Ultrasonographic evaluation of early knee osteoarthritis
Review Article

Ultrasonographic evaluation of early knee osteoarthritis

Nobuyuki Kumahashi1, Suguru Kuwata1, Hiroshi Takuwa1,2, Yuji Uchio1

1Department of Orthopaedics, Shimane University, Izumo, Japan; 2Laboratory for Bone and Joint Diseases, RIKEN Center for Integrative Medical Sciences (IMS), Tokyo, Japan

Contributions: (I) Conception and design: N Kumahashi, Y Uchio; (II) Administrative support: None; (III) Provision of study materials or patients: None; (IV) Collection and assembly of data: N Kumahashi; (V) Data analysis and interpretation: N Kumahashi; (VI) Manuscript writing: All authors; (VII) Final approval of manuscript: All authors.

Correspondence to: Nobuyuki Kumahashi. Department of Orthopaedics, Shimane University, School of Medicine, 89-1 Enya-cho, Izumo, Shimane, 693-8501, Japan. Email: n-kuma@med.shimane-u.ac.jp.

Abstract: Ultrasonography (US) is non-invasive, fast, easy-to-use and safe. The quality and resolution of US has been improved in recent years. Recently, in orthopaedics, US measurements have been used for the diagnosis of osteoarthritis (OA), which can provide the basis for proper early treatment. In this chapter, we review the published articles about US evaluation of knee OA, especially focused on cartilage, including probe selection, probe direction, evaluation site, knee position, measurement of thickness, echogenesis and superficial sharpness of cartilage with histological findings and accuracy to detect early OA changes.

Keywords: Ultrasound (US); cartilage; early osteoarthritis; knee

Received: 14 August 2018; Accepted: 27 August 2018; Published: 14 September 2018.

doi: 10.21037/aoj.2018.08.05


Knee osteoarthritis (OA) is common in elderly adults and is often seen in the outpatient clinic. The severity of clinical symptoms of knee OA is not reflected the radiological stage (1,2) and knee pain is associated with poor health status and psychological distress (3). However, OA is now diagnosed by radiological grading, which is based only on the joint space width. It is important to detect OA in its early stages to prevent it from progressing. However, a convenient method with rapid evaluation has not been reported.

Ultrasonography (US) is widely used in many medical fields and non-invasive US can be employed to evaluate cartilage, meniscus, tendon, joint effusion, bursitis and synovitis, as well as bony structures related to OA such as osteophytes and subchondral bone (4-23). US examination has been reported to be useful for evaluation of cartilage quantity, which cannot be detected by X-ray (4,5,9,11,21,23). It also has been reported that OA initiates superficially, and previous reports have shown the correlation between echoic superficial findings and histological score in the detection of early OA (8,10,11,13,17,19). However, optimization of the measurement protocol of US to detect early OA has not been established.

In this chapter, we review US evaluation of the cartilage and other components to reveal its potential to detect early knee OA. The aim of this review is to evaluate the effectiveness of US diagnosis of early OA, and to explore the associations between US findings such as echogenesis and superficial sharpness with histological findings in early knee OA.

Patients and methods

Study selection

We focused on US evaluation of the knee, especially cartilage and other OA characteristic findings, and chose 20 articles that provided the information about the sample site/species, sample number, probe selection, probe direction, knee position, evaluation site, assessment of the cartilage, and accuracy of the US diagnosis.

All items were reviewed in the selected articles and are summarized in Table 1.

Table 1
Table 1 Summary of US examination of all articles
Full table

Sample selection

All studies included animal and human subjects. Rabbit and bovine knees were widely used models in cartilage research. In human beings, OA patients who underwent total knee arthroplasty (TKA) were primary subjects. Fresh cadavers, osteochondral plugs or punched samples were also examined.


US evaluation

Evaluation site

In the knee, the medial and lateral condyle, intercondylar notch area, and patellar cartilage were examined. In addition, medial meniscal displacement, synovitis and Baker cysts were checked.

Probe selection

Linear and spherical type probes were used in non-invasive US. The small transducer was used for intra-articular evaluation using US.

Probe frequency

In the orthopaedic area, 3–5 MHz frequency is used to assess the deep structures and 7–10 MHz frequency is used to evaluate more superficial areas under the skin. 0.1–1 MHz has been suggested as the most useful for bone characterization. The low frequency of conventional US may lack the resolution necessary for cartilage and has limited accuracy and sensitivity to detect cartilage degeneration. Previous reports have shown that higher frequencies have the potential to detect cartilage degeneration with ultrasound transducers arranged close to the cartilage (11-13,19). The drawback of using a high frequency probe is high attenuation, which inevitably prevents analysis of subchondral bone.

In this series, probe frequency was 5 to 55 and 5–15 MHz was most often used.

Internal or external US

External US is a safe and non-invasive approach to the knee joints, but minimization of the effect of overlying tissue layers is needed. Internal US is more accurate because there is no interference with soft tissues, but it is an invasive procedure.

Probe direction to the knee

The direction of the probe to the femoral condyle is limited due to the patella. Suprapatellar, transverse or longitudinal sagittal plane approaches have been used.

The transverse direction is most often used to check the femoral cartilage.

The longitudinal sagittal approach to the distal end of the midline medial and lateral condyle is conducted when can be used with the patient sitting on the table in the deep flexed position. The ultrasound beam is kept perpendicular to the surface of the distal end of the femur along the midline (9).

There has been only one article that compared the suprapatellar transverse and longitudinal sagittal plane approaches to check the accuracy of cartilage thickness by validation with MRI (9). Yoon reported that the longitudinal sagittal plane of OA patients was more accurate than the transverse image in measuring cartilage thickness of the medial femoral condyle when compared to MRI.

Knee position

The knee position for cartilage measurement using US has been controversial because the shadow of the patella makes it difficult to determine the weight-bearing area of the cartilage. Therefore, it is important to assess how much of the weight-bearing femoral cartilage can be detected with US.

For the medial and lateral condyles and the intercondylar cartilage, the maximally flexed position was used most often by sweeping the cartilage from proximal to distal with the probe in transverse position in this series of studies. Lee reported that the correlation between in vivo and in vitro US grading of OA patients and maximum angle of knee flexion less than 120 degrees was not reliable, and they recommended the knee full flexed position in measurement by US (8).

Diagnostic study between arthroscopic or macroscopic and US findings

Three articles compared the US findings and arthroscopic or macroscopic findings of different characteristics of the articular surfaces of the human knee joint.

Kuroki reported that US signal intensity of cartilage, not signal duration or interval between signals, is most helpful to differentiate various OA grades of OA patients using the International Cartilage Repair Society (ICRS) scoring system. It is especially useful to distinguish grade 0 (normal) from grade 1 cartilage (nearly normal) (7).

Kaleva reported that the intra-articular arthroscopic US method (IAUS, 40 MHz) could provide a diagnostic method based on different characteristics of the articular surfaces of the human knee joint for evaluation of the severity of articular cartilage lesions, ICRS grade, and early osteoarthritic changes. There was no correlation between roughness index of the cartilage surface (URI) and ICRS grade. The IAUS score was one rank higher than the ICRS grade estimated during conventional arthroscopy. The IAUS method can distinguish ICRS grade 0 (normal) and grade 1 (almost normal) based on the characteristics of the cartilage surface. However, this method could only be performed invasively (12).

Saarakkala reported that US gradings based on interface sharpness and echogenicity of the cartilage in the transverse plane of the human femoral condyle in patients with pain had a strong correlation with arthroscopic changes of cartilage at least for the medial condyle and the sulcus area (15). The cartilage was subjectively evaluated using five classifications based on the US findings of the condition of the superficial layers (sharp interface or irregular interface) and the intensity of cartilage (monotonous anechoic or hyper echogenicity) as follows: grade 0 (sharp interface or monotonous echogenicity), grade 1 (loss of sharp interface and increased echogenicity), grade 2A (loss of sharp interface and increased echogenicity, less than 50% local thinning of cartilage), grade 2B (irregular interface and hyper echogenicity, more than 50% but less than 100% local thinning), and grade 3 (100% local loss of the cartilage).

The diagnostic accuracy of the cartilage surface of the medial and lateral condyles and sulcus area was evaluated. Sensitivity (83%) was good only in the medial femoral condyle. Specificity was good for the femoral sulcus and lateral condyle. A positive predictive value (88–100%) was strong at all sites. The negative predictive value (24–46%) was low. The diagnostic odds ratio and a general indicator of test performance of US varied between 5.0 and 13. They indicated that negative findings do not rule out degenerative changes.

From these findings it was concluded that in vivo intra-articular US examination could reflect more accurate macroscopic findings than that of being found with arthroscopy, indicating that US examination has the potential to detect early OA of the knee, especially changes to the superficial layer of the cartilage.

Diagnostic study between histological and US findings

Six articles compared US findings and histological findings.

Niu et al. reported that high frequency US measurements (55 MHz) can reflect changes in the URI and ultrasound reflection coefficients (R) of the cartilage samples from a rabbit model with different OA grades (OARSI grade), especially in early and mild OA (13). URI could be used to distinguish early OA and mild OA but had no ability to accurately distinguish grade 1 from grade 2. This study suggests that these two US acoustic parameters have the potential to become objective criteria in OA grading.

Nishitani et al. reported that ultrasound (10 MHz) can detect macroscopically undetectable changes in OA reflecting histological and biochemical degeneration in both rabbit and human OA samples, respectively (17). The URI was significantly different from nearly normal, early OA and moderate OA, and was correlated with Mankin scores. Parameters derived from fourier-transform infrared spectroscopy (FTIR) reflected collagen and proteoglycan content in the superficial zone.

Wang reported the feasibility of high frequency (40 MHz) ultrasound for simultaneous assessment of the surface condition, acoustic parameters, thickness and mechanical properties of articular cartilage of fresh bovine cartilage in vitro (11). They showed a significant relationship between the surface integrity and URI. The ultrasound reflection intensity is sensitive to early histological changes of OA in vitro.

Lee et al. reported significant correlations between both in vivo US findings (5–12 MHz) and in vitro gradings (sharpness of the superficial margin, and clarity and thickness of the cartilage band) with histologic gradings of distal medial and lateral femoral condyles according to Huang’s criteria (8).

Aula et al. showed that 5 MHz ultrasound could provide diagnostically valuable information on both cartilage and bone (10). They showed the feasibility for simultaneous measurement of the acoustic properties of articular cartilage and subchondral bone of bovine osteochondral samples by pulse-echo ultrasound at 5 MHz, which could quantify both cartilage and subchondral bone changes. The relationship between the US parameters (backscattered signal from the internal cartilage and from subchondral bone) and properties of cartilage and bone were not as strong as earlier studies focusing only either on bone or cartilage.

Männicke et al. reported that the combinations of high frequency and US-based parameters derived from surface reflections and signals backscattered from the cartilage matrix (40 MHz) exhibit potential to characterize the very early degeneration stages of articular cartilage of human punched samples evaluated by a modified Mankin score (19). The surface parameters [integrated reflection amplitude (IRC) and URI] used in this study are related to cartilage softening and surface fibrillation. They suggested that both minimally invasive arthroscopic US and high frequency transcutaneous US could conceivably be used for in vivo application.

Based on these findings, US examination, especially high frequency US in vitro, correlates with histological change and can detect superficial microscopic histological changes that cannot be detected macroscopically. Further studies are needed for optimization of the ultrasound frequency and measurement geometry in clinical application.

Diagnostic study between other findings of knee OA and US findings

There were seven reports about critical US findings reflecting OA changes.

Radiological assessment

Naredo et al. reported the comparative study with clinical and radiographic assessment that knee effusion and protrusion of the medial meniscus with displacement of the medial collateral ligament measured with US were associated with knee mechanical pain and pain at rest between the symptomatic and asymptomatic OA patients (6). In addition, protrusion of the medial meniscus may contribute to the radiologic medial FT space narrowing.

Riecke et al. reported that five separate domains including morphological changes and inflammation in the medial and lateral compartment, and effusion were examined with US and US score displayed substantial reliability and reproducibility in comparison with standing radiographs of the knees (18).

Pain assessment

Wu et al. reported that medial meniscus compartment synovitis and suprapatellar effusion measured with US were positively linearly associated with knee pain at motion with equal radiographic grades of OA in both knees (14).

Medial meniscal and other assessment

Kawaguchi et al. reported that medial meniscus displacement (MRD) measured with US increased with weight bearing, with a close association between extraarticular displacement of the medial meniscus and progression of OA (16).

Yanagisawa et al. reported that the longitudinal US image in knee extension in both the standing and supine positions had a high reliability and excellent diagnostic accuracy for OA of the knee based on the joint space, peripheral osteophytes and MRD, and could accurately detect the early degeneration of the knee joint (20,22).

Bevers et al. reported the prospective and explorative study that six US features including effusion, synovial proliferation, infrapatellar bursitis, meniscus displacement protrusion, Baker’s cyst and cartilage thickness were investigated in 55 OA patients receiving standardized multimodal treatment for 1 year. The prevalence of inflammatory features like effusion and synovial hypertrophy show a fluctuating pattern in time in contrast to meniscus displacement protrusion, Baker’s cyst (21).

Clinical relevance between US findings and muscle power

Berger et al. reported that KL grade was limited and not significant as a severity stratification method to show differences in knee extension force. The KL radiological classification has not been associated with OA symptoms and muscle power (1). On the other hand, there has been one report of the relationship between cartilage thickness measured by US and muscle power showing that femoral cartilage thicknesses were positively correlated with isometric quadriceps and hamstring muscle strength at baseline and after three months of a muscle strengthening program (23).


The US evaluation of knee cartilage has the potential to detect both in vitro and in vivo superficial cartilage abnormalities and characterize very early degenerative stages of articular cartilage before macroscopic changes. US is a promising technique for clinical screening of degenerative changes of articular cartilage and detection of the early stages of OA in vivo. Further studies are needed to elucidate the diagnostic value and potential limitations of US and to reveal the most appropriate combination of invasive or non-invasive probe selection with other conditions when used clinically.




Conflicts of Interest: The authors have no conflicts of interest to declare.


  1. Berger MJ, Kean CO, Goela A, et al. Disease severity and knee extensor force in knee osteoarthritis: data from the Osteoarthritis Initiative. Arthritis Care Res (Hoboken) 2012;64:729-34. [Crossref] [PubMed]
  2. O'Reilly SC, Jones A, Muir KR, et al. Quadriceps weakness in knee osteoarthritis: the effect on pain and disability. Ann Rheum Dis 1998;57:588-94. [Crossref] [PubMed]
  3. O'Reilly SC, Muir KR, Doherty M. Knee pain and disability in the Nottingham community: association with poor health status and psychological distress. Br J Rheumatol 1998;37:870-3. [Crossref] [PubMed]
  4. Aisen AM, McCune WJ, MacGuire A, et al. Sonographic evaluation of the cartilage of the knee. Radiology 1984;153:781-4. [Crossref] [PubMed]
  5. Mathiesen O, Konradsen L, Torp-Pedersen S, et al. Ultrasonography and articular cartilage defects in the knee: an in vitro evaluation of the accuracy of cartilage thickness and defect size assessment. Knee Surg Sports Traumatol Arthrosc 2004;12:440-3. [Crossref] [PubMed]
  6. Naredo E, Cabero F, Palop MJ, et al. Ultrasonographic findings in knee osteoarthritis: a comparative study with clinical and radiographic assessment. Osteoarthritis Cartilage 2005;13:568-74. [Crossref] [PubMed]
  7. Kuroki H, Nakagawa Y, Mori K, et al. Ultrasound properties of articular cartilage in the tibio-femoral joint in knee osteoarthritis: relation to clinical assessment (International Cartilage Repair Society grade). Arthritis Res Ther 2008;10:R78. [Crossref] [PubMed]
  8. Lee CL, Huang MH, Chai CY, et al. The validity of in vivo ultrasonographic grading of osteoarthritic femoral condylar cartilage: a comparison with in vitro ultrasonographic and histologic gradings. Osteoarthritis Cartilage 2008;16:352-8. [Crossref] [PubMed]
  9. Yoon CH, Kim HS, Ju JH, et al. Validity of the sonographic longitudinal sagittal image for assessment of the cartilage thickness in the knee osteoarthritis. Clin Rheumatol 2008;27:1507-16. [Crossref] [PubMed]
  10. Aula AS, Töyräs J, Tiitu V, et al. Simultaneous ultrasound measurement of articular cartilage and subchondral bone. Osteoarthritis Cartilage 2010;18:1570-6. [Crossref] [PubMed]
  11. Wang SZ, Huang YP, Saarakkala S, et al. Quantitative assessment of articular cartilage with morphologic, acoustic and mechanical properties obtained using high-frequency ultrasound. Ultrasound Med Biol 2010;36:512-27. [Crossref] [PubMed]
  12. Kaleva E, Virén T, Saarakkala S, et al. Arthroscopic Ultrasound Assessment of Articular Cartilage in the Human Knee Joint: A Potential Diagnostic Method. Cartilage 2011;2:246-53. [Crossref] [PubMed]
  13. Niu HJ, Wang Q, Wang YX, et al. Ultrasonic reflection coefficient and surface roughness index of OA articular cartilage: relation to pathological assessment. BMC Musculoskelet Disord 2012;13:34. [Crossref] [PubMed]
  14. Wu PT, Shao CJ, Wu KC, et al. Pain in patients with equal radiographic grades of osteoarthritis in both knees: the value of gray scale ultrasound. Osteoarthritis Cartilage 2012;20:1507-13. [Crossref] [PubMed]
  15. Saarakkala S, Waris P, Waris V, et al. Diagnostic performance of knee ultrasonography for detecting degenerative changes of articular cartilage. Osteoarthritis Cartilage 2012;20:376-81. [Crossref] [PubMed]
  16. Kawaguchi K, Enokida M, Otsuki R, et al. Ultrasonographic evaluation of medial radial displacement of the medial meniscus in knee osteoarthritis. Arthritis Rheum 2012;64:173-80. [Crossref] [PubMed]
  17. Nishitani K, Kobayashi M, Kuroki H, et al. Ultrasound can detect macroscopically undetectable changes in osteoarthritis reflecting the superficial histological and biochemical degeneration: ex vivo study of rabbit and human cartilage. PLoS One 2014;9. [Crossref] [PubMed]
  18. Riecke BF, Christensen R, Torp-Pedersen S, et al. An ultrasound score for knee osteoarthritis: a cross-sectional validation study. Osteoarthritis Cartilage 2014;22:1675-91. [Crossref] [PubMed]
  19. Männicke N, Schöne M, Oelze M, et al. Articular cartilage degeneration classification by means of high-frequency ultrasound. Osteoarthritis Cartilage 2014;22:1577-82. [Crossref] [PubMed]
  20. Yanagisawa S, Ohsawa T, Saito K, et al. Morphological evaluation and diagnosis of medial type osteoarthritis of the knee using ultrasound. J Orthop Sci 2014;19:270-4. [Crossref] [PubMed]
  21. Bevers K, Bijlsma JW, Vriezekolk JE, et al. The course of ultrasonographic abnormalities in knee osteoarthritis: 1 year follow up. Osteoarthritis Cartilage 2014;22:1651-6. [Crossref] [PubMed]
  22. Yanagisawa S, Kobayashi T, Ohsawa T, et al. Sonographic evaluation of the medial aspect of the knee in a population-based cohort study. J Clin Ultrasound 2015;43:318-21. [Crossref] [PubMed]
  23. Tuna S, Balcı N, Özçakar L. The relationship between femoral cartilage thickness and muscle strength in knee osteoarthritis. Clin Rheumatol 2016;35:2073-7. [Crossref] [PubMed]
doi: 10.21037/aoj.2018.08.05
Cite this article as: Kumahashi N, Kuwata S, Takuwa H, Uchio Y. Ultrasonographic evaluation of early knee osteoarthritis. Ann Joint 2018;3:75.