Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

Prospective Analysis of Hip Arthroscopy With 2-Year Follow-up

Prospective Analysis of Hip Arthroscopy With 2-Year Follow-up - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related

Prospective Analysis of Hip Arthroscopy
With 2-Year Follow-up
J. W. Thomas Byrd, M.D., and Kay S. Jones, M.S.N., R.N.
Purpose: Numerous indications, but little outcome data, have been reported for hip arthroscopy. The
purpose of this prospective study is to report the 2-year results of hip arthroscopy performed on a
consecutive series of patients for a variety of disorders. Type of Study: Case series. Materials and
Methods: There were 38 procedures performed on 35 patients who have achieved 2-year follow-up.
All patients were assessed with a modified Harris hip score (pain and function) preoperatively and
postoperatively at 1, 3, 6, 12, and 24 months or until a subsequent procedure was performed.
Variables studied included age, sex, diagnosis, duration of symptoms, onset of symptoms, center-
edge angle, Workers’ Compensation, and pending litigation. Results: Follow-up was obtained on all
patients. The median score improved from 57 to 85 points. This included 10 cases (9 patients) who
underwent a subsequent procedure at an average of 10 months (6 total hip arthroplasty, 1 core
decompression, 3 second arthroscopy) with an index score of 54 compared with 52 at the time of the
second procedure. The median improvement for the following diagnoses was: loose body (34), labral
lesion (27), synovitis (26), chondral injury (18), arthritis (14), and avascular necrosis (211). Of the
variables studied, the most statistically significant finding was that older men with longer duration of
symptoms did worse. Two complications occurred in 1 patient: partial neuropraxia of the lateral
femoral cutaneous nerve and focal myositis ossificans along the anterior portal tract. Conclusions:
Hip arthroscopy can be performed for a variety of conditions (except end-stage avascular necrosis)
with reasonable expectations of success and an acceptable complication rate. This is the first report
to quantitate the results of hip arthroscopy for a heterogeneous population. Key Words: Hip
rthroscopic surgery of the hip is a well-estab-
lished technique. The indications have been well
formulated for both diagnostic and interventional pur-
These indications do continue to evolve as
both skills and technology mature. The basic princi-
ples for carrying out the technique have been de-
Modest variations exist depending on
whether the procedure is performed in the supine or
the lateral position, and arthroscopy has even been
reported without distraction of the joint.
Despite the accepted nature of arthroscopy and the
growing amount of literature to support the indica-
tions, there have been few data reported on the results
of the procedure. Most articles provide anecdotal in-
formation or case reports, or define only a small subset
of the population. This is the first report to quantitate
the results of hip arthroscopy in a consecutive series of
patients for a heterogeneous group of indications.
In March 1993, the authors began tracking all pa-
tients undergoing hip arthroscopy. As of December
1997, data on 121 consecutive cases had been gath-
ered. All procedures were performed supine using
distraction on a modified fracture table as previously
reported by the senior author.
The substance of
From the Southern Sports Medicine & Orthopaedic Center, and
the Department of Orthopaedics and Rehabilitation, Vanderbilt
University School of Medicine, Nashville, Tennessee, U.S.A.
Address correspondence and reprint requests to J. W. Thomas
Byrd, M.D., Southern Sports Medicine & Orthopaedic Center,
2021 Church St, Second Floor, Nashville, TN 37203, U.S.A.
2000 by the Arthroscopy Association of North America
doi: 10.1053/jars.2000.7683
578 Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 16, No 6 (September), 2000: pp 578–587

this study includes those patients who have achieved
2-year follow-up and encompasses 38 procedures per-
formed on 35 patients.
All patients were assessed using a modified Harris
hip score (Table 1). This included an assessment
based on pain (44 points) and function (47 points). A
multiplier of 1.1 provides a total possible score of 100.
The elements of deformity (4 points) and range of
motion (5 points) from the original Harris hip score
were deleted because neither of these are principal
indications for arthroscopy. The score was recorded
preoperatively, and then postoperatively at 1, 3, 6, 12,
and 24 months, or until a subsequent procedure was
performed and the patient dropped out.
The variables studied are outlined in Table 2. Age
and sex were recorded. The diagnosis was noted and a
scheme developed (Table 3) to accommodate for clas-
sification of the various diagnoses that may be en-
countered. Patients may have 1 or more diagnoses.
Duration of symptoms before arthroscopy was re-
corded. The onset of symptoms was noted and defined
TABLE 1. Modified Harris Hip Scoring System
Pain (points)
44 None/ignores
40 Slight, occasional, no compromise in activity
30 Mild, no effect on ordinary activity, pain after activity, uses aspirin
20 Moderate, tolerable, makes concessions, occasional codeine
10 Marked, serious limitations
0 Totally disabled
Function: Gait
11 None
8 Slight
5 Moderate
0 Severe
0 Unable to walk
11 None
7 Cane, long walks
5 Cane, full time
4 Crutch
2 2 canes
0 2 crutches
0 Unable to walk
Distance Walked
11 Unlimited
8 6 blocks
5 2-3 blocks
2 Indoors only
0 Bed and chair
Functional Activities
4 Normally
2 Normally with banister
1 Any method
0 Not able
4 With ease
2 With difficulty
0 Unable
5 Any chair, 1 hour
3 High chair,
⁄2 hour
0 Unable to sit,
⁄2 hour, any chair
Public Transportation
1 Able to enter public transportation
0 Unable to use public transportation
Total Points
3 1.1
Total Score
NOTE. The Harris hip score includes 91 points for pain and function and 9 points for range of motion
and deformity. Arthroscopy is principally indicated for pain and function. Consequently, the section for
range of motion and deformity has been deleted. The multiplier (1.1) is used to give a total possible score
of 100.
TABLE 2. Variables Studied
Duration of symptoms
Onset of symptoms
CE angle
Worker’s compensation
Pending litigation

as follows: traumatic involved a major injury such as
an auto accident, fall from height, hip dislocation;
acute signified a twisting episode or other well-de-
fined event precipitating the acute onset of symptoms;
and insidious meant that there was no injury or
precipitating event, but simply the gradual onset of
worsening symptoms. The center-edge (CE) angle
was recorded on the anteroposterior radiograph in all
cases. If it was a Workers’ Compensation case or if
there was pending litigation, this was noted. Compli-
cations were also recorded.
Statistical Methods
To estimate the overall change in score at 24
months for the entire group, dropouts were assigned a
score lower than the minimum score of those who had
a complete 24-month follow-up and the median score
was used as the measure of central tendency of the
score and change from baseline. The change in score
from baseline for those who had complete follow-up
was adjusted for baseline score [100pD in score/(100-
Baseline Score)] and compared with patient factors
using linear regression analysis. Exploratory graphic
analysis was used to justify the use of linear regres-
sion. This led to using a log transformation of dura-
tions of symptoms in the regression analysis because
of the distribution of this value. A factor was consid-
ered related to the change in score if the coefficient in
the regression analysis was significantly (P , .05)
different than zero.
Univariate analysis of the relation of dropout status
and patient factors was performed using the Fisher exact
test for categorical variables and t tests for continuous
variables. Multiple variable analysis was carried out us-
ing logistic regression analysis to look at combinations
of factors that might predict dropout status.
TABLE 3. Diagnoses for Hip Arthroscopy
Arthritic disorders
Rheumatoid arthritis
Inflammatory arthritis
Osteoarthritis: primary, secondary to inverted labrum, post-
traumatic, secondary to synovial chondromatosis,
secondary to Perthes disease, secondary to dysplasia,
secondary to slipped capital femoral epiphysis
Calcium pyrophosphate disease
Dysplastic disease of the hip (CE angle ,20 )
Borderline dysplastic disease of the hip (CE angle 20 -25 )
Perthes disease
Avascular necrosis of the femoral head
Stage: I, II, III, IV, V, VI
Articular surface: intact, fragmented
Synovial chondromatosis
Total hip replacement
Free fragments
Inflammatory process
Soft tissue impingement
Loosening: acetabular component, femoral component, both
Loose bodies
Avascular necrosis
Synovial chondromatosis
Foreign body
Osteochondritis dissecans
Etiology: rheumatoid arthritis, synovial chondromatosis, gout;
calcium pyrophosphate disease (inflammatory), chemical
induced, idiopathic, traumatic, pigmented villonodular
synovitis, other
Pattern: focal (pulvinar), diffuse
Ligamentum teres damage
Complete rupture
Partial rupture
Degenerate ligament
Chondral damage
Acute traumatic
Chronic traumatic
Grade: I, II, III, IV
Location: femoral head, acetabulum, femoral head and
Labral pathology
Etiology: traumatic, degenerative, idiopathic, congenital,
acetabular dysplasia
Morphology: radial flap, radial fibrillated, peripheral
longitudinal, inverted, unstable
Location: anterior, posterior, lateral, anterolateral
Osteochondritis dissecans
Grade: stable—intact articular surface, fragmented articular
surface; unstable
Perthes disease
With limited range of motion
Without limited range of motion
Not impinging

Follow-up was achieved in 100% of patients (38
cases in 35 patients). Overall, the median score im-
proved 28 points from a preoperative value of 57 to 85
postoperatively. These results included 10 cases (9
patients) who underwent a subsequent procedure at an
average of 10 months. For this group of 10 cases, the
median preoperative score was 54 compared with 52
at the time of the second procedure. Six patients
underwent total hip arthroplasty: 2 for osteoarthritis, 1
for inflammatory arthritis, 1 for rheumatoid arthritis,
and 2 for stage V avascular necrosis (AVN). One
patient with previously documented AVN underwent
arthroscopy for removal of a large loose body and
subsequently underwent core decompression. One pa-
tient, after removal of loose bodies for synovial chon-
dromatosis, developed symptoms from an area of my-
ositis ossificans along the anterior portal tract. At the
time of excision of this mass, second-look arthroscopy
was performed to rule out recurrent intra-articular
disease. One patient underwent 2 subsequent arthro-
scopic procedures for recurrent intra-articular symp-
Among the 38 procedures were 20 left and 18 right
hips. Of the 35 patients, 18 were male and 17 were
female; the average age was 38 years (range, 15 to 84
years). The diagnoses are listed in Table 4. The dura-
tion of symptoms ranged from 1 to 156 months with
an average of 21 months. The onset was traumatic in
7 cases, acute in 15, and insidious in 16 cases. The CE
angle ranged from 19 to 48 with an average of 32 .
Four cases involved Workers’ Compensation claims
and 4 cases had pending litigation.
Statistical Results
The median scores for all procedures are outlined in
Fig 1. The change in score over 24 months for those
who had complete follow-up was significantly related
to their baseline score (Fig 2). The lower the baseline
score, the higher the change in score. There was no
significant difference in the overall change in score
adjusted for baseline between male and female pa-
For female patients, there was no significant rela-
tionship between change in score adjusted for baseline
and any other factor. There was a mild decreasing
trend with respect to duration of symptoms (P 5 .08).
For male patients, there was a significant relation-
ship between age and change in score adjusted for
baseline (P 5 .03) and, if 1 outlier were removed, this
relationship was quite strong (P 5 .002) (Fig 3). There
was also an interaction between age and duration of
symptoms (P 5 .04), indicating that the smallest in-
crease in scores occurred for older men with longer
duration of symptoms.
Among the 38 procedures, 10 patients (26%)
dropped out before completion of the 24-month pe-
riod. The general time course of those followed-up to
TABLE 4. Diagnoses for Hip Arthroscopy
Labral pathology (23)
Chondral damage (15)
Arthritic disorder (9)
Synovitis (9)
Loose bodies (6)
Avascular necrosis (4)
Perthes disease (2)
Synovial chondromatosis (1)
Ligamentum teres damage (1)
FIGURE 1. The graph plots the median score over the period of the
FIGURE 2. An inverse linear relationship is noted between the
change in score from baseline and the baseline score.

completion and those who dropped out is shown in
Figs 4 and 5. Dropout status was not related to any
patient factor.
Results for the most common diagnoses are out-
lined in Fig 6. To summarize, the median improve-
ment for the following diagnoses were: loose body
(34), labral lesion (27), synovitis (26), chondral
injury (18), arthritis (14), and AVN (211). Labral
lesions were also subcategorized as isolated lesions
or occurring in conjunction with accompanying
chondral damage (Fig 7).
The results based on onset of symptoms are detailed
in Fig 8. While the traumatic group seemed to fare
much better than the acute, which was slightly better
than the insidious group, these differences were not
statistically significant.
Results based on whether it was a Workers’ Com-
pensation case or whether there was pending litigation
are shown in Fig 9. Although both of these groups
seemed to respond much better than the group that did
not involve Workers’ Compensation or pending liti-
gation, these findings, again, were not statistically
significant. There was no correlation of results with
CE angle and no trend was noted (Fig 10).
Two complications occurred in 1 patient who un-
derwent removal of multiple loose bodies associated
with synovial chondromatosis. Most of the loose bod-
ies were removed with a grasper introduced along the
anterior portal tract. Postoperatively, the patient had
partial neuropraxia of the lateral femoral cutaneous
nerve characterized by a small patchy area of reduced
sensation in the lateral aspect of the thigh. He also
subsequently developed a focal area of myositis ossi-
ficans within the portal tract. There was some local
irritation from the palpable bony mass because the
patient was quite thin. He later underwent an unevent-
ful excision of the mass as well as a second-look
arthroscopy to rule out recurrent disease.
For group analysis purposes, use of the median
score was selected because it most accurately reflected
the patient groups, incorporating those patients who
FIGURE 4. The individual
scores are plotted for those
cases followed to completion.
The numbers on the plot lines
correspond with the patients’
ID numbers.
FIGURE 3. The relationship is plotted for male patients of the
change in score adjusted for baseline with age.

dropped out during the 24-month period. To delete
these cases from the study at the time they dropped out
would have artificially made the subsequent results
appear better over the period of the study. Thus, by
assuming that the result of a patient in the drop-out
group would be lower than the lowest score for those
who completed the study, the median score provided a
fair and accurate assessment.
As noted in the results, there was a significant
correlation between the amount of improvement and
the patient’s baseline score. Specifically, the lower the
baseline score, the higher the change in score. This
simply reflects that the better the patient’s preopera-
tive score, the less room there was for improvement.
Thus, the score adjusted for baseline was used [100pD
in score/(100-Baseline Score)] for statistical analysis.
This basically reflects the percentage of improvement.
However, for clinical reporting purposes, this is a busy
formula and thus only the modified Harris hip score is
used for reporting clinical results.
The Harris hip rating system is an imperfect instru-
ment for use in this population because the system was
designed to assess hip arthroplasty. Nonetheless, it is
a recognized and accepted method of quantitating
subjective measurements. The group can be roughly
classified using Harris’ original scheme (90-100 ex-
cellent, 80-90 good, 70-80 fair, below 70 poor). Over-
all, for the group completing 2-year follow-up (n 5
28), the preoperative rating included 25 poor, 2 fair,
and 1 good, compared with postoperative results of 5
poor, 2 fair, 5 good, and 16 excellent results. For those
FIGURE 5. The individual
scores are plotted for those pa-
tients who underwent a subse-
quent procedure. The numbers
on the plot lines correspond
with the patients’ ID numbers.
FIGURE 6. Results are plotted for specific diagnoses.
FIGURE 7. Results for labral pathology are subcategorized for
isolated labral damage and cases with labral and chondral damage.

patients who dropped out (n 5 10), there were 9 good
and 1 poor preoperative ratings compared with 8 poor
and 2 good postoperative ratings at the time of drop
For the diagnosis-based classification, the number
of cases is too small and the categories too numerous
to determine statistical differences between the
groups. However, knowledge of the results for a given
diagnosis aids the surgeon in counseling patients on
the projected response to arthroscopic intervention.
The classification scheme presented here (Table 3)
distinguishes major groups as well as detailed sub-
groups. While the numbers presently are small, it is
felt that this system will be useful in the future for
further defining the clinical relevance of diagnosis-
based results.
Loose Bodies
Removal of symptomatic loose bodies is recognized
to be one of the most gratifying of all arthroscopic
procedures. Its role in the hip has been previously
documented in the literature and is supported by the
results of this study with the greatest symptomatic
improvement (34 points).
Labral Lesions
Arthroscopic resection (partial labrectomy) for la-
bral pathology also resulted in symptomatic improve-
ment (27 points). Labral debridement in the presence
of arthritic changes has previously been noted to be
associated with poorer outcomes and this is why labral
lesions were subcategorized as isolated lesions or oc-
curring in conjunction with accompanying chondral
damage. Isolated lesions did better (31 points) than
those with associated articular surface injury (18
points). Although these differences may not have been
statistically significant, the observation is certainly
consistent with those of previous investigators regard-
ing poorer outcomes with accompanying chondral
Arthroscopic debridement was performed for syno-
vial disease and for synovial disease with coexisting
intra-articular pathology. For the group, significant
symptomatic improvement was noted (26 points), re-
flecting that arthroscopy does have a place in the
treatment of synovial disease of the hip as it has been
documented for other joints. It should be noted that
this was a diverse group, including 2 cases each of
rheumatoid arthritis, chemically induced synovitis, in-
flammatory synovitis, and traumatic synovitis, and 1
case of synovial chondromatosis.
FIGURE 8. Results are defined based on the onset of symptoms.
FIGURE 9. Results are distinguished based on whether Workers’
Compensation was involved or litigation was pending.
FIGURE 10. Results are subcategorized based on the CE angle.

Chondral Injury
Patients with chondral injury, as a group, showed an
overall improvement of 18 points. This included a
spectrum of pathology from acute isolated fractures of
the articular surface to diffuse erosive lesions from
degenerative disease. The numbers are still too small
for subclassification to be of use, but over time, it may
provide a better predictive index for defining response
to treatment.
Arthritis is not an arthroscopic diagnosis. It does not
define the extent of articular surface damage, the sta-
tus of the synovium, or accompanying degenerative
changes to either the labrum or ligamentum teres.
Nonetheless, the question is often asked, “How do
patients with arthritis do?” Consequently, arthritis as a
category has been included. It is redundant and over-
laps with other categories but is an accurate clinical
classification. Arthroscopic debridement for arthritis
showed a modest response with an overall improve-
ment of 14 points.
Four patients with AVN of the femoral head under-
went arthroscopy. Three of these had stage V disease
and underwent arthroscopy as a palliative procedure to
try to postpone consideration of total hip arthroplasty.
The results were poor: 2 patients subsequently went
on to hip replacement within 5 months, and 1 is living
with his symptoms, although they have worsened over
time (218 points at 2-year follow-up). A fourth pa-
tient with stage II AVN underwent removal of an
accompanying loose body with symptomatic improve-
ment of 20 points, although she subsequently under-
went core decompression to address the lesion in her
femoral head.
Consequently, in the authors’ opinion, arthroscopy
as a palliative procedure for end-stage AVN is con-
traindicated. However, AVN itself is not necessarily a
contraindication to arthroscopy. Other reports support
the role of arthroscopy, both as a staging procedure
and as a method of addressing coexistent intra-artic-
ular pathology for patients being considered for pro-
cedures to revascularize the femoral head.
Onset of Symptoms
Patients with traumatic onset of symptoms seemed
to do best, and those with acute onset did better than
those with insidious onset. Although the differences
were not statistically significant, it is wise for the
surgeon to keep in mind that patients with a gradual
onset of symptoms and no precipitating event may not
fare as well in their response to surgery. Certainly, a
traumatic or acute event does not assure a good out-
come, nor does an insidious onset preclude a good
We postulate that, with a major traumatic event, the
patient is likely to have had an otherwise healthy joint.
Thus, arthroscopic treatment of a traumatic lesion may
result in more predictable symptomatic improvement.
Conversely, gradual onset of symptoms in the absence
of an injury suggests either a degenerative process, or
some other inherent predisposition to intra-articular
pathology, and a less predictable response. Intermedi-
ate between the 2 is the onset of symptoms resulting
from an acute event such as a twisting injury. While
the event may be sufficient to explain a damaged
structure, the authors speculate that, for many of these
individuals, the injury would have been less likely to
occur in an otherwise healthy joint.
While there was no significance with relationship to
age among the female patients, 1 of the few corollaries
was noted among male patients with increasing age.
Older men showed less improvement. There was also
an interaction with duration of symptoms, which is
discussed below.
Duration of Symptoms
As noted, longer duration of symptoms among male
patients correlated with less successful outcomes.
There was also an interaction of age and duration of
symptoms that was greater than could be explained
solely by the additive effect of the 2 influencing fac-
This certainly suggests that chronic disorders (at
least among men) may respond less reliably to arthro-
scopic intervention. This must still be kept in a clinical
perspective for each individual patient. More recent
onset of symptoms should not lower the threshold for
recommending surgery, whereas long-standing dura-
tion of symptoms may not necessarily preclude a
successful outcome.
It is best for the surgeon to always take a thoughtful,
conservative approach to considering arthroscopic in-
tervention of the hip. Although it is apparent that
arthroscopy offers little for some chronic disorders,
there are also many instances of relatively recent and
often acute onset of hip pain that, given time, will

resolve spontaneously. This is often attributed to tran-
sient synovitis or to other ill-defined but self-limited
CE Angle
CE angle is a radiographic measure of dysplasia. By
the definition used here, only 1 patient had dysplasia
(CE angle, ,20 ) and 7 patients were grouped as
borderline dysplasia (CE angle, 20 to 25 ). By sta-
tistical analysis, there was no correlation between CE
angle and outcome. There was also no trend noted
toward poorer results with smaller angles (Fig 10).
Dysplasia is often cited as a harbinger of poor results
and perhaps even a contraindication to arthroscopy,
but is not supported here. Nonetheless, it is prudent to
take a practical approach to the overall clinical assess-
ment, recognizing that there are many factors involved
in the symptoms associated with the dysplastic hip.
Workers’ Compensation and Pending Litigation
Interestingly, as a group, those cases involving
Workers’ Compensation and those involving pending
litigation did better than those that did not involve
either of these circumstances. Although this was not
statistically significant, it certainly indicates that ar-
throscopy is not contraindicated in these groups and
that a successful outcome can be achieved.
Subsequent Procedure
Undergoing an unplanned subsequent procedure
constitutes a poor result, but in some cases it may not
necessarily mean a clinical failure. Four patients with
degenerative arthritis (2 osteoarthritis, 1 inflammatory
arthritis, and 1 rheumatoid arthritis) subsequently un-
derwent total hip replacement at an average of 12
months. However, in 3 of these cases, because the
radiographs were essentially normal, the disease had
gone unrecognized. The advanced disease, evident
arthroscopically, provided important information in
the subsequent decision to perform total hip arthro-
plasty. Also, 1 35-year-old patient with rheumatoid
arthritis remained improved from his arthroscopic pro-
cedure (21 points) but subsequently chose to undergo
a hip replacement 14 months postoperatively based on
his observation of function in other patients with a hip
replacement. In addition, 3 subsequent procedures (2
total hip arthroplasty, 1 core decompression) were
performed on patients with AVN. These were dis-
cussed previously in this section.
One patient, discussed below, underwent second-
look arthroscopy at the time of excision of an area of
myositis ossificans. The hip joint was clear, and his
residual symptoms were independent of any intra-
articular pathology (74 preoperative and 73 postoper-
ative points at the time of subsequent procedure).
The final case represents a patient who sustained an
acute twisting injury to her hip and underwent arthro-
scopy for a torn portion of the anterior labrum. After
initial improvement, her symptoms deteriorated and a
second arthroscopic procedure defined a partial rup-
ture of the ligamentum teres, which was debrided.
Again, a period of initial improvement was followed
by deterioration in her symptoms, which ultimately
responded to a third and final arthroscopy with more
aggressive resection of the anterior labrum. The short-
comings of this case show how we are still learning to
precisely define the nature of intra-articular pathology
as well as the cause of intra-articular mechanical hip
As noted, 2 complications occurred in 1 patient,
each illustrating an important principle. A previous
anatomic study has shown that branches of the lateral
femoral cutaneous nerve will always pass close to the
anterior portal.
Laceration of the nerve can be
avoided by careful attention to the details of the tech-
nique, but 1 of these branches will always be vulner-
able to traction neuropraxia with vigorous instrumen-
tation from this portal. In this example, once the portal
tract had been developed, hand instruments were in-
troduced along the path to remove multiple loose
bodies resulting from synovial chondromatosis. It is
easy to understand how 1 of the branches of the nerve
lying close to this area could have been contused
during the course of the procedure. The sensitivity of
the lateral femoral cutaneous nerve to compression
injury is also well acknowledged from meralgia par-
esthetica. In this case, as only small branches of the
nerve are usually involved, the patchy area of reduced
sensation is fairly small.
However, in this case, the patient also developed an
area of myositis ossificans along the anterior portal
tract with the lesion measuring approximately 2.25 3
1 3 1.50 cm. The patient was quite thin and the mass
had the mechanical effect of causing pain and irrita-
tion around the anterior aspect of the joint. In fact, it
was hard to distinguish whether it was solely the
lesion that was irritating the front of his hip, or
whether he may have developed symptoms from re-
sidual intra-articular disease. The lesion was excised
uneventfully 23 months after the index procedure, and

a second-look arthroscopy confirmed that the hip joint
was still clear. Since then, the patient has continued to
have resolution of his symptoms.
Myositis ossificans is a well-recognized complica-
tion associated with open hip surgery. It is rarely
considered as a potential complication of arthroscopy.
Thus, this case is an important reminder that, even
though the likelihood of complications in arthroscopy
is quite low, it is not negligible.
Arthroscopic surgery of the hip is a well-established
technique. This report substantiates the benefits of this
procedure for a variety of disorders. Over time, con-
tinuation of the model proposed here will allow us to
better define results and, consequently, the indications
for this technique. It should be noted that, although
arthroscopy may have a role in addressing coexistent
pathology associated with AVN, as a palliative proce-
dure for end-stage disease, it is uniformly unsuccessful.
The complications associated with this procedure,
while not insignificant, are in the authors’ opinion
acceptable for the merits of the procedure in properly
selected patients. This article represents the senior
author’s early experience in hip arthroscopy and, it is
hoped, a point lower on the learning curve than our
present understanding. Potential limitations associated
with these early cases included reliability of investi-
gative studies, ability to properly select patients, un-
derstanding and interpretation of intra-articular pa-
thology as viewed arthroscopically, and the skills and
technology for carrying out operative arthroscopy.
Each of these limitations should be lessened over
time. We have only begun to scratch the surface in
fully defining the indications, contraindications, tech-
niques, and complications associated with this proce-
dure. This article and others like it should serve as an
impetus and model for further investigative study.
Acknowledgment: The authors acknowledge the in-
valuable assistance of George Reed in the statistical analysis
and Sharon Simmons in preparation of the manuscript.
1. Byrd JWT. Indications and contraindications. In: Byrd JWT,
ed. Operative hip arthroscopy. New York: Thieme, 1998;7-24.
2. Sampson TG, Glick JM. Indications and surgical treatment of
hip pathology. In: McGinty J, Caspari R, Jackson R, Poehling
G, eds. Operative arthroscopy. Ed 2. New York: Raven, 1995;
3. Byrd JWT. Hip arthroscopy utilizing the supine position. Ar-
throscopy 1994;10:275-280.
4. Byrd JWT. The supine position. In: Byrd JWT, ed. Operative
hip arthroscopy. New York: Thieme, 1998;123-138.
5. Glick JM, Sampson TG, Gordon RB, Behr JT, Schmidt E. Hip
arthroscopy by the lateral approach. Arthroscopy 1987;3:4-12.
6. Sampson TG, Farjo L. Hip arthroscopy by the lateral approach.
In: Byrd JWT, ed. Operative hip arthroscopy. New York:
Thieme, 1998;105-122.
7. Klapper RC, Dorfmann H, Boyer T. Hip arthroscopy without
traction. In: Byrd JWT, ed. Operative hip arthroscopy. New
York: Thieme, 1998;139-152.
8. Byrd JWT. Hip arthroscopy: The supine position. In: McGinty
J, Caspari R, Jackson R, Poehling G, eds. Operative arthros-
copy. Ed 2. New York: Raven, 1995;1091-1099.
9. Byrd JWT. Diagnostic and operative arthroscopy of the hip.
In: Andrews J, Timmerman L, eds. Diagnostic and operative
arthroscopy. Philadelphia, WB Saunders, 1997;209-224.
10. Byrd JWT. Hip arthroscopy: Principles and application. An-
dover, MA: Smith & Nephew Endoscopy, 1996.
11. Byrd JWT. Hip arthroscopy for post-traumatic loose fragments
in the young active adult: Three case reports. Clin Sport Med
12. Byrd JWT. Arthroscopy of select hip lesions. In: Byrd JWT, ed.
Operative hip arthroscopy. New York: Thieme, 1998;153-170.
13. Farjo LA, Glick JM, Sampson TG. Hip arthroscopy for ace-
tabular labrum tears. Arthroscopy 1997;13:409 (abstr).
14. Hunter DM, Ruch DS. Hip arthroscopy. J South Orthop Assoc
15. Byrd JWT, Pappas JN, Pedley MJ. Hip arthroscopy: An ana-
tomic study of portal placement and relationship to the extra-
articular structures. Arthroscopy 1995;11:418-423.
16. Byrd JWT: Complications associated with hip arthroscopy. In:
Byrd JWT, ed. Operative hip arthroscopy. New York: Thieme,