Acetabular Rim Reduction for the Treatment of
Femoroacetabular Impingement Correlates With Preoperative
and Postoperative Center-Edge Angle
Marc J. Philippon, M.D., Andrew B. Wolff, M.D., Karen K. Briggs, M.P.H.,
Chad T. Zehms, M.D., and David A. Kuppersmith, B.S.
Purpose: The purpose of this study was to quantify the change in degrees in the center-edge (CE)
angle for each millimeter of acetabular rim resected in hips undergoing arthroscopic acetabular rim
trimming. Methods: Preoperative and postoperative CE angle and millimeters of rim reduction were
prospectively collected in 58 hips that underwent arthroscopic rim reduction. There were 35 women
and 23 men. The mean age was 32 years. The inclusion criterion was hip arthroscopy for femoro-
acetabular impingement in patients without dysplastic hips. Two orthopaedic surgeons made inde-
pendent measurements of the CE angle on preoperative and postoperative anteroposterior pelvis
radiographs. To determine the amount of rim reduction intraoperatively, the lunate surface was
measured with an arthroscopic ruler at the 12-o’clock position before and after rim trimming. The rim
trimming was performed by a single surgeon using a 5.5-mm motorized bur. Results: For the 58 hips
included in this study, the mean rim reduction performed was 3.2 mm (range, 1 to 9 mm). The mean
change in CE angle was 3.9° (range, 0° to 17°). All numbers were normally distributed. By use of
a regression model, the change in the CE angle could be determined by the following formula:
Change in CE angle ? 1.8 ? (0.64 ? rim reduction in millimeters). The interobserver intraclass
correlation coefficient for radiographic measurement of the CE angle was 0.92 (95% confidence
interval, 0.87 to 0.95), indicating excellent interobserver reliability. Conclusions: The amount of
change in the CE angle can be estimated by the amount of bony resection performed at the 12-o’clock
position on the lunate surface in the arthroscopic treatment of femoroacetabular impingement. We
found that 1 mm of bony resection equals 2.4° of change in the CE angle and 5 mm of bony resection
equals 5° of change in the CE angle. Level of Evidence: Level II, diagnostic study.
thology,1,2with increased awareness of the etiology of
hip pain and an ever-expanding amount of literature
ip arthroscopy has become an increasingly used
tool for the treatment of intra-articular hip pa-
on the results of various treatments performed arthros-
copically.3-8The rationale for acetabular rim trimming
is 3-fold: (1) to directly address the offending pathol-
ogy causing the impingement as in pincer-type or
mixed pincer-cam–type femoroacetabular impinge-
ment (FAI), (2) to resect areas of grade IV chondral
damage, and (3) to protect repaired damaged labrum
from further impingement.1,2
In patients with FAI, acetabular overcoverage can
be treated with acetabular rim trimming either ar-
throscopically or through an open procedure in an
attempt to restore the normal anatomy of the femo-
roacetabular articulation.1,2,8-11That pincer-type or
mixed cam-pincer–type FAI can cause labral tears,
chondral damage, and significant disability is well doc-
umented.1-15Excellent results of arthroscopic acetabular
rim trimming in conjunction with other procedures, such
From the Steadman Philippon Research Institute (M.J.P.,
K.K.B., D.A.K.), Vail, Colorado; Nirschl Orthopaedic Center
(A.B.W.), Arlington, Virginia; and Naval Health Clinic (C.T.Z.),
Great Lakes, Illinois, U.S.A.
Recipient of the 2009 Arthroscopy Association of North America
Fellows Essay Award.
Supported by Smith & Nephew, Andover, MA.
Received March 31, 2009; accepted November 4, 2009.
Address correspondence and reprint requests to Marc J. Phil-
ippon, M.D., Steadman Hawkins Research Foundation, Attention:
Clinical Research, 181 W Meadow Dr, Ste 1000, Vail, CO 81657,
U.S.A. E-mail: email@example.com
© 2010 by the Arthroscopy Association of North America
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 26, No 6 (June), 2010: pp 757-761
as osteochondroplasty of the femoral head-neck junction
and labral repair to treat these disorders, have been
Acetabular rim trimming not only can function to
correct the retroversion or coxa profunda causing
pincer- or mixed-type FAI, but it can also protect
compromised or repaired labral tissue and remove
areas of chondral damage.1,2Often, chondral defects
are seen at the lateral margin of the acetabular rim at
the chondral-labral junction. This can be seen as a
result of carpet delamination of the articular cartilage
from labral detachment and subsequent wear, repeti-
tive and/or irregular abutment from cam-type FAI, or
traumatic injury.1,7-11It has been our practice to elimi-
nate these areas at the chondral-labral junction through
acetabular rim trimming where possible.
However, it is difficult to ascertain the limits of
resection of the acetabular rim to avoid what amounts
to iatrogenic acetabular dysplasia. In 1939 Wiberg12
described the center-edge (CE) angle as an assessment
of acetabular dysplasia, noting that CE angles greater
than 25° were normal, angles between 24° and 20°
were borderline normal, and angles less than 20° were
pathologic. This range of normal was subsequently
verified by Fredensborg16in patients aged 15 to 80
years. In Wiberg’s study of 44 patients aged 13 to 60
years, all hips in the pathologic range went on to
develop osteoarthritis in the 4- to 28-year follow-up
period, with the rate of development of osteoarthritis,
as well as the likelihood of subluxation, correlating
with lower CE angles.12A similar study by Cooper-
man et al.14could not replicate the direct correlation of
rate of progression of osteoarthritis and CE angle or
other radiographic measurements of acetabular dys-
plasia, but they reported that most of the 32 hips in
their study followed for 22 years with CE angles less
than 20° went on to develop osteoarthritis.
A 1995 study by Murphy et al.15from Bern, Swit-
zerland, followed the contralateral hips of 286 patients
who had a total hip arthroplasty for acetabular dys-
plasia. They found that no patient in whom the hip
functioned well until the age of 65 years had a CE
angle less than 16°.
More recently, hips with pathologic CE angles have
been associated with labral tears.17In a 2006 study by
Guevara et al.,17the hips of 99 patients with unilateral
surgically confirmed labral tears were compared with the
contralateral hips of 85 of these patients without evi-
dence of labral tears. They reported that the prevalence
of CE angles less than 20° was significantly higher in the
hips with labral tears than in the contralateral hips with-
out labral tears (42% v 16%, P ? .01).
To minimize the risk of creating an undercon-
strained hip or a hip likely to have progression of
osteoarthritis, it may be important to be able to cor-
relate the amount of resection performed with preop-
erative and postoperative radiographic imaging. A
useful and practical predictor of future hip pathology
stemming from acetabular insufficiency is the CE
angle.12The CE angle has been shown to be a predic-
tor of future hip arthrosis.12-15
The purpose of this study was to determine whether
the change in CE angle effected by arthroscopic ace-
tabular rim trimming could be estimated by intraop-
erative measurements. Our hypothesis was that the
change in CE angle from preoperative to postoperative
radiographic measurements could be determined from
the change in the measurement of the lunate surface at
the 12-o’clock position at the time of arthroscopic
acetabular rim trimming.
Between August 2007 and April 2008, 58 patients
were prospectively enrolled in this institutional review
board–approved study. Patients presented for the
treatment of debilitating hip pain. The mean age at the
time of arthroscopy was 32 years (range, 18 to 61
years). There were 23 men and 35 women. There were
22 left hips and 36 right hips. The mean body mass
index was 23 (range, 18 to 32).
Inclusion criteria were hip arthroscopy for the treat-
ment of FAI and rim reduction at the time of hip ar-
throscopy. Exclusion criteria were dysplastic hips, hips
that did not undergo arthroscopic acetabular rim trim-
ming, and hips with inadequate preoperative and/or post-
operative anteroposterior (AP) pelvis radiographs. Inad-
equate films were defined by greater than 2 to 3 cm of
distance between the sacrococcygeal joint and the pelvic
brim and/or increased rotation as evidence by greater
than 1 to 2 cm of lateral shift from the midline in the
distance from the sacrococcygeal joint and the pel-
vic brim. Preoperative AP pelvis radiographs were
evaluated for the CE angle. The CE angle was
measured as the angle subtended between a perpendic-
ular line from the center of the femoral head and the
lateral margin of the acetabulum (Fig 1). This was mea-
sured both preoperatively and postoperatively.
The acetabular depth was measured intraopera-
tively. A graduated intraoperative ruler was used (Fig
2). The measurement was taken by a single surgeon
from the superior aspect of the cotyloid fossa to ap-
proximately the 12-o’clock position of the lunate sur-
face, opposite the transverse acetabular ligament. This
M. J. PHILIPPON ET AL.
measurement was recorded in a prospective database
both preoperatively and postoperatively.
Two orthopaedic surgeons made independent mea-
surements of the CE angles to determine interobserver
reliability of the CE angle. The rim trimming was
performed by a single surgeon using a 5.5-mm mo-
Comparisons between continuous variables were per-
formed by use of the Pearson correlation coefficient.
Comparison of continuous variables by binary categori-
cal variables was performed by use of the independent-
samples t test. A linear regression model with backward
stepwise selection was used to determine the equation
that expressed the dependent variable in terms of con-
stants and the independent variable. The coefficient of
multiple determination (adjusted r2) was used to indicate
how much of the variability in the change in CE angle
was accounted for by the rim reduction in the final
multiple linear regression model. Statistical analysis was
performed by use of SPSS software (version 10.0; SPSS,
The mean preoperative CE angle was 36.4° (range,
25° to 51°) for observer 1. For observer 2, the mean
preoperative CE angle was 36.7° (range, 22° to 51°).
The mean postoperative CE angle was 32.3° (range,
23° to 49°) for observer 1. For observer 2, the mean
postoperative CE angle was 33.1° (range, 22° to 44°).
The interobserver intraclass correlation coefficient for
radiographic measurement of the CE angle was 0.92
(95% confidence interval, 0.87 to 0.95), indicating
excellent interobserver reliability.
The mean preoperative 12-o’clock acetabular depth
was 25 mm (range, 12 to 41 mm). The mean postop-
erative 12-o’clock acetabular depth was 21.8 mm
(range, 11 to 34 mm). The mean rim reduction per-
formed was 3.2 mm (range, 1 to 9 mm). The mean
change in CE angle was 3.9° (range, 0° to 17°) (Table
1). The change in CE angle correlated with the reduc-
tion of acetabular rim (r ? 0.4, P ? .004) (Fig 3).
On the basis of multivariate linear regression, dif-
ference in rim reduction predicted change in CE angle
(r2? 0.3, P ? .003). From this regression, the fol-
lowing equation was obtained to estimate the relation:
Change in CE angle ? 1.8 ? (0.64 ? rim reduction in
at superior aspect of cotyloid fossa (arrow) to approximately 12-
o’clock position of lunate surface, opposite transverse acetabular
Intraoperative view of acetabular depth measurement
angle measurement (A) preoperatively and (B) postoperatively.
The preoperative CE angle was 40°, and the postoperative angle
AP view of hip from a pelvis radiograph showing CE
ACETABULAR RIM AND CENTER-EDGE ANGLE
millimeters). For 1 mm of rim reduction, 2.4° of
reduction in CE angle can be expected (Table 1).
This study showed that the change in CE angle
could be estimated from millimeters of acetabular
bony resection. Furthermore, this study showed that
the CE angle is a reliable measurement with excellent
inter-rater reliability as a radiographic diagnostic tool
for the measurement of acetabular coverage.
Although CE angle is not a perfect means of assessing
acetabular coverage, there is certainly a range below
which the hip is at a higher risk of developing osteoar-
thritis, labral tears, and even instability.12-17It is our
practice to use the guidelines set forth by Wiberg12to
identify dysplastic hips as well as a guide for the amount
of rim resection possible in any particular hip. Thus rim
trimming is not performed in patients with CE angles
with CE angles approaching 25°. However, these num-
bers are not considered in isolation. Other measurements
should also be taken into consideration, including the
weight-bearing surface angle, Sharp’s angle, acetabular
width-to-depth ratio, and femoral head extrusion index,
as well as evidence of lateral or superior subluxation of
the femoral head.17-20False profile views of the pelvis
can also be helpful and have been reported to be more
sensitive than AP pelvis views for the diagnosis of dys-
plasia.13In patients in whom acetabular coverage is of
particular concern, computed tomography scan with or
without 3-dimensional reconstruction can be helpful.
is not part of our routine practice. Thus, although there
are perhaps more complete means of assessing the
amount of acetabular coverage of a hip, the CE angle on
an AP pelvis radiograph is simple and reproducible, with
relatively low radiation exposure and a relatively long
track record of reports in the literature.
The findings in this study suggest that changes from
preoperative to postoperative CE angles on the AP pelvis
radiograph can be predicted by the amount of acetabular
rim trimming performed at the 12-o’clock position. The
regression model performed in this study found that the
following: Change in CE angle ? 1.8 ? (0.64 ? rim
reduction in millimeters). Put more simply, the first mil-
limeter of rim trimming will decrease the CE angle by
approximately 2.4°, with each additional millimeter re-
sected decreasing the CE angle by 0.6 mm to a change of
5° for a 5-mm rim trimming.
The utility of this finding lies in its simple applica-
tion for clinical use. With a carefully measured pre-
operative CE angle on an AP pelvis radiograph, the
surgeon can use these numbers as a rough guideline as
to how the amount of acetabular rim trimming will
affect the postoperative CE angle. This is of para-
mount importance in the treatment of hips in which a
large rim trimming would be desirable. For example,
in a patient with a preoperative CE angle of 35°, a
chondral defect of 5 mm in width at the chondral-
labral junction could be safely eliminated by a 5-mm
rim trimming because the CE angle would be pre-
dicted to decrease to 30°.
As outlined previously, the CE angle is by no means
a flawless predictor of future hip pathology. This
formula provides a method to use the CE angle to
estimate the amount of rim that can be removed. There
are several other limitations to this study. Among
these are variations in the quality of the AP pelvis
radiographs. Although we excluded patients with AP
pelvis radiographs in which the distance from the
lated with reduction of acetabular rim. A circle represents 1 patient,
with each additional cross bar indicating an additional patient at the
data point. The y-intercept is at 1.8 mm, as shown in the formula.
The line, which the formula is based on, is an estimation of the
relation between the change in CE angle and reduction in acetab-
Statistical graph showing change in CE angle corre-
Estimated Changes in CE Angle by Millimeter
of Rim Reduction
Millimeter of Rim Reduction Change in CE Angle
M. J. PHILIPPON ET AL.
pelvic brim to the sacrococcygeal junction was not
between 2 and 3 cm, this still allows for a fairly wide
variation of angles of radiographic projection. Indeed,
a recent report suggests that pelvic tilt can cause
significant variations in determinations of acetabular
angles.21Another limitation of this study is that the
rim reduction was only measured at the 12-o’clock
position on the lunate surface. This only measures 1
aspect of the acetabular rim resection. Although the
vast majority of rim reductions performed at our in-
stitution include the 12-o’clock position, it is some-
times the case that more resection is performed in
areas posterior or, more commonly, anterior to the
12-o’clock position, as in cases of pincer-type FAI
caused by acetabular retroversion. Furthermore, a re-
cent study by Zumstein et al.22has shown the relative
inaccuracy of intraoperative arthroscopic estimation
of the positioning of the resection on the clock face of
the acetabular rim especially for more posteriorly
based rim trimmings. This possible inaccuracy of our
position on the clock face at precisely the 12-o’clock
position was mitigated by our consistent measurement
of the rim trimming directly superior to the superior
aspect of the cotyloid fossa. Similarly, the AP radio-
graphs and the CE angle measurements of these ra-
diographs may not reflect more anteriorly based rim
trimmings because they do not project as the lateral-
most aspect of the acetabulum by which the CE angle
is measured. This likely accounts for some of the data
points in which a significant rim resection was per-
formed with little change in the CE angle.
The findings of this study showed that the amount
of change in the CE angle can be estimated by the
amount of bony resection performed at the 12-o’clock
position on the lunate surface in the arthroscopic treat-
ment of femoroacetabular pincer impingement. We
found that 1 mm of bony resection equals 2.4° of
change in the CE angle and 5 mm of bony resection
equals 5° of change in the CE angle. The following
formula can be used: Change in CE angle ? 1.8 ?
(0.64 ? rim reduction in millimeters).
1. Philippon MJ, Stubbs AJ, Schenker ML, Maxwell RB, Ganz
R, Leunig M. Arthroscopic management of femoroacetabular
impingement: Osteoplasty technique and literature review.
Am J Sports Med 2007;35:1571-1580.
2. Philippon MJ, Schenker ML. A new method for acetabular rim
trimming and labral repair. Clin Sports Med 2006;25:293-297.
3. Philippon MJ, Briggs KK, Kuppersmith DA, Hines SL, Max-
well RB. Outcomes following hip arthroscopy with microfrac-
ture. Arthroscopy 2007;23:e11 (Abstr).
4. Philippon MJ, Briggs KK, Hines SL, Kuppersmith DA, Max-
well RB. Early results of labral repair. Arthroscopy 2007;23:
5. Larson CM, Giveans MR. Arthroscopic management of femo-
roacetabular impingement: Early outcomes measures. Arthros-
6. Philippon MJ, Schenker ML, Briggs KK, Kuppersmith DA,
Maxwell RB, Stubbs AJ. Revision hip arthroscopy. Am J
Sports Med 2007;35:1918-1921.
7. Philippon MJ, Schenker ML, Briggs KK, Kuppersmith DA.
Femoroacetabular impingement in 45 professional athletes:
Associated pathologies and return to sport following arthro-
scopic decompression. Knee Surg Sports Traumatol Arthrosc
8. Ganz R, Parvizi J, Beck M, Leunig M, Notzli H, Siebenrock
KA. Femoroacetabular impingement: A cause for osteoarthri-
tis of the hip. Clin Orthop Relat Res 2003:112-120.
9. Lavigne M, Parvizi J, Beck M, Siebenrock KA, Ganz R,
Leunig M. Anterior femoroacetabular impingement: Part I.
Techniques of joint preserving surgery. Clin Orthop Relat Res
10. Beck M, Leunig M, Parvizi J, Boutier V, Wyss D, Ganz R.
Anterior femoroacetabular impingement: Part II. Midterm re-
sults of surgical treatment. Clin Orthop Relat Res 2004:67-73.
11. Beck M, Kalhor M, Leunig M, Ganz R. Hip morphology
influences the pattern of damage to the acetabular cartilage:
Femoroacetabular impingement as a cause of early osteoar-
thritis of the hip. Bone Joint Surg Br 2005;87:1012-1018.
12. Wiberg G. Studies on dysplastic acetabula and congenital
subluxation of the hip joint. Acta Chir Scand 1939;83:58.
13. Harris WH. Etiology of osteoarthritis of the hip. Clin Orthop
Relat Res 1986:20-33.
14. Cooperman DR, Wallensten R, Stulberg SD. Acetabular dys-
plasia in the adult. Clin Orthop Relat Res 1983:79-85.
15. Murphy SB, Ganz R, Muller ME. The prognosis in untreated
dysplasia of the hip. A study of radiographic factors that
predict the outcome. J Bone Joint Surg Am 1995;77:985-989.
16. Fredensborg N. The CE angle of normal hips. Acta Orthop
17. Guevara CJ, Pietrobon R, Carothers JT, Olson SA, Vail TP.
Comprehensive morphologic evaluation of the hip in patients
with symptomatic labral tear. Clin Orthop Relat Res 2006;453:
18. Sharp IK. Acetabular dysplasia: The acetabular angle. J Bone
Joint Surg Br 1961;43:268-272.
19. Li PL, Ganz R. Morphologic features of congenital acetabular
dysplasia: One in six is retroverted. Clin Orthop Relat Res
20. Delaunay S, Dussault RG, Kaplan PA, Alford BA. Radio-
graphic measurements of dysplastic adult hips. Skeletal Radiol
21. Richards PJ, Pattison JM, Belcher J, Decann RW, Anderson S,
Wynn-Jones C. A new tilt on pelvic radiographs: A pilot study.
Skeletal Radiol 2009;38:113-122.
22. Zumstein M, Hahn F, Sukthankar A, Sussmann PS, Dora C.
How accurately can the acetabular rim be trimmed in hip
arthroscopy for pincer-type femoral acetabular impinge-
ment: A cadaveric investigation. Arthroscopy 2009;25:164-
ACETABULAR RIM AND CENTER-EDGE ANGLE