CT and MRI as Diagnostic and Management Decision Tools for First Time Lateral Patellar Dislocations: A Cross-Sectional, Retrospective Study

Abstract

Background

Following first-time lateral patellar dislocation (FTLPD), most patients are treated conservatively, although 50% of patients will have recurrent dislocations. Typically, radiographs followed by CT and/or MRI are used to assist the clinician in determining treatment strategy and, combined with clinical findings, intraarticular free bodies (CT/MRI), significant medial patellofemoral ligament (MPFL) tear (MRI) and lateral displacement of the patella (CT) form relative indications for surgery.

Methods

Radiographs, MRI and CT knee studies of 34 patients after lateral patellar dislocation (26 FTLPD) were evaluated for intraarticular free bodies, patellar/trochlear fracture, lateral femoral condyle compression, MPFL tear, tibial tuberosity-trochlear groove (TT-TG) distance, and surgery indications. Free bodies and fractures were also evaluated on knee radiographs. FTLPD was analyzed as a subgroup. Surgical indications were compared between imaging modalities.

Results

Among FTLPD (26 patients); free bodies were identified in 13 and 19 patients using MRI and CT respectively, compared with 5 patients on radiographs; this was statistically significant. In 8 cases surgery was indicated based on MPFL tear (MRI) combined with lateral patellar displacement (CT). When MRI and CT results were combined, 21 of 26 patients had imaging indications for surgery compared to 13 and 19 patients based on the MRI or CT alone, respectively.

CT was statistically better than MRI alone or MRI with radiographs in identifying patients requiring surgery.

Conclusion

An MRI or CT study is warranted to determine the need for surgery. A second imaging study (different from the first) should be considered, if surgical indication was not established from the initial study or clinical presentation.

Background

Lateral patellar dislocation constitutes 2%–3% of the intraarticular traumatic knees injuries and is the second leading cause of intraarticular hemorrhage (9%–16%) in young people [1–3].

Most of the times the patella does not stay dislocated and active patellar reduction is required in only 20% of cases [4]. Pain and swelling may mask the correct diagnosis, and 50–75% of cases may be missed [1, 5]. Therefore, imaging has a major role in diagnosis and in severity assessment.

In the long term, acute patellar dislocation can lead to instability of the joint with recurrent dislocations, pain, reduced function and degenerative changes. When dislocation occurs, a tear of the medial supporting soft tissues, mostly the medial patellofemoral ligament (MPFL) is observed [1, 2]. In addition, due to patellar/trochlear impact, fracture of both structures may occur and create free intraarticular fragments.

There are known risk factors for patellar instability and patellar dislocation. A large Tibial Tuberosity-Trochlear Groove (TT-TG) distance (more than 15–20 mm) is considered a risk factor for patellar instability and currently computed tomography (CT) is the gold standard for this measurement [6, 7]. Other risk factors include age, sex, family history and specific imaging measurements including trochlear dysplasia, patellar height and TT-TG/patellar length ratio [8–11].

First time patellar dislocations can be treated conservatively or surgically. Indications or relative indications for surgery based on imaging studies following first dislocation include: presence of intraarticular free fragments which need to be repositioned to the affected bone or be removed and/or a massive MPFL tear with lateral patellar displacement (compared to the contralateral knee) [3, 12]. Recurrent patellar dislocations are also an indication for surgery [3, 7, 13, 14]. Following knee radiographs, usually performed in the acute setting, computed tomography (CT) or magnetic resonance imaging (MRI) are the two imaging modalities used to diagnose and characterize patellar dislocation.

Currently there is no consensus regarding the preferable imaging modality (CT or MRI), the benefit of either modality compared to knee radiographs, or the additive value of both modalities together. A retrospective study evaluating CT and MRI studies of patients following lateral patellar dislocation was performed to evaluate the additive value of CT and MRI in patellar dislocations for characterization of the injury and to aid in surgical decision-making. To the best of our knowledge no previous data has been published on this subject.

Methods

Institutional review board approval for the evaluation of imaging studies and medical charts of subjects following patellar dislocation between the years 2012–2017 was obtained. Patients who had patellar dislocation imaging studies including both CT and MRI with a maximum time interval of one month between the studies were included in the study. Patients that did not meet these criteria, had extensive knee injury, prior knee surgery, or another known pathology (i.e., tumor, infection etc.) were excluded.

Radiograph Evaluation

AP and lateral knee radiographs of the affected knee performed minimum one week or less following patellar dislocation were evaluated for the presence of intraarticular fragments or fracture. The absence of knee radiographs was not considered an exclusion criterion. Radiographs were evaluated by a sports medicine fellowship trained orthopedic surgeon.

CT and MRI Evaluation

MRI examinations were all performed on two 3.0 T MRI units (Ingenia, Philips Medical Systems, Eindhoven, The Netherlands). The MRI protocol included the following sequences and sequence parameters:

Coronal T1-w (TR = 620 ms, TE = 12 ms, FOV = 150) and PD with fat suppression (TR = 5200 ms, TE = 32 ms, FOV = 150). Sagittal PD with and without fat suppression (TR = 5200 ms, TE = 32 ms, FOV = 150). Axial T2-w with fat suppression, (TR = 4000 ms, TE = 60 ms, FOV = 140).

All CT examinations were carried out on two 64-slices CT scanners with a slice thickness of 0.6 mm (ICT 956, Brilliance, Philips Medical Systems, Eindhoven, The Netherlands; VCT LightSpeed, GE Healthcare, Milwaukee, WI, USA). Images were reformatted in multiple planes and evaluated in a bone and soft tissues algorithm.

All CT and MRI studies were separately evaluated by a single senior musculoskeletal radiologist, with at least a 2 week time interval between evaluations. In each study, the following parameters were recorded: free intraarticular fragments; fragments’ location, origin (femur or patella) and size; patellar fracture and location (medial/lateral facet, central fracture); femoral fracture; cartilage/bone and cartilage lateral femoral condyle (LFC) compression fracture and TT-TG distance. Bone fragments which are part of the MPFL avulsion fracture mechanism (located on the medial side of the patella) were not registered as intraarticular fragments. The prevalence of each parameter was compared between the imaging modalities.

MPFL tear (patellar or femoral attachments) and its grade (minimal, massive, etc.) were evaluated on MRI study. A medial patella femoral ligament tear was considered minimal if less than 50% of the MPFL insertion area to the patella or the femur was detached; moderate if the tear consisted of about 50%; and massive if most of the MPFL insertion area to the patella or the femur was detached or was detached entirely (Fig. 1).

Fig. 1.

MRI of a 14 y/o female after first time patellar dislocation with massive tear of the MPFL femoral insertion

The position of the patella on CT axial scan was compared to the contralateral knee. Anterior condylar line was established first, and then perpendicular line was drawn up to the lateral border of the patella. Another perpendicular line was drawn from the highest part of the medial femoral condyle and the distance between the lines was measured [15]. (Fig. 2).

Fig. 2.

CT of a 14 y/o female after first time lateral patellar dislocation of the right knee. The patellar on the right side is laterally subluxated compared to the left knee

Because the goal of our study was to compare CT and MRI, additional risk factors for patellar dislocation including patellar height and trochlear dysplasia were not evaluated; previous studies have shown no difference between CT and MRI in evaluating these parameters [13, 16].

The subgroup of first-time patellar dislocation was evaluated for surgical treatment indications based on radiographic findings: Intraarticular free fragments which need to be repositioned to the affected bone or removed on CT, MRI (size ≥ 5 mm) or radiographs (Fig. 3 + 4), or massive MPFL tear on MRI combined with lateral displacement of the patella on CT compared to the healthy contralateral knee [3, 12]. These imaging surgical indications were compared between CT versus MRI, and between CT versus MRI combined with radiographs (Fig. 4).

Fig. 3.

CT of a 19 y/o male after first time lateral patellar dislocation with osteochondral fracture of the patellar dome and medial facet

Fig. 4.

15 y/o female after first time patellar dislocation with large free fragment in the lateral gutter, Osteochondral fracture of the medial facet of the patella

Statistical Analysis

Analysis was performed for the entire study cohort and for the subgroup of patients with a first incidence of patellar dislocation. Student t-test was used to compare the CT versus MRI findings and CT versus MRI combined with radiographs. A p-value below 0.05 was considered statistically significant.

Results

One-hundred and sixty-three cases of patients with lateral patellar dislocation were surveyed. Thirty-eight patients had both CT and MRI studies performed one month or less apart. Four patients were excluded due to extensive knee injuries. 34 patients met the inclusion criteria, 21 males and 13 females, with an average age of 25.2 ± 10 years (range from 12 to 45), and average time interval between imaging studies of 6.2 ± 7.1 days. Out of those, 26 patients had first time dislocation, 17 males and 9 females, with an average age of 22.6 ± 9.7 years (range from 14 to 42) and average time interval between imaging studies of 5.8 ± 6.7 days.

Imaging Surgical Indications for First Time Dislocations (N = 26)

Nineteen and 13 patients were indicated for surgery due to intraarticular fragment on CT and MRI (≥ 5 mm), respectively (p = 0.01). All 13 fragments identified on MRI were also identified on CT.

Nineteen (73%) out the first time dislocators cohort had knee radiographs. Intraarticular fragments were detected in 5 radiographs (26%). Those 5 cases were detected on CT, however, 2 out of the 5 were not detected on MRI. Significantly more free fragments were detected on CT than with the combination of MRI and radiographs (CT = 19, MRI and radiographs = 15, p = 0.04).

Eight patients were diagnosed with a massive MPFL tear on MRI and the same 8 were identified as having lateral patellar displacement on CT.

Out of the first dislocators cohort, 19 (73%) subjects had an indication for surgery based on CT and 13 (50%) based on MRI (p = 0.01). Combining both imaging modalities, 21 (81%) subjects had an indication for surgery.

By performing an MRI study following an initial CT scan for patients who did not have an indication for surgery based on the CT alone, 2 (10%) additional patients were diagnosed as candidates for surgery (p = 0.16). Both patients were indicated for surgery due to massive MPFL tear and lateral displacement of the patella compared to the contralateral side.

By performing a CT scan following an initial MRI, 8 (38%) additional patients were diagnosed as candidates for surgery (p = 0.002). Six of the 8 patients were indicated for surgery based on large free bony fragments and 2 patients were indicated based on the combined findings of massive MPFL tear with lateral displacement of the patella compared to the contralateral side.

Adding a CT examination following an initial radiograph and MRI, 4 (27%) additional patients were diagnosed as candidates for surgery (P = 0.04). Two of the patients were indicated for surgery based on large free bony fragments and 2 patients were indicated based on the combined findings of massive MPFL tear with lateral displacement of the patella compared to the contralateral side.

Post Patellar Dislocation Findings on MRI and CT for the Entire Cohort (34 Patients)

Data regarding intraarticular fragments incidence and characteristics on CT and MRI for all 34 patients (first and recurrent dislocation) is presented in Table 1.

Table 1.

Free intraarticular fragments detection on CT and MRI

		CT	MRI	p All	p 1st dislocation
				(N = 34)	(N = 26)
Fragment origin: patella	No fragments	13	9
	One fragments	9	21
	Two or more	12	4
	Total	21	25	0.324	0.252
Size (largest fragment)		3.07 mm (41.4±)	2.88 mm (5.5±)	0.84	0.65
Fragment origin: femur	No fragments	17	22
	One fragments	8	9
	Two or more	9	3
	Total	17	12	0.023*	0.412
Size (largest fragment)		5.98 mm (92.8±)	5.46 mm (32.9±)	0.55	0.6
Any fragment (patella or femur)	Total	30	27	0.26	0.42

CT computed tomography, MRI magnetic resonance imaging

Overall, no statistically significant difference between the prevalence and/or measured fragment size between CT and MRI was detected when taking into account fragments of any dimension originating from both patella and femur for the entire cohort and for first time dislocations. However, significantly more femoral originating fragments were detected on CT compared to MRI for the entire cohort (17, 12, p = 0.023). Similarly, more fragments were detected on CT compared to MRI in first time dislocations but here, without statistical significance. Five (17%) fragments detected on CT were not seen on MRI and 2 (7%) cartilaginous fragments detected on MRI were not seen on CT. MRI detected more fragments originating from the patella.

Compression of the lateral femoral condyle was detected significantly more on MRI than on CT (Entire cohort; CT = 5, MRI = 17, p = 0.0006, first dislocations; CT = 4, MRI = 11, p = 0.03). There was no statistically significant difference regarding patellar fracture (Entire cohort; CT = 25, MRI = 30) and trochlear fracture (Entire cohort; CT = 17, MRI = 15).

The average TT-TG distance measured on CT (16.2 mm, SD = 4.3) was significantly greater than on MRI based measurement (13.1, SD = 3.9, p = 0.002).

A single patient had a fracture in the lateral tibial plateau on both CT and MRI. One other patient had a fracture in the posterior lateral femoral condyle on both studies. Lateral collateral ligament tear was identified in that patient on MRI. No other injuries were identified for other structures (menisci, cruciate ligaments, etc.).

Discussion

In the current study, we aimed to evaluate the contribution of MRI and CT studies to the surgical decision-making algorithm in patients with first time patellar dislocation. We also explored the contribution of each imaging modality in the detection of post lateral patellar dislocations findings such as free fragments and fractures.

The inherent imaging characteristic advantages of each imaging modality dictated the results of the additive value of MRI and CT in imaging the post dislocated patellar knee. Intraarticular free bodies originating from the femur were detected significantly more on CT compared to MRI. On MRI, the fragments are usually edematous with high signal intensity on fluid sensitive sequences and are thus difficult to distinguish from the high signal intensity joint fluid. This is not the case with CT studies where acute fracture fragments are distinctive and more apparent.

Conversely, significantly more lateral femoral condyle compression injuries were detected by MRI. This of course resulted by the superb tissue contrast resolution of MRI enabling the detection of the edematous compressed bone. Detection of lateral femoral condyle compression fractures is a significant finding in the diagnosis of lateral patellar dislocation [2, 5]. Similarly, medial patellofemoral ligament tears can only be reliably detected on MRI and not on CT. Once the diagnosis of lateral patellar dislocation is established, the presence of free fragments and their origin guides surgical decision-making in both recurrent and lateral patellar dislocation.

When combining the CT and MRI indications for surgery, 81% of first time lateral patellar dislocation patients were allocated to surgery, while 38% of them would have been missed by MRI alone and 10% by CT scan alone. MRI is considered by many the imaging modality of choice for knee evaluation after patellar dislocation [1, 2, 4, 22, 23]. It is however clear from our results that large free intraarticular fragments can be easily missed on MRI leading to erroneous conservative versus surgical management. Moreover, the addition of knee radiographs to the MRI evaluation may still lead to incorrect management in 23% of our cohort.

However, detection of MPFL massive tears allocated an additional 8% of patients to surgery that would have otherwise been missed by CT alone. Therefore, MRI may have value in patellar dislocation management for cases where surgical indication is not clearly identified with CT.

We propose here a brief of the management algorithm that is customary at our institute. When a patient arrives to the hospital with a LPD, initial clinical evaluation and X-ray imaging are conducted. If no fragments can be seen on the X-ray, we do a CT scan. If an osteochondral fracture is observed and there is an indication for surgery (such as large free fragment) the patient is admitted, and we also do an MRI in search for other pathologies (such as massive MPFL tear and additional findings like meniscus or ACL tear) and to aid us to plan the required surgery. If there is no indication for surgery based on the CT scan, the patient can continue an ambulatory management. If a patient with a recent LPD already did an MRI scan and there is no indication for surgery based on the MRI findings, we add a CT scan in search for free fragments and a decision for surgery is made.

A patients arrives to the ER:

A patients with LPD with an MRI scan:

CT is considered the gold standard for measuring TT-TG distance [6] and a distance larger than 15–20 mm is considered a risk factor for patellar instability [6, 7]. In the current study the average TT-TG distance measured with CT was significantly larger than with MRI (16.2 mm, 13.1 mm respectively, p = 0.002). While one study reported similar findings [17], other studies shown no difference between modalities [6], potentially resulting from the highly variable TT-TG measurements in different patient positioning and flexion/extension angles [18, 19].

There is paucity of data in the literature regarding concomitant injuries such as menisci and ligament injuries (cruciate ligaments, collateral ligaments) with lateral patellar dislocation.

Studies have reported these concomitant injuries to occur in 4–11% of lateral patellar dislocation cases [20, 21]. In the current study, only one patient (3%) was diagnosed with a concomitant injury (lateral collateral ligament tear).

The current study is not without limitations. First, it is a retrospective study with a relatively small cohort. However, even with the limited size, our data shows that CT has statistically significant advantages over the MRI in detecting large free fragments. Second, no prognostic data were evaluated. The authors are well aware that surgical indications for first time lateral patellar dislocation patients are based on a combination of imaging and clinical findings, nevertheless, the aim of the current study was to compare between CT and MRI regarding the contribution of each imaging modality to surgical decision-making algorithm. Third, the current study cohort average age was 25 years with a majority of males (62%), while the reported patellar dislocation patients are generally younger (10–17 years) with female predominance [1, 4, 13]. It is possible that the inclusion criteria of both CT and MRI lead to a selection bias.

Conclusions

MRI and CT studies each have distinct advantages in the evaluation of patients with patellar dislocation. MRI is more reliable in diagnosing traumatic soft tissue injuries, bone compression and cartilaginous intraarticular fragments, while bony fragments originating from the femur are more frequently detected on CT. Therefore, the use of both MRI and CT is recommended as part of the evaluation and decision-making in patients after patellar dislocation.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability statement

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Informed consent

Informed consent was not obtained as this it a retrospective study. Data was gathered from medical records and anonymity was kept during the whole work process. We received an IRB approval.       Data availability statement: The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.