Knee landmarks detection via deep learning for automatic imaging evaluation of trochlear dysplasia and patellar height.

OBJECTIVES: To develop and validate a deep learning-based approach to automatically measure the patellofemoral instability (PFI) indices related to patellar height and trochlear dysplasia in knee magnetic resonance imaging (MRI) scans. METHODS: A total of 763 knee MRI slices from 95 patients were included in the study, and 3393 anatomical landmarks were annotated for measuring sulcus angle (SA), trochlear facet asymmetry (TFA), trochlear groove depth (TGD) and lateral trochlear inclination (LTI) to assess trochlear dysplasia, and Insall-Salvati index (ISI), modified Insall-Salvati index (MISI), Caton Deschamps index (CDI) and patellotrochlear index (PTI) to assess patellar height. A U-Net based network was implemented to predict the landmarks' locations. The successful detection rate (SDR) and the mean absolute error (MAE) evaluation metrics were used to evaluate the performance of the network. The intraclass correlation coefficient (ICC) was also used to evaluate the reliability of the proposed framework to measure the mentioned PFI indices. RESULTS: The developed models achieved good accuracy in predicting the landmarks' locations, with a maximum value for the MAE of 1.38 ± 0.76 mm. The results show that LTI, TGD, ISI, CDI and PTI can be measured with excellent reliability (ICC > 0.9), and SA, TFA and MISI can be measured with good reliability (ICC > 0.75), with the proposed framework. CONCLUSIONS: This study proposes a reliable approach with promising applicability for automatic patellar height and trochlear dysplasia assessment, assisting the radiologists in their clinical practice. CLINICAL RELEVANCE STATEMENT: The objective knee landmarks detection on MRI images provided by artificial intelligence may improve the reproducibility and reliability of the imaging evaluation of trochlear anatomy and patellar height, assisting radiologists in their clinical practice in the patellofemoral instability assessment. KEY POINTS: • Imaging evaluation of patellofemoral instability is subjective and vulnerable to substantial intra and interobserver variability. • Patellar height and trochlear dysplasia are reliably assessed in MRI by means of artificial intelligence (AI). • The developed AI framework provides an objective evaluation of patellar height and trochlear dysplasia enhancing the clinical practice of the radiologists.

Imaging evaluation of patellofemoral joint instability: a review.

The multifactorial origin of anterior knee pain in patellofemoral joint disorders leads to a demanding diagnostic process. Patellofemoral misalignment is pointed out as one of the main causes of anterior knee pain. The main anatomical risk factors of patellofemoral instability addressed in the literature are trochlear dysplasia, abnormal patellar height, and excessive tibial tubercle-trochlear groove distance. Diagnostic imaging of the patellofemoral joint has a fundamental role in assessing these predisposing factors of instability. Extensive work is found in the literature regarding the assessment of patellofemoral instability, encompassing several metrics to quantify its severity. Nevertheless, this process is not well established and standardized, resulting in some variability and inconsistencies. The significant amount of scattered information regarding the patellofemoral indices to assess the instability has led to this issue. This review was conducted to collect all this information and describe the main insights of each patellofemoral index presented in the literature. Five distinct categories were created to organize the patellofemoral instability indices: trochlear dysplasia, patellar height, patellar lateralization, patellar tilt, and tibial tubercle lateralization.

Osteochondral transplantation using autografts from the upper tibio-fibular joint for the treatment of knee cartilage lesions.

PURPOSE: Treatment of large cartilage lesions of the knee in weight-bearing areas is still a controversy and challenging topic. Autologous osteochondral mosaicplasty has proven to be a valid option for treatment but donor site morbidity with most frequently used autografts remains a source of concern. This study aims to assess clinical results and safety profile of autologous osteochondral graft from the upper tibio-fibular joint applied to reconstruct symptomatic osteochondral lesions of the knee. METHODS: Thirty-one patients (22 men and 9 women) with grade 4 cartilage lesions in the knee were operated by mosaicplasty technique using autologous osteochondral graft from the upper tibio-fibular joint, between 1998 and 2006. Clinical assessment included visual analog scale (VAS) for pain and Lysholm score. All patients were evaluated by MRI pre- and post-operatively regarding joint congruency as good, fair (inferior to 1 mm incongruence), and poor (incongruence higher than 1 mm registered in any frame). Donor zone status was evaluated according to specific protocol considering upper tibio-fibular joint instability, pain, neurological complications, lateral collateral ligament insufficiency, or ankle complaints. RESULTS: Mean age at surgery was 30.1 years (SD 12.2). In respect to lesion sites, 22 were located in weight-bearing area of medial femoral condyle, 7 in lateral femoral condyle, 1 in trochlea, and 1 in patella. Mean follow-up was 110.1 months (SD 23.2). Mean area of lesion was 3.3 cm2 (SD 1.7), and a variable number of cylinders were used, mean 2.5 (SD 1.3). Mean VAS score improved from 47.1 (SD 10.1) to 20.0 (SD 11.5); p = 0.00. Similarly, mean Lysholm score increased from 45.7 (SD 4.5) to 85.3 (SD 7.0); p = 0.00. The level of patient satisfaction was evaluated, and 28 patients declared to be satisfied/very satisfied and would do surgery again, while 3 declared as unsatisfied with the procedure and would not submit to surgery again. These three patients had lower clinical scores and kept complaints related to the original problem but unrelated to donor zone. MRI score significantly improved at 18-24 months comparing with pre-operative (p = 0.004). No radiographic or clinical complications related to donor zone with implication in activity were registered. CONCLUSIONS: This work corroborates that mosaicplasty technique using autologous osteochondral graft from the upper tibio-fibular joint is effective to treat osteochondral defects in the knee joint. No relevant complications related to donor zone were registered. LEVEL OF EVIDENCE: Case series, Level IV.

Empirical antibiotic therapy in prosthetic joint infections.

When dealing with prosthetic joint infections (PJI) there is often the need to start antibiotic therapy without having identified the underlying pathogen. Under these circumstances there is no consensus regarding which antibiotic to use. We aimed to produce local recommendations for empirical antibiotic treatment of PJI by describing the microbiological spectrum involved and respective antibiotic susceptibility profile. We examined the records of 75 consecutive patients that underwent surgery for prosthetic joint infection from July 2001 to December 2008. There were 49 women and 26 men with an average age of 63 years. Ninety culture results were available from 41 hips and 34 knee replacements. Staphylococcus sp. was present in most infections (72.8%) regardless of surgical site or classification. The prevalence of methicillin-resistance among staphylococci was 64.2% with no relevant difference between sub-groups. Vancomycin is 100% effective against most commonly isolated Gram positives. Gram negative pathogens were present in about 15% of all cases, especially in haematogenous and chronic infections. Carbapenems and aminoglycosides are the most effective antibiotics against these pathogens. Our results suggest that in acute post-operative infections, treatment should start with vancomycin. In chronic and haematogeneous infections, vancomycin in combination with carbapenems appears to be an effective regimen. Treatment should be adjusted as soon as preliminary or definitive microbiology results are available.