Total hip arthroplasty in patients less than 40 years old with avascular necrosis of the femoral head. A 5 to 19-year follow-up study.

Abstract

We evaluated 34 total hip arthroplasties (THA) for avascular necrosis of the femoral head in 26 patients who were younger than 40 years at the time of the index operation. The average age at the time of THA was 283 (17-38) years. The mean follow-up was 10.9 (5-19) years. We used 6 cemented and 28 cementless acetabular components and a cemented polished tapered stem in all hips. The mean Charnley modified Merle d' Aubigné Postel score was 8.9 points preoperatively and 16.6 points at final follow-up. Two cemented metal-backed cups were revised because of aseptic loosening at 10 years, one cemented polyethylene cup failed at 12 years and 2 more all polyethylene cups had pending failure. The metal-backed cemented cups and the all polyethylene cups had a longer follow-up than the uncemented cups. Therefore and because of the proven unfavourable long-term outcome of the cemented metal-backed Exeter cup we are not able to draw any firm conclusions regarding the preferred choice between cemented and uncemented cups in this particular group of patients. On the femoral site none of the stems had to be revised. Our results show that a cemented polished tapered femoral component has an excellent survival in primary total hip replacement in young patients with avascular necrosis of the femoral head.

Full text

Page 1

We evaluated 34 total hip arthroplasties (THA) for
avascular necrosis of the femoral head in 26 patients
who were younger than 40 years at the time of the
index operation. The average age at the time of THA
was 28.3 (17-38) years. The mean follow-up was 10.9
(5-19) years. We used 6 cemented and 28 cementless
acetabular components and a cemented polished
tapered stem in all hips. The mean Charnley modified
Merle d’ Aubigné Postel score was 8.9 points pre -
operatively and 16.6 points at final follow-up. Two
cemented metal-backed cups were revised because of
aseptic loosening at 10 years, one cemented polyethyl-
ene cup failed at 12 years and 2 more all polyethylene
cups had pending failure. The metal-backed cement-
ed cups and the all polyethylene cups had a longer
follow-up than the uncemented cups. Therefore and
because of the proven unfavourable long term out-
come of the cemented metal-backed ExeterTMcup we
are not able to draw any firm conclusions regarding
the preferred choice between cemented and unce-
mented cups in this particular group of patients. On
the femoral site none of the stems had to be revised.
Our results show that a cemented polished tapered
femoral component has an excellent survival in pri-
mary total hip replacement in young patients with
avascular necrosis of the femoral head.
Keywords : total hip arthroplasty ; avascular necrosis ;
young patients ; cemented polished stems.
INTRODUCTION
Avascular necrosis (AVn) of the femoral head is
one of the main causes of early end-stage degener-
ative arthritis of the hip in adolescents and young
adults. in developed countries AVn is estimated to
be responsible for 5 to 12% of total hip arthroplas-
ties (THA) (13). The condition may have multiple
aetiologies, including traumatic and non-traumatic
factors (13). The use of steroids and alcohol abuse
account for 90% of the atraumatic causes (27).
Ultimately THA remains the only helpful solution
in advanced stages of AVn, but in the long run the
intervention is associated with higher failure rates
compared with THA in an older population (29,30).
No benefits or funds were received in support of this studyActa Orthopædica Belgica, Vol. 77 - 1 - 2011
Acta Orthop. Belg., 2011, 77, 53-60
Total hip arthroplasty in patients less than 40 years old with avascular necrosis
of the femoral head
A 5 to 19-year follow-up study
Jean-Pierre SiMOn, Pieter BERgER, Johan BELLEMAnS
From the University Hospital Pellenberg, Belgium
ORIGINAL STUDY
I Jean-Pierre Simon, MD, PhD, Professor in Orthopaedics.
I Pieter Berger, MD, Orthopaedic Resident.
I Johan Bellemans, MD, PhD, Professor in Orthopaedics and
Chairman.
Orthopaedic Department, University Hospital Pellenberg,
Leuven, Belgium.
Correspondence : Jean-Pierre Simon, University Hospital
Pellenberg, Weligerveld 1, B-3212 Leuven, Belgium.
E-mail : jean-pierre.simon@uz.kuleuven.ac.be
© 2011, Acta Orthopædica Belgica.

Page 2

Moreover it is believed that there is a difference in
prognosis of THA performed for various aetiologies
of AVn (11,18). Likewise the results of THA for
AVn are less satisfactory compared with THA for
other sources of osteoarthritis (3,6,29,33). The debate
concerning the optimal fixation method of the
implants and the best possible bearing combination
for this young group of patients remains
open (3,28,29,33,34). This study evaluates the clinical
and radiographic results of THA at a medium term
follow-up in patients with AVn of the femoral head
who were younger than 40 years at the time of the
index operation.
MATERIALS AND METHODS
The study was approved by the hospital ethics com-
mittee. Using the institutional database for THA we iden-
tified 31 patients younger than 40 years who received a
THA for AVn of the femoral head (40 hips). Because the
minimum cut off for follow-up was set at 5 years,
5 patients were excluded, leaving 26 patients with
34 THAs for review. There were 13 male and 13 female
patients. The mean age at the time of index surgery was
32 years for male patients, and 24.5 years for female
patients .The are range was from 17 to 38 years. The pri-
mary causes leading to AVn were steroid intake in
13 patients (19 hips), alcohol abuse in 2 patients (3 hips),
trauma in 1 patient (1 hip), AVn following surgery for
slipped capital epiphysis in 4 patients (4 hips), surgery
for an aneurysmal bone cyst in 1 patient (1 hip) and idio-
pathic in 2 patients (3 hips). AVn was also seen after a
combined femoral and pelvic osteotomy for Perthes dis-
ease in 2 patients (2 hips) and repeated surgery following
septic arthritis in 1 patient (1 hip). Three patients had a
previous core decompression of both hips. All operations
were done through a posterolateral approach in a clean
air enclosure using total body exhaust suits. Three types
of uncemented acetabular components were used : 18
MorscherTMcups with a conventional polyethylene
28 mm internal bearing (Zimmer inc, Warsaw, in for-
merly Sulzer Orthopedics, Switzerland), 3 MetasulTM
cups with a 28 mm metal internal bearing (Zimmer inc,
Warsaw, in formerly Sulzer Orthopedics, Switzerland)
and 7 TrilogyTM(Zimmer inc, Warsaw, in, USA) cups
with a 28 mm conventional polyethylene bearing. in the
MorscherTMcup the polyethylene is welded to the exter-
nal metal mesh in order to avoid micromovements and
wear between the mesh and the polyethylene. The same
principle applies to the MetasulTMcup in which the poly-
ethylene is sandwiched and welded between the metal
external mesh and the metal internal bearing. The
TrilogyTM cup bears an internal exchangeable polyethyl-
ene liner. in addition 6 ExeterTMcemented metal backed
acetabular components (Stryker inc, Caen, France) with a
26 mm internal diameter conventional polyethylene were
used. Ten femoral components were polished triple
tapered cemented MS-30TM(Zimmer inc, Warsaw, in for-
merly Sulzer Orthopedics, Switzerland) stems and 24
were polished double tapered cemented ExeterTM(Stryker
inc, Caen, France) stems. Cementing technique for the
femoral component included retrograde injection with a
cement gun of hand mixed antibiotic cement (Simplex
PTMStryker inc) followed by pressurization of the cement
with a SorbothaneTMseal. Post-operatively immediate full
weight bearing was allowed.
Follow-up was at 6 weeks, 3 months and 1 year post-
operatively. Thereafter clinical and radiographic evalua-
tion was performed every 2 to 3 years. The Charnley
modification of the Merle d’Aubigné-Postel (CMA)
grading system was used for clinical evaluation preoper-
atively and during follow-up examination (5,26). Patients
were divided into functional categories A, B or C.
Radiographic assessment was performed by evalua-
tion and comparison of serial anteroposterior radi-
ographs of the pelvis and anteroposterior and lateral radi-
ographs of the hip taken at each visit. The appearance
and evolution of radiolucent lines between the 6 weeks
and follow-up radiographs was evaluated. The classifica-
tion of DeLee and Charnley was used for the cup and the
method of gruen et al to assess the femoral site (9,12).
grading of heterotopic ossification was performed
according to Brooker et al (4). A Kaplan-Meier survival
curve was produced in the analysis of the failure free
rate. Two patients were lost to follow-up at 6 and 9 years
respectively. Revision of both the femoral and acetabular
component were taken as endpoint for the overall sur-
vival probability of the total hip replacement. in addition
survival of both components was calculated separately.
For the evaluation of the clinical scores the Wilcoxon
signed rank test was used. The effects of gender, cemen-
tation, core decompression and functional group were
investigated by use of a Proportional Odds Model. The
significance level was set at p < 0.05.
RESULTS
The mean follow-up was 10.9 years (range : 5-
19, SD 3.9). The median follow-up was 12 years
with 5 implants having 12 years follow-up,
54
J.-P. SiMOn, P. BERgER, J. BELLEMAnS
Acta Orthopædica Belgica, Vol. 77 - 1 - 2011

Page 3

13 implants having more than 12 years follow-up
and 16 implants having less than 12 years follow-up.
Functional outcome
The mean preoperative CMA score for the group
as a whole was 8.9 points. When subdivided in
functional category the mean preoperative CMA
score was 9.3 points for patients with unilateral hip
involvement, 9.1 points for patients with bilateral
hip involvement and 7.9 points for patients with
other conditions affecting gait. The overall post -
operative CMA score for the entire group was
16.6 points, which is a significant improvement of
more than 75% compared with the overall preoper-
ative CMA score (p < 0.001) (Fig. 1). There was no
statistically significant difference in the postopera-
tive CMA score amongst the 3 functional categories
(p = 0.4) (Table i). Similarly, no statistically signif-
icant differences in postoperative CMA scores were
found between males and females (p = 0.9) and
between hips that had previous core decompression
and hips that had not (p = 0.1). The postoperative
CMA score was not different for patients who had a
cemented cup or a cementless cup. When classified
according to aetiology there were no statistically
significant differences in postoperative CMA score
(Table ii).
Survival analysis
Kaplan-Meier survivorship analysis, with revi-
sion of either component as end-point for failure,
demonstrated an overall survival of the THAs of
95% at 10 and 90% at 15 and 19 years post -
operatively. The survival rate of the stems was
100% at 10, 15 and 19 years after the index opera-
tion (Fig. 2). The overall survival rate of the cups
was 100% during the first 9 years. Due to aseptic
loosening of 2 cemented metal-backed cups in the
tenth year, this value dropped to 95% after 10 years
and further to 90% after 12 years due to an
additional all polyethylene cemented cup failure
(Fig. 3). When subdividing the cups between
cemented cups and cementless cups the former had
a survival rate of 100% throughout the whole
Acta Orthopædica Belgica, Vol. 77 - 1 - 2011
AVASCULAR nECROSiS OF THE FEMORAL HEAD
55
Fig. 1. — Charnley modification of the Merle d’Aubigné-
Postel (CMA) pre- and postoperative scores for pain, mobility
and function.
Table i. — Preoperative and postoperative values at latest follow-up of the Charnley modification of the Merle d’Aubigné-Postel
(CMA) grading system. CMA-score for sex and functional group
SexnTHABilateral THAMean Age
(years)
CMA pre-op (S.D.)CMA post-op (S.D.)
Male
Female
13
13
17
17
4
4
31.9
24.6
8.9 (1.8)
8.8 (1.7)
16.9 (1.2)
16.3 (1.1)
group
A
B
C
9
11
6
9
16
9
0
5
3
23.9
30.5
28.8
9.3 (1.6)
9.1 (1.5)
7.9 (2.0)
16.6 (1.1)
16.8 (0.9)
16.2 (1.6)
Total26 348 28.38.9 (1.7)16.6 (1.2)

Page 4

period, while the latter had a drop from 100% to
60% in the tenth year.
Radiographic findings
Only one stem was associated with a radiolucent
line in zone 1 and 2 at 5 years follow-up and it was
mildly progressive at 16 years follow-up (Table iii).
This progression was not sufficient to conclude that
there was any loosening. Calcar resorption in zone
7 was not seen. in 97% of the femoral implants no
radiolucent lines could be detected in the stem-
cement and bone-cement interface at the time of last
follow-up. Two cemented acetabular components
showed definite loosening with a radiolucent line
extending in all zones and progressive migration at
56
J.-P. SiMOn, P. BERgER, J. BELLEMAnS
Acta Orthopædica Belgica, Vol. 77 - 1 - 2011
Table ii. — Causes of AVn leading to THA. There were no significant differences in post-op Charnley modification of the Merle
d’Aubigné-Postel (CMA) values for various aetiologies of AVn
Causen Hips CMA post-op (S.D.)
Steroids131916.3 (1.2)
Trauma11 18.0 (0)
Slipped Epiphysis4417.0 (0.8)
Alcohol23 15.3 (1.5)
LCP2216.5 (0.7)
idiopathic2315.0 (1.0)
Others22 17.0 (0)
Total26 34 16.6 (1.2)
Fig. 3. — Survival analysis of the acetabular component with
95% confidence interval.
Fig. 2. — Survival analysis of the femoral component : 100%
at 19 years follow-up. The upper limit coincides with 100%
survival.

Page 5

10 years of follow-up. This occurred in a patient
who had undergone a single stage bilateral THA.
Both cups were revised to a cementless porous
coated cup. All 4 remaining cemented cups had
radiolucencies at the last follow-up. in addition 3 of
these cups showed mild proximal migration indicat-
ing probable loosening. Two Morscher press-fit
porous cups showed a radiolucent line in zone i.
none of the cementless implants showed migration
or other evidence of probable or definite loosening.
Asymmetric cup wear was seen in 3 cups. no dislo-
cation or loosening had occurred at the last follow-
up.
Complications and revisions
Brooker grade ii ossifications were seen in
2 hips. These lesions were associated with mild
impairment of movement. One Brooker grade iii
occurred and this resulted in pain and restricted
active and passive movement (Table iii).
in one patient redness and swelling appeared in
the scar 6 years after the index operation. This was
treated with a thorough wound debridement. At
latest follow-up there still remained suspicion of a
deep low-grade infection of the implant.
none of the patients developed a deep venous
thrombosis or pulmonary embolism.
no dislocations or periprosthetic fractures were
seen. in 2 patients occasionally a feeling of sublux-
ation was noted. Since all of the femoral compo-
nents survived, no femoral revisions occurred. Two
metal-backed cemented femoral components need-
ed revision after 10 years. Since the 4 remaining
cemented cups showed radiolucent lines and 3 cups
were proximally migrating the imperative conclu-
sion is that revision is almost unavoidable for these
implants.
DISCUSSION
The main concern for THA in juvenile and young
patients is the almost absolute certainty for future
revision. Preservation of the bone stock is therefore
one of the primary concerns in this age group. To
overcome these problems, more durable bearing
materials have been developed. Large diameter
metal-on-metal resurfacing arthroplasty has been
advocated as a bone sparing procedure for young
active patients. With more than ten years follow-up
there is, at the present time, insufficient evidence to
determine whether modern metal-on-metal resurfac-
ing offers advantages over standard total hip arthro-
plasty (16,25). in addition the occurrence of high
metal ion concentrations after implantation of large
diameter metal-on-metal articulations raises serious
concerns for their use in women during child bear-
ing age. A further worry is the reported incidence of
symptomatic and asymptomatic pseudotumours (23).
Finally higher revision rates have been reported
among patients who received modern metal-on-
metal resurfacing arthroplasty (16,25).
The mediocre results for conventional THA in
young patients are mainly due to aseptic loosening
Acta Orthopædica Belgica, Vol. 77 - 1 - 2011
AVASCULAR nECROSiS OF THE FEMORAL HEAD
57
Table iii. — Radiographic results for the acetabular and femoral component and incidence of heterotopic ossification
Radiolucencies acetabular componentHips
Zone i
Probable loosening : proximal migration without progression
Definite loosening : All zones + progressive migration + tilt
Zones ii and iii
5
2
3
1
Radiolucencies femoral component
Zone 1 + 21
Heterotopic ossifications
grade i
grade ii
grade iii
3
2
1