Abstract
Background: The hip joint is a common location for metastatic
disease. Actual as well as impending fractures at this site are frequently due
to mechanical instability after tumor invasion and are usually treated
surgically with hip arthroplasty. The objective of this study was to analyze
survival and influences on survival after hip arthroplasty for metastatic hip
disease.
Methods: Two hundred and ninety-nine patients who had undergone a
total of 306 hemiarthroplasty or total hip arthroplasty procedures for
treatment of a pathologic or an impending pathologic hip fracture between 1969
and 1996 at our institution were included in this study. Data that had been
acquired prospectively within the total joint registry of our institution were
reviewed retrospectively.
Results: The median duration of survival after the arthroplasty was
8.6 months. The duration of survival was significantly associated with the
site of the fracture, location of the primary tumor, and time from the
diagnosis of the primary tumor to the surgery for the fracture (p = 0.05).
The time from the diagnosis to the arthroplasty was a significant independent
predictor of survival.
Conclusions: Patients undergoing hip arthroplasty for metastatic
disease have a limited life expectancy, with only 40% (120) of the 299
patients in our series still alive at one year after the surgery. By
identifying prognostic factors regarding life expectancy, this study provides
surgeons and oncologists with information with which to weigh risks and
benefits of hip arthroplasty for individual patients preoperatively.
Level of Evidence: Prognostic study, Level II-1
(retrospective study). See Instructions to Authors for a complete description
of levels of evidence.
Metastatic bone disease is about forty times more common than are primary
bone tumors1.
Certain tumor types, such as breast, prostate, kidney, lung, and thyroid
cancers, frequently metastasize to
bone2. The frequency
of pathologic fractures is thought to increase over time along with the
increases in survival due to improvements in treatment of patients with
cancer3. The most
common site of metastasis to the long bones is the proximal part of the
femur1. The goals of
surgical treatment of structurally important metastatic disease about the hip
are fracture stabilization, pain reduction, and restoration of function. The
durability of the construct should exceed the expected lifetime of the
patient, and it should allow immediate weight-bearing if possible.
Several studies have focused on the outcome of treatment of pathologic hip
fractures with various types of implants. The results suggest that joint
replacement offers an excellent means with which to stabilize these types of
fractures4-9.
A few authors have studied the survival times of a limited number of patients
after hip replacement for the treatment of a pathologic or an impending
pathologic fracture. These authors found that the life expectancy of the
patients rarely exceeds a few years and that many patients die within the
first two years after the
surgery5,6,9-11.
Therefore, the two objectives of our study were to determine the actual
survival times of patients undergoing hip replacement for the treatment of a
pathologic or an impending pathologic fracture of the hip and to determine
which factors are associated with a longer survival time.
Two hundred and ninety-nine patients in whom a total hip arthroplasty or a
partial hip arthroplasty had been performed for the primary treatment of a
pathologic or an impending pathologic hip fracture caused by metastatic
disease between 1969 and 1996 (to allow for a minimum duration of follow-up of
five years) were included in this study, which was approved by our
institutional review board. All patients who met these criteria were
identified in the total joint registry of our institution. Patients treated
with a hip replacement at our institution are routinely evaluated in person,
with standardized letters, or with telephone questionnaires prospectively.
These evaluations are performed at two to three months after the arthroplasty
and at one year, two years, five years, and each five-year interval
subsequently until revision or death, end points for which the accuracy of the
database was shown to be
>95%12.
The database allowed the abstraction of patient age, height, and weight at
the time of the arthroplasty; gender; date of the surgery; side and site of
the fracture; surgical approach; implant type; and surgical complications. A
chart review was performed to identify the location of the primary tumor,
tumor cell type, duration of the neoplastic disease prior to the hip
replacement, and fracture stage (actual or impending) and to confirm the site
of the metastatic lesion.
A total of 306 hip arthroplasties performed between 1969 and 1996 in 299
patients with the above-mentioned criteria were identified. Seven of the
patients had bilateral arthroplasty of the hip performed for the treatment of
metastatic disease. All actual or impending acetabular, capital femoral and
femoral neck, and intertrochanteric and subtrochanteric fractures of
pathologic origin for which the arthroplasty was the primary treatment were
included. Patients were excluded from the study when an attempt to fix the
fracture by open reduction and internal fixation as the primary procedure had
failed.
The fractures were categorized as acetabular, femoral head/neck, or
intertrochanteric/subtrochanteric. When both the acetabulum and the femur were
involved, the fracture was categorized as acetabular. Involvement of the
acetabulum usually limits the surgical options (such as the use of bipolar
components) more than does femoral involvement and may require a more
extensive reconstructive procedure. The implant types were categorized as
bipolar, unipolar, total hip, tumor, and customized prostheses. Postoperative
dislocations, local infections, and periprosthetic fractures were considered
to be major operative complications.
For assessment of the duration of patient survival, the dates of the
operations were grouped as 1969 to 1975, 1976 to 1982, 1983 to 1989, and 1990
to 1996. The height and weight of the patient at the time of the surgery were
available for 173 arthroplasties. The body mass index of the patients was
calculated by dividing the weight in kilograms by the height in meters squared
(kg/m2). The body mass index was categorized as <20, 20 to 25,
or >25.
Population Demographics
The mean age of the patients at the time of the arthroplasty was 62.6 years
(median, sixty-three years; standard deviation, 12.5 years; range, twenty-four
to ninety-three years). Two hundred and two (66%) of the arthroplasties were
performed in women and 104 (34%), in men. Forty-five arthroplasties were done
from 1969 to 1975; sixty-one, from 1976 to 1982; ninety-two, from 1983 to
1989; and 108, from 1990 to 1996. The mean and median durations from the
surgery to the last follow-up examination were seventeen and 8.6 months
(range, less than one month to nineteen years) for the entire group of
arthroplasties in the study and 5.7 and 5.1 years (range, zero days to
nineteen years) for people who were alive at the time of this review. The
distribution of patients according to the surgery date intervals, sides and
sites of the fracture, surgical approaches, fracture stages, implant types,
and complications is shown in the Appendix. The postoperative complication
rate was low, with only eleven arthroplasties (3.6%) followed by dislocation;
five (1.7%), by infection; and eleven (3.6%), by periprosthetic fracture.
Statistical Methods
Overall survival was estimated with use of the Kaplan-Meier survival method
as a function of time from the hip
arthroplasty13.
Survival estimates were also calculated for discrete risk factors with the
Kaplan-Meier method. Associations with risk factors as well as the final
multiple variable model were assessed with use of the Cox proportional hazards
regression model14.
A time-dependent analysis was used to assess whether the occurrence of a major
operative complication was associated with survival. Potential risk factors
that were found to be significant in the univariate analysis were considered
in the multiple variable model. The alpha level was set at 0.05 for
significance.
The end point of the study was the death of the patient. The deaths of 289
patients were documented by the end of the observation time, nine patients
were lost to follow-up, and one patient was still alive with continuing hip
function.
Two sets of analyses were done for impending fractures. One calculation
treated the impending fractures as a separate fracture-site category (the
fourth category). The other calculation included the impending fractures in
one of the three categories depending on the location of the metastatic
involvement of the bone (acetabular, femoral head/neck, or
intertrochanteric/subtrochanteric).
Breast cancer was the most frequent primary tumor (147 arthroplasties,
48.0%), followed by prostate cancer (forty-four arthroplasties, 14.4%), lung
cancer (twenty-seven arthroplasties, 8.8%), renal cell carcinoma (twenty-six
arthroplasties, 8.5%), and cancer of unknown primary origin (eighteen
arthroplasties, 5.9%). In terms of histologic tumor-cell types, it was only
feasible to distinguish between adenocarcinoma (241 arthroplasties, 78.8%) and
non-adenocarcinoma (sixty-five arthroplasties, 21.2%) since the individual
numbers of cell types within the non-adenocarcinoma group were very small. The
numbers of patients according to the primary tumor site are displayed in the
Appendix. The mean and median of the body mass index were 26.6 and 26.3,
respectively (standard deviation, 5.83; range, 16 to 56.6).
Survival Time
The median duration of survival after the arthroplasties for the pathologic
or impending pathologic fractures at the hip was 8.6 months (range, zero days
to nineteen years). The survival rate (and 95% confidence interval) was 78.6%
(74.2% to 83.4%) at three months after the surgery, 60.2% (54.9% to 66.0%) at
six months, 40% (35.0% to 46.0%) at one year, 21.5% (17.3% to 26.7%) at two
years, and 6% (3.8% to 9.4% at five years)
(Fig. 1). Only seventeen
patients were still being followed after five years.
Fracture Site and Survival
The median duration of survival was 271 days for the patients with a
femoral head/neck fracture, 183 days for those with an
intertrochanteric/subtrochanteric fracture, and 474 days for those with an
acetabular fracture (Fig. 2).
The patients with an acetabular fracture survived for significantly longer
than did those with a femoral head/neck fracture (p < 0.01; hazard ratio =
0.6 [0.4 to 0.8]), and those with a femoral fracture survived for a
significantly shorter duration than did those with an acetabular fracture
(femoral head/neck fractures: p < 0.01, hazard ratio = 1.7 [1.2 to 2.5];
intertrochanteric/subtrochanteric fractures: p < 0.01, hazard ratio = 1.9
[1.2 to 2.9]).
Impending Compared with Actual Fractures
The median duration of survival after the arthroplasty was 327 days for the
patients with an impending fracture and 235.5 days for those with an actual
fracture. Interestingly, with the numbers available, survival was not
significantly longer in the group with an impending fracture (p = 0.12, hazard
ratio = 0.8 [0.6 to 1.1]).
Type of Primary Tumor
The median survival time was 221.5 days for the patients with a primary
kidney tumor, 162 days for those with a lung tumor, 302.5 days for those with
a prostate tumor, 421 days for those with a breast tumor, and 114 days for the
remaining patients (Fig. 3).
Comparison of the patients with breast cancer with those of the patients with
each of the other types of tumors revealed that patients with kidney cancer,
lung cancer, and "other" cancer types had a significantly poorer
duration of survival (p = 0.05). Patients with prostate cancer showed a
tendency for a shorter duration of survival, but this difference did not reach
significance (p = 0.1).
Time from Diagnosis of Primary Tumor to Surgery for Fracture
The median time from the diagnosis of the tumor to the surgery for the
pathologic or impending pathologic fracture of the hip was twenty-six months,
with a range of zero days to twenty-nine years. The time was considered to be
zero for all patients in whom the tumor was diagnosed perioperatively. The
patients with a shorter interval between the diagnosis of the primary tumor
and the surgery for the fracture or impending fracture had a poorer prognosis
(p < 0.01), with a hazard ratio of 0.97 (0.96 to 0.99) per year
(Fig. 4).
Multiple Variable Model
The age of the patient, the operative approach, the date (seven-year
interval) of the surgery, the histologic tumor-cell type (adenocarcinoma or
non-adenocarcinoma), the occurrence of a complication, and the body mass index
were not significantly associated with survival (p > 0.05). The final
multiple variable regression model, adjusted for age, revealed the time from
the primary tumor diagnosis to the hip arthroplasty to be an independent
predictor of patient survival after surgery (p = 0.05, hazard ratio =
0.98).
In this study, we analyzed the survival time of patients who had had a hip
arthroplasty for a pathologic or an impending pathologic hip fracture caused
by metastatic disease. We did not review the systemic chemotherapies, hormonal
therapies, or local radiation therapies prior to the surgery. However, the
influence of those treatments was likely limited, since the vast majority of
patients had advanced metastatic cancer for which no further specific
oncologic treatment was promising. The fact that the overall survival time of
our patients did not increase between 1969 and 1996 supports this
assumption.
Initially, differences in survival between genders, favoring women, were
detected, but those differences disappeared when gender-related tumors (breast
and prostate) were removed from the calculation. Bauer and Wedin reported that
patients with metastatic breast carcinoma had a longer survival time than did
patients with prostate or lung
carcinoma11. We
also found that patients with metastatic breast cancer survived for
significantly longer than did patients with kidney or lung cancer. Our
observation that metastatic renal carcinoma is a negative prognostic factor
contrasts with the findings in a study of patients from the Stockholm
region11. However,
the number of patients included in the Swedish study was not large enough to
show statistical significance. The five-year survival rate of our patients
with metastatic renal cell carcinoma (5%) was lower than that observed by Jung
et al.15 (14%).
However, the five-year survival in that study was analyzed according to the
diagnosis of metastasis in all appendicular bones and not according to the
date of surgery to treat a pathologic fracture about the hip. Therefore, our
patients probably had more advanced disease.
Surprisingly, patients with an acetabular tumor had a better survival
prognosis than did patients with a femoral lesion in our study. One could
argue that a tumor has to be more aggressive to spread to the more distant
femoral site or that the difference in survival between the patients with an
acetabular tumor and those with a femoral tumor represents an anatomic
chronologic bias, meaning that the acetabulum is involved by metastasis
earlier in the course of metastatic disease than is the femur.
The shorter the time interval from the diagnosis of the primary tumor to
the surgery for the pathologic or impending pathologic fracture, the more
limited was the survival. It appears that cancers that have a more aggressive
nature, and therefore metastasize to the bone earlier and cause the pathologic
fracture earlier, have a poorer prognosis. Thus, one has to be cautious about
assuming that patients with a long disease history and then a fracture will
not survive long. The long disease history likely reflects a less aggressive
tumor.
In addition to the factors that were found to influence survival in our
study, the factors that did not have an influence are equally worthy of note.
Some of the parameters with no impact are surprising and contradict common
assumptions.
In agreement with the findings in a Swedish
study11, our data
showed no difference in survival attributable to the age at surgery.
Therefore, clinicians should be cautious about using age as a criterion for
decisions regarding surgical intervention.
Postoperative complications also did not alter the overall survival, even
though one might expect that multiple reoperations (for dislocations, for
example) could compromise survival. Therefore dislocations, periprosthetic
fractures, and infections should be addressed just as intensely in this
patient population as they are in the general population.
Operative treatment of impending fractures has been recommended to avoid
the graver consequences of an actual fracture. Published guidelines regarding
when to perform such prophylactic surgery have differed. A defect that
involves >50% of the cortex or is larger than 2.5 cm, as seen on
radiographs, has been thought to be at high risk for fracture in a
weight-bearing
bone16,17.
However, those prediction criteria were challenged by Keene et al., who did
not find a strong relationship between the size of the lesion and the risk of
fracture18. Mirels
developed a score based on the size of the lesion, radiographic features,
anatomic site, and
pain19. Such
approaches are only somewhat helpful, and the decision regarding whether to
fix an impending fracture is based mainly on the clinical judgment of the
surgeon. We did not find a significant difference in survival between patients
treated for an actual fracture and those treated for an impending fracture.
This was surprising since one might expect the group with an impending
fracture to have benefited from the bias introduced by the additional time
between the primary tumor diagnosis and the pathologic fracture. It is
conceivable, however, that the group with an impending fracture did not
represent the same population as did the patients with an actual fracture, and
this potential difference could account for the unexpected similarity in
survival.
According to Damron and Sim, the first principle of surgical treatment of
metastatic disease is appropriate patient selection, with the time needed for
recovery from the surgery not exceeding the anticipated survival of the
patient20. We hope
that our study provides useful parameters with which surgeons and oncologists
can estimate survival after a pathologic or an impending pathologic fracture
of the hip and thus provide the best treatment for each patient.
Tables showing the clinical features of the hips in this study and patient
survival according to tumor type are 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).
Harrington KD.Orthopaedic management of metastatic bone disease. St.
Louis: Mosby; 1988.
1988
Buckwalter JA, Brandser EA.
Metastatic disease of the skeleton. Am Fam Physician.1997;55:
1761-8.551761
1997
[PubMed]
Hage WD, Aboulafia AJ, Aboulafia
DM. Incidence, location, and diagnostic evaluation of metastatic bone
disease. Orthop Clin North Am.2000;31:
515-28.31515
2000
[PubMed][CrossRef]
Habermann ET, Sachs R, Stern RE,
Hirsh DM, Anderson WJ Jr. The pathology and treatment of metastatic
disease of the femur. Clin Orthop.1982;169:
70-82.16970
1982
[PubMed]
Lane JM, Sculco TP, Zolan S.
Treatment of pathological fractures of the hip by endoprosthetic replacement.
J Bone Joint Surg Am.1980;62:
954-9.62954
1980
[PubMed]
Harrington KD, Johnston JO, Turner
RH, Green DL. The use of methylmethacrylate as an adjunct in the internal
fixation of malignant neoplastic fractures. J Bone Joint Surg
Am.1972;54:
1665-76.541665
1972
Algan SM, Horowitz SM. Surgical
treatment of pathologic hip lesions in patients with metastatic disease.
Clin Orthop.1996;332:
223-31.332223
1996
[PubMed][CrossRef]
Clarke HD, Damron TA, Sim FH.
Head and neck replacement endoprosthesis for pathologic proximal femoral
lesions. Clin Orthop.1998;353:
210-7.353210
1998
[PubMed][CrossRef]
Marco RA, Sheth DS, Boland PJ, Wunder
JS, Siegel JA, Healey JH. Functional and oncological outcome of acetabular
reconstruction for the treatment of metastatic disease. J Bone
Joint Surg Am.2000;82:
642-51.82642
2000
Wedin R, Bauer HC, Rutqvist LE.
Surgical treatment for skeletal breast cancer metastases: a population-based
study of 641 patients. Cancer.2001;92:
257-62.92257
2001
[PubMed][CrossRef]
Bauer HC, Wedin R. Survival after
surgery for spinal and extremity metastases. Prognostication in 241 patients.
Acta Orthop Scand.1995;66:
143-6.66143
1995
[PubMed][CrossRef]
McGrory BJ, Morrey BF, Rand JA,
Ilstrup DM. Correlation of patient questionnaire responses and physician
history in grading clinical outcome following hip and knee arthroplasty. A
prospective study of 201 joint arthroplasties. J
Arthroplasty.1996;11:
47-57.1147
1996
[CrossRef]
Kaplan EL, Meier P. Nonparametric
estimation from incomplete observations. J Am Stat
Assoc.1958;53:
457-81.53457
1958
[CrossRef]
Cox DR. Regression models and
life-tables [with discussion]. J Roy Statist Soc Series
B.1972;34:
187-202.34187
1972
Jung ST, Ghert MA, Harrelson JM,
Scully SP. Treatment of osseous metastases in patients with renal cell
carcinoma. Clin Orthop.2003;409:
223-31.409223
2003
[PubMed][CrossRef]
McBroom RJ, Cheal EJ, Hayes WC.
Strength reductions from metastatic cortical defects in long bones.
J Orthop Res.1988;6:
369-78.6369
1988
[PubMed][CrossRef]
Leggon RE, Lindsey RW, Panjabi
MM. Strength reduction and the effects of treatment of long bones with
diaphyseal defects involving 50% of the cortex. J Orthop
Res.1988;6:
540-6.6540
1988
[CrossRef]
Keene JS, Sellinger DS, McBeath AA,
Engber WD. Metastatic breast cancer in the femur. A search for the lesion
at risk of fracture. Clin Orthop.1986;203:
282-8.203282
1986
[PubMed]
Mirels H. Metastatic disease in
long bones. A proposed scoring system for diagnosing impending pathologic
fractures. Clin Orthop.1989;249:
256-64.249256
1989
[PubMed]
Damron TA, Sim FH. Operative
treatment for metastatic disease of the pelvis and the proximal end of the
femur. J Bone Joint Surg Am.2000;82:
114-26.82114
2000