EMET Education
3D classifications
Kellgren-Lawrence classification

Classification · Knee osteoarthritis

Kellgren-Lawrence Classification: 3D Visualization

The five radiographic grades of osteoarthritis, from 0 to IV, shown on a 3D model of the knee. With the original article, reliability data and free infographics for teaching and publication.

Animate in 3D Download the infographic

The 3D model opens in your browser, at the grade shown, with no sign-up. Infographics are free to reuse under CC BY 4.0.

Infographic: the five Kellgren-Lawrence grades on a 3D knee seen from the front in extension, from grade 0 (normal cartilage) to grade IV (medial joint space closed, bone exposed and collapsed), with a zoom on the medial joint line under each grade. — Free to reuse under CC BY 4.0. Download in high definition.

1957Published by Kellgren and Lawrence in the Annals of the Rheumatic Diseases
0–IVFive grades, from no osteoarthritis to severe osteoarthritis
12,640Citations of the original article (OpenAlex, October 4, 2026)
KL ≥ 2The usual definition of radiographic osteoarthritis in epidemiological studies

The five grades

From grade 0 to grade IV, in 3D

Each image is a rendering of the EMET Education knee model at that grade: anterior view, knee in extension, patella shown transparent and menisci hidden so that the joint line can be seen. Wear predominates in the medial femorotibial compartment, the most often involved, with lesser lateral and patellofemoral involvement at the higher grades. The 3D shows what the radiograph infers: cartilage loss behind the narrowing, osteophytes, then bone exposure and collapse.

0None

Radiographic criteria
No radiographic feature of osteoarthritis.
In the 3D model
Intact articular cartilage, no osteophyte.
Animate grade 0 in 3D
IDoubtful

Radiographic criteria
Doubtful joint space narrowing, possible osteophytic lipping.
In the 3D model
Medial compartment: cartilage thinned by about a fifth in the weight-bearing zone, dull surface, early osteophytes about 1 mm high. Lateral and patellofemoral compartments intact.
Animate grade I in 3D
IIMinimal

Radiographic criteria
Definite osteophytes, possible joint space narrowing.
In the 3D model
Medial compartment: cartilage thinned by half in the weight-bearing zone, osteophytes about 2.5 mm at its margins. Early changes in the lateral compartment.
Animate grade II in 3D
IIIModerate

Radiographic criteria
Multiple moderate osteophytes, definite joint space narrowing, some sclerosis, possible deformity of the bone ends.
In the 3D model
Medial compartment: cartilage almost gone in the center of the weight-bearing zone, subchondral bone exposed, osteophytes about 4 mm, narrowed joint space. Lateral and patellofemoral compartments at an early stage.
Animate grade III in 3D
IVSevere

Radiographic criteria
Large osteophytes, marked joint space narrowing, severe sclerosis, definite deformity of the bone ends.
In the 3D model
Medial compartment: bone exposed over a wide area and collapsed (about 2 mm on the tibia, 1 mm on the femur), osteophytes about 6 mm, joint space closed. Lateral compartment worn by half, early patellofemoral wear.
Animate grade IV in 3D

Schematic rendering, for teaching: the model illustrates each grade on a single knee, it does not reproduce a radiograph. Criteria reworded from the original description; the 1957 figures and the 1963 atlas are not reproduced.

The original article

Kellgren and Lawrence, 1957

J. H. Kellgren and J. S. Lawrence published Radiological assessment of osteo-arthrosis in the Annals of the Rheumatic Diseases in 1957 (16:494–502, doi:10.1136/ard.16.4.494). Their aim was epidemiological: to grade, reproducibly, the radiographs of population samples.

The grading rests on the radiographic features they took as evidence of osteoarthritis:

osteophytes at the joint margins or, in the knee, on the tibial spines;
periarticular ossicles, mainly around the interphalangeal joints;
narrowing of the joint cartilage with sclerosis of the subchondral bone;
small pseudocystic areas with sclerotic walls in the subchondral bone;
altered shape of the bone ends.

Each joint received a global grade from 0 to 4: none, doubtful, minimal, moderate, severe. The scale was adopted by the World Health Organization in 1961 as the radiological definition of osteoarthritis for epidemiological studies, and illustrated in the Atlas of Standard Radiographs of Arthritis (1963). As of October 4, 2026, OpenAlex counts 12,640 citations of the original article.

Reliability

How reproducible is the grade?

Interobserver agreement varies widely between studies, from fair to almost perfect, depending on the description used, the radiographic view and the readers' experience.

Interobserver reliability of the Kellgren-Lawrence grade for the knee, as compiled by Alarcón Perico et al., 2024 (reference 6). Coefficients differ between studies and are not directly comparable.
Study	Coefficient	Interobserver agreement
Wing et al., 2021	Intraclass correlation coefficient	0.85
Günther & Sun, 1999	Intraclass correlation coefficient	0.81
Köse et al., 2018	Cohen's kappa	0.47
Schiphof et al., 2011	Cohen's kappa	0.41
Wright et al. (MARS Group), 2014	Intraclass correlation coefficient	0.38
Gonçalves et al., 2016	Krippendorff's alpha	0.33

Two practical findings: experienced readers agree more often with themselves than trainees do, and the radiographic view matters. In the MARS cohort, the six radiographic scales studied, Kellgren-Lawrence included, were more reproducible on a 45° posteroanterior flexion weight-bearing radiograph (Rosenberg view; pooled ICC 0.63) than on an anteroposterior radiograph (0.55), and correlated better with the cartilage damage seen at arthroscopy.

Variants and related scales

Five descriptions, several scales

The wording of the grades has drifted since 1957. Schiphof et al. found five different descriptions of the Kellgren-Lawrence grades in the major cohort studies; for grade 2, one replaced "definite osteophytes and possible narrowing" with "definite osteophyte, unimpaired joint space", another with "minimal osteophytes, possible narrowing, cysts and sclerosis". On 3,071 people of the Rotterdam Study, these variants changed mostly the lowest cut-off (grade ≥ 1), and little the usual cut-offs (≥ 2 and ≥ 3).

When reporting a grade, say which description you used. Other scales grade different things:

References 1, 6, 9, 10 and 11.
Scale	What it grades	Typical use
Kellgren-Lawrence (1957)	A global grade per joint (0 to IV), combining osteophytes, narrowing, sclerosis and deformity.	Epidemiology, trial eligibility, everyday reporting.
Ahlbäck (1968)	Tibiofemoral narrowing up to obliteration of the joint space, then bone attrition.	Advanced disease, before surgery.
OARSI atlas (2007)	Each feature separately (joint space narrowing, osteophytes) in each compartment, on a reference atlas.	Research: follow-up of individual features.
IKDC	Remaining joint space, from A (normal) to D (severely abnormal), on a weight-bearing radiograph.	Sports and knee surgery; best balance of reliability and arthroscopic correlation in the MARS study.
Compartment-specific Kellgren-Lawrence (2024)	A Kellgren-Lawrence grade for each compartment, patellofemoral included.	Describing which compartments are involved (interobserver α 0.51 to 0.56).

Use

In research and in clinical practice

In research

Kellgren-Lawrence grade 2 or higher is the usual definition of radiographic knee osteoarthritis in epidemiological studies, and the grade is a common eligibility criterion or outcome in cohorts and trials, and a reference label for automated grading of radiographs. Its main limit there is the one Schiphof described: unless the description is stated, two studies "using Kellgren-Lawrence" may not grade the same way.

In clinical practice

The grade summarizes the structural severity of the joint; it does not measure pain or disability. Radiographic and clinical osteoarthritis are discordant: in a systematic review, 15% to 76% of people with knee pain had radiographic osteoarthritis, and 15% to 81% of people with radiographic osteoarthritis had pain. The radiograph should not be read in isolation.

In practice, a total knee replacement is usually discussed for advanced structural disease, grade III or IV, when pain and functional limitation persist despite nonoperative treatment; the grade is one element of that decision, not the decision.

Grading pitfalls

Five pitfalls when grading a knee

Grade on a weight-bearing radiograph. Narrowing is underestimated without load. The 45° posteroanterior flexion view (Rosenberg) shows posterior wear of the condyles better and gave more reproducible grades than the anteroposterior view in the MARS study.
Do not dismiss "doubtful" osteophytes. In a 10-year follow-up of women, 62% of knees graded 1 for a doubtful osteophyte developed definite osteophytic osteoarthritis, against 22% of knees without any sign (Hart and Spector, 2003).
State the description you use, especially for grade 2, whose wording differs most between studies.
Say what you graded. The original grade is global, for the whole joint. Grading the worst compartment, or each compartment, is a different measure; the patellofemoral joint needs a lateral or axial view and is not covered by the frontal grade.
Keep the grade on the radiograph. Kellgren-Lawrence is a radiographic scale: it does not apply to MRI, which sees cartilage, menisci and bone marrow directly.

Self-test

Which grade?

Five views of the same knee from the medial side, in random order. Find the grade of each one; the answer comes with the criteria.

Knee in 3D seen from the medial side, at an unknown Kellgren-Lawrence grade, for the self-test.

Which Kellgren-Lawrence grade is this?

Download and cite

Free infographics, for teaching and publication

Use them in a course, a presentation, a social media post or a scientific article, under the Creative Commons Attribution 4.0 license: free reuse, including commercial, provided EMET Education is credited with a link to this page.

Portrait infographic of the five Kellgren-Lawrence grades, one row per grade.

Portrait, 1080 × 1350LinkedIn, Instagram, a slide in portrait. Download · PNG

Landscape infographic of the five Kellgren-Lawrence grades, side by side.

Landscape, 1920 × 1080Presentations, courses, websites. Download · PNG

High-definition landscape infographic of the five Kellgren-Lawrence grades.

High definition, 3840 × 2160Print and figures for publication. Download · PNG

Cite this figure

Ready to paste. For a journal figure, add the original reference (Kellgren and Lawrence, 1957).

Vancouver

EMET Education. Kellgren-Lawrence classification of knee osteoarthritis: 3D visualization [Internet]. 2026 [cited]. Available from: https://www.emet-education.com/en/classifications/kellgren-lawrence/. CC BY 4.0.

APA

EMET Education. (2026). Kellgren-Lawrence classification of knee osteoarthritis: 3D visualization [Infographic]. https://www.emet-education.com/en/classifications/kellgren-lawrence/ (CC BY 4.0)

HTML

<a href="https://www.emet-education.com/en/classifications/kellgren-lawrence/"><img src="https://www.emet-education.com/en/classifications/kellgren-lawrence/img/infographic.webp" alt="Kellgren-Lawrence classification in 3D" width="960"></a><br>Figure: <a href="https://www.emet-education.com/en/classifications/kellgren-lawrence/">EMET Education</a>, CC BY 4.0

Questions

Frequently asked questions

What is the difference between Kellgren-Lawrence grade 2 and grade 3?

Grade 2 has definite osteophytes with at most possible narrowing of the joint space. Grade 3 adds definite narrowing, with multiple moderate osteophytes, some subchondral sclerosis and possible deformity of the bone ends. In practice, narrowing that becomes definite marks the change from 2 to 3.

Which grade defines radiographic knee osteoarthritis?

Grade 2 or higher (KL ≥ 2) in most epidemiological studies. Grade 1 remains debated: its "doubtful" osteophytes are associated with later osteoarthritis, but its definition varies most between descriptions.

Which radiograph should be graded?

A weight-bearing radiograph. The 45° posteroanterior flexion view (Rosenberg) shows the posterior part of the condyles, which an extension view misses, and gave more reproducible grades than the anteroposterior view in a multicenter study.

Does the Kellgren-Lawrence grade correlate with pain?

Only loosely. Depending on the study, 15% to 81% of people with radiographic knee osteoarthritis report pain, and 15% to 76% of people with knee pain have radiographic osteoarthritis. The grade describes structure; it must be read with the clinical picture.

Is the classification only for the knee?

No. Kellgren and Lawrence graded several joints, and the scale is still used for the hip and the hand. This page shows the knee, the joint where it is most used.

Can I reuse these images in a course or an article?

Yes. The images and infographics on this page are under the Creative Commons Attribution 4.0 license: free reuse, including commercial, with credit to EMET Education and a link to this page. The "Cite this figure" box gives ready-to-paste citations.

How were the 3D images made?

They are renderings of the EMET Education knee osteoarthritis model, set to each grade: the model thins the cartilage, grows the osteophytes and collapses the bone according to the grade, compartment by compartment (here medial-predominant), and the femur rests on the tibia, so the wear sets the varus tilt. The patella is shown transparent and the menisci hidden so that the joint line can be seen. It is a schematic model for teaching and patient explanation; it does not reproduce a radiograph.

Sources

Author and references

Page written by Dr. Mikaël Chelli, orthopedic surgeon. Last reviewed October 4, 2026.

Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957;16(4):494–502. doi:10.1136/ard.16.4.494
Kellgren JH, Jeffrey MR, Ball J, editors. The epidemiology of chronic rheumatism. Vol 2: Atlas of standard radiographs of arthritis. Oxford: Blackwell Scientific Publications; 1963.
Kohn MD, Sassoon AA, Fernando ND. Classifications in brief: Kellgren-Lawrence classification of osteoarthritis. Clin Orthop Relat Res. 2016;474(8):1886–93. doi:10.1007/s11999-016-4732-4
Schiphof D, Boers M, Bierma-Zeinstra SM. Differences in descriptions of Kellgren and Lawrence grades of knee osteoarthritis. Ann Rheum Dis. 2008;67(7):1034–6. doi:10.1136/ard.2007.079020
Schiphof D, de Klerk BM, Kerkhof HJ, et al. Impact of different descriptions of the Kellgren and Lawrence classification criteria on the diagnosis of knee osteoarthritis. Ann Rheum Dis. 2011;70(8):1422–7. doi:10.1136/ard.2010.147520
Alarcón Perico D, Camacho Uribe A, Jaimes Niño S, et al. A proposed modification to the Kellgren and Lawrence classification for knee osteoarthritis using a compartment-specific approach. J Exp Orthop. 2024;11(1):e12008. doi:10.1002/jeo2.12008
Wright RW; MARS Group. Osteoarthritis classification scales: interobserver reliability and arthroscopic correlation. J Bone Joint Surg Am. 2014;96(14):1145–51. doi:10.2106/JBJS.M.00929
Rosenberg TD, Paulos LE, Parker RD, et al. The forty-five-degree posteroanterior flexion weight-bearing radiograph of the knee. J Bone Joint Surg Am. 1988;70(10):1479–83. doi:10.2106/00004623-198870100-00006
Ahlbäck S. Osteoarthrosis of the knee. A radiographic investigation. Acta Radiol Diagn (Stockh). 1968;Suppl 277:7–72.
Altman RD, Gold GE. Atlas of individual radiographic features in osteoarthritis, revised. Osteoarthritis Cartilage. 2007;15 Suppl A:A1–56. doi:10.1016/j.joca.2006.11.009
Felson DT, McAlindon TE, Anderson JJ, et al. Defining radiographic osteoarthritis for the whole knee. Osteoarthritis Cartilage. 1997;5(4):241–50. doi:10.1016/S1063-4584(97)80020-9
Hart DJ, Spector TD. Kellgren & Lawrence grade 1 osteophytes in the knee: doubtful or definite? Osteoarthritis Cartilage. 2003;11(2):149–50. doi:10.1053/joca.2002.0853
Bedson J, Croft PR. The discordance between clinical and radiographic knee osteoarthritis: a systematic search and summary of the literature. BMC Musculoskelet Disord. 2008;9:116. doi:10.1186/1471-2474-9-116

3D models

Show the grade to your patient

The images on this page come from an interactive model. In clinic, set the grade and the compartments from your patient's radiograph, then show the treatment discussed.