Journal of Orthopaedic Surgery 2000, 8(1): 3337

Evaluation of Letournel and Judet classification

of acetabular fracture with plain radiographs
and three-dimensional computerized
tomographic scan
Boonyarak Visutipol, Pornchai Chobtangsin, Bunyat Ketmalasiri, Narongchai Pattarabanjird and
Namchai Varodompun
Institute of Orthopaedics, Lerdsin General Hospital, Bangkok, Thailand.

ABSTRACT
Letournal and Judet classification of acetabular fracture
is widely used. The classification is based on the
identification of fracture lines on plain radiographs.
Three-dimensional CT scan was claimed to give a
better view of the fracture line.
Our study showed that intraobserver
reproducibility and interobserver reliability were
almost the same when classification was done by using
plain radiographs and 3D-CT scan. And 3D-CT scan
did not increase either the interobserver reliability or
the intraobserver reproducibility in classifying the
fracture.
Key words: Letournal and Judet classification, acetabular
fracture, three dimensional CT scan, intraobserver
reproducibility, interobserver reliability

INTRODUCTION
Letournel and Judet classification (L-J classification)5,10
of the acetabular fracture is widely used (Fig. 1). Plain

radiographs in AP, iliac and obturator oblique5,13 are

commonly used to evaluate the fracture lines in each
area of the pelvis and acetabulum. However, due to
the complexity of the fracture, the accuracy of the
diagnosis is in doubt.
It is claimed that two-dimensional computerized
tomography (2D-CT scan) (Fig. 2) gives more detail of
the fracture lines and might give more accuracy in
classification.6,7,11 However, there are some difficulties
because a lot of images have to be viewed. Modern
three-dimensional computerized tomography (3D-CT
scan) using a software program to create threedimensional images from 2D-CT scan is said to give
better details of the fracture lines and gives more
accuracy in classifying the fracture (Fig. 3).1,2,4
One way to measure the accuracy and efficiency
of a classification is to measure the intraobserver
reproducibility and interobserver reliability, which is
the measurement of the agreement of one observer at
different periods of time and among different
observers, respectively.
To our knowledge, there is no study which
compares the use of plain radiographs and 3D-CT scan
to classify acetabular fracture in terms of intra and
interobserver variation.

Address correspondence and reprint requests to: Dr Boonyarak Visutipol, Lerdsin General Hospital, Silom Road, Bangkok, 10500
Thailand.

34

Journal of Orthopaedic Surgery

Boonyarak Visutipol et al.

Figure 1

L-J classification which was divided into 10 subtypes.

Figure 2

2D-CT scan

Vol. 8 No. 1, June 2000

Evaluation of Letournel and Judet classification of acetabular fracture

35

interval. The serial numbers of the radiographs were

changed before the second review.
The 3D-CT scans, which were created from 3 mm
cut interval 2D-CT scans, from another set of 20
patients, were reviewed in the same manner as the
plain radiographs.
Then intraobserver reproducibility and
interobserver reliability were evaluated by using the
Kappa test introduced by Cohen. 3 The Kappa
coefficient (K), 3 which is the level of agreement
corrected by proportion of agreement by chance, was
calculated using an SPSS 7.5 software program and
interpreted by using Landis and Koch guidelines:9
values of less than 0.00 indicate poor agreement; 0.00
to 0.20, slight agreement; 0.21 to 0.40, fair agreement;
0.41 to 0.60, moderate agreement; 0.61 to 0.80,
substantial agreement; and more than 0.80, excellent
agreement.

RESULTS

Figure 3
views.

3D-CT scan: The images can be created in different

The objectives of this study were:

1. To find the intraobserver reproducibility and
interobserver reliability of L-J classification with
the use of plain radiographs (AP, iliac and
obturator oblique) and 3D-CT scan.
2. To evaluate if 3D-CT scan could increase the
intraobserver reproducibility and interobserver
reliability from the use of plain radiographs.

MATERIALS

AND

METHODS

20 sets of plain radiographs (AP, iliac and obturator

oblique) from 20 patients who had fractures of the
acetabulum were studied. The names, hospital number
and date of X-ray were covered. The radiographs were
marked with serial numbers from 1 to 20.
These radiographs were reviewed by 5 orthopaedic
surgeons. Before the review, all the observers were
supplied with information and diagrams of the L-J
classification of the acetabulum. The sets of plain
radiograph were reviewed twice with a 2-month

From our study, intraobserver reproducibility of the

plain radiographs and 3D-CT scan were almost the
same (0.42 for plain radiographs and 0.44 for 3D-CT
scan), and both were in a moderate level of agreement
(Table 1). Interobserver reliability in both groups were
0.24, which was in the fair level of agreement. Therefore
3D-CT scan did not help increasing the reliability (Table
2).
Table 1
Intraobserver reproducibility
Kappa coefficients for intraobserver reproducibility
Observer

K plain radiograph

K 3D-CT scan

0.33

0.47

0.59

0.38

0.24

0.46

0.56

0.59

0.39

0.30

mean Kappa for plain radiographs = 0.42 (0.24 to 0.59)

mean Kappa for 3D-CT scan
= 0.44 (0.30 to 0.59)

Table 2
Interobserver reliability
Kappa coefficients for interobserver reliability
Plain radiography

mean K 0.24 (0.15 to 0.37)

3D-CT scan

mean K 0.24 (0.30 to 0.46)

36

Journal of Orthopaedic Surgery

Boonyarak Visutipol et al.

DISCUSSION
Plain radiography in AP and both oblique views of
the pelvis, as described by Judet et al.,8 has long served
as the standard for diagnosis of acetabular fractures.
Modern computerized scan in 2 dimensions has been
suggested to be superior in the detection of fracture
involving the sacrum, quadrilateral surface, acetabular
roof, posterior acetabular lip and the abnormality of
the joint space.6,12 The disadvantage is that many
images have to be viewed. 3D-CT scan, which can be
rotated in different views and some parts can be erased
for a better vision, was claimed to be more useful in
characterizing complex displacement of the
acetabulum.1,2
Our study was not designed to evaluate the
efficacy of the 3D-CT scan in identifying the fracture
lines. The objective of our study was to evaluate the
level of agreement of one observer and among different
observers in classifying acetabular fracture using L-J
classification.
From our study, moderate intraobserver
reproducibility (K = 0.42) and fair interobserver
reliability (K = 0.24) were observed when the
classification was done using plain radiographs. When
the classification was done using 3D-CT scan, the
intraobserver reproducibility increased slightly but
remained in the moderate level of agreement (K = 0.42)
while the interobserver reliability remained the same
(K = 0.24). We think the reasons why both the
intraobserver reproducibility and interobserver
reliability were low in L-J classification is because the
classification is too complicated (there are 10 types)
and too difficult to differentiate one type from the other
especially in complex fracture groups. Though 3D-CT

scan might have given a better detail of the fracture

line, it did not help to improve either the intraobserver
reproducibility or the interobserver reliability. This
suggested that even when the fracture lines are more
clearly viewed, the classification is so complicated that
an accurate classification cannot be made.
Among the observers, observer number 4 who is
specialized in trauma seems to have the highest Kappa
both in plain radiographs and 3D-CT scan. This
suggests that the level of experience might be an
important factor in determining the L-J classification.
The effect of experience on determining L-J
classification needs further study. However, even with
the most experienced surgeon, the Kappa coefficient
was still in the moderate level of agreement which,
again, reflects the difficulty of the classification.
The question is whether it is really necessary to
differentiate one subtype from another in complex
fracture groups when all of them require an extensive
approach and the prognosis depends on the
comminution and the degree of reduction. Thus,
simplification of the classification may be helpful in
improving the intraobserver reproducibility and
interobserver reliability.

CONCLUSION
Intraobserver reproducibility and interobserver
reliability were found to be low in L-J classification
and 3D-CT scan did not seem to increase them. The
results of this study suggest that Letournel and Judet
classification might be too complicated and too difficult
for the average orthopaedic surgeon.

REFERENCES
1. Billet FPJ, Schmitt WGH, Gay B. Computed tomography in traumatology with special regard to the advances of threedimensional. Arch Othop Trauma Surg 1992, 111:1317.
2. Bruk DL Jr, Mears DC, Kennedy WH. Three-dimensional computed tomography of acetabular fractures. Radiology 1985,
155:1836.
3. Dunn G. Design and analysis of reliability studies: The statistical evaluation of measurement errors. London: Edward Arnold,
1989.
4. Gautsch TL, Johnson EE, Seeger LL. True three dimensional stereographic display of 3D-reconstructed CT scans of the pelvis
and acetabulum. Clin Orthop 1994, 305:13851.
5. Guyton JL. Fracture of hip, acetabulum, and pelvis. In: Crenshaw AH, ed. Campbells Operative Orthopaedics. 9th ed. St
Louis: CV Mosby, 1998, 223452.
6. Harley JD, Mack LA, Winquist RA. CT of acetabular fractures. AJR 1982, 138:4137.
7. Hayes WC, Balkisson RAA. Current concepts in imaging of the pelvis and hip. Orthop Clin North Am 1997, 28:61741.
8. Judet R, Judet J, Letournel E. Fracture of the acetabulum: Classification and surgical approaches for open reduction: Preliminary
Report. J Bone Joint Surg [Am] 1964, 46:161546.
9. Landis JR, Koch GG. The measurement of observers agreement for categorical data. Biometrics 1977, 33:15974.
10. Letournel E. Aectabular fractures: Classification and management. Clin Orthop 1980, 151:81123.

Vol. 8 No. 1, June 2000

Evaluation of Letournel and Judet classification of acetabular fracture

37

11. Rafii M, Firooznia H, Golimber C, Waugh T Jr, Naidick D. The impact of CT in clinical management of pelvic and acetabular
fractures. Clin Orthop 1983, 178:22835.
12. Shirhoda A, Brashear HR, Staab EV. Computed tomography of acetabular fractures. Radiology 1980, 134:6838.
13. Tile M. Fracture of the acetabulum. In: Rockwood CA, ed. Rockwood and Greens Fractures in Adults, 4th ed. Philadelphia:
Lippincott-Raven Publishers, 1996, 161758.