Accuracy of clinical examination and magnetic resonance in assessment of chondral lesion of knee joint by arthroscopy

Introduction: The knee joint has a unique anatomical structure in the human body. The localization between the two longest bones in the human body – femur and tibia – makes it prone to injuries, trauma, and other pathologies. Clinical examination of the joint is still the primary method in evaluating the condition of the patient's knee. The study aims to determine the diagnostic accuracy of clinical examination and magnetic resonance (MR) in assessing chondral lesions of knee joint using arthroscopy as a reference standard. Patients and methods: The examination was conducted on 94 patients (58 males and 36 females) with knee injuries. Clinical examination indicated a primary chondral lesion of knee cartilage in eight patients (five men and three women), with an average age of 45.75. Besides the clinical examination, the diagnostics were performed using MR imaging by Siemens of 0.5 Tesla, and arthroscopy was performed using Storz arthroscope. Results: Our research has generated the following values of clinical and MR results for chondral lesions: Sensitivity (Se) = 12.5%, Specificity (Sp): could not be calculated, Positive Predictive Value (PPV) = 100%, Negative Predictive Value (NPV) = 0% and Accuracy (ACC) = 12.5%. The accuracy of clinical and intraoperative results for chondral lesion was: Se =100%, Sp: could not be calculated, PPV = 100%, NPV: could not be calculated, and ACC = 100%. MR imaging and arthroscopy findings of chondral lesion showed: Se = 100%, Sp = 0%, PPV = 12.5%, NPV: could not be calculated and ACC = 12.5%. In comparing the clinical sign and MR and intraoperative result, Positive Predictive Value for patients with chondral lesion was maximal (100%), while comparing MR with the intraoperative result, Positive Predictive Value was 12.5%. In comparison between clinical sign and intraoperative results, the accuracy for patients with chondral lesion was 100%, while comparing the clinical sign with MR result and MR with the intraoperative result, the accuracy was 12.5%. Conclusion: Our examinations have shown that MR examination is not currently as valid for diagnosing injury of chondral cartilage of knee as the medical community or patients have anticipated it.


Introduction
The knee joint has a unique anatomical structure in the human body that, together with localization between the two longest bones in the human body -femur and tibia -makes it prone to injuries, trauma, and other pathologies. [1]. Joint surfaces of condyle and femur trochlea, tibia, and patella plateau are pre-covered with specialized connective tissue -hyaline cartilage -of different thicknesses. The joint hyaline cartilage has a complex and unique structure, and its integrity is crucial for the proper functioning of the joint. The cartilage is exceptionally smooth on articulated surfaces, elastic, capable of sustaining high pressure generated even during regular movements within the joint. It is primarily composed of chondrocytes and proteoglycans. The macroscopic and microscopic analysis helped conclude that the cartilage composition is not the same in every segment. Still, it is layered, depending on the collagen fibers orientation and chondrocytes density. [2] Clinical examination of the joint is still the primary method in evaluating the condition of the patient's knee. The clinical examination is commenced with collecting the anamnestic information, informing on symptoms and description of means of injuring, inspection, palpation, and static and dynamic evaluation of the joint's condition. Clinical tests: Fründ's "fear test, "crepitation," and "Patellar apprehension test "are applied for clinical assessment of the state of the knee cartilage. [1][2][3] The following method in the diagnostic protocol is most frequently the standard radiography of the knee joint in two directions. Suppose both clinical and radiography signs indicate the lesion of cartilage. In that case, it is necessary to do magnetic resonance (MR) because MR has been proven to be an excellent diagnostic method for evaluating articulated cartilage. Multi-Detector Computed Tomography, with higher spatial resolution, may have a more significant role in diagnosing chondral damages. [2-4] The aim of the study is to determine the diagnostic accuracy of clinical examination and MR in assessing chondral lesions of the knee joint using arthroscopy as a reference standard.

Patients and Methods
The study included patients treated from a knee injury at polyclinic "MEDICAL Centre" in Travnik between June 1st, 2016, and June 1st, 2018. A total of 94 patients (58 men and 36 women) were examined. Eight patients (five men and three women) had a clinically primary chondral lesion. The youngest patient was eight, while the oldest one was 71-year-old. The mean age was 45.75, with a median of 54.5. Two patients were injured in February, and the others were injured in January, March, April, May, August, and November, respectively. Seven patients with chondral lesions injured their right leg, and one patient injured their left leg. All eight patients suffered from a knee injury and had clinically positive clinical signs: Fründ's fear test "and "crepitation." Five patients had MR diagnostics and arthroscopy performed at the same month of injury occurrence. Two patients performed the following month, while one patient arthroscopy was completed two months following the injury. The MR imaging device was Siemens 0.5 Tesla (T). Arthroscopy was used as the "golden standard" using Storz arthroscope. Arthroscopies performed under local anesthesia with analgosedation went uneventful. Table 1 shows the calculations of sensitivity, specificity, positive predictive values, negative predictive values, and test accuracy.  Sensitivity represents the true positive rate. It is calculated as a proportion of True Positive cases concerning the total number of ill patients using the following formula: (1) Sensitivity is the measure of the test accuracy related to the population of patients with the condition and represents the test's ability to identify those patients who have the condition. Specificity is the accuracy in negative samples. It is calculated as a proportion of True negative patients in relation to the total number of healthy patients using the following formula: (2) Detecting early changes in cartilage and determining the degree of cartilaginous defects without osteoarthritis is essential in deciding the treatment and prognosis of the outcome of the treatment. [6][7][8] Although many radiology methods today make the presentation of the joint cartilage possible, only MR ensures the visualization of the outlines of the entire cartilage of the joint unit possible and its inner structure and associated elements subchondral bone.
[4] The majority of the population suffers from cartilage diseases and defects of various causes that generate miscellaneous symptomatology. MR has shown great potential in assessing the condition of the articulated cartilage because it is possible to reach high contrast between the articulated cartilage and the surrounding structures.
[5] MR is beneficial in diagnosing osteochondral defects and osteochondritis dissecans presenting with a more or more minor defect on subchondral bone. The display of isolated cartilage abnormality as in chondromalacia of inflammatory or degenerative arthritis poses a huge issue. [5] Proper interpretation of MR results depends directly on the spatial resolution of the examination performed and the signal-to-noise ratio. The cartilage structure is small compared with spatial resolution of the entire image. Intrachondral variations are too small to be displayed clearly due to the "partial volume effect". As a consequence of the complex structure of cartilage, the MR signal generated from the very cartilage is difficult to quantify because it depends mostly on parameters applied during the imaging. So far, the achievement is that structural and physiological characteristics of cartilage are adequately displayed on the devices of higher magnetic field strength (3T, but on the systems of 1.5T too). This provides enough data with a high spatial resolution that enables visualization of internal cartilage structure [6]. In our research, we used a magnetic field of 0.

Conclusion
This study has shown weak sensitivity and specificity of preoperative MR diagnostics in detecting cartilage lesions of the knee. High rates of false diagnosis have been noted, suggesting that MR of 0.5 Tesla is not very precise in assessing chondral lesions of knee joint cartilage. We conclude that MR examination is currently not valid for diagnosing the injury of the chondral cartilage of the knee, as the medical community or the patient might anticipate.