Research Presentation Session

RPS 1110 - Knee

Lectures

1
RPS 1110 - Distal femoral cortical irregularity (DFCI): increased prevalence in competitive young alpine skiers

RPS 1110 - Distal femoral cortical irregularity (DFCI): increased prevalence in competitive young alpine skiers

05:51C. Stern, Chur / CH

Purpose:

Tumour-like cortical irregularities at the posterior distal femur are common incidental findings in adolescents, but their origin is still debated. The purpose was to investigate the prevalence of irregularities at the tendinous attachment of the medial/lateral head of the gastrocnemius (MHG/LHG) and of the adductor magnus (AM) muscle in competitive young alpine skiers compared to young adults of the same age.

Methods and materials:

Knee MRI examinations of 105 competitive young alpine skiers (age 13-16 years) and a control group (matched for gender, age, and extremity) of 105 adolescent patients were evaluated by two radiologists for the presence of cortical lesions at the femoral attachment site of the MHG, LHG, and AM tendon. Lesion diameters were measured on sagittal fat-suppressed MR images. Statistical analysis included Person's Chi-square, Student’s t-test, and kappa statistics.

Results:

Cortical lesions were found in 58.1% of competitive alpine skiers (61/105) compared to 26.7% in the control group (28/105)(p<0.001). In 2 skiers, more than one lesion per individual was found versus none for the control. Lesion distribution for competitive athletes and the control group were 60/63 (95%) and 26/28 (93%) lesions at the MHG, 3/63 (5%) and 1/28 (3.5%) at the LHG, and 0/63 (0%) and 1/28 (3.5%) at the AM attachment site, respectively. Inter-reader agreement was almost perfect (k=0.87). No substantial difference was observed for the average size of MHG lesions in athletes (3.7 mm) versus the control group (3.4 mm) (p=0.321).

Conclusion:

A cortical irregularity of the distal femur at the tendon attachment sites is a frequent incidental finding on MRI, with a significantly increased prevalence in competitive young alpine skiers, supporting the theory of a stress-related origin. These irregularities should not be mistaken for malignancy.

Limitations:

A retrospective control group.

Ethics committee approval

Ethics committee approval obtained.

Funding:

No funding was received for this work.

2
RPS 1110 - Evaluation of medial meniscal extrusion using weight-bearing ultrasound: correlation with MRI and meniscal tears

RPS 1110 - Evaluation of medial meniscal extrusion using weight-bearing ultrasound: correlation with MRI and meniscal tears

05:17A. Falkowski, Zürich / CH

Purpose:

To compare medial meniscal extrusion as seen on weight-bearing ultrasound compared with MRI and meniscal tears.

Methods and materials:

Patients obtaining a routine knee MRI were prospectively evaluated with supine and weight-bearing ultrasound (US) of the medial meniscus. The position of the outer boundary of the medial meniscus on US images and MRI was measured relative to the tibia by a fellowship-trained musculoskeletal radiologist. A correlation was made to the presence or absence of a meniscal tear and statistical significance was calculated via a Student t-test.

Results:

50 knees from 49 subjects (23 male, 26 female; mean age 44±15 years) were included (18 right, 32 left; one bilateral). The mean medial meniscal extrusion on supine US was 1.3 mm (range -1.5-3.6 mm), with no significant difference compared with MRI (p=0.21), which increased to 2.1 mm on weight-bearing US. In the 38% (19/50) of subjects with meniscal tears on MRI, the mean medial meniscal extrusion on weight-bearing US was 2.13 mm (range 0-4.4 mm) with a change between supine and weight-bearing of 0.63 mm (range -1.8-2.7 mm). In the 62% (31/50) of subjects with no meniscal tear, the mean medial meniscal extrusion on weight-bearing US was 2.05 mm (range 0-3.8 mm) with a change between supine and weight-bearing of 0.87 mm (range -0.1-2.2 mm) and no significant difference between subjects with and without tear (p=0.805 and p=0.413).

Conclusion:

Supine US was comparable with supine MRI for the assessment of medial meniscal extrusion. The presence of a meniscal tear did not result in increased medial meniscal extrusion on weight-bearing US compared with no meniscal tear.

Limitations:

The reading of the images was performed by one reader.

Ethics committee approval

IRB-approved study with informed consent.

Funding:

No funding was received for this work.

3
RPS 1110 - The assessment of medial meniscus extrusion on ultrasound using MRI as a reference standard

RPS 1110 - The assessment of medial meniscus extrusion on ultrasound using MRI as a reference standard

05:40R. Zeitoun, Cairo / EG

Purpose:

To assess medial meniscus extrusion using US on supine and standing (weight-bearing) positions as a sign of the presence of underlying meniscal tears. The study was done using MRI as a reference standard for the detection of tears.

Methods and materials:

This observational analytic prospective study included 103 patients (48 females and 55 males), aged between 18-70 years. The US measured values of the medial meniscal extrusion on supine and standing positions were correlated to the MRI diagnosis of tears. The included patients were classified into two groups according to the presence or abscence of meniscal tears on MRI. Statistical analysis was performed for the obtained measurements for both groups and P<0.05 was considered significant. ROC curve was done to reach a cut-off value for extrusion as a sign of medial meniscus tear.

Results:

45 patients (43.68%) had medial meniscal tears (horizontal, vertical, branching, radial, root, and bucket-handle tears). The measured extrusion on supine and standing positions, as well as the difference between them in the two groups, all showed a significant P=0.001. The extrusion on standing position with a cutoff value of >3.3 mm showed a specificity (96.55%), positive predictive value (93.8%), and accuracy (83.5%) for the detection of a medial meniscus tear.

Conclusion:

Measuring medial meniscus extrusion using ultrasonography, especially in a standing position, is a good sign to predict tears.

Limitations:

The sample size and the number of patients with different subtypes of tears were not large enough for a powerful conclusion to correlate the value of extrusion and the subtype of tear.

Ethics committee approval

The study has been approved by the Ethical Committee of Faculty of Medicine, Cairo University, in compliance with the Helsinki declaration.

Funding:

No funding was received for this work.

4
RPS 1110 - The correlation between the anatomical variations of the knee joint and fat-pad pathologies and patellar tilt

RPS 1110 - The correlation between the anatomical variations of the knee joint and fat-pad pathologies and patellar tilt

06:10A. Cilengir, İzmir / TR

Purpose:

The correlation between patellar tilt, oedema in the superolateral portion of the infrapatellar fat pad, and chondromalacia was described in the literature. However, there is not enough information about other variations that may cause knee joint pathologies. A correlation with quadriceps patellar tendon angle (QPA) was not described. Our aim was to describe the correlation between knee joint variations and analyse accompanying fat-pad, cartilage, and tendon pathologies.

Methods and materials:

A retrospective analysis of age, gender, side, femoral sulcus angle (SA), tibial tubercule-trochlear groove distance (TTTG), patellar tendon length (PTL), patellar height (PH), Insall-Salvati ratio (ISR), lateral trochlear inclination angle (LTI), trochlear facet asymmetry (TFA), trochlear depth (TD), trochlear dysplasia (TDis), suprapatellar-infrapatellar-prefemoral fat-pad oedema (SPE-IPE-PFE), suprapatellar effusion (SE), patellar and femoral chondromalacia (PC and FC), and quadriceps and patellar tendinopathy (QT and PT) on 406 knee MRIs with patellar tilt (PTi) was performed. The relationship between PTi, QPA, and other parameters were evaluated. 40 knee MRIs without PTi were selected as a control group.

Results:

The presence of SPE-IPE-PFE, SE, PC, and FC was more frequent in the PTi group; SA, TT-TG, PTL, ISR, LTI, TFA, and TD values were significantly different between the two groups. The distribution of PTi was significantly related to TDis, PC, QT, SA, TT-TG, LTI, and TD in the PTi group. The distribution of QPA was significantly associated with TDis, IPE, PFE, QT, SA, and LTI in the PTi group. We calculated 10 degrees of PTi as a cut-off value for a superolateral portion of the infrapatellar fat-pad oedema.

Conclusion:
Many anatomical variations of the knee joint, including QPA, are associated with fat-pad, cartilage, and tendon pathologies.
Limitations:

The restrospective design.

Ethics committee approval

This study has approval from an institutional review board.

Funding:

No funding was received for this work.

5
RPS 1110 - Attachment type of the posterior meniscofemoral ligament and clinical significance

RPS 1110 - Attachment type of the posterior meniscofemoral ligament and clinical significance

06:44S. Ham, Seoul / KR

Purpose:

To investigate the correlation between the attachment type of the posterior meniscofemoral ligament (pMFL) and the incidence of lateral meniscal tears.

Methods and materials:

We retrospectively evaluated 191 patients who underwent knee MRIs. We assessed the attachment type of the pMFL (high vs low) and measured the thickness of the pMFL. Then we evaluated the presence of a meniscal tear or discoid meniscus. We assessed if there was a significant relationship between the mean thickness of the pMFL and the incidence of meniscal tears or a discoid meniscus according to the type of pMFL using independent t-tests. We also assessed the significance of the relationship between the mean thickness of the pMFL and the presence or absence of a meniscal tear or discoid meniscus using Pearson’s Chi-square test.

Results:

High type pMFLs were significantly thicker than low type pMFLs (p<0.001). There was no significant difference in the incidence of meniscal tears according to the type of pMFL. There was also no significant difference in pMFL thickness between patients with meniscal tears and those without meniscal tears.

Conclusion:

High type pMFLs tend to be thicker than low type pMFLs, but the thickness is irrelevant to the incidence of meniscus injury.

Limitations:

We could not completely exclude the possibility of selection bias and confounding factors because this was a retrospective study. We only included symptomatic patients who had knee MRI data, which could also contribute to a selection bias. We did not correlate pMFL results to the clinical outcomes.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

6
RPS 1110 - Volume and quantitative dynamic contrast-enhanced MR blood perfusion parameters of the infrapatellar fat-pad and their relationship with oedema and effusion in patients with patellofemoral pain

RPS 1110 - Volume and quantitative dynamic contrast-enhanced MR blood perfusion parameters of the infrapatellar fat-pad and their relationship with oedema and effusion in patients with patellofemoral pain

06:02R. van der Heijden, Rotterdam / NL

Purpose:

The infrapatellar fat-pad (IPFP) is implied as a source of pain in knee osteoarthritis (OA) and patellofemoral pain (PFP), a supposed precursor of OA. Inflammation or increased IPFP volume may have causative effects. The aim of this study was to quantitatively evaluate dynamic contrast-enhanced (DCE) MRI parameters as a surrogate measure of inflammation and volume of the IPFP, and explore their relationship with IPFP oedema and joint effusion in PFP patients.

Methods and materials:

PFP patients and healthy control subjects underwent 3T MRI comprising non-fat saturated FSPGR and DCE-MRI. Motion was corrected by image registration. The IPFP was delineated on the FSPGR sequence using Horos software. Volume was calculated and quantitative perfusion parameters (Ktrans, Kep, Ve, and Vp) were extracted by fitting the extended Tofts’ pharmacokinetic model. Differences in volume and DCE-MRI parameters between patients and controls were tested by linear regression analyses adjusted for confounders of sex, age, BMI, and sports participation. The same applied to IPFP oedema and effusion.

Results:

43 controls and 35 PFP patients were included. Volume and perfusion parameters were not statistically significantly different between groups. Knees with effusion showed a higher perfusion of the IPFP.

Conclusion:

The IPFP has been implied as a possible source of knee pain, but its blood perfusion as a surrogate measure of inflammation and volume do not seem to play a role in patellofemoral pain. PFP patients with effusion showed higher perfusion, indicating inflammation, in contrast to controls.

Limitations:

Due to the large intersubject variability, small differences in perfusion or volume could have been undetected. But given the small effect sizes, these would not be clinically relevant.

Ethics committee approval

Approved by the institutional review board and written informed consent obtained.

Funding:

Partly funded by ESSR, RSNA, Erasmus University, and Dutch Artritis Foundation.

7
RPS 1110 - The importance of being minocycline chlorhydrate. Sclerosant acting antibiotic versus corticosteroids to treat symptomatic Baker cysts: a prospective study

RPS 1110 - The importance of being minocycline chlorhydrate. Sclerosant acting antibiotic versus corticosteroids to treat symptomatic Baker cysts: a prospective study

05:50I. Percivale, Novara / IT

Purpose:

A prospective study about the efficiency and safety on the use of a sclerosant antibiotic (minocycline chlorhydrate) versus steroids (triamcinolone acetonide) to treat patients with symptomatic Baker cysts.

Methods and materials:

We prospectively enrolled 66 patients, randomised into two cohorts; cyst drainage followed by antibiotic injection and cyst drainage followed by steroids injection. 58 patients underwent 6 month clinical and radiologic follow-up. The primary end-point was a volume reduction of more than 40% with respect to the cyst native volume. Secondary end-points were duration, pain reduction (VAS), and cyst permanence (volume reduction less than 40%).

Results:

The primary end-point has been reached in the 66.7 % of patients treated with antibiotics (mean volume reduction: 54%) and in the 35.7% of patients treated with steroids (mean volume reduction: 69%). Cyst permanence was seen in the 65% of cases treated with steroids and in the 34% of cases treated with antibiotics. We achieved a mean pain reduction (VAS) of 3.2 points. The procedural duration was higher in patients treated with antibiotics (15.5 min) compared to corticosteroids (8.6 min).

Conclusion:

Both treatments are safe (no major or minor adverse events) but antibiotic injection is more efficient considering the primary end-point. Despite an increased procedural time, patients with antibiotic treatment have a lower rate of cyst permanence.

Limitations:

The limited number of patients and short follow-up.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

8
RPS 1110 - Collective intelligence has increased diagnostic performance compared to expert radiologists in the evaluation of knee MRI

RPS 1110 - Collective intelligence has increased diagnostic performance compared to expert radiologists in the evaluation of knee MRI

06:46C. Messina, Milan / IT

Purpose:

As is widely known, the traditional diagnostic approach, which involves a single physician, can result in worrying error rates. Collective intelligence has recently been considered as a possible strategy to pool individual physicians' diagnoses and reduce misdiagnosis. Our aim is to present an investigation about the accuracy of these techniques on a real-world experiment of MR assessment.

Methods and materials:

We asked 13 radiologists from a tertiary orthopaedic institution to identify abnormal exams among 417 knee MR images from the Standford MRNet dataset1 (which is a low-resolution imaging dataset used to train machine-learning predictive models). Radiologists were also asked to assess their confidence on their decisions and the perceived complexity of each case. We assessed each radiologist's accuracy in comparison with the accuracy rates obtained by leveraging collective intelligence techniques.

Results:

The radiologists' panel obtained an average accuracy of 0.82-0.01 (95% confidence interval). Taking the majority annotation for each exam as the gold standard, we obtained an accuracy of 0.86. Weighing the annotations of the radiologists' by their confidence resulted in an accuracy of 0.87; considering only the most accurate raters, or selecting the most surprisingly popular annotation, we obtained an accuracy of 0.88, in either case. All collective intelligence techniques increased diagnostic accuracy in the cases for which the radiologists exhibited a greater misdiagnoses rate, on low-confidence and high-complexity cases.

Conclusion:

Collective intelligence techniques were associated with increased diagnostic accuracy compared with the average of the individual radiologists and the most accurate ones. These promising results suggest that these techniques merit further consideration for applications in the ground-truthing of machine-learning models and, potentially, in clinical settings to reduce diagnostic error.

Limitations:

n/a

Ethics committee approval

n/a

Funding:

No funding was received for this work.

9
RPS 1110 - Agreement between cartilage morphology on MRI and weight-bearing CT and radiographs respectively for visualising patellofemoral OA features in the MOST study

RPS 1110 - Agreement between cartilage morphology on MRI and weight-bearing CT and radiographs respectively for visualising patellofemoral OA features in the MOST study

06:50N. Segal, Kansas City / US

Purpose:

The patellofemoral joint is frequently affected by osteoarthritis (PFOA) and is incompletely imaged on radiographs. Weight-bearing CT (WBCT) could offer advantages for visualisation. This study determined the agreement between the detection of cartilage damage on MRI and PFOA on WBCT and radiographs.

Methods and materials:

The right knees of 60 people who did not have findings of PFOA at a prior clinic visit were randomly selected. WBCT and radiographs were read for an OARSI JSN score (0=none, 1=minimal, 2=moderate, and 3=total joint-space loss), and MRI was read for a MOAKS cartilage score (size and % of full-thickness) by 2 musculoskeletal radiologists, each with more than 10 years experience, who were blinded to the subject. Weighted kappa coefficients were calculated for agreement between modalities. Inter-reader reliability was assessed with weighted kappa statistics.

Results:

The mean±SD age and BMI for the 60 participants (66.7% women) were 67.6±9.8 years and 30.0±5.3 kg/m2, respectively. There was a fair-to-moderate agreement between PFOA findings on WBCT and semi-quantitative scores of PF cartilage on MRI, while an agreement was none-to-slight between radiographic PFOA readings and MRI readings. Inter-rater reliability for the WBCT JSW {kappa=0.60 (0.48, 0.72)} and MRI MOAKS-CM {kappa=0.70 (0.61, 0.79)} readings were good.

Conclusion:

PFOA scoring on WBCT agrees with MRI substantially better than does scoring on XR. At the same relative radiation level as radiographs (<0.10 mSv), WBCT imaging holds the potential to improve understanding of the weight-bearing patellofemoral joints beyond what is possible with radiographs.

Limitations:

Radiologists were provided with written instructions, but no in-person training in the scoring systems used and no consensus readings were completed, potentially attenuating inter-rater agreement.

Ethics committee approval

Following an IRB-approved informed consent process, all participants provided written informed consent.

Funding:

NIH R01AR071648, U01AG18832, and U01AG19069.

PEP Subscription Required

This course is only accessible for ESR Premium Education Package subscribers.