Abstract
Background: The technique of extracortical bone-bridging and
ingrowth fixation with a porous coating over the shoulder region of the
implant and augmentation by autogenous bone-grafting was introduced to improve
the longevity of implant fixation. The potential advantages of this technique
are that new-bone formation across the bone-prosthesis junction may share
stress and may prevent osteolysis by sealing off this critical region against
the infiltration of wear particles. The objectives of this study were to
examine the prevalence of stem-loosening with use of the extracortical
bone-bridging and ingrowth technique, the amount of bone formation over the
porous-coated region of this prosthesis, and the characteristics of bone
formation over the porous-coated region and adjacent bone.
Methods: Forty-three patients who had prosthetic reconstruction with
the extracortical bone-bridging and ingrowth technique from 1976 to 1990 were
included in this retrospective study. The mean length of follow-up was 9.7
years (range, two to twenty-one years). All but one patient were managed with
autogenous bone graft; five, with allograft and autograft; and one, with
allograft only. Extracortical bone formation was measured over a 2-cm length
of the porous-coated region of the prosthesis in four zones (the medial and
lateral aspects on anteroposterior radiographs and the anterior and posterior
aspects on lateral radiographs) and was reported as the percentage of the
total length (8 cm) covered by extracortical bone with a thickness of >1
mm. The Spearman rank coefficient was used to assess the correlation between
pairs of continuous variables.
Results: The final average percentage of the porous-coated region
that was covered by extracortical bone formation was 76% ± 34% for all
patients and anatomical sites of reconstruction. Use of bone cement was
associated with less bone formation (p = 0.04), and this value remained lower
at the final measurement (p = 0.06). One stem had aseptic loosening, but no
sign of osteolysis was found. The radiographic appearance of the bone
formation had stabilized at two years of follow-up. All patients with
allograft augmentation had greater bone formation. The amount of extracortical
bone formation did not differ in relation to the type of porous coating,
anatomical sites, pathological disorder, sex or age of the patient, or length
of reconstruction.
Conclusions: As shown by the low prevalence of stem-loosening (two
of fifty-six stems or one of forty-three patients), the use of the
extracortical bone-bridging and ingrowth fixation technique is associated with
improved stem fixation in segmental bone-replacement prostheses applied for
limb salvage. In the demanding biomechanical environment and with the risk of
stress and particle-related bone resorption, the extracortical bone-bridging
and ingrowth fixation is an attractive method to provide long-lasting implant
fixation.
Level of Evidence: Therapeutic study, Level IV (case
series [no, or historical, control group]). See Instructions to Authors for a
complete description of levels of evidence.
Advances in musculoskeletal oncology have resulted in increased survival
and improved limb-salvage procedures. Efforts to develop durable
reconstructive procedures to match the improved life expectancy of this
patient population continue. Prosthetic replacement is currently the most
commonly used reconstruction procedure after tumor resection. However, aseptic
loosening of the implant, fracture of the components, and the potential
biologic reaction to metallic and polyethylene wear particles remain long-term
problems1, with a
35% to 40% failure rate at five to ten years of
follow-up2,3.
A "composite" fixation technique, defined as extracortical
bone-bridging and ingrowth
fixation4-6,
was developed to improve long-term implant fixation. The use of autogenous
bone-grafting to induce the formation of extracortical bone-bridging across
the porous-coated implant shoulder and host bone followed the well-established
osteogenic principle of the use of the onlay bone graft introduced by
Phemister7 in 1947.
In addition to the clinical
trial5,6
of the extracortical bone-bridging and ingrowth implant fixation technique, a
number of animal experiments were carried out to study the effect of this
prosthetic fixation
method6,8,9.
Theoretical analysis and bench tests were performed to maximize its
biomechanical advantages and to provide implant design
guidelines2,5,10.
In segmental bone and joint replacement, substantial bending moments during
load-bearing can create severe tensile stresses at the stem fixation
site11. In addition
to load-sharing by the extracortical bone formation, this collar of connective
tissue (Fig. 1) confines the
metallic wear particles within the capsule and reduces the risk of osteolysis,
which can lead to bone resorption and
stem-loosening1,12,13.
Other benefits of this fixation method are that (1) cement provides secure
initial fixation to enhance bone-graft incorporation and allows early
weight-bearing and range of motion, (2) a solid stem increases implant fatigue
strength, (3) fewer stem sizes facilitate the design of a modular system, and
(4) the implant is easier to remove and revise, if necessary.
The main goal of this retrospective study of a consecutive series of
patients from one institution with minimal variation in implant design was to
assess the efficacy of the extracortical bone-bridging and ingrowth fixation
technique in providing long-term stable stem fixation. In addition to analysis
of the prevalence of stem-loosening after use of the technique, the amount of
bone formation over the porous-coated region of the prosthesis and the
characteristics of bone formation over the porous-coated region and adjacent
bone were evaluated.
Study Group
From 1976 to 1990, fifty-nine patients at the Mayo Clinic underwent
limb-salvage surgery with segmental replacement prostheses with use of the
extracortical bone-bridging and ingrowth fixation technique. When patients
with less than two years of follow-up were excluded, forty-three patients
remained in the series. There were twenty-six women and seventeen men. The
mean age was thirty-six years (range, fourteen to seventy-four years) at the
time of surgery. Thirty-four patients had the operation because of a
neoplastic condition and nine, because of a non-neoplastic condition. The mean
age of the patients who had a tumor was thirty-one years, and the mean age of
those without a tumor was fifty-four years. The femur was affected in
thirty-eight patients (twelve had involvement of the proximal part; three, the
diaphysis; and twenty-three, the distal aspect); the tibial diaphysis, in two;
and the proximal part of the humerus, in three.
Nineteen patients had reconstruction after a primary tumor resection,
fifteen had revision of a prosthesis implanted after a previous tumor
resection, and nine had a revision of a prosthesis after a procedure that was
not tumor-related (see Appendix). In the tumor group, the diagnosis included
osteosarcoma (seventeen patients), giant-cell tumor (seven), chondrosarcoma
(five), Ewing sarcoma (two), and malignant fibrous histiocytoma,
dermatofibroma, and lymphoma in one patient each. In the group without a
tumor, the operations were done for the treatment of congenital hip dysplasia
(four patients), Gorham disease (two), and degenerative joint disease, trauma,
and rheumatoid arthritis in one patient each.
Because this retrospective review spanned fifteen years, two types of
porous-coated segmental implants were used. One system (Styker/Howmedica,
Allendale, New Jersey) was based on a cobalt-chrome-molybdenum cast material
with a beaded porous surface, and the other system (Zimmer, Warsar, Indiana)
used titanium alloy (Ti-6Al-4V) with pure titanium fiber-metal porous
coating2,5,10.
Both systems contained a porous coating at the shoulder of the segmental
portion of the prosthesis to enhance extracortical bone-bridging. Prosthetic
components were used as a single (custom-made) or modular system
(Fig. 2). Bone cement was used
for stem fixation in all but seven patients who had a press-fitted
porous-coated stem. Two of these seven patients had metal plates applied to
achieve added implant stability. Titanium-fiber-metal-coated prostheses were
used in twenty-six patients, and cobalt-chrome-molybdenum implants with beaded
porous coating were used in seventeen. All but one patient had autogenous
iliac bone grafts in particle or strip form applied at the bone-prosthesis
junction, five patients were managed with a combination of autogenous iliac
bone and banked allogeneic bone, and one patient had allograft only.
The duration of follow-up was calculated from the time of surgery to the
time of the latest evaluation. The data on the series were updated with a
review of the clinical charts and the most recent radiographs. Patients who
were followed elsewhere had radiographs sent to our institution for review.
Information on all patients who had the operation more than ten years before
the time of the review revealed that twenty-five were alive, four had died,
and two were lost to follow-up (at 5.5 and 6.5 years). The mean length of
follow-up was 9.7 years (range, two to twenty years).
Clinical and Radiographic Analysis
Clinical data, including implant-related complications, were assessed. In
addition, measurements of the implant-bone junctions were carried out on both
the sequential anteroposterior and mediolateral radiographs that clearly
visualized the amount and extent of new-bone formation around the
porous-coated prosthetic shoulder region. The amount of extracortical bone
formation was measured along the porous-coated shoulder region for a length of
2.0 cm from the implant-bone junction line in four zones (the medial and
lateral aspects on anteroposterior radiographs and the anterior and posterior
apsects on lateral radiographs). The length of the new bone formation in this
region with a thickness of >1 mm that was directly juxtaposed to the porous
implant surface was measured and expressed as a percentage of the total length
(8 cm) of the four zones (Fig.
3). If one side of the 2.0-cm length in the critical zone in one
projection was covered with new bone, it was counted as 25% coverage of the
total length of the porous-coated region. Therefore, when all four zones were
covered with bone, the amount of bone formation over the porous-coated region
was 100%. In each zone, partial coverage of the defined length was also
included. In intercalary reconstruction or in knee joint arthrodesis, bone
formation at both the proximal and the distal site was determined. For the
proximal part of the femur, the fixation site of the intercalary prosthesis
was classified as distal and, for the distal part of the femur, the fixation
site (of the intercalary prosthesis) was defined as proximal in the overall
analysis.
Statistical Analysis
Data on extracortical bone formation were analyzed for each of four groups:
(1) all patients—the average of the proximal and distal measurements,
(2) patients with both proximal and distal measurements—the proximal
measurements only, (3) patients with both proximal and distal
measurements— the distal measurements only, and (4) patients with only
one measurement at each time in the sequential follow-up. The sex, age at the
time of surgery, anatomical site of the reconstruction, type of neoplasia, use
of chemotherapy, length of implant, type of porous coating, use of bone
cement, and duration of follow-up were univariately assessed for associations
with extracortical bone formation at two years and at the time of the last
follow-up, and the change in extracortical bone formation between the two-year
and the last follow-up evaluation was determined. A significance level of p
< 0.05 was used for all tests.
The Spearman rank coefficient was used to assess the correlation between
pairs of continuous variables. The Wilcoxon rank-sum test or the
Kruskal-Wallis test was used to test for group differences in continuous
measurements. The changes in extracortical bone formation between the two-year
and the final follow-up evaluation and the differences between the proximal
and distal graft sites in patients who had diaphyseal bone-segment replacement
were assessed by the Wilcoxon signed-rank test.
Stem-loosening occurred in only two of the fifty-six stems implanted in the
forty-three patients. Both stems that became loose were in one patient (Case
19) who had had a knee arthrodesis; no extracortical bone had formed around
either the proximal or the distal stem because of the poor grafting technique.
After revision and repeat grafting, extracortical bone formation was 0%
proximally and 100% distally with no loosening at sixty months
postoperatively. Stem-loosening did not occur in any other patient.
The quality of extracortical bone formation was usually uniform along its
length and had the appearance of cortical bone radiographically. There were
radiolucent zones between the extracortical new bone and the porous surface of
the implant body, which mimicked the intracortical marrow space in many cases.
Bone formation was invariably associated with a continuous layer of new bone
extending from the adjacent cortex over the full length of the porous region
of the prosthesis. In segmental bone replacement involving the femur, the
extent of new-bone formation was more pronounced in the medial and posterior
aspects of the long bone.
Although the main objective of this study was to evaluate the outcome of
stem fixation with use of the extracortical bone-bridging and ingrowth
fixation technique, a radiographic assessment of the segmental prosthesis and
its interface with bone was also carried out. Most patients had no radiolucent
lines around the stem, but nine patients had incomplete radiolucent lines of
<2 mm in thickness, and three patients had radiolucent lines of >2 mm in
thickness but not completely around the stem. None of these radiolucent lines
led to loosening of the implant, possibly because of successful extracortical
bone formation.
The final total percentage of the length of the porous-coated region that
was covered with extracortical bone formation for all patients and anatomical
sites of reconstruction was 76% ± 34%. No significant difference in the
amount of new-bone formation was found with respect to the anatomical site of
reconstruction. All patients with allograft augmentation had greater
extracortical bone formation. Extracortical bone formation did not differ in
relation to the type of porous coating, anatomical site, pathological
disorder, sex or age of the patient, or the length of the reconstruction.
Since the amount and quality of bone-grafting were difficult to ascertain, we
did not attempt to correlate grafting technique and bone formation. Instead,
the amount of bone formation at the time of the two-year follow-up was used to
compare the subsequent changes in bone formation as a function of time.
Since thirteen patients had either diaphyseal bone replacement or a knee
arthrodesis that contained two stems, they were analyzed separately to see
whether the presence of both the proximal and the distal site would have an
effect on the outcome. The remaining thirty patients had a single measurement
of extracortical bone-bridging at each time-point. With the numbers available,
no significant difference was found between extracortical bone formation at
the proximal and distal sites among the patients with two stems with respect
to both the two-year and final follow-up results. In addition to the separate
study of the proximal and the distal extracortical bone formation, these
measurements were averaged for each patient and time-point and were included
in the correlation of the single measurement of extracortical bone formation
with other variables (described below). When the extent of extracortical bone
formation was investigated according to different anatomical locations, no
significant differences were found, with the numbers available
(Table I). There appeared to be
less bone formation in proximal humeral reconstructions, but the number of
cases was too small to be certain.
The time-related change in extracortical bone formation between the
two-year and final follow-up results was studied, and these values were highly
correlated in all patients for all proximal, distal, and averaged measurements
(see Appendix). Thus, the extracortical bone formation at the two-year
evaluation predicted the fate of stem fixation at the long-term follow-up
evaluation. In addition, all patients had a significant increase in
extracortical bone formation between the two-year and the final follow-up
review (p < 0.01). Of all of the patients who had extracortical bone
formation of 100% at two years, only one had a measured decrease in bone
formation at the time of the final follow-up; however, the amount of the
decrease (5%) was thought to be insignificant. Of the patients who had bone
formation of >80% at two years, two had a measured decrease in bone
formation, but the reductions were all =22%. Therefore, the quality and
amount of bone formation at two years did not change substantially with time,
and in some instances it improved with time.
Chemotherapy had no effect on the extracortical bone formation.
Furthermore, age, sex, pathological disorder, and factors related to implant
length, type of porous coating, and treatment of a neoplastic or
non-neoplastic condition also had an insignificant effect on bone formation
measurements.
The effect of bone cement on extracortical bone formation was studied in
detail. In the proximal measurement group, the use of bone cement was
associated with lower extracortical bone formation (p = 0.04). However, no
significant association was found between the use of bone cement and
extracortical bone formation among all patients (the average value was used
for the group with two measurements) or within the distal measurement group
(Table II). No association
could be studied in the group that had a single measurement because all of
those patients had bone cement for the initial implant fixation.
Complications
There were thirteen complications in the current series, and they included
prosthetic loosening, joint dislocation, implant fracture, bone fracture, and
deep infection (see Appendix). Only one patient (Case 19) had stem-loosening
(both the proximal and the distal stem), which occurred within the first year
of follow-up and was caused by a poor grafting technique without the use of
bone cement for the initial fixation. After revision and repeat bone-grafting,
the stems remained stable after more than five years of follow-up. All other
prosthetic loosenings involved the nonsegmental replacement side of the joint;
four patients had loosening of the acetabular component and one had loosening
of the tibial component, which was secondary to osteolysis.
One patient (Case 27) had a deep infection around a distal femoral
prosthesis that was used to salvage a failed allograft, and he required an
above-the-knee amputation. The retrieved implant-bone specimen showed
excellent extracortical bone formation
(Fig. 4).
Another patient (Case 22) who had a knee arthrodesis with use of a titanium
prosthesis with porous-coated press-fit stems had fractures of both the
proximal and the distal stem (Fig.
5). The proximal stem fractured first, but revision was not
necessary because abundant extracortical bone had bridged and united the
femoral cortex and prosthesis shoulder. Subsequently, the distal stem
fractured, and solid bone-bridging obviated the need for revision. The factors
contributing to the fractures were the small stem diameter, the less rigid
material used, and the high bending stress that occurred at the site of the
knee fusion during activities. This patient had no other complication after
eight years of follow-up.
Segmental endoprosthesis fixation has been shown to be a useful
reconstruction method after resection of a tumor and as a salvage procedure in
selected patients with a failed implant associated with massive bone loss.
However, concern for long-term stem fixation has stimulated advancements in
implant design to improve fixation and extend the longevity of the
components14.
Extracortical bone-bridging and ingrowth fixation is a technique to elicit
bone formation over the junction between the extracortical surface of the bone
and the adjacent porous-coated shoulder of a
prosthesis2. The
potential advantages of successful extracortical bone-bridging and ingrowth
fixation include enhanced implant fixation, gradual transfer of stresses
across the bone-prosthesis junction, and improved support of the stem-body
portion of the
prosthesis15.
Furthermore, extracortical bone formation, by forming a so-called purse string
of osseous or fibrous tissue at the junction between diaphyseal bone and the
shoulder of the prosthesis, may prevent the migration of microparticles from
the joint into the cement-bone
interface13.
With the low prevalence of stem-loosening (two of fifty-six stems; one of
forty-three patients) in the present series of patients with long-term
follow-up, extracortical bone formation, as measured radiographically,
fulfilled its intended objectives. Moreover, this fixation method seems to
accomplish the purse-string effect of sealing off possible wear debris, since
no stem-loosening occurred secondary to osteolysis in all the stems fixed by
the extracortical bone-bridging and ingrowth method. The improved results with
allograft augmentation seem to indicate that a better grafting technique
definitely enhances the quality of the extracortical bone formation. The study
by Virolainen et
al.16 provides
support for the need for continuous improvement in the grafting technique to
enhance fixation.
In this series, the survival rate of the implants, as determined by stem
fixation, was similar regardless of the anatomical region of the
reconstruction. The low prevalence of stem-loosening associated with the
extracortical bone-bridging and ingrowth technique is of interest because the
patients were relatively young and would have been expected to have a higher
predisposition to prosthetic loosening than would older patients. The
radiographic assessment of implant stems in our patients demonstrated that
loosening was not a factor in these large segmental reconstructions. Analysis
revealed that the total length of extracortical bone formation involved
=25% of the porous-coated shoulder regions of the prosthesis in
thirty-seven of the forty-three patients and =50% of the implant shoulder
region in forty-five (80%) of fifty-six stems. It is reassuring that the
extracortical bone was maintained for more than a decade after implantation.
In three patients, extracortical bone formation involved <25% of the
implant shoulder region and some degree of bone resorption was evident.
Although the reasons for this result are unclear, placement and secured
fixation of sufficient graft around the entire circumference of the
porous-coated shoulder of the prosthesis may be important for the development
and maintenance of substantial bone formation.
While the limitations of this retrospective study do not allow us to
conclude that the extracortical bone-bridging and ingrowth fixation technique
is directly responsible for extended longevity in the megaprostheses, the low
prevalence of stem-loosening in this series is remarkable. Moreover, we are
encouraged by the fact that once bone formation occurs and has been maintained
for two years, it appears to be active in the load-sharing function at
host-prosthesis junctions even with a long period of follow-up. The results
support the notion that the strength of fixation at the bone-prosthesis
interface is enhanced by extracortical bone formation and provides a strong
rationale for further research.
A table showing specific data on all study patients is available with the
electronic versions of this article, on our web site at
(go to the article citation and click on "Supplementary Material")
and on our quarterly CD-ROM (call our subscription department, at
781-449-9780, to order the CD-ROM).
Collier JP, Surprenant VA, Jensen RE,
Mayor MB. Corrosion at the interface of cobalt-alloy heads on
titanium-alloy stems. Clin Orthop.
1991;271:
305-12.271305
1991
[PubMed]
Chao EY, Sim FH. Composite
fixation of salvage prostheses for the hip and knee. Clin
Orthop. 1992;276:
91-101.27691
1992
Peabody TD, Eckardt JJ.
Complications of prosthetic reconstructions. In: Simn MA, Springfield D,
editors. Surgery for bone and soft-tissue tumors.
Philadelphia: Lippincott-Raven; 1998. p
467-79.467
1998
Chao EY. A composite fixation
principle for modular segmental defect replacement (SDR) prostheses.
Orthop Clin North Am.
1989;20:
439-53.20439
1989
[PubMed]
Chao EY, Sim FH. Modular
prosthetic system for segmental bone and joint replacement after tumor
resection. Orthopedics.
1985;8:
641-51.8641
1985
[PubMed]
Chin HC, Frassica FJ, Markel MD,
Frassica DA, Sim FH, Chao EY. The effects of therapeutic doses of
irradiation on experimental bone graft incorporation over a porous-coated
segmental defect endoprosthesis. Clin Orthop.1993;289:
254-66.289254
1993
[PubMed]
Phemister DB. Treatment of
ununited fractures by onlay bone grafts without screw or tie fixation and
without breaking down of the fibrous union. J Bone Joint
Surg. 1947;29:
946-60.29946
1947
Nakao Y, Hirasawa Y, Dueland T, Chao
EY. [Natural history and biologic mechanism of the transition from grafted
bone to extra-cortical bone bridge over porous-coated segmental prostheses].
Nippon Seikeigeka Gakkai Zasshi.
1992;66: 38-49.
Japanese.6638
1992
[PubMed]
Okada Y, Suka T, Sim FH, Gorski JP,
Chao EY. Comparison of replacement prostheses for segmental defects of
bone. Different porous coatings for extracortical fixation. J Bone
Joint Surg Am. 1988;70:
160-72.70160
1988
Chao EY. Optimal design in tumor
prostheses: application of extracortical bone bridging and ingrowth fixation
principle. In: Uchida A, Ono K, editors. Recent advances in
musculoskeletal oncology. New York: Springer; 1992.
p 133-46.133
1992
Taylor SJ, Perry JS, Meswania JM,
Donaldson N, Walker PS, Cannon SR. Telemetry of forces from proximal
femoral replacements and relevance to fixation. J
Biomech. 1997;30:
225-34.30225
1997
[CrossRef]
Chao EY, Frassica F, Pritchard DJ,
Moyer TP. Metal ion release in patients with porous coated megaprostheses.
Trans Orthop Res Soc.
1995;20:
731.20731
1995
Ward WG, Johnston KS, Dorey FJ,
Eckardt JJ. Extramedullary porous coating to prevent diaphyseal osteolysis
and radiolucent lines around proximal tibial replacements. A preliminary
report. J Bone Joint Surg Am. 1993;
75: 976-87.75976
1993
[PubMed]
Sim FH, Chao EY. Prosthetic
replacement of the knee and a large segment of the femur or tibia.
J Bone Joint Surg Am.
1979;61:
887-92.61887
1979
[PubMed]
Abudu A, Carter SR, Grimer RJ.
The outcome and functional results of diaphyseal endoprostheses after tumour
excision. J Bone Joint Surg Br.
1996;78:
652-7.78652
1996
[PubMed]
Virolainen P, Inoue N, Nagao M,
Ohnishi I, Frassica F, Chao EY. Autogenous onlay grafting for enhancement
of extracortical tissue formation over porous-coated segmental replacement
prostheses. J Bone Joint Surg Am. 1999;
81: 493-9.81493
1999
[PubMed]