Vol. 5, Issue 9 : Medical Imaging of Osteoarthritis

Medical Imaging of Osteoarthritis of the Knee

By Jack Clement, M.D., Ph.D, Medical Director, Musculoskeletal Imaging , Intrinsic Imaging, LLC,
San Antonio (TX), Bolton (MA),
Volume 6, Issue 9


Arthritis is broadly defined as the inflammation of one or more small or large joints. While there are a number of types of arthritis, the most common form is osteoarthritis, also termed degenerative joint disease. Osteoarthritis is the number one cause of disability in the United States. According to the Center for Disease Control (CDC), in 2010-2012, 50% of adults reported having a physician diagnosis of arthritis and nearly 50% of all people will develop symptomatic knee arthritis by the age of 85, with the risk increasing to 66% in obese patients. Furthermore, the monetary costs of arthritis are staggering with US$ 80 billion dollars spent on direct medical care for patients with arthritis in 2003 and US$ 47 billion dollars in consequent productivity losses during that time.

An estimated 800,000 knee replacements were performed last year and this number is projected to increase to 3.5 million procedures by 2030. These costs are only projected to increase with the aging American demographic. As such, there is considerable biotechnology research aimed at developing novel therapies, including device, biologics, and disease modifying osteoarthritis drugs (DMOADs) for the more effective treatment of osteoarthritis and its progression.

Medical Imaging plays a central role in the assessment of the severity and evolution of osteoarthritis. This article aims to provide guidance on the strengths and limitations of different imaging modalities used to assess the progression of osteoarthritis of the knee and the application of these imaging modalities to clinical trials that rely on imaging endpoints.


The pathophysiology of osteoarthritis is poorly understood, but has as its common endpoint the destruction of articular cartilage which is the smooth shock absorbing tissue on the end of bones where joints form. Predisposing factors include age, obesity, prior trauma or surgery, and genetic factors. Osteoarthritis may evolve primarily as damage to articular cartilage with slow eroding of the chondral surface, eventually exposing underlying bone. This leads to instability of the joint and a chronic underlying joint inflammation (synovitis) that produces pain, stiffness, swelling, and limited range of motion. Chronic joint inflammation and joint instability then leads to degeneration of periarticular soft tissues crucial for stabilizing the joint such as ligaments, menisci, and tendons. Alternatively, osteoarthritis may begin as injury to jointstructural stabilizing elements leading to chronic joint micro instability and increasing stress on articular cartilage leading to its more rapid dissolution. In reality, both of these processes likely occur simultaneously in a synergistic fashion but to variable degrees in each patient.

Due to the incidence and severity of knee joint osteoarthritis, it is the most common joint targeted for the investigation of novel osteoarthritis therapies. Additionally, the cartilage within the knee is the thickest and most easily imaged articular cartilage among all joints. This allows imaging to serve as an endpoint in the assessment of the success of DMOADs and other treatments in the preservation of articular cartilage quality and integrity.

Historically, the frontal knee radiograph is the hallmark imaging examination in the diagnosis of the presence and severity of knee joint osteoarthritis. Two primary findings are used to assess its severity, specifically joint space narrowing and osteophyte formation. The loss of articular cartilage manifests as narrowing of the radiolucent joint space between the femur and tibia, either medially or laterally. Of important note is that articular cartilage cannot be directly seen on the radiograph. Its thickness is inferred by assessing the degree of space separating the subchondral bone plates. This is an important limitation to the use of radiographs to assess cartilage integrity, as a normal joint space distance on a radiograph may be present in a knee with no articular cartilage remaining due to anatomic positioning and alignment or other structural variations.

Osteophytes are also a hallmark of chondral loss. These peripheral bone outgrowths occur when the shock absorbing qualities of articular cartilage are diminished and mechanical forces are transferred to the subchondral bone. The bone responds to this increased stress by proliferating and growing, forming peripheral spurs and bulky bony outgrowths termed osteophytes. The size and degree of osteophyte formation correlates with the degree of cartilage loss along the involved bones. As osteoarthritis progresses, there is increased bone formation underneath the subchondral bone plate, termed sclerosis, and slow progression of joint deformity

Radiographic Osteoarthritis Assessments

The Kellgren and Lawrence radiographic grading system is the most commonly used scale for assessing the severity of osteoarthritis. This categorical scale is based upon the twin assessments of joint space narrowing (cartilage loss) and osteophyte formation with sclerosis and deformity of joint contour important in later stages of osteoarthritis. This scale is used near universally in assessing knee osteoarthritis in clinical trials with the criteria of joint space narrowing the most important observation within each grade. This system has also been adopted by the World Health Organization as the reference standard for cross-sectional and longitudinal osteoarthritis epidemiological studies.

Though widely adopted, the Kellgren Lawrence (K/L) scale does have some significant limitations. Of key importance, though progression of joint space narrowing is the primary outcome by which clinical trials have assessed pharmacologic or biologic efficacy, there is actually poor correlation between radiographic joint space narrowing, often a late finding in osteoarthritis, and clinical symptoms. Additionally, the rate of joint space narrowing has been shown to be variable caused by differences in patient characteristics, clinical status, and inconsistent radiographic positioning during serial examinations.

The K/L grading system is also not able to distinguish the medial from the lateral compartment and does not evaluate the patellofemoral joint. To mitigate positioning effects, many research trials often employ a Synaflexer frame to obtain weight-bearing knee PA radiographs with the knee positioned in 20-degree flexion and 5-degree foot external rotation.

The K/L scale has also been criticized because the distance between each grading level is not equivalent to the degree of progression of osteoarthritis. Thus, progression from grade 0 to 1, the point at which intervention is most crucial, does not have the same clinical importance as progression from grade 2 to 3.

To provide an additional means to assess osteoarthritis, the Osteoarthritis Research Society International (OARSI) refined the K/L radiographic grading system for knee arthritis. Much like the K/L system, the OARSI grading system relies on evaluation of osteophytes and joint space narrowing, but extends the evaluation of joint space narrowing to both the medial and lateral compartments of the knee and the evaluation of osteophytes along both the femoral condyles and tibial plateaus rather than treating them as a collective entity. This is beneficial because it provides a more global assessment of knee joint arthritis. Within this scoring system, the degree of osteophytosis and joint space narrowing is separately graded from 0 (normal) to 3 (severe). OARSI has also provided an imaging atlas to aid in standardizing assessment among different radiologists.

Though radiographic osteoarthritis grading systems have value, perhaps their most important criticism is the fact that the most important structural abnormality they attempt to quantify, articular cartilage thickness and integrity, is inferred rather than directly visualized. Joint space narrowing serves as a surrogate for cartilage loss. Joint spacing may be near normal in a knee with completely denuded cartilage. Additionally, only cartilage along the weight bearing portions of the knee joint, parallel to the radiographic beam, is assessed. Patellofemoral cartilage and cartilage along the anterior and posterior portions of the knee, important in flexion and extension movements, are not assessed in weight bearing anterior views of the knee. The inability to directly visualize the pattern of articular cartilage loss, especially in the early stages of osteoarthritis when intervention can achieve the greatest results, is a severe limitation to radiographic based systems as the sole criteria for assessing biologic efficacy.

Magnetic Resonance Imaging Assessments

Magnetic Resonance Imaging (MRI) has begun to augment plain film radiography in the imaging assessment of the presence and progression of osteoarthritis. MRI is a tomographic technique that provides superior soft tissue contrast with direct visualization of all joint structures including articular cartilage, menisci, subchondral bone, and ligaments.

MRI is a widely available imaging technique with standardized imaging planes and sequences for assessing knee osteoarthritis. Additionally, MRI is very sensitive to early changes of knee osteoarthritis such as chondral softening, fissuring, and early bone marrow changes that are not present on radiographs.

MRI systems suitable for knee imaging include 1.0T extremity systems, standard 1.5T MRI magnets, and high field 3.0T MRI open bore systems with the higher field magnets capable of the best spatial and contrast resolution for visualizing articular cartilage.

MRI also shows future promise in the detection of very early stage cartilage degeneration. T2 cartilage mapping looks at early cartilage ultrastructural changes seen in early degeneration. This leads to subtle increases in cartilage water content which can be quantified on heavily T2 weighted fluid sequences. It allows the detection of cartilage abnormalities before chondral loss. T2 mapping is also used in the clinical setting to follow chondral healing in the setting of microfracture and chondrocyte implantation. It is not commonly used in clinical trial studies of knee arthritis.

Several semi-quantitative scoring methods have been devised in an attempt to quantify the MR findings in knee osteoarthritis. These scoring systems all focus on assessing the integrity of articular cartilage, but also combine cartilage assessment with whole joint evaluation to get a more complete picture of the process of knee degeneration.

The most comprehensive scoring system for assessing knee osteoarthritis is termed the Whole-Organ Magnetic Resonance Imaging Score (WORMS). This scoring system quantifies several of the structural changes of knee osteoarthritis including cartilage signal and morphology, subarticular bone abnormality (bone marrow edema), subarticular bone cysts, bone loss or attrition, osteophytes, meniscal pathology, ligament pathology, and synovitis.

The WORMS scoring method subdivides the knee into a total of 14 articular and subarticular regions allowing detailed regional analysis of both early and late stage changes of knee osteoarthritis. Similar to radiographic scoring systems, these MRI scoring systems all lean heavily on cartilage integrity and osteophytosis as important determinates of the total osteoarthritis score. Using the WORMS criteria, cartilage and osteophyte assessment compromise approximately 55% of the total score.

The strength of MR scoring systems is further evidenced by their ability to quantify very early changes in chondral degeneration including surface cartilage loss and bone marrow edema, findings invisible to radiographic evaluation.


Medical Imaging plays a central role in the assessment of knee arthritis in clinical trials. Radiography remains the first line of imaging, as it is affordable and well established over many decades. The Kellgren Lawrence and more recently OARSI scoring systems provide predictable, meaningful data and should be included in all clinical trials addressing knee osteoarthritis. Radiography plays an important role in long term trials that span many years or decades, the time scale over which the changes seen on radiographic scoring systems occur.

MRI is a relatively new imaging modality in the setting of knee osteoarthritis clinical trials, but with exquisite soft tissue contrast and the ability to directly visualize early chondral abnormalities and whole organ disease, it is eclipsing radiography in importance. As interventions for knee osteoarthritis aim at preventing or slowing disease progression at very early stages as opposed to mere symptomatic relief, MRI and the WORMS criteria will play a central role in the evaluation of the success of new pharmacologics and biologics whose effects are expected to occur over months to years.

John Clement - Intrinsic Imaging Radiologist

John Clement – Intrinsic Imaging Radiologist

About the Author

Dr. Clement is a board-certified fellowship trained radiologist at Intrinsic Imaging. He is a graduate of Tulane University in New Orleans, Louisiana. He completed M.D. and Ph.D. degrees at Baylor College of Medicine in Houston, TX where his training was sponsored by the National Institutes of Health. Dr. Clement has co-authored peer reviewed, original research articles in Science, New England Journal of Medicine, and Neuron. After completion of his graduate and medical training, Dr. Clement then completed a diagnostic radiology residency at Baylor College of Medicine. He then further sub-specialized in musculoskeletal radiology at Harvard Medical School’s Massachusetts General Hospital in Boston, Massachusetts. He is a board-certified radiologist with specialized training in orthopedic radiology and interventional pain management.