1 Introduction

Patellar dislocation, a common knee injury that occurs when the patella is displaced from the femoral trochlea, predominately affects young and active patients in the second or third decade and is estimated to compose up to 3% of all knee injuries.^1–6 Typically, the management of a first-time lateral patellar dislocation (LPD) is conservative with surgical treatment reserved for those with displaced osteochondral fragments, bony avulsion of the MPFL, or meniscal tear.^7,8 There has been mixed evidence that demonstrates that the repair of the medial patellofemoral ligament (MPFL) in the setting of LPD results in a reduction of the redislocation rate.^9,10 However, those who have ever had a LPD are at greater risk for redislocation, with three times the risk for recurrent dislocation for female patients.^11,12

The MPFL's primary role is to prevent lateral translation of the patella from the femoral trochlea, and it is assisted by the patellotibial and patellomeniscal ligament complex.^13,14 Hemarthrosis and medial soft tissue injury to these structures; are almost universal in LPD.^4,13,15 The dislocation mechanism can also result in other soft tissue injuries, such as injury to the menisci.^4,16,17 Concomitant meniscal injury in the setting of an LPD may necessitate surgical intervention, so proper radiological evaluation of the knee is critical in the setting of LPD. Prevalence of meniscal injury after LPD has been reported in several MRI-based studies, and ranges from as low as 8.2% to as high as 29%.^13,15–23

In light of the frequent occurrence of meniscal changes observed on MRI following a first-time LPD, this study aims to assess the accuracy of these MRI findings. Specifically, we will compare MRI interpretations of the lateral meniscal signal against confirmatory arthroscopic imaging to determine the accuracy of the former in diagnosing meniscal injuries associated with first-time LPD. Although meniscal injury diagnosed by MRI has been studied extensively in the setting of LPD, from our understanding, there have been no investigations that arthroscopically confirm the presence of these reported meniscal injuries. We hypothesize that changes in the lateral meniscal MRI signal are not indicative of actual structural meniscal injuries but may be due to other pathological soft-tissue changes associated with the injury event.

2 Materials and methods

An Institutional Review Board approved retrospective review was performed using the Nuance mPower database to query all knee MRI impressions at a single imaging facility with keywords “lateral patellar dislocation” between the years 2012 and 2022. All cases were reviewed for evidence of LPD determined by the presence of effusion, disruption of the medial soft tissue stabilizers of the patella, bone edema on the lateral femoral condyle and medial patella, lateral patellar tilt or subluxation, and/or osteochondral injuries.

To assess radiographic evidence of meniscal injury, the menisci were evaluated for abnormal high signal intensity and graded according to the grading scale described by Lotysch et al. and later modified by Dillon et al.^24,25 Grade 1 signal was identified as a rounded or amorphous signal in the meniscus that did not disrupt an articular surface. Grade 2a signal was identified as a linear signal that did not disrupt an articular surface, 2b as an abnormal signal that contacted the articular surface in one image, and 2c as an abnormal wedge-shaped signal that did not disrupt the articular surface. Lastly, a grade 3 signal was identified as a signal that disrupted an articular surface in 2 or more frames. The imaging in the study was reviewed by one fellowship-trained musculoskeletal radiologist with over twenty years of experience and a musculoskeletal radiology fellow. Both were blinded to the timepoints of the imaging studies. Discrepancy between grading was resolved by consensus.

Each chart was then reviewed to extract the following variables: age, sex, laterality, approximate date of injury, date of MRI, date of surgery, and operative report. Surgeries included in the review took place at a large academic institution and were performed by two fellowship-trained orthopaedic sports medicine surgeons. In all surgeries, the operating surgeon used a probe to check for undersurface tear of the lateral meniscus. Radiologists were blinded to surgical arthroscopic imaging and operative reports. All quantitative data is presented as median (range). All categorical data is reported as a percentage (%).

3 Results

The results of our retrospective search yielded 355 cases for review. Of those cases, we excluded records without any clinic note at our institution,²⁴ records where the review of MRI was not consistent with LPD or duplicate chart was found (136), and records where patients did not have surgery at our institution or had concomitant anterior cruciate ligament (ACL), multi-ligament, or patellar tendon repairs (151), yielding 44 cases for our study (see Fig. 1).

Close

Fig. 1 Consort Diagram depicting the selection criteria, grading, and arthroscopic findings.

Export to PPT

Between 2017 and 2022, 44 cases of LPD with correlated arthroscopic imaging from one large academic institution were identified in 42 patients (2 patients had bilateral cases at asynchronous times). Table 1 highlights the sex distribution, laterality of the injury, as well as timing between the date of injury to the initial MRI, clinic visit, and timing between the imaging and surgery.

All cases demonstrated meniscal signals in the anterior horn of the lateral meniscus. Table 2 highlights our findings between the observed signal changes on MRI when compared to arthroscopic findings. There were zero cases with tear of the lateral meniscus upon review of the operative reports and arthroscopic images.

4 Discussion

The most important findings of the study are that in the setting of first-time LPD, T2 signal abnormalities in the lateral meniscus are not representative of structural changes when viewed arthroscopically. This is in line with our hypothesis that these changes as seen on MRI are not structural, but possibly a sequelae of the mechanism of injury on the surrounding soft tissue structures.

Multiple studies have concluded that the incidence of meniscal tear on MRI after LPD is between 8.2% and 29%, however, these were secondary outcomes in those studies, many do not distinguish between medial and lateral meniscal findings, none reported the grade of meniscal signal found, and none were able to confirm these tears via arthroscopy, the reference standard.^{15–19,21,23,25,26} In Guerrero et al., the authors found a 21% incidence of meniscal tear in their patient population but mentioned in their discussion the importance of additional arthroscopic study as their identified meniscal signal may not correlate arthroscopically with meniscal tear.¹⁶

MRI is thought to be a highly reliable method of determining the presence of meniscal tear, with published sensitivity as high as 96.4% in the medial meniscus and 82% in the lateral meniscus.^26,27 In their review of classifications of signal changes in the meniscus, Dillon et al. reviewed arthroscopic data and demonstrated that tears were present in 22/23 (96%) of the menisci where meniscal signal extended to the articular surface in two contiguous images (grade 3), and 4/5 (80%) of grade 2c menisci with a wedge-shaped area of the increased signal were considered abnormal.²⁵ Given these findings and the known accuracy of MRI, many studies in the literature have reported T2 signal in the meniscus as a meniscal tear without arthroscopic confirmation (see Fig. 2).

Importantly, there is one study by Sanders et al. that correlates soft tissue injury on MRI after LPD with arthroscopic data, however, this was in 14 patients and the purpose was to describe the accuracy of MRI in diagnosing the extent and location of the MPFL injury in order to direct surgical management.¹³ The present study focuses on diagnosing and grading MRI signals in the lateral meniscus and correlating this signal arthroscopically. We found that out of 44 cases of LPD in which 27 cases demonstrated grade 2a or above signal changes in the lateral meniscus, there were no meniscal tears when evaluated via arthroscopy. Specifically, we had 10 cases of grade 3 signal change in the lateral meniscus, where Dillon et al. arthroscopically confirmed a 96% accuracy in diagnosing meniscal tear, and in each case, there was no arthroscopic evidence of tear. In the three examples shown in Fig. 2, the linear signal in the lateral meniscus that extends into the articular surface mimics a tear, however, upon arthroscopic review, there were no visible tears. These findings suggest that the identified signal in the lateral meniscus in the setting of LPD is not due to an associated meniscal tear.

Close

Fig. 2 A-F. Six images of MRI and their corresponding arthroscopic imaging A. T2-weighted MRI of left knee, demonstrating grade 3 signal change in the anterior horn of the lateral meniscus. B. Paired intra-operative arthroscopic image of the left knee lateral meniscus in MRI shown in Fig. 2A. C. T2-weighted MRI of left knee, demonstrating grade 3 signal change in the anterior horn of the lateral meniscus. D. Paired intra-operative arthroscopic image of the left knee lateral meniscus in MRI shown in Fig. 2C. E. T2-weighted MRI of right knee, demonstrating grade 3 signal change in the anterior horn of the lateral meniscus. F. Paired intra-operative arthroscopic image of the right knee lateral meniscus in MRI shown in Fig. 2E.

Export to PPT

The concept of acute trauma to the knee causing meniscal injury that appears as a signal on MRI without a discrete tear has been discussed in the literature.^28,29 Cothran et al. described a study population of 6 patients, 5 of them with ACL tears and meniscal signal suggestive of a tear in the posterior horn of the medial meniscus in which no patients had meniscal tears on arthroscopy.²⁸ They found in one of the operative reports that their orthopaedic surgeon described an area of abnormal injection and vascularity in the periphery of the meniscus, however, the meniscal appearance was not described in the remainder of their reports.²⁸ Additionally, in 2 patients who presented for a follow-up MRI, the previously found meniscal signal had resolved.²⁸ This led the authors to propose the idea of a transient injury to the meniscus as a result of the compressive force it receives during the mechanism of trauma that caused an ACL tear, however, they note that in all 6 patients, the meniscal signal was more globular or amorphous than linear.²⁸ In contrast, as shown in Fig. 2, our cases demonstrate a linear signal, particularly in our grade 3 menisci in which there is contact with the articular surface.

This transient meniscal injury, or contusion as labeled by Cothran et al., could certainly be misinterpreted as a tear given that the signal may contact the articular surface.²⁸ The authors further discuss that a potential cause of meniscal signal on MRI may be due to a release of fluid and blood derivatives from a compressive injury to the vascular zone of the meniscus.²⁸ Although the inner third of the meniscus is relatively avascular with a surrounding perimeniscal capillary plexus, the periphery, and roots receive a ample blood supply through endoligamentous vessels that form capillary loops.³⁰ One proposed explanation for the findings in these studies is that a compressive injury to the meniscus causes increased permeability of these vessels, causing fluid to invaginate between the fibers of the meniscus and show T2 signal along the fibers, which could be interpreted as a tear. Our study was unable to review repeat MRI after injury to assess for resolution of this signal, and the long-term implications of a meniscal contusion have yet to be determined.

Further complicating the diagnostic landscape of meniscal injury are the frequent occurrences of signal alterations in the anterior horn of the lateral meniscus. Notably, Shepherd's 2002 study on a sample population of 947 asymptomatic patients indicated a 74% false positive rate for these findings, raising questions about their clinical relevance.³¹ The rate of incidental findings in this study may suggest that these alterations are not always indicative of an injury related to the lateral patellar dislocation but could represent a more benign variation or potentially physiologic variation. In our study, while we observed linear signal alterations, we cannot definitively correlate these findings to the lateral patellar dislocations without knowing more about the overall presence in a non-injured population. It is important to consider the incidence of these alterations in asymptomatic individuals, as highlighted by Shepherd et al., to avoid overdiagnosis and the consequent potential for unnecessary interventions.

Our presented data may help improve the interpretation accuracy of associated soft tissue injuries in preoperative evaluation after LPD. A recent Cochrane meta-analysis of randomized controlled trials stated that although there is some evidence of lower rates of short-term re-dislocation after surgical management,³² the quality of available evidence and long-term follow-up is insufficient to support a current change in practice.³³ Additionally, complication rates after surgical intervention have been reported as high as 26%, with little difference in long-term outcomes between surgical and non-surgical management.^34,35 The current consensus in the literature is for conservative management after first-time LPD unless there is an obvious osteochondral fracture or loose intra-articular fragment, recurrent instability after conservative management with bracing, physical therapy, and activity modification, or there is an alternate or associated injury that requires surgical intervention.^{1–4,33,36,37} Although the aim of this paper is not to discuss indications for surgery, it is important to note that incorrect diagnosis of meniscal tear as an associated injury after LPD may provide the impetus for surgeons and patients to consider surgical intervention more heavily when it would not otherwise have been indicated.

Potential limitations of this study include its retrospective nature and relatively small patient population after narrowing to cases with arthroscopic data and operative reports that could be accessed at our institution. Given our study population in the urban setting, it is likely that many patients pursued orthopaedic care outside our institution, which further limited our available data. Additionally, there are some patients in our study with a prolonged time to MRI and time to operative intervention, however there was no difference in outcome between these patients and those who received care much sooner. In our study, we were unable to evaluate for possible resolution of the lateral meniscal signal on repeat MRI after injury, which could further support the hypothesis of the meniscal contusion and is an area available for further investigation.

In summary, to our knowledge, this is the first study to provide grading of meniscal signal with arthroscopic correlation in patients with knee trauma resulting in LPD. In this retrospective review of 44 cases of LPD, we could not find a single case of lateral meniscal tear at arthroscopy, even with grade 2 and grade 3 signals in the lateral meniscus on MRI. Therefore, in the setting of LPD, unless there is MRI evidence of complete full-thickness rupture of the anterior horn of the lateral meniscus, all T2 signal abnormalities in the lateral meniscus likely do not indicate a meniscal tear. This may improve the accuracy of identification of associated soft-tissue injuries after LPD and assist surgeons in deciding which patients will require surgical intervention.

5 Conclusion

MRI findings of signal alterations in the lateral meniscus post-LPD may not indicate an actual tear. This could guide surgical decision-making in primary LPD management.

Ethical statement

The study protocol was reviewed and approved by the Institutional Review Board (IRB) of The University of Miami Miller School of Medicine (IRB approval number: 20220801).

Patient confidentiality and privacy were strictly maintained throughout the study. All patient data were anonymized and de-identified prior to analysis to ensure that no personal identifiers were accessible. Informed consent was waived by the IRB due to the retrospective nature of the study, as the research involved no more than minimal risk to the subjects, and the waiver did not adversely affect the rights and welfare of the subjects.

No additional interventions or procedures were performed on patients as part of this study. The data used were collected as part of routine clinical care and were stored securely in accordance with institutional data protection policies.

Funding statement

The authors would like to acknowledge that no external funding was received for the completion of this study. The research, data collection, analysis, and manuscript preparation were carried out solely through the authors' own resources and institutional support. No financial support or grants were provided by any funding agencies, commercial entities, or non-profit organizations for this work.

Consent statement

Due to the retrospective nature of the study, obtaining informed consent from patients or their guardians was not feasible. The IRB of The University of Miami Miller School of Medicine granted a waiver of informed consent as the research involved no more than minimal risk to the subjects, and the waiver did not adversely affect the rights and welfare of the subjects. However, we confirm that all necessary permissions and approvals were obtained for accessing and using patient data.

No additional interventions or procedures were performed on patients as part of this study. The data used were collected as part of routine clinical care and were stored securely in accordance with institutional data protection policies.

CRediT authorship contribution statement

Griffin Harris: Writing – original draft, investigation, formal analysis, Nikhil Patel: Writing – review & editing, Gilberto O. Lobaton: Writing – review and editing, Clementina Cabrera: Investigation, Daniel Quintero: Statistical Analysis, Michael Baraga: Supervision, Methodology, Jean Jose: Supervision, Investigation, Methodology, Conceptualization.