Introduction

In December of 1993, A.H. Burstein⁵ in his Editorial "Fracture Classification Systems: Do They Work and are They Useful?" stated that a fracture classification must be both functional and useful. To be functional the classification system must have a high degree of "...intraobserver reliability or repeatability and interobserver reliability". To be useful the system should "...help the surgeon to choose an appropriate method of treatment for each and every fracture ... and should provide the surgeon with a reasonably precise estimation of the outcome of that treatment".

We agree that the initial AO Classification¹ which relied on the observer having to make three choices between three possible answers while determining the fracture Type, Group and Subgroup, led a times to confusion and disappointments, particularly when surgeon attempted to go beyond the fracture Type⁶. In order to overcome these difficulties, we developed in 1995 a binary system of questions. These binary questions take the surgeon in a logical progression to the right answer. The reader will find in both pamphlets a detailed description of this binary system of questions. If one is not able to arrive at a clear choice when posing the questions between the only two available options, then it is clear that one does not have enough information to classify the fracture. This suggests the need for further X-ray projections or even a CT scan to obtain the necessary information.

The Comprehensive Classification of Fractures = CCF² was used in the 3rd ed. of the Manual of Internal Fixation³ as the basis for the choice of suggested treatment.

The CCF has other important advantages:

it is a classification system which is not based on regional anatomical features but one which can be applied to all fractures of all bones
in most instances it indicates the severity of the fracture and the likely outcome of its treatment
its system of questions with suggested possible answers in determining the Subgroups and Qualifications will suggest fracture detail to be looked for at surgery
it is a stimulus to learning and teaching and to the evaluation of outcome of fractures treated by both closed and surgical methods
it is useful for research and for publication
the new computer program⁴ based on the CCF for long bones as well as for fractures of the pelvis and acetabulum with CT and three dimensional reconstruction will be of great help to all centres treating complex musculoskeletal trauma

The format of "The Comprehensive Classification of Fractures" has changed. The contents is almost identical. We have come to the conclusion, however, that the interpretation of X-rays and schematic representations of fractures can vary from one observer to the other. We have even discovered that the same observer after a lapse of time can come to a different diagnosis on the basis of the same X-ray.

We hope that the new format which we have developed will overcome these inter- and intra-observer errors and that with this new system all surgeons will be able to classify with consistency all fracture Types and Groups. This has come about because the determination of the fracture Type and Group is now accomplished with a binary system of questions each with only two options for an answer instead of the old system where the surgeon had to choose one from three schematic representations of the fracture Types and Groups.

If the choice between two options cannot be made, then it might be necessary to consult the glossary for an exact definition. If this does not yield an answer, order a new X-ray and rarely a CT scan. In most cases, however, it is not difficult to establish the correct fracture Type and Group in order to initiate the correct treatment.