Kellgren-Lawrence Grading

For Medical Students

What It Shows

The Kellgren-Lawrence system assigns a knee an ordinal grade of radiographic osteoarthritis from the features visible on a plain film. 5.3.17 covers joint space narrowing.

The features it combines are osteophyte, joint space narrowing, sclerosis and bony contour change. 5.3.18 covers osteophytes.

It matters because it is the scale almost all clinical and research communication uses, and its properties are not what its use implies. 5.3.20 covers the alignment changes.

How It Is Acquired

Assess the features on a standing anteroposterior view, which is what the system was defined on. 5.3.1 covers the anteroposterior view.

Combine them into a single grade, which is the system's design: one number for the whole compartment picture. 5.3.19 covers subchondral sclerosis.

Note that osteophyte carries the early grades, so a knee with narrowing and no osteophyte sits awkwardly in the scale. 5.3.18 covers osteophytes.

Grade each knee rather than each compartment, since the system is applied per knee. 5.3.17 covers joint space narrowing.

No per grade criteria are reproduced on this page, because the published wording is a quantity this environment cannot check. 5.1.7 covers evidence based imaging.

What Abnormal Looks Like

Marginal osteophyte, which the scale weights earliest. 5.3.18 covers osteophytes.

Joint space narrowing in a compartment. 5.3.17 covers joint space narrowing.

Subchondral sclerosis at the loaded surface. 5.3.19 covers subchondral sclerosis.

Bony contour deformity and a change in alignment at the severe end. 5.3.20 covers the alignment changes.

MRCS Level

How It Is Acquired

In a viva the expected answer names the system, says what it combines, and names its two main limitations.

The expected points are the standing view, the combined features, the osteophyte weighting, the per knee application, and the ordinal nature.

Say that the grades are ordinal, so the step between two grades is not a fixed amount of disease. 5.3.17 covers joint space narrowing.

Say that it was defined on extended films, which the Rosenberg view is not. 5.3.5 covers the Rosenberg view.

Say that the grade correlates loosely with symptoms. 5.1.2 covers clinical examination versus imaging.

What Abnormal Looks Like

The reasoning expected is that an ordinal scale ranks without measuring, so averaging grades, treating them as an interval scale or computing a change score all misuse it. 5.1.7 covers evidence based imaging.

Combining several features into one number also means two knees with the same grade can differ in which feature produced it, which matters when the decision is compartment specific. 5.3.18 covers osteophytes.

The system grades a knee and surgery is planned per compartment, so a grade is the wrong unit for most operative decisions. 5.3.17 covers joint space narrowing.

Being defined on extended films means it systematically misses the early flexed contact area loss the Rosenberg view shows. 5.3.5 covers the Rosenberg view.

How Good It Is

The grading system is the best established communication tool in radiographic osteoarthritis and a poor measurement of it.

Its best contribution is a shared vocabulary that makes populations comparable. 5.21.5 covers the osteoarthritis imaging literature.

It is limited by being ordinal, by combining features, by grading a knee rather than a compartment, and by having been defined on a view that misses early loss. 5.3.5 covers the Rosenberg view.

No figures appear for accuracy on this page, under decision 028, and no grade criteria or grade numbers are reproduced either: the published wording and its thresholds are quantities this environment cannot check.

FRCS Level

How Good It Is

At decision level the grade is useful for describing a population and almost useless for choosing an operation.

A compartment by compartment assessment, with the joint space and the alignment stated, is what an operative decision needs. 5.3.17 covers joint space narrowing.

Unicompartmental and osteotomy decisions depend on the other compartments being preserved, which a single knee grade does not report. 5.16.5 covers osteotomy planning.

A high grade in an asymptomatic knee and a low grade in a disabled one are both common, so the grade cannot carry the indication. 5.1.2 covers clinical examination versus imaging.

The grade's value in practice is as shorthand in correspondence and in audit, and saying so is more honest than using it as a threshold. 5.20.6 covers reporting osteoarthritis.

Where It Misleads

The commonest error is using a grade as an operative threshold. 5.1.2 covers clinical examination versus imaging.

A knee grade is used where a compartment assessment was needed. 5.3.17 covers joint space narrowing.

Grades from extended films are compared with grades from flexed views. 5.3.5 covers the Rosenberg view.

Grade changes are treated as a quantity of progression. 5.21.5 covers the osteoarthritis imaging literature.

A grade is quoted without the view it was assessed on. 5.20.8 covers common reporting errors.

Fellowship Level

Where It Misleads

At fellowship level this system is a mid twentieth century population tool doing work it was never designed for.

It was built to classify populations in epidemiological surveys, and it is used to grade individuals for clinical decisions, which is a different task with different requirements. 5.21.5 covers the osteoarthritis imaging literature.

Several modified versions exist and are not interchangeable, so a grade without its version is ambiguous, and the version is almost never stated. 5.20.1 covers structured reporting.

Reader agreement on the grade is moderate at best and is worst in the middle of the scale, which is where most clinical decisions sit. 5.20.8 covers common reporting errors.

Automated grading reproduces the scale faithfully, including its ordinal nature and its feature combining, so automation improves consistency without improving validity. 5.18.8 covers artificial intelligence in imaging.

The defensible practice is to describe the compartments and the alignment and to give the grade as a convenience, which inverts the usual emphasis. 5.20.6 covers reporting osteoarthritis.