (awake vs. nonawake) and extent of injury (partial vs. complete rupture) have an impact on diagnostic accuracy. The current literature did not contain sufficient data to calculate pooled specificity; therefore, no clear recommendation regarding diagnostic accuracy of the physical examination for ACL insufficient knees could be made. Given the advances in the resolution of MRI and concomitant capability for diagnosing ACL ruptures, the possibility of conducting a diagnostic accuracy study for physical examination of ACL ruptures is now available and could greatly improve the understanding of the true accuracy of these physical diagnostic tests. Acknowledgments A special thanks to four collaborators to this systematic review: Louis-Philippe Baisi, Nicole Friel, Hiba Abdul Mannan and Zakia Islam.
References 1. Anderson AF, Lipscomb AB (1989) Preoperative instrumented testing of anterior and posterior knee laxity. Am J Sports Med 17(3):387–392 2. Araki D, Kuroda R, Kubo S, Nagamune K, Hoshino Y, Nishimoto K, Takayama K, Matsushita T, Tei K, Yamaguchi M, Kurosaka M (2011) The use of an electromagnetic measurement system for
13
Knee Surg Sports Traumatol Arthrosc anterior tibial displacement during the Lachman test. Arthroscopy 27(6):792–802 3. Araki D, Kuroda R, Matsushita T, Matsumoto T, Kubo S, Nagamune K, Kurosaka M (2013) Biomechanical analysis of the knee with partial anterior cruciate ligament disruption: quantitative evaluation using an electromagnetic measurement system. Arthroscopy 29(6):1053–1062 4. Araujo PH, Ahlden M, Hoshino Y, Muller B, Moloney G, Fu FH, Musahl V (2012) Comparison of three non-invasive quantitative measurement systems for the pivot shift test. Knee Surg Sports Traumatol Arthrosc 20(4):692–697 5. Beldame J, Bertiaux S, Roussignol X, Lefebvre B, Adam J-M, Mouilhade F, Dujardin F (2011) Laxity measurements using stress radiography to assess anterior cruciate ligament tears. Orthop Traumatol Surg Res 97(1):34–43 6. Benjaminse A, Gokeler A, van der Schans CP (2006) Clinical diagnosis of an anterior cruciate ligament rupture: a meta-analysis. J Orthop Sports Phys Ther 36(5):267–288 7. Braunstein EM (1982) Anterior cruciate ligament injuries: a comparison of arthrographic and physical diagnosis. Am J Roentgenol 138(3):423–425 8. Cooperman JM, Riddle DL, Rothstein JM (1990) Reliability and validity of judgments of the integrity of the anterior cruciate ligament of the knee using the Lachman’s test. Phys Ther 70(4):225–233 9. Dejour D, Ntagiopoulos PG, Saggin PR, Panisset J-C (2013) The diagnostic value of clinical tests, magnetic resonance imaging, and instrumented laxity in the differentiation of complete versus partial anterior cruciate ligament tears. Arthroscopy 29(3):491–499 10. Dhillon AK, Oday A-D, Servant CT (2013) Diagnostic accuracy of ACL tears according to tear morphology. Acta Orthop Belg 79(1):76–82 11. Espregueira-Mendes J, Pereira H, Sevivas N, Passos C, Vasconcelos JC, Monteiro A, Oliveira JM, Reis RL (2012) Assessment of rotatory laxity in anterior cruciate ligament-deficient knees using magnetic resonance imaging with Porto-knee testing device. Knee Surg Sports Traumatol Arthrosc 20(4):671–678 12. Galway HR, MacIntosh DL (1980) The lateral pivot shift: a symptom and sign of anterior cruciate ligament insufficiency. Clin Orthop Relat Res 147:45–50 13. Graham GP, Johnson S, Dent CM, Fairclough JA (1991) Comparison of clinical tests and the KT1000 in the diagnosis of anterior cruciate ligament rupture. Br J Sports Med 25(2):96–97 14. Griffin LY, Agel J, Albohm MJ, Arendt EA, Dick RW, Garrett WE, Garrick JG, Hewett TE, Huston L, Ireland ML et al (2000) Noncontact anterior cruciate ligament injuries: risk factors and prevention strategies. J Am Acad Orthop Surg 8(3):141–150 15. Hardaker WT, Garret WE Jr, Bassett FH (1990) Evaluation of acute traumatic hemarthrosis of the knee joint. South Med J 83(6):640–644 16. Hughston JC, Andrews JR, Cross MJ, Moschi A (1976) Classification of knee ligament instabilities Part 1: the medial compartment and cruciate ligaments. J Bone Joint Surg Am 58(A2):159–172 17. Jackson JL, O’Malley PG, Kroenke K (2003) Evaluation of acute knee pain in primary care. Ann Intern Med 139(7):575–588 18. Jonsson T, Althoff BO, Peterson L, Renström P (1982) Clinical diagnosis of ruptures of the anterior cruciate ligament: a comparative study of the Lachman test and the anterior drawer sign. Am J Sports Med 10(2):100–102 19. Junkin DM Jr, Johnson DL, Fu FH, Miller MD, Willenborg M, Fanelli GC, Wascher DC (2009) Knee ligaments injuries. In: Kibler WB (ed) Orthopaedic knowledge update: sports medicine 4, 4th edn. Rosemont, Illinois, pp 135–154
Knee Surg Sports Traumatol Arthrosc 20. Katz JW, Fingeroth RJ (1986) The diagnostic accuracy of ruptures of the anterior cruciate ligament comparing the Lachman test, the anterior drawer sign, and the pivot shift test in acute and chronic knee injuries. Am J Sports Med 14(1):88–91 21. Kirchhoff C, Brucker PU, Imhoff AB (2013) Evolving concepts in tunnel placement for ACL reconstruction. In: Johnson DH (ed) Operative arthroscopy, 4th edn. Pennsylvania, Philadelphia, pp 772–777 22. Lange T, Freiberg A, Dröge P, Lützner J, Schmitt J, Kopkow C (2014) The reliability of physical examination tests for the diagnosis of anterior cruciate ligament rupture: a systematic review. Man Ther. doi:10.1016/j.math.2014.11.003 23. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, Clarke M, Devereaux PJ, Kleijnen J, Moher D (2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 62(10):e1–e34. doi:10.1371/journal.pmed.1000100 24. Malanga GA, Andrus S, Nadler SF, McLean J (2003) Physical examination of the knee: a review of the original test description and scientific validity of common orthopedic tests. Arch Phys Med Rehabil 84(4):592–603 25. Miller MD (2011) Single-bundle anterior cruciate ligament repair. In: Wiesel SW (ed) Operative techniques in orthopaedic surgery, 1st edn. Pennsylvania, Philadelphia, pp 342–349 26. Miller RH, Azar FM (2013) Knee injuries. In: Canale ST, Beaty JH (eds) Campbell’s operative orthopaedics, 12th edn. Pennsylvania, Philadelphia, pp 2052–2211 27. Oei EHG, Nikken JJ, Verstijnen ACM, Ginai AZ, Myriam Hunink MG (2003) MR imaging of the menisci and cruciate ligaments: a systematic review. Radiology 226(3):837–848 28. Panisset J-C, Duraffour H, Vasconcelos W, Colombet P, Javois C, Potel J-F, Dejour D (2008) Analyses clinique, radiologique et arthroscopique de la rupture du LCA: étude prospective de 418 cas. Rev Chir Orthop Reparatrice Appar Mot 94(8 supp):362–368 29. Peeler J, Leiter J, MacDonald P (2010) Accuracy and reliability of anterior cruciate ligament clinical examination in a multidisciplinary sports medicine setting. Clin J Sport Med 20(2):80–85 30. Sampson MJ, Jackson MP, Moran CJ, Moran R, Eustace SJ, Shine S (2008) Three Tesla MRI for the diagnosis of meniscal
and anterior cruciate ligament pathology: a comparison to arthroscopic findings. Clin Radiol 63(10):1106–1111 31. Schaefer FKW, Schaefer PJ, Brossmann J, Frahm C, Muhle C, Hilgert RE, Heller M, Jahnke T (2006) Value of fat-suppressed PD-weighted TSE-sequences for detection of anterior and posterior cruciate ligament lesions: comparison to arthroscopy. Eur J Radiol 58(3):411–415 32. Scholten R, Opstelten W, van der Plas CG, Bijl D, Deville W, Bouter LM et al (2003) Accuracy of physical diagnostic tests for assessing ruptures of the anterior cruciate ligament: a meta-analysis. J Fam Pract 52(9):689–694 33. Schub DL, Altahawi F, Meisel AF, Winalski C, Parker RD, Saluan PM (2012) Accuracy of 3-Tesla magnetic resonance imaging for the diagnosis of intra-articular knee injuries in children and teenagers. J Pediatr Orthop 32(8):765–769 34. Solomon DH, Simel DL, Bates DW, Katz JN, Schaffer JL (2001) Does this patient have a torn meniscus or ligament of the knee? Value of the physical examination. JAMA 286(13):1610–1620 35. Sutton KM, Bullock JM (2013) Anterior cruciate ligament rupture: differences between males and females. J Am Acad Orthop Surg 21(1):41–50 36. Swain MS, Henschke N, Kamper SJ, Downie AS, Koes BW, Maher CG (2014) Accuracy of clinical tests in the diagnosis of anterior cruciate ligament injury: a systematic review. Chiropr Man Ther 22:25. doi:10.1186/s12998-014-0025-8 37. Thomas S, Pullagura M, Robinson E, Cohen A, Banaszkiewicz P (2007) The value of magnetic resonance imaging in our current management of ACL and meniscal injuries. Knee Surg Sports Traumatol Arthrosc 15(5):533–536 38. Tsai K-J, Chiang H, Jiang C-C (2004) Magnetic resonance imaging of anterior cruciate ligament rupture. BMC Musculoskelet Disord. 5:21. doi:10.1186/1471-2474-5-21 39. Wagemakers HP, Luijsterburg PA, Boks SS, Heintjes EM, Berger MY, Verhaar JA, Koes BW, Bierma-Zeinstra SM (2010) Diagnostic accuracy of history taking and physical examination for assessing anterior cruciate ligament lesions of the knee in primary care. Arch Phys Med Rehabil 91(9):1452–1459 40. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J (2003) The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 3:25
13
