Afifty-nine-year-old woman with seropositive rheumatoid arthritis and
osteoporosis underwent an elective total hip replacement with use of a
cemented Muller prosthesis (JRI, London, United Kingdom). Aseptic loosening of
the stem occurred three years later, at which time the patient underwent
revision of the stem with a long-stem cemented implant (JRI). Routine
intraoperative culture samples were negative, and the postoperative course was
unremarkable.
Seven years later, the patient sustained an atraumatic unstable
periprosthetic fracture (that is, a Vancouver type-B2
fracture4) of the
shaft of the femur at the distal end of the prosthesis
(Fig. 1). The stem was revised
with use of an uncemented fully hydroxyapatite-coated stem (JRI), which did
not extend distal to the fracture site. It was decided during the procedure
that sufficient mechanical stability at the fracture site was afforded by the
use of cerclage wires.
Two months postoperatively, while rising from a chair, the patient
sustained an additional stable periprosthetic fracture of the femoral shaft
just distal to the tip of the prosthesis (a Vancouver type-B1
fracture4). The
fracture was reduced and was fixed with use of autogenous iliac-crest bone
graft at the fracture site and a Dall-Miles cable-and-plate fixation system
(Stryker-Howmedica-Osteonics, East Rutherford, New Jersey). The femoral
prosthesis was retained. Four months later, the position of the fracture and
hardware was satisfactory and callus was seen radiographically
(Fig. 2-A).
The patient was fully weight-bearing and independent with a cane until she
presented two years later with new pain in the limb and a fever. The white
blood-cell count was 21.2 × 109/L, the erythrocyte
sedimentation rate was 90 mm/hr, and the C-reactive protein level was 221
µg/L. Radiographs demonstrated loosening of the plate and loss of position
at the fracture site (Fig.
2-B). Intraoperatively, an infected nonunion was confirmed on the
basis of the presence of an abscess at the fracture site. All metal was
removed, including the securely bonded hydroxyapatite-coated stem. Multiple
samples were taken for culture, and a thorough débridement was
performed. No organism grew on culture. The patient was managed with
intravenous antibiotic medication and longitudinal skeletal traction for eight
weeks, until she was free of infection as demonstrated by normalization of
temperature, a white blood-cell count of 8.0 × 109/L, an
erythrocyte sedimentation rate of 18 mm/hr, and a C-reactive protein level of
<4 µg/L. Radiographically, the remaining femoral bone was ectatic. The
proximal two-thirds of the femur showed evidence of extensive combined
segmental and cavitary bone loss, and discontinuity could be seen in the
middle part of the shaft (Fig.
3). The femoral abnormalities were categorized as grade 6 and the
bone quality was categorized as level 3 according to the classification system
described by the American Academy of Orthopaedic
Surgeons5.
The femur was reconstructed with use of a long, extensively
hydroxyapatite-coated titanium-alloy stem that was distally locked with two
uncoated screws (Cannulok; Orthodesign, Dorset, United Kingdom). No bone graft
was used. The patient was instructed to walk with toe-touch weight-bearing. At
four months postoperatively, radiographs demonstrated fracture union, with
new-bone ingrowth in all areas of the prosthesis, and the patient was walking
independently with the aid of a walker. Three years after the reconstruction,
the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)
score was 58 points and the Harris hip score was 73 points.
When faced with the combination of periprosthetic fracture, extensive bone
loss, and infection, our goal was to eradicate the infection and then to
choose a reconstructive option that would permit rapid functional
rehabilitation, stable fixation, and reconstitution of bone stock. Our
approach permitted fracture stabilization, fracture union, and sufficient
stability for weight-bearing.
Theoretically, a hydroxyapatite coating facilitates the formation of a
rapid biological bond between the host bone and the implant. Over time, living
bone replaces the hydroxyapatite, and the new bone grows directly onto the
prosthesis with no intervening fibrous tissue
layer6-8.
Furukawa et al.9
implanted hydroxyapatite-coated titanium rods in the distal part of the femur
in fifty rabbits and observed histological evidence of direct osseointegration
in all animals, indicating that this new-bone formation could enhance
stability between the bone and the implant at the site of a fracture repair.
In a similar study, Tisdel et
al.10 showed that
the pull-out strength of hydroxyapatite-coated rods was consistently greater
than that of uncoated implants and concluded that the use of this surface
could enhance component fixation in unfavorable clinical circumstances. Lassus
et al.11
successfully used hydroxyapatite-based bone graft for the treatment of two
chronically infected tibial nonunions in one patient who had sustained a
bilateral fracture, and they reported no recurrence of infection in either
limb. Low reinfection rates have been reported in association with the use of
hydroxyapatite-coated stems for revision total hip arthroplasty following
infection.
In our patient, who had extensive bone loss after multiple revisions, the
last of which had been performed for the treatment of an infected nonunion of
a periprosthetic fracture, stable fixation was achieved with use of a
long-stemmed hydroxyapatite-coated implant that bypassed the fracture site by
4.5 femoral diameters and distal metaphyseal fixation was achieved with use of
interlocking screws.
In our view, a sufficiently long interlocking stem provided a mechanically
stable final construct. Since we did not make use of bone graft at the
fracture site, we believe that it was the hydroxyapatite coating that
facilitated fracture-healing in the presence of the stable mechanical
construct. At the most recent radiographic follow-up, sixteen months
post-operatively, we saw new-bone growth extending along the
hydroxyapatite-coated stem and bridging the gaps in the femoral cortex
(Fig. 4).