Classification system
AO/OTA Classification
AO/OTA Fracture and Dislocation Classification
The AO/OTA classification gives every fracture an alphanumeric code in which the first digit is the bone, the second the segment and the letter the fracture type — extra-articular, partial articular or complete articular in an end segment, and simple, wedge or multifragmentary in a shaft — so that 32-A is a simple femoral shaft fracture and 43-C a complete articular pilon fracture.
What the AO/OTA category means
AO/OTA categories describe different patterns, not increasing severity — a higher-lettered or higher-numbered category is not necessarily a worse problem. The groups below are descriptive.
- Type A1–1
Simple in a shaft; extra-articular in an end segment.
- Type B2–2
Wedge in a shaft; partial articular in an end segment.
- Type C3–3
Multifragmentary in a shaft; complete articular in an end segment.
Published criteria and reliability
The published cut-offs that define this instrument, and the measured agreement between observers using it. Each is reported with the population it was derived in.
| Type | Value | Subscale | Population | Follow-up | Source |
|---|---|---|---|---|---|
| Criterion | Segment boundaries are set by the square-of-the-width rule: each end segment occupies a square whose side equals the widest part of the epiphysis | — | Long bone fractures coded from plain radiographs | — | Meinberg EG 2018 |
| Criterion | Diaphyseal types: A simple (two fragments), B wedge (main fragments contact after reduction), C multifragmentary (no contact after reduction) | — | Diaphyseal fractures of the humerus, forearm, femur, tibia and fibula | — | Meinberg EG 2018 |
| Criterion | End segment types: A extra-articular, B partial articular, C complete articular | — | Proximal and distal segment fractures, excluding the proximal femur, proximal humerus and malleolar segment | — | Meinberg EG 2018 |
| Criterion | Agreement is substantial at the type (letter) level and falls progressively at the group and subgroup levels | — | Surgeons and trainees coding fractures from radiographs | — | Audigé L 2005 |
Criterion is a published cut-off, definition or performance figure belonging to a classification or decision rule — it is not a change score and should not be read as one.
How the score is calculated
Stated in full so the calculation can be independently checked, and reproduced by hand if needed.
Each fracture is given a code built from a location and a morphology. The location is two characters: the bone (1 humerus, 2 radius and ulna, 3 femur, 4 tibia and fibula, with further numbers for the scapula, clavicle, patella, pelvis, spine, hand and foot) followed by the segment (1 proximal, 2 diaphyseal, 3 distal, and 4 for the malleolar segment of the tibia and fibula). Segments are separated by the square-of-the-width rule: each end segment occupies a square whose side equals the widest part of that epiphysis, and the diaphysis is what lies between. The morphology is then a type letter, and the letter means different things in different segments. In a diaphyseal segment, A is a simple fracture with two fragments, B is a wedge fracture in which the main fragments still touch once reduced, and C is a multifragmentary fracture in which they do not. In an end segment, A is extra-articular, B is partial articular with part of the joint surface still attached to the shaft, and C is complete articular with the whole joint surface separated from it. Three segments depart from that: in the proximal femur (31) A is trochanteric, B is the neck and C is the head; in the proximal humerus (11) A is extra-articular unifocal, B extra-articular bifocal and C articular; and in the malleolar segment (44) the letters follow the Danis-Weber levels, A infrasyndesmotic, B transsyndesmotic and C suprasyndesmotic. Below the type, the compendium subdivides into groups and subgroups — 32-A1, 32-A2 and so on — and adds universal modifiers for open injury, soft-tissue status, bone loss, segmental involvement and side. Because the forearm has two bones, the 2018 revision writes the bone into the code as 2R or 2U, so a distal radius fracture is 2R3 rather than 23.
Missing data
A code is built from all three elements; a bone and segment without a type is a location, not a classification. Where the imaging does not allow the type to be determined — commonly the B versus C decision in an articular fracture on plain films — obtain the CT rather than guessing, because that letter is what determines the fixation strategy.
Instrument profile
- Developer
- Maurice E. Müller and the AO Foundation, merged with the Orthopaedic Trauma Association classification; current version edited by Eric G. Meinberg, Julie Agel, Craig S. Roberts, Matthew D. Karam and James F. Kellam
- Institution
- AO Foundation, Switzerland, and the Orthopaedic Trauma Association, United States
- First published
- 1990
- Custodian
- AO Foundation and the Orthopaedic Trauma Association, who jointly publish and revise the compendium (last revised 2018)
- Items
- 3
- Response format
- Three selections: bone, segment and type letter
- Score range
- 1 to 3
- Direction
- Categories are not ranked by severity
- Completion time
- About 1 minutes
- Regions
- General Health
- Conditions
- Long bone fracture, Femoral shaft fracture, Tibial shaft fracture, Distal radius fracture, Proximal humerus fracture, Pilon fracture, Ankle fracture, Femoral neck fracture
- Procedures
- Intramedullary nailing, Open reduction and internal fixation, External fixation, Arthroplasty for fracture
- Validity
- The system is a description rather than a prognosis: it encodes where the fracture is and how the bone is broken, which is what registries, trials and implant studies need in order to compare like with like. It does not encode the soft-tissue injury or the patient, which is why the universal modifiers exist alongside it.
- Reliability
- Reliability is good at the bone, segment and type level — the level built here — and degrades at the group and subgroup level, where distinctions become fine enough that experienced observers disagree. Trials that report only the type letter are reporting the reliable part of the system.
When to use the AO/OTA
Well suited to
- Recording a fracture in a registry, a trial or an operation note in a form that another unit will read identically.
- Comparing implant or technique series, where the type letter is the minimum stratification needed for the comparison to mean anything.
- Teaching fracture morphology, because the letters encode the decisions — is the joint involved, do the main fragments touch — that drive fixation.
Limitations and poor fits
- As a treatment algorithm. The code describes the bone; the decision depends on the soft tissues, the patient and the surgeon’s experience, none of which it contains.
- In place of the specific classification a region already has, where that classification carries information the AO/OTA letter does not — Garden or Pauwels for the femoral neck, Schatzker for the tibial plateau, Sanders for the calcaneus, Neer for the proximal humerus.
- Coding to group or subgroup level from memory or from a summary such as this page: use the compendium, because several groups were revised in 2018.
- Paediatric fractures, which are coded by the separate AO paediatric comprehensive classification of long bone fractures.
AO/OTA compared with other instruments
Instruments measuring a related construct, with the practical difference between them.
| Instrument | Items | Range | What it adds |
|---|---|---|---|
| Garden | 2 | 1–4 | The Garden classification grades intracapsular femoral neck fractures into four types by the completeness of the fracture line and the degree of displacement on an anteroposterior radiograph, where types I and II are non-displaced and types III and IV are displaced. |
| Schatzker | 1 | 1–6 | The Schatzker classification sorts tibial plateau fractures into six types by the plateau involved and the fracture pattern, where types I to III involve the lateral plateau and carry a better prognosis, and types IV to VI are higher-energy injuries with medial, bicondylar or metaphyseal-diaphyseal involvement. |
| Neer | 5 | 1–4 | The Neer classification sorts proximal humerus fractures into one-, two-, three- and four-part injuries by counting how many of the four anatomical segments — articular surface, greater tuberosity, lesser tuberosity and shaft — are displaced more than 1 cm or angulated more than 45°, with separate categories for fracture-dislocations and articular surface fractures. |
| Weber | 3 | 1–3 | The Danis-Weber classification sorts ankle fractures into three types by the level of the fibular fracture relative to the tibiotalar joint line — type A below the syndesmosis, type B at it, and type C above it — because the height of the fibular fracture predicts how much of the syndesmosis is torn. |
| Gustilo-Anderson | 4 | 1–5 | The Gustilo-Anderson classification grades open fractures from type I to type IIIC by wound size, degree of soft tissue damage, contamination and vascular injury, and it is the system that drives antibiotic choice, debridement planning and the decision on soft tissue coverage. |
Copyright and permitted use
Published openly because it is the question every clinician and researcher asks before using an instrument, and almost nobody answers it.
The classification system — the coding grammar of bone, segment and type — is published in the peer-reviewed literature and is used universally in fracture description without a licensing programme. The compendium document itself, its diagrams and its full group and subgroup catalogue are copyright the AO Foundation and the Orthopaedic Trauma Association; this page reproduces the coding structure and the type definitions, not the compendium.
References
- Meinberg EG, Agel J, Roberts CS, Karam MD, Kellam JF. Fracture and Dislocation Classification Compendium — 2018. Journal of Orthopaedic Trauma. 2018.
- Marsh JL, Slongo TF, Agel J, et al.. Fracture and dislocation classification compendium — 2007: Orthopaedic Trauma Association classification, database and outcomes committee. Journal of Orthopaedic Trauma. 2007.
- Audigé L, Bhandari M, Hanson B, Kellam J. A concept for the validation of fracture classifications. Journal of Orthopaedic Trauma. 2005.
Editorial status
This page has not yet been signed off by a named clinical reviewer. The scoring logic is stated in full above so it can be independently verified, and every threshold is traceable to its source. See our editorial methodology.