Surgeons and scientists have once again produced a large amount of data
related to outcomes, biomaterials, surgical techniques, the treatment of
complications, and socioeconomic issues related to total hip arthroplasty
surgery. One recent debate in hip surgery is on the most efficacious treatment
for displaced intracapsular hip fractures. Following the style of our previous
Specialty Updates, this review is organized into seven sections: (1) primary
surgery, (2) surface replacement, (3) revision surgery, (4) biomaterials and
prosthetic design, (5) complications, (6) practice management, and (7)
treatment of intracapsular hip fractures.
Minimal Incision Surgery
Total hip arthroplasty performed with use of smaller tissue-preserving
techniques received intense initial popularity and scrutiny. There are now
more data documenting differences among the various techniques. Moreover, the
complication rate associated with certain techniques may be too high for
patient safety.
Pagnano et al. performed staged bilateral total hip arthroplasties in
twenty-six patients. One hip was treated with use of the two-incision
technique, and the contralateral hip was treated with use of a mini-posterior
approach. Sixteen of the twenty-six patients reported greater satisfaction on
the side that had been treated with the miniposterior approach; the reasons
cited included earlier recovery (50%), cosmetic appearance (25%), and a
combination of reasons (25%). Eight patients preferred the side treated with
the two-incision technique; all cited early recovery as the reason. Two
patients had no preference. Sharkey and Hozack conducted a prospective,
randomized study in which the standard approach was compared with the
miniincision approach. They found no difference between the groups with regard
to intraoperative blood loss, transfusions, operative time, or time to
hospital discharge. Moreover, there was no difference with regard to patient
outcome measurements such as the Harris hip score, the Short Form-36, and a
validated rehabilitation improvement test at three to six months. The authors
also evaluated the efficacy of intense rehabilitation and pain-management
protocols in both groups. There was better functional improvement, a shorter
hospital stay, higher patient satisfaction, and lower analgesia requirements
in the group of patients who had received the intense preoperative and
postoperative rehabilitation and pain-management protocol. This superior
outcome was independent of incision size. These data suggest that accelerated
discharge and functional return following primary total hip arthroplasty was
not a function of incision size but rather was a function of many psychosocial
variables.
Murphy and Tannast reported superior (p < 0.0001) recovery and
functional improvement at six weeks following a superior capsulotomy
mini-incision approach. However, there was no difference between the
mini-incision and standard-incision groups at twelve weeks. Hildebrand et al.
conducted gait analysis at three, six, and twelve weeks after total hip
arthroplasty in patients who had been managed with three different
minimal-incision approaches: a two-incision approach, a one-incision
anterolateral approach, and a one-incision posterior approach. All patients
demonstrated improvement in the Harris hip and WOMAC scores. There was a
faster increase in these parameters in association with the two-incision
approach at three (p < 0.01) and twelve weeks (p < 0.05). There was no
difference in range of motion or abductor strength among the groups. More
patients in the two-incision group were able to walk without support at three
weeks as compared with those in the other two groups (p < 0.01). Moreover,
the stride length, cadence, and speed were all superior in the two-incision
group (p < 0.01). Another group of investigators from Japan conducted a
prospective, randomized study in which gait analysis was used to assess
differences in muscle recovery. The strength of selected muscle groups
(abductor, extensor, and flexor) was best in association with the
mini-posterior approach in contrast with the two-incision, mini-anterolateral,
and standard posterior approaches. The differences were apparent at one month
and persisted for as long as six months after surgery. Different
minimal-incision surgical approaches may influence the temporal events of
functional return following total hip arthroplasty. Patients should be
properly counseled with regard to realistic expectations.
One proposed advantage of minimal-incision total hip arthroplasty is
reduced cost as a result of accelerated discharge. Greidanus et al. conducted
a prospective, randomized trial in which a two-incision anterior approach
(sixty-six total hip arthroplasties) was compared with a one-incision anterior
approach (ninety-nine total hip arthroplasties). Hospital stay and narcotic
use were lower in the two-incision group (p < 0.05). However, this was
offset by greater resource utilization in the operating room (time and
fluoroscopy use) with the two-incision approach. Another proposed advantage of
minimal-incision surgery is reduced blood loss. Vaughn et al. found no
difference in acute blood loss or laboratory measures of blood levels between
two groups of 100 hips that were treated with either a minimal two-incision
approach or a standard one-incision approach. In fact, more patients required
allogeneic blood transfusions in the two-incision group (p > 0.05). The
authors also reported an average 50% (thirty-minute) increase in operative
time in the two-incision group. Last, thirteen complications occurred,
including ten intraoperative fractures, two dislocations, and one instance of
stem loosening in the two-incision group. In contrast, only one complication
occurred in the standard-incision group. There was no difference with regard
to operative time, blood loss, and complications between the first fifty and
second fifty total hip arthroplasties in the mini-incision group.
Bal et al.1
reported on early complications associated with the two-incision technique in
a study of eighty-nine consecutive total hip arthroplasties. These results
were compared with those for ninety-six total hip arthroplasties that had been
performed by the same surgeon one year previously with use of a mini-lateral
incision. There was no difference in operative time. Blood loss was greater in
the two-incision group (p < 0.0001). The duration of hospital stay was
shorter in the two-incision group (p = 0.001). Complications occurred in 42%
of the patients in the two-incision group, compared with 6% of those in the
mini-lateral group (p < 0.0001). Reoperations were required in 10% of the
patients in the two-incision group because of fracture, loosening,
dislocation, and/or wound complications. However, there was significant
improvement with greater experience. When the first forty and the second
forty-eight total hip arthroplasties performed with the two-incision technique
were compared, there was a reduction in the rates of overall complications (p
= 0.02) and revision surgery (p = 0.04). However, there was no difference
between the early and late groups with regard to the prevalence of injury to
the lateral femoral cutaneous nerve. The two-incision technique is a
technically demanding procedure. The surgeon, although quite experienced with
minimal-incision total hip arthroplasty, still experienced these
complications, particularly during the early phase of the learning curve.
Fixation Without Cement
Stem
Morshed et al. conducted a meta-analysis with use of stratified and
regression methods. Twenty studies were included. There was no difference
between fixation with and without cement when revision of either component (p
= 0.88) or both components (p = 0.30) was used as the end point. Subgroup
analysis of studies not restricted to younger patients (those less than
fifty-five years old) demonstrated a 3.8% (95% confidence interval, 0.7% to
6.9%) advantage in association with cement fixation. There was a trend toward
superior results in younger patients who were managed with cementless
fixation. Evidence-based guidelines to support any specific fixation method
will require much larger and more rigorously designed studies.
Two stem designs have gained extensive clinical application over the past
decade: (1) extensively coated stems with straight geometry and (2) proximally
coated stems with tapered geometry. Engh et al. followed 223 total hip
arthroplasties that had been performed with an extensively coated stem. The
mean age at the time of surgery was fifty-five years. The latest evaluation
was conducted at a mean of 18.6 years for 137 living patients (143 hips). Only
three stems had been revised because of aseptic loosening, and the rate of
stem survival was 97.5% at twenty years. Callaghan et al. reported the
outcomes associated with a second-generation extensively coated stem in a
study of seventy-nine hips that had six to eight years of follow-up. The mean
age of the patients was fifty-five years. No stem was revised. With regard to
fixation status, 92% of the stems were classified as bone-ingrown and 8% were
classified as fibrous-stable. The prevalence of thigh pain was 6%. These data
were significantly superior (p < 0.05) to those for first-generation
proximally coated stems that had been inserted by the same surgeon and
evaluated after a comparable follow-up interval, for which the rates of
revision, loosening, and thigh pain were 3%, 5%, and 13%, respectively. Jasty
et al. reported the minimum five-year results of 101 consecutive total hip
arthroplasties that had been performed with use of a proximally coated
(fiber-metal) tapered stem. The mean age of the patients was fifty-five years.
No stem was revised. All stems were bone-ingrown, and the prevalence of thigh
pain was 0%. No femoral osteolysis was observed.
Hydroxyapatite Coating
The debate on the proposed benefits of hydroxyapatite coating has
continued. Lombardi et al. reported on 192 hips that had been treated with
stems of identical geometry and surface texture (tapered, proximally coated)
either with hydroxyapatite (sixty-one hips) or without hydroxyapatite (129
hips). The mean duration of follow-up was 11.6 years. Only one stem (a
non-hydroxyapatite-coated stem) was revised. There was no difference between
the groups with regard to the Harris hip score or the rate of revision.
Capello et al. reported the minimum fifteen-year results of 156 total hip
arthroplasties that had been performed with use of another
hydroxyapatite-coated tapered stem. The mean age of the patients at the time
of surgery was 51.5 years. The mean Harris hip score was 92, with 91% of the
patients reporting no pain or limp at the time of the final follow-up.
Revision because of loosening was necessary in only one hip. The fifteen-year
stem survival rate was 99.1%, and the combined revision rate for loosening and
osteolysis was 1.8%. McCarthy and Lee reported the five-year follow-up data
for 114 total hip arthroplasties that had been performed with stems of
identical geometry and surface texture either with hydroxyapatite coating
(seventy-five hips) or without hydroxyapatite coating (thirty-nine hips). No
stem was revised or was loose. There was no difference (p > 0.05) in the
radiographic features of bone remodeling around either stem type. An
interesting finding was the higher short-term (six-week) overall Harris hip
score (82 compared with 73) and the higher pain score (42 compared with 36) in
the hydroxyapatite-coated group. There was no difference between the groups
with regard to the Harris hip score at two years or five years. The authors
suggested a potential benefit of hydroxyapatite-coated stems in selected
patients with high expectations of an early return to high-demand and
high-level function following surgery.
Cup
Moore et al.2
proposed radiographic signs of osseointegration of cementless cups.
Radiographic evaluation was done for 119 hips that had adequate serial
radiographs after primary arthroplasty and before revision surgery.
Osseointegration was confirmed with use of mechanical testing at the time of
revision. The authors proposed five radiographic signs of stable cup fixation:
(1) the absence of radiolucent lines, (2) the presence of a superolateral
buttress, (3) medial bone stress-shielding, (4) radial trabeculae, and (5) an
inferomedial buttress. Ninety-seven percent of the cups with three or more of
these signs were determined to be stable at the time of revision surgery. In
contrast, 83% of the cups with two or fewer signs were loose.
Callaghan et al. reported ten to eighteen-year results of 118 consecutive
hip replacements that had been performed with use of a first-generation
cementless cup in patients younger than fifty years old. Three cups were
revised because of loosening, and four were revised because of osteolysis
without loosening. In addition, fifteen revisions were performed for liner
exchange. Pelvic osteolysis was observed in 15% of the hips. The mean linear
wear rate was 0.2 mm/year. Fixation durability was fivefold superior to that
associated with cemented cups (p < 0.01) that had been inserted by the same
surgeon in a comparable patient population. Cup fixation durability has been a
principal limitation in patients with previous pelvic and acetabular
fractures. Lombardi et al. reported on sixty-three cementless cups that had
been inserted for the treatment of posttraumatic arthritis following
thirty-four acetabular, twenty-three pelvic, and six combination fractures.
After a mean duration of follow-up of seven years, eight revisions had been
performed because of wear and/or loosening. The ten-year rate of cup survival
was 88.1% with revision as the end point. Another challenging population
includes patients who have had previous pelvic irradiation. Lewallen et al.
used a new cup design made of tantalum trabecular metal in this setting. The
series included thirteen cups that were reviewed after a mean duration of
follow-up of twenty-nine months. The mean radiation dose was 5260 cGy. The
mean interval between radiation therapy and total hip arthroplasty was
thirteen years. No cup was revised or loose. Surgeons from the Rothman
Institute also reported durable cementless cup fixation in association with
the use of a hemispherical design with supplemental screws in patients who had
had previous pelvic irradiation for the treatment of prostate cancer. These
results provided promise for improved cup fixation in this extremely difficult
clinical setting.
Cup liners with an elevated rim were initially introduced to reduce
instability. Recent data reflect the risk of impingement associated with these
liners. Comfort et al. analyzed a community-based total hip arthroplasty
registry that included 3070 hips. The procedures had been performed by
fifty-three surgeons in five hospitals over thirteen years. The liners were
grouped into three categories: (1) no elevation (17%), (2) 10° of
elevation (69%), or (3) >10° of elevation (14%). One hundred and
seventy liners were revised. The overall rate of liner survival at 13.3 years
was 94.5%. There was a difference in the rate of survival (p = 0.02) between
flat liners (98.1%) and liners with an elevation (93.7% for liners with
10° of elevation and 93.5% for those with >10° of elevation). All
hips that were revised because of recurrent dislocations or shell loosening
initially had an elevated liner. As a consequence of these findings, the rate
of use of elevated liners in that community decreased from a high of 99% in
1996 to 50% in 2004. That report reflects the utility of a joint registry in
altering clinical practice.
Fixation with Cement
The use of cement fixation has decreased considerably. Collis et al.
provided additional data validating the superior durability of a polished stem
surface texture when inserted with cement. Those authors reported on 1031
consecutive cemented stems with a polished texture that were inserted over ten
years. There were four different stem geometries as the designs evolved over
the decade. Six stems had been revised because of aseptic loosening or
osteolysis after two to twelve years of follow-up. Debonding always occurred
at the bone-cement interface in the failed cases. There was no difference
among the four stem geometries. The revision rate for polished stems was
fivefold lower than that for stems with a more roughened texture that had been
inserted by the same surgeon and followed for a comparable period of time.
Surface replacement initially was introduced in the 1970s; however, its
durability was poor and thus it was abandoned by most surgeons. There has been
a resurgence of surface replacement over the past decade because of
improvements in design, bearing surface, and instrumentation. Surface
replacement has become the fastest-growing segment of all total hip
arthroplasties worldwide. Ritter reported his experience with a
first-generation surface replacement. Sixty-five hips were treated between
1977 and 1981. Sixty-six percent of the procedures failed at some time between
three months and twenty-two years (mean, 12.3 years). Femoral failure occurred
in 37% of the hips, and cup failure occurred in 29% of the hips. Among the
femoral failures, 50% were due to femoral neck fracture. The author
specifically commented on the eight femoral failures that occurred after a
minimum of ten years. Every hip showed narrowing at the neck as a result of
stress-shielding under the shell. The failure rate for these surface
replacements was 5.4 times greater than the rate for total hip arthroplasties
that had been performed by the same surgeon with use of stemmed designs over
the same time-period. These data clearly do not reflect the improvements with
the current designs, bearing surfaces, and surgical techniques. They should
serve as reference against which all newer designs can be measured.
There is now more than eight years of clinical experience worldwide with
contemporary metal-on-metal surface replacement. The longest and most
extensive clinical experience has been with the Birmingham Surface Replacement
(BSR) system. Daniel et
al.3 reported that
only one (0. 2%) of 446 hips that had been treated with this system had been
revised after a mean duration of follow-up of 3.3 years. Nearly 90% of the
entire patient group were engaged in high-demand jobs or recreational
activities. Treacy et
al.4 reported a 99%
survival rate after a minimum duration of follow-up of five years in a study
of 144 consecutive hips that were treated with the same implant system.
Femoral loosening has been a major mechanism of failure following surface
replacement. Amstutz et al. reviewed the difference in outcomes between the
first 300 surface replacements (Group 1) and the second 300 surface
replacements (Group 2) that they performed. The mean age for the entire cohort
was forty-nine years, and 74% of the patients were male. The mean duration of
follow-up was seven years for Group 1 and 4.6 years for Group 2. Survival
analysis demonstrated a difference (p = 0.015) between Groups 1 and 2.
Overall, twenty-five hips (4.17%) were converted to a stemmed hip replacement,
and two-thirds of the conversions were performed because of femoral loosening.
Metaphyseal radiolucencies were apparent in 59% of the failed hips by
twenty-four months and in 100% by fifty months. The authors identified the
most critical technical factors for success as being meticulous femoral neck
bone preparation, a dry bone bed, circumferential fixation holes, and
cementing of the stem. They also recommended avoiding bone-grafting because
bone-grafting would reduce the area of cement fixation to host bone. Amstutz
also reported a 1% rate of femoral neck fracture in his personal experience of
800 surface replacements. He recommended (1) reaming the neck at 140°, (2)
avoiding notching the lateral femoral neck cortex, and (3) abandoning surface
replacement in cases in which large bone defects are present at the head-neck
junction.
Vail compared surface replacement (forty-five hips) with stemmed total hip
arthroplasty (123 hips). There was greater improvement in the Harris hip score
among the patients managed with resurfacing (p < 0.001). Moreover, there
were also differences in activity (p < 0.001), function (p < 0.001), and
range of motion (p < 0.001) that favored surface replacement. There was no
difference between the groups with regard to complications. Mont et al.
conducted gait analysis in a study of fifteen patients who had been managed
with a standard total hip arthroplasty, fifteen patients who had been managed
with surface replacement, and eleven volunteers with hip osteoarthritis. All
patients who had had an arthroplasty (including those managed with total hip
arthroplasty and those managed with surface replacement) had a Harris hip
score of >90 twelve months after surgery. Patients who had been managed
with resurfacing had the best hip kinematics (abductor and extensor moments)
and the best walking velocity, even when compared with the volunteers who had
osteoarthritis. These data further validate the clinical efficacy of hip
resurfacing in selected patients.
Fixation failure has been correlated with specific anatomical
characteristics of the proximal part of the femur. Schmalzried et
al.5 proposed a
radiographic grading scale based on four principal characteristics: (1) bone
density, (2) bone shape, (3) biomechanics, and (4) focal defects. Hips were
categorized into five grades on the basis of the number of unfavorable
characteristics. Lower grades were associated with a better clinical outcome.
The authors concluded that proper patient selection and precise surgical
technique are critical in order to avoid complications and to achieve
durability after surface replacement.
Kurtz et al.6
reported a relatively constant revision burden of 17.5% for total hip
arthroplasty from 1991 to 2002, with revisions accounting for approximately
20% of all total hip arthroplasty cost expenditures. The investigators
constructed a model to predict the future rate of revision total hip
arthroplasty. They projected 48,000 revisions in 2010, more than 67,000 in
2020, and nearly 97,000 in 2030 (a 137% increase from 2005). Similarly, they
projected a greater than twofold increase in primary total hip arthroplasties
(from 240,000 to 570,000) from 2005 to 2030. They also projected that the
number of practicing orthopaedic surgeons in the United States would increase
from approximately 20,000 in 2010 to 26,000 in 2030. On the basis of their
numbers, the mean case load for each surgeon would continue to increase from
decade to decade. Moreover, the increasing volume of revisions will
undoubtedly pose a challenge to resource utilization and cost containment.
Many elderly patients (those who are more than eighty years old) require
revision total hip arthroplasty. Parvizi et al. reviewed 170 revision total
hip arthroplasties that had been done in octogenarians at a tertiary
high-volume arthroplasty center over an eight-year period. There was
significant improvement in function and pain relief according to the SF-36
scores after revision (p < 0.05). There was a 16.5% rate of orthopaedic
complications and a 15.3% rate of major medical complications in the first six
months after revision. Approximately half (48.1%) of the patients had died by
a mean of 4.6 years after revision. The authors concluded that revision total
hip arthroplasty is excellent for improving quality of life despite a high
rate of complications in this patient population.
Stem
Iorio, Healy, and Presutti conducted a prospective, surgeon-randomized,
matched-cohort study in which femoral revisions that had been performed with a
cementless, modular, straight-stem femoral component with flutes were compared
with those that had been performed with a cemented component. All patients had
minimal (Paprosky type-I or II) bone deficiency. Each surgeon performed
forty-three consecutive revisions with use of the same technique. The mean
duration of follow-up was seven years. There was no difference between the
groups with regard to the rates of stem rerevision (p = 0.56) or overall
reoperation (p = 0.29). Although it is being used with less frequency, cement
fixation may be effective in selected cases of femoral revision. One of the
most common femoral revision techniques is to use distal fixation in the
diaphysis with either an extensively-coated straight or curved cylindrical
stem or a tapered, straight stem with flutes. The fixation stability of both
stem designs is dependent on sufficient stem-canal engagement. Paprosky and
Sporer reported on the use of 9 and 10-in (22.9 and 25.4-cm) fully coated
cylindrical stems for revision arthroplasty in a series of patients with
Paprosky type-IIIB deficiencies (characterized by no metaphyseal support and a
diaphyseal segment measuring <4 cm) or type-IV deficiencies (characterized
by an insufficient diaphyseal segment). The mean duration of follow-up was 4.2
years. The rate of mechanical failure was 0% for hips with type-IIIB
deficiencies when the canal diameter was <19 mm, 18% for hips with
type-IIIB deficiencies when the canal diameter was >19 mm, and 37.5% for
hips with type-IV deficiencies. In contrast, there were no failures after
revisions that had been performed with impaction grafting (ten hips) or with a
tapered stem with flutes (ten hips) in the presence of similar bone
deficiencies. These data reflect the limitations of extensively-coated
cylindrical stems when femoral bone is deficient.
Another way to address major segmental bone deficiencies is with an
allograft-prosthesis composite. Christie et al. reported the results for
twelve hips after a mean of eight years of follow-up. All allografts healed to
the host. Seven reconstructions had failed after a mean duration of follow-up
of nine years. There was graft resorption in all hips. Most importantly, the
surgeons found the allograft segment to be inadequate to support a new
stem.
Some investigators have recommended revision with use of a total femoral
replacement in the most extreme clinical situations. Friesecke et
al.7 presented the
results of 100 consecutive total femoral replacements that had been performed
over 7.5 years. The mean operative time was nearly four hours. Seventy-seven
percent of the procedures were performed because of the failure of a total hip
arthroplasty, whereas the others were performed because of knee-related issues
or periprosthetic fractures. The rate of complications (including infection,
dislocation, material failure, patellar problems, and peroneal nerve palsy)
was 32%. The overall functional rating for both the hip and the knee was
satisfactory in most patients. However, the mean duration of follow-up was
short (six years) and the durability of this salvage option remains to be
documented with longer follow-up.
Cup
Berry presented a further update on eighty-nine revisions that were
performed with use of a jumbo cup with multiple screws and a fiber-metal
surface texture. Particulate graft was used in 61% of the procedures, and
structural graft was used in 10%. The mean duration of follow-up was twelve
years (minimum, ten years). Ten cups were rerevised because of loosening (five
cups), infection (four), or instability (one). This technique is excellent
(with a reported success rate of 93.3%), provided that there is adequate bone
support and initial cup stability. Another technique that has been used to
address extreme acetabular defects such as pelvic discontinuity is the use of
custom-designed flanged cups. DeBoer et al. followed twenty-one consecutive
revisions that had been performed with use of such a design. The mean duration
of follow-up was ten years. Definite healing of the discontinuity was evident
in eighteen of twenty-one hips. No cup migration was observed. No cup was
revised. Despite the durable fixation and healing of the bone defects, the
mean Harris hip score was <90 and 71% of the patients still required
walking aids.
The tantalum trabecular metal cup was introduced into clinical use in 1997.
Unger et al.8
evaluated sixty consecutive revisions that were performed with use of this
type of cup, with most (92%) of the cups not having any supplemental screws.
The mean duration of follow-up was just short of four years. One hip was
revised because of cup loosening, 75% of the hips had no radiolucencies around
the cup, and 17.5% of the hips had new nonprogressive radiolucencies around
the cup. No cup migrated. Sporer and Paprosky reported on the use of a modular
trabecular metal cup with multiple screws to treat major acetabular
deficiencies. They inserted the cups with trabecular metal augments to
supplement peripheral, and occasionally central, defects in thirty-four
patients with a Paprosky type-IIIA defect (proximal and medial migration). The
mean duration of follow-up was 3.1 years. All cups remained stable. One hip
required revision to a constrained liner because of instability. The authors
also reported on the use of a similar technique for thirteen patients with a
Paprosky type-IIIB defect (pelvic discontinuity). The mean duration of
follow-up was 2.6 years, and no cup was loose. Potential concerns included
fretting between the shell and the augments and the failure of bone
ingrowth.
Wear and Osteolysis
Osteolysis remains the principal clinical issue that can adversely affect
the durability and function of any total hip arthroplasty. Engh et al.
reviewed the plain radiographs and computed tomography scans of ninety-five
hips after a minimum duration of follow-up of five years. Plain radiographs
underestimated the total prevalence of pelvic osteolysis as compared with the
findings on computed tomography scans. Most of the underestimation was in the
detection of smaller lesions that were perhaps not of clinical relevance.
There was a strong correlation (r2 = 0.81) between the
two-dimensional size and the three-dimensional volume estimate of the lesions.
Engh concluded that the clinical relevance of pelvic osteolysis can be
reliably estimated on plain radiographs. This is important with regard to
surveillance costs in general community practice, where sophisticated software
and a high-resolution computed tomography scanner may not be readily available
or affordable.
One of the most frequent revision procedures is articulation exchange with
bone-grafting of osteolytic areas. Puri et al. compared the preoperative and
postoperative computed tomography scans for fifteen patients who underwent
such a revision. The mean size of the lesions decreased from 6.38
cm2 preoperatively to 2.94 cm2 at two years
postoperatively (p < 0.0001). Graft incorporation was seen in thirteen of
the fifteen cases, and no cup became loose. The authors concluded that modular
liner exchange with bone-grafting is an effective technique with which to
restore bone stock and to maintain cup stability.
Highly Cross-Linked Polyethylene
Highly cross-linked polyethylene has received tremendous focus since its
introduction into clinical use. It has evolved into the most frequently used
bearing surface for total hip arthroplasty. However, clinical wear
measurements generally have not matched those reported initially from
laboratory testing. McCalden et al. reported wear measurements in a
prospective, randomized trial of 100 hips that were treated with either
conventional or highly cross-linked polyethylene. Both groups were treated
with identical implant designs and fixation techniques. There was no
difference between the groups with regard to demographic characteristics or
clinical outcome measures. The two-dimensional wear rates were 0.14 and 0.09
mm/year for the conventional and highly cross-linked polyethylene groups,
respectively. The three-dimensional wear rates were 0.15 and 0.11 mm/year for
the conventional and highly cross-linked groups, respectively. These
differences represented 36% and 27% reductions for two and three-dimensional
wear, respectively. Engh et al. reported intermediate-term wear measurements
for more than 200 hips that were randomized to total hip arthroplasty with use
of either conventional or highly cross-linked polyethylene. All femoral heads
were 28 mm. The mean linear wear rates were 0.20 and 0.02 mm/year for the
conventional and highly cross-linked polyethylene groups, respectively. Geller
et al. reported wear measurements for seventy-five total hip arthroplasties
after a minimum duration of follow-up of five years. The polyethylene liners
had been made with use of an electron beam process. No osteolysis was observed
in the femur or pelvis. There was no difference in the wear rate between 28-mm
heads (0.01 mm/year) and 32-mm heads (0.06 mm/year). Longer follow-up will be
necessary to confirm that these lower wear rates continue to correlate with a
reduced prevalence of osteolysis and implant fixation failure.
Retrieval analysis of failed and well-functioning implants has demonstrated
a high frequency of neck-liner impingement. Impingement can cause wear and
even fracture of the liner rim. Usrey et al. reported that sixty-eight (60%)
of 113 retrieved liners showed evidence of impingement at the rim, with
moderate to severe damage being seen in association with 32% of the liners.
They specifically examined the liners for backside wear, which was evident in
association with 31% of the liners. Backside wear was evident in 61%
(twenty-two) of the thirty-six liners that showed impingement. In contrast,
backside wear was evident in only 16% (twelve) of the seventy-three liners
that showed no impingement. This difference was significant (p < 0.0001).
There was also a significant difference between the groups with regard to the
wear rate (0.33 mm/year for liners with impingement, compared with 0.19
mm/year for those without impingement; p = 0.03). Currier et al. reported
fatigue fractures of highly cross-linked polyethylene liners secondary to
oxidation. They analyzed seven retrieved liners made of highly cross-linked
gamma-irradiated polyethylene that was annealed below melting temperature. Six
liners were found to have reduced mechanical properties due to oxidation. In
fact, five showed fatigue damage (cracking and delamination) at the liner rim.
These changes were evident after only two years of in situ use. Researchers
from the same institution reported rim fractures in three of eleven retrieved
liners made of another highly cross-linked polyethylene. In this group, there
was no excessive oxidation of the material itself in the areas of rim
fracture. The damage always occurred in the thinnest portion of the liner and
in cups that had been placed in a more vertical position (that is, in greater
abduction). These findings suggest that highly cross-linked polyethylene
liners are subject to failure mechanisms that are not dissimilar to those of
conventional polyethylene. Kurtz et
al.9 analyzed the
mechanical properties of twelve retrieved cup liners that were made of highly
cross-linked gamma-irradiated polyethylene. All liners had an identical design
with regard to shell geometry and locking mechanism. All liners were revised
for reasons other than wear. The mean time in situ was 1.9 years. Maximum
oxidation occurred near the unloaded bearing surface at a mean depth of 0.9
mm. The ultimate strength properties in unworn areas were within 10% of the
initial ultimate strength of never-implanted liners. Strength properties were
lower in unworn areas than in worn areas (p = 0.01). There was no correlation
between the duration of implantation and the strength properties of either the
worn or unworn surfaces. There were variations within each liner as a function
of sampling. Additional retrieval studies will be needed to fully determine
the material properties after long-term clinical use.
No correlation has been made between wear reduction and clinical
osteolysis. One variable is the host's biological response to the wear debris
generated from highly cross-linked polyethylene. Minoda et al. analyzed wear
particles of highly cross-linked polyethylene that were retrieved from the
pseudoapsule at the time of revision surgery. They found fewer particles (5.33
× 107 particles/g) than the previously proposed critical
particle load for osteolysis (1 × 1010 particles/g). The
particle geometry was generally smaller (0.66 µm) and rounder (1.44 ×
1.37 µm) than that of particles of conventional polyethylene that were
analyzed in the same laboratory. However, the particle dimensions were larger
than what had been documented from in vitro testing. These data reflect the
particle characteristics from one particular material but do suggest that in
vivo wear of this material is different from in vitro wear testing.
Quantitative assessments of patient activity have been found to correlate
with wear following total hip arthroplasty. The single greatest limitation of
activity assessment with use of pedometers and other methods is patient
compliance. Silva et
al.10 compared the
use of a four-day assessment protocol with that of a longer assessment
protocol lasting seven days or more. The mean activity was 5464 steps/day
during the four-day collection period, compared with 5737 steps/day during the
longer collection period (a 5.0% difference). Patients were categorized into
several activity levels on the basis of the pedometer reading. There was high
agreement among the activity categories between the two data collection
methods. This is an important finding as the use of a shorter collection
period will ensure greater patient compliance. This form of activity
assessment does not account for seasonal variations, nor does it allow for
distinction of types of activities.
Ceramic Bearings
Intermediate-term clinical data are now available for two series of total
hip arthroplasties performed with use of ceramic-on-ceramic bearings. Capello
et al., in a study of 257 hips with a minimum duration of follow-up of five
years, reported no fracture or failure of the ceramic bearing surfaces. The
rate of femoral osteolysis was 0.6%, in contrast with the rate of 22.1% in a
comparable series of hips with metal-polyethylene bearings and identical cup
and stem designs. There was no difference between the two groups with regard
to the Harris hip score. Murphy reported on 141 hips after a mean duration of
follow-up of four years. There was no fracture or bearing failure and no
femoral or pelvic osteolysis. Toni et al. performed 3746 total hip
arthroplasties with use of a ceramic-on-ceramic bearing between 1990 and 2004.
They reported four head fractures and ten liner rim chips. No head fractures
occurred in hips that had been treated with a head that was 32 mm or larger.
The survival rate for this ceramic articulation was 98.2% at fifteen
years.
There have been reports of squeaking in hips with a ceramic-on-ceramic
bearing. O'Toole et al. conducted a detailed review of 2397 total hip
arthroplasties that had been performed with ceramic-on-ceramic bearings.
Seventeen patients (0.7%) reported squeaking. The mean time to the onset of
the squeaking was fourteen months. Compared with patients without squeaking,
those who reported squeaking were younger (mean age, fifty-six years), heavier
(mean weight, 90 kg), and taller (mean height, 180 cm). Eight patients had
squeaking only when the hip was in flexion (such as when picking up something
from the floor); these patients did not associate squeaking with walking,
whereas the other patients did. Computed tomography scans were made to assess
component position. There was no significant difference between the groups
with regard to cup abduction or anteversion. Toni et al. reported on the fluid
analysis of aspirates from squeaking ceramic-on-ceramic hips. The presence of
ceramic particles established the diagnosis of articulation damage leading to
squeaking and to revision surgery. The authors recommend using computed
tomography scans to assess component position as suboptimal position could
result in impingement, creating ceramic chips or even a bearing fracture.
Metal-on-Metal Bearings
Controversies remain with regard to the safety of metal-on-metal bearing
surfaces, particularly with regard to metal ion release and the potential for
hypersensitivity reactions. Jacobs et al. reported additional data in their
ongoing prospective longitudinal study examining the serum and urine metal ion
levels in patients with such a bearing surface. The cohort included twenty-two
patients with surface replacements and fifteen patients with total hip
replacements. All components and bearing surfaces were made by the same
manufacturer. Serum and urine samples were collected at specific time
intervals after surgery for as long as thirty-six months. The ion levels were
greater in both groups after surgery compared with preoperative levels, and,
at four months after surgery, the serum chromium and cobalt levels were
greater in the surface replacement group than in the total hip arthroplasty
group. The authors concluded that the larger head sizes (range, 46 to 52 mm)
used for surface replacement did not result in greater metal ion
concentrations when compared with the head size used for total hip
arthroplasty (32 mm). Rasquinha et al. reported on steady-state serum metal
ion levels in patients with four different bearing surfaces after a minimum
duration of follow-up of five years. Forty total hip arthroplasties were
randomized into four groups: (1) cemented metal-polyethylene, (2) cementless
metal-polyethylene, (3) cementless ceramic-polyethylene, and (4) cementless
metal-metal. There were no differences among the four groups with regard to
clinical and radiographic outcomes. The highest cobalt and chromium levels
were found in the metal-metal group. These data suggest lower corrosion and
release of metal ions at the head-neck taper junction in association with
ceramic heads as compared with metal (cobalt-chromium) heads.
One of the proposed advantages of hard-on-hard bearings is a reduction in
osteolysis. Park et
al.11, in a study
of 169 total hip arthroplasties that had been performed with a contemporary
metal-on-metal bearing, reported ten cases of osteolysis (prevalence, 5.9%)
after two years of follow-up. The patients with early osteolysis had a
significantly higher prevalence of hypersensitivity to cobalt than control
patients did (p = 0.03). Two patients underwent reoperation, and periarticular
tissue was retrieved. Histologic examination showed a perivascular infiltrate
of lymphocytes. Immunophenotyping demonstrated CD3-positive T cells and
CD68-positive macrophages, and immunohistochemical analysis demonstrated the
presence of bone-resorbing cytokines. These findings suggest that delayed
hypersensitivity may be the cause of early osteolysis in some patients
following metal-on-metal total hip arthroplasty.
The increasing use of metal-on-metal bearings in younger patients has
raised the concern of potential adverse effects, particularly on child-bearing
females. McMinn et al. collected whole blood from the mother and umbilical
cord after five deliveries. Four of the five patients had a unilateral
metal-on-metal surface replacement, and the other patient had a bilateral
surface replacement. The levels of chromium and cobalt recorded in the cord
blood were an average of 50% lower than those recorded in the maternal blood.
There also was an excellent correlation between cord and maternal blood with
regard to the levels of trace elements. It is thus possible to follow maternal
blood ion levels during pregnancy as a monitor of fetal safety. These data
clearly underscore the importance of continued surveillance in patients with
metal-on-metal bearings.
Retrieval analysis of a well-functioning metal-on-metal total hip
replacement has been reported in the past. Clarke et
al.12 reported the
results of retrieval analysis of a McKee-Farrar total hip replacement after
thirty years. They specifically looked for surface roundness and roughness.
The deviation from perfect roundness for the head and cup was 15.6 µm and
15.1 µm, respectively. Surface roughness was 0.015 µm for the head and
0.012 µm for the cup. Histologic examination did not demonstrate any
histiocytes or plasma cells. Cobalt and chromium were not identified in the
tissues. No systemic trace-element analysis was done prior to the patient's
death.
Infection
Multiple laboratory tests and imaging modalities are currently used to
diagnose periprosthetic infection. Deirmengian et al. analyzed gene expression
patterns in white blood cells isolated from joints with implants. They found
distinctively different patterns of gene expression between infected and
noninfected cases. They also reported different patterns between joints with
infection and those with inflammatory arthritis (gout). The differences were
highly significant (p < 0.001). The most important finding was that
abnormal gene expression was not found in cells isolated from noninfected
joints. It is hoped that with further refinement, synovial fluid analysis for
gene expression may provide increased sensitivity and specificity in
comparison with the tests that are currently available. Parvizi et al.
reported on fluorodeoxyglucose-positron emission tomography (FDG-PET) imaging
as a noninvasive diagnostic modality for periprosthetic infection. They
evaluated eighty-three patients who underwent revision surgery. Infection was
correctly diagnosed with FDG-PET in twenty-one of twenty-three cases
(specificity, 91.3%). Infection was correctly excluded in sixty-two of
sixty-three cases (specificity, 98.4%). The positive and negative predictive
values of this imaging technique were 95.5% and 96.9%, respectively. These
values were far superior to those for indium-111 white blood-cell
scintigraphy.
Cahill et al. compared the quality of life for thirty-four patients who had
an infection with that for sixty-two patients who did not have an infection.
They used a variety of outcome measurement instruments, including the WOMAC,
SF-36, and Assessment of Quality of Life (AQoL) instruments. Functional
outcome and health-related quality of life parameters were significantly
poorer in those who had an infection (p < 0.05). The treatment of infection
is a burden to both hospitals and surgeons. Bozic and Ries analyzed resource
utilization at a tertiary teaching hospital. They collected data on
twenty-five consecutive revision total hip arthroplasties for two-stage
reimplantation, twenty-five revisions for reasons other than infection, and
twenty-five primary total hip arthroplasties over an eighteen-month period.
Revision for the treatment of infection was associated with a longer operative
time, more blood loss, a higher rate of complications, and a longer hospital
stay. Moreover, revision for the treatment of infection resulted in higher
costs for the inpatient stay and higher outpatient charges during the twelve
months following the revision procedure. The referral rate for infection at
the site of total hip arthroplasty had increased significantly at that center
over the previous five years, while there had been no change in the referral
rate for revisions for reasons other than infection during the same period.
This trend could reflect the technical and financial disincentives for
hospitals and surgeons to provide treatment for patients with infection at the
site of a total hip arthroplasty.
The most effective treatment protocol for an infection at the site of a
total hip arthroplasty is a two-stage revision. Surgeons from the Mayo Clinic
reviewed the results of 169 consecutive total hip arthroplasties that had been
performed with use of a two-stage protocol over a period of ten years (from
1988 to 1998). Cement fixation was used in 72% of the procedures. The mean
duration of follow-up was six years. The rate of rerevision was 15% overall,
with rates of 9.5% for reinfection, 5% for aseptic loosening, and 0.5% for
periprosthetic fracture. Reinfection was correlated neither with the initial
pathogen nor with the type of fixation.
One of the most difficult clinical situations is having unexpected positive
intraoperative cultures following revision total hip arthroplasty. There is no
consensus with regard to the best method for the treatment of this problem.
Berend et al. reported seven positive cultures after 103 revisions in which
the preoperative workup had been negative for infection. They chose to treat
each patient with intravenous antibiotics for six weeks without additional
surgery. None of the patients had development of a clinical infection or
fixation failure.
Dislocation
Dislocation remains one of the most frequent reasons for early total hip
revision. Hoeffel et al. reviewed the data on seventy-six revisions for
recurrent dislocation that had been collected in a community-based registry of
3210 total hip arthroplasties from 1991 to 2003. The mean time from the index
total hip arthroplasty to revision was 1.1 years. The techniques that were
used included liner/head exchange (50%), shell revision (42%), stem/shell
revision (5%), and insertion of a constrained liner (3%). Redislocation
occurred in 16% of the hips, with half of those requiring another revision.
The lowest redislocation rate was associated with complete shell revision
(4.5%), and the worst rate was associated with liner/head exchange (10.5%).
The rate of survival of the revised hips was 97.1% at twelve years. One newer
option for the treatment of recurrent dislocation is the use of a
large-diameter femoral head. This improves the head-neck ratio and increases
the offset and the range of motion. Paprosky and Sporer reported the clinical
results for forty-seven hips that had undergone revision for the treatment of
recurrent dislocation. Seventeen hips had received a 36-mm head, eight had
received a 40-mm head, and twenty-two had received a larger tripolar component
with a >40-mm articulation. The mean duration of follow-up was three years.
Two redislocations occurred; both were successfully treated by changing the
shell position.
Periprosthetic Femoral Fracture
Periprosthetic femoral fractures are complex and difficult to treat. Their
incidence will continue to increase as more patients are living longer with
higher activity levels. In addition, the increased incidence of revision
surgery could also contribute to a higher rate of periprosthetic fractures.
Lindahl et al.13
reported on 1049 fractures, recorded in the Swedish Hip Registry, that had
occurred between 1979 and 2000. The overall incidence was 0.4% for primary
total hip arthroplasty and 2.1% for revision total hip arthroplasty. The mean
time from the procedure to the fracture was 7.4 years for primary total hip
arthroplasty and 3.9 years for revision total hip arthroplasty. The ten-year
rate of implant survival was 69.9% following the treatment of a periprosthetic
fracture, compared with 90.5% for hips without a periprosthetic fracture.
Broaching of the femur is a critically important step in terms of sizing,
stability, and fixation durability of the stem. Intraoperative fracture of the
proximal part of the femur is the most frequent complication associated with
minimal-incision techniques. There is definitely a risk when there is a
geometric mismatch between the broach and the femur as occurs in association
with the use of tapered and wedge-shaped implant designs. Markel et al.
evaluated the impaction forces on the broaches and the stem implant with use
of instruments with strain gauges. They operated on seventeen patients using a
flat, tapered, wedge-shaped design. The mean number of impactions for canal
preparation and insertion of the stem was 111. A wide range of impaction
forces was recorded, with the highest approaching 9000 lb. An important
finding was that the maximum force used for the last broach was nearly twofold
greater than the force used to insert the stem. Therefore, there is greater
risk of a calcar fracture in association with the use of the final broach than
there is in association with the insertion of the stem itself. Data such as
these may be of use for future instrument design and surgical training. Berend
et al. reviewed the results of 3089 total hip arthroplasties that had been
performed with cement (2295 hips) or without cement (794 hips) at a single
center over fourteen years. Calcar fractures occurred in 2.7% of the hips. The
fracture rate was higher for cementless stems (6.9%) than for cemented stems
(0.8%). Regression analysis identified the anterolateral approach and female
gender as risk factors. A fracture did not have an adverse effect on fixation
durability for cementless stems but led to inferior results for cemented stems
(rate of survival of stem fixation, 97.1% compared with 93.3%) after a mean
seven years of follow-up.
At the present time, there is no agreement with regard to the most
effective treatment method for periprosthetic fracture. This is in part due to
the many different patterns of fractures and the need to consider implant
fixation stability. Contemporary fracture fixation methods have focused on
tissue-preserving techniques. Ricci et
al.14 reported the
results of treatment of fifty consecutive Vancouver type-B1 fractures
(fractures around a stable implant) with use of indirect reduction and a
single lateral plate inserted with tissue-preserving techniques. No bone graft
was used. After a mean duration of follow-up of twenty-four months, all
fractures had healed. There was one deep infection, and 73% of the patients
returned to their prefracture level of functional capacity.
Nerve Palsy
Nerve palsy is a rare but devastating complication of total hip
arthroplasty. Farrell et
al.15 reviewed the
results of 27,004 total hip arthroplasties that had been performed at one
institution over thirty years. Forty-seven patients with postoperative motor
nerve palsy were identified (prevalence, 0.17%). There were twenty-nine
complete palsies (including sixteen peroneal, eleven sciatic, and two femoral
palsies) and eighteen incomplete palsies (including fourteen peroneal, three
sciatic, and one femoral palsy). Risk factors included hip dysplasia,
posttraumatic arthritis, the posterior approach, and lengthening of >1.1
cm. Thirty-six percent of the patients with a complete palsy recovered fully
at a mean of twenty-one months. Seventy percent of the patients with an
incomplete palsy recovered fully. Fortunately, only five patients required
medication for the treatment of neurogenic pain. These data are useful for
preoperative patient counseling, particularly for patients with dysplasia and
posttraumatic arthritis.
Epidemiology
More evidence has become available in support of earlier surgical
intervention for patients in need of total joint arthroplasty. Lavernia et al.
prospectively followed sixty-three total hip arthroplasties and sixty-nine
total knee arthroplasties for three consecutive years. Comprehensive
evaluation was performed with use of the WOMAC, the SF-36, and the Quality of
Well-Being (QWB) scale. The patients were divided into two categories: (1)
severe functional impairment ("waited too long") and (2) mild
impairment. Significant improvement was seen in both groups after surgery (p
< 0.05); however, all outcome measures were inferior in patients with
severe impairment even after three years.
Disparity of care among different patient groups is an area of focus in
public health policy. Lau et al. reported significant (p < 0.05)
differences in the rates of total hip arthroplasty among racial groups in the
United States. The mean rates (per 100,000 individuals) were 72.2 for whites,
32.7 for blacks, 16.8 for Hispanics, 13.5 for Asians, and 8.7 for Native
Americans. This disparity could be related to access to care, cultural
barriers, and/or socioeconomic issues. Regardless of the reasons, the elderly
Hispanic and Asian populations are projected to grow by sixfold to sevenfold
in the next five decades, compared with a projected twofold increase among
elderly non-Hispanic whites. As the elderly minority population expands in the
United States, the unique challenges facing the delivery of health care to
these population segments must be addressed.
Safety of Hip Arthroplasty
In recent years, several authors have correlated surgical volume with
clinical outcome. Doro analyzed the results of total hip arthroplasty from the
Nationwide Inpatient Sample between 1988 and 2000. These data were based on a
random sample of 20% of all patient admissions in the United States. Hospitals
were categorized into either the highest quartile (more than 146 primary and
more than forty-six revision arthroplasties per year) or the lowest quartile
(fewer than forty primary and fewer than ten revision arthroplasties per
year). In-hospital mortality was the principal outcome measured. There was a
significant difference between high-volume and low-volume hospitals with
regard to the rate of mortality following primary arthroplasty (0.16% compared
with 0.29%; p < 0.01). There also was a significant difference in mortality
following revision surgery as compared with primary surgery (0.48% compared
with 1.2%; p < 0.01). The difference was especially profound in older
patients, with the data showing a 7.5-fold greater mortality for patients more
than seventy-five years of age undergoing primary surgery in low-volume
hospitals.
There has been increased interest in reducing hospital stay for a variety
of reasons. Parvizi et al. prospectively evaluated the medical complications
in a study of 2411 patients undergoing 2631 hip and knee arthroplasties and
found that 32.4% of the patients had at least one complication. Twenty-one
percent of these were considered to be major complications, including death.
Only 22% of the major complications occurred on the day of the operation,
whereas 66% had become apparent by three days. The authors concluded that
early discharge protocols may only be safe for selected patients.
Pain Management
Multimodal perioperative pain management has been implemented to facilitate
accelerated rehabilitation. Peters et al. evaluated the efficacy of a protocol
involving the use of preoperative and postoperative oral narcotics and COX-2
inhibitors, spinal anesthesia, and either femoral nerve block (for patients
managed with total knee arthroplasty) or local anesthetic infiltration (for
patients managed with total hip arthroplasty). Two cohorts of fifty
consecutive hip and knee arthroplasties were evaluated before and after
institution of the protocol. There was a significant (p < 0.005) decrease
in narcotic use and length of hospital stay and an increase in distance walked
during therapy for patients managed with the protocol. Shore et al. performed
a prospective, randomized study to evaluate the efficacy of periarticular
injection with use of a combination of ropivacaine, ketorolac, epimorphine,
and epinephrine in 128 patients undergoing total knee and total hip
arthroplasty. They found significant (p < 0.01) reductions in intravenous
narcotic use and in visual analog pain scores in patients who received the
injections. Parvataneni et al. also reported reduction of narcotic use and
physical therapy services following the institution of a local injection
protocol. More importantly, 90% of the patients reached functional and
activity milestones earlier than those who did not receive injections and 74%
described their recovery as "easy" on a standardized
questionnaire. Newer pain-management protocols coupled with tissue-preserving
surgical techniques may provide accelerated functional return, at least in
selected patients.
Inpatient Rehabilitation
There have been some recent policy changes with regard to inpatient
rehabilitation following total joint arthroplasty. Costs associated with
inpatient rehabilitation have not been widely studied. Lavernia et al.
conducted a cost analysis of 136 consecutive patients following primary total
hip or knee arthroplasty. The mean cost (and standard deviation) was $10,751
± $598 for the rehabilitation unit stay and $2393 ± $92 for
home-care services. This difference was significant (p < 0.001). The
authors projected that the annual rehabilitation costs for total joint
arthroplasty in the United States would exceed $3.4 billion. Vincent et al.
examined functional improvement and costs for inpatient rehabilitation
following 331 primary and revision total hip arthroplasties. Patients managed
with primary total hip arthroplasty had a shorter length of stay than patients
managed with revision total hip arthroplasty, and the degree of improvement in
functional independence was inferior following revision total hip
arthroplasty. The total hospital cost was also higher in the revision group.
The worst outcome and highest costs were found for patients who underwent
revision total hip arthroplasty for the treatment of infection.
Total Hip Arthroplasty in Specific Patient Populations
Obese Patients
There have been several reports in the high-profile lay press on the
perceived reluctance of surgeons to operate on obese patients. Turgeon et al.
reviewed resource utilization in a study of 1343 patients who underwent total
hip arthroplasty in two community hospitals. Obesity was stratified according
to body mass index, with the highest category being a body mass index of
>35. The most obese patients were 2.3 times more likely to have a hospital
stay in excess of five days and 2.6 times more likely to require inpatient
rehabilitation. There was no difference among the groups with regard to
operative times or blood transfusions. An important finding was that patients
with a body mass index of >25 (all obese categories) had lower Harris hip
scores at one year than did nonobese patients. Dalury evaluated forty-six
patients (fifty-seven total hip arthroplasties) with a body mass index of
>35 after a mean of 5.8 years. The mean Harris hip score improved from 36
to 84. The most common complication was wound drainage (eight hips). There
were no deep infections, dislocations, leg-length inequalities, or venous
thromboemboli. No cup was revised. One stem was revised because of loosening,
and another was loose according to radiographic criteria. A technical
challenge was that 17% of the cups were placed with a high abduction angle,
probably because of exposure limitations. These data will facilitate the
formulation of practice guidelines and modifications in surgical techniques to
minimize complications in the obese patient population.
Young Patients
Lombardi et al. reported on 247 hip arthroplasties that were performed with
use of a tapered titanium-alloy stem design with a plasma-sprayed porous
surface in patients who were less than forty years old. The mean age at the
time of surgery was 34.8 years, and the mean body mass index was 29
kg/m2. The mean duration of follow-up was 5.8 years, with 125 hips
having a minimum of five years of follow-up and fifty-one hips having a
minimum of ten years of follow-up. Three stems were revised because of
loosening (two) or thigh pain (one). The rate of stem survival was estimated
to be 98.8% for up to eighteen years. Archibeck et al. reported on 100
consecutive total hip arthroplasties that were performed with use of a
proximally-coated straight stem in patients who were less than fifty years
old. The mean duration of follow-up was nine years. The result was rated as
good or excellent in 84% of the hips. No stem was revised for loosening. There
were twelve revisions, including six for wear or osteolysis, five for
dislocation, and one for infection. The rate of stem survival at ten years was
98%.
Controversy has existed with regard to the best surgical method with which
to treat a displaced femoral neck fracture. Several recent reports have
presented data in support of total hip arthroplasty for the treatment of such
fractures in the elderly. Keating et
al.16 conducted a
prospective, randomized trial to evaluate the differences in efficacy,
complications, and cost among three treatment options for the treatment of
displaced femoral neck fractures in the elderly. The investigators randomized
207 patients to internal fixation, hemiarthroplasty, or primary total hip
arthroplasty. The rate of reoperation was 36% for the internal fixation group,
compared with 5% for the hemiarthroplasty group and 9% for the total hip
arthroplasty group. The internal fixation group also had inferior hip and
functional scores. Patients who had been managed with total hip arthroplasty
had slightly better hip and functional scores at twenty-four months than did
those who had been managed with hemiarthroplasty. The overall cost over two
years was higher for the internal fixation group, principally because of a
higher rate of reoperation. There was no difference between the
hemiarthroplasty and total hip arthroplasty groups with regard to cost.
Hoffman et al. reviewed data from thirteen prospective, randomized trials
comparing internal fixation and arthroplasty for the treatment of displaced
femoral neck fractures. They compared quality-adjusted life-years (QALYs) and
cost. The best functional outcome was attained with total hip arthroplasty
(4.5 QALYs), followed by hemiarthroplasty (3.6 QALYs) and internal fixation
(2.6 QALYs). Cost-effectiveness analysis showed that hemiarthroplasty had the
lowest cost per procedure ($22,208), followed by total hip arthroplasty
($23,551) and internal fixation ($29,733). The incremental cost-effectiveness
of total hip arthroplasty over hemiarthroplasty was $2400/QALY. Blomfeldt et
al.17 performed a
prospective, randomized trial of 102 consecutive patients in which internal
fixation was compared with total hip arthroplasty for the treatment of
displaced femoral neck fractures. Patients were followed for four years. The
reoperation rate was 4% for the total hip arthroplasty group and 47% for the
internal fixation group. Hip function was better and the decline in
health-related quality of life parameters was less in the total hip
arthroplasty group at four, twelve, and twenty-four months. The difference was
not significant by forty-eight months.
The editorial staff of The Journal reviewed a large number of
recently published research studies related to the musculoskeletal system that
received a Level of Evidence grade of I. Over 100 medical journals were
reviewed to identify these articles, which all have high-quality study design.
In addition to articles cited already in this Update, six level-I articles
were identified that were relevant to total hip arthroplasty. A list of those
titles is appended to this review after the standard bibliography. We have
provided a brief commentary about each of the articles to help guide your
further reading, in an evidence-based fashion, in this subspecialty area.
Bhandari M, Bajammal S, Guyatt GH, Griffith L, Busse JW, Schunemann H,
Einhorn TA. Effect of bisphosphonates on periprosthetic bone mineral
density after total joint arthroplasty. A meta-analysis. J Bone Joint Surg
Am. 2005;87:293-301.
Six studies (296 patients) were analyzed to determine whether
bisphosphonates have an effect on periprosthetic bone loss in the hip and
knee. All studies were randomized, controlled trials in which dual-energy
x-ray absorptiometry scans were used to measure periprosthetic bone mineral
density after total joint arthroplasty. Significantly less periprosthetic bone
loss had occurred in patients managed with bisphosphonates than in control
patients at three, six, and twelve months (p < 0.01). Bone loss was
prevented more effectively around cemented than around cementless implants. A
larger effect was also seen following total knee arthroplasty than following
total hip arthroplasty. These studies focused on the results of the dualenergy
x-ray absorptiometry scans alone and not on clinical data such as patient
outcomes, side effects, the periprosthetic fracture rate, or the rate of
revision. Additional analysis is necessary before general recommendations can
be made.
Mundy GM, Birtwistle SJ, Power RA. The effect of iron
supplementation on the level of haemoglobin after lower limb arthroplasty.
J Bone Joint Surg Br. 2005;87:213-7.
One hundred and twenty patients were randomized to treatment with ferrous
sulfate tablets or placebo for three weeks following primary total hip or knee
arthroplasty. Ninety-nine patients completed the study. The dose of ferrous
sulfate was 200 mg three times per day. There was no difference in demographic
characteristics or preoperative laboratory values between the groups. The
hemoglobin levels were measured preoperatively, on Days 1 and 5 after surgery,
and during Weeks 3 and 6. Patients with preexisting anemia (as indicated by a
hemoglobin level of <13 for men and of <11.5 for women) were excluded.
The hemoglobin levels dropped by an average of 30% immediately after surgery
and recovered to an average of 85% at three weeks after surgery, independent
of iron therapy. No significant difference was seen between the two groups in
terms of hemoglobin levels after surgery. However, there was a slight trend
toward increased hemoglobin levels at six weeks in the iron-supplementation
group. Ten percent of the patients were noncompliant because of
gastrointestinal side effects. On the basis of these results, the routine
administration of iron supplementation following total joint arthroplasty does
not appear to provide any added benefit with regard to recovery of the
hemoglobin level.
Chimento GF, Pavone V, Sharrock N, Kahn B, Cahill J, Sculco TP.
Minimally invasive total hip arthroplasty: a prospective randomized study.
J Arthroplasty. 2005;20:139-44.
Sixty patients were randomized to receive either a minimally invasive total
hip arthroplasty (twenty-eight patients) or a standard-incision total hip
arthroplasty (thirty-two patients). The surgical approach was decided by
drawing a card immediately preoperatively to determine whether an 8-cm or
15-cm incision was to be used. A posterolateral approach was used, with repair
of the short external rotators at time of closure. Patients with a body mass
index of >30 and those who had had previous hip surgery were excluded.
There was no significant difference between the patients in terms of age,
weight, body mass index, or hip score. The minimally invasive surgery group
had significantly less blood loss intraoperatively (p = 0.003) and less total
blood loss (p = 0.009) when adding drain output. There was no difference
between the groups in terms of operative time, transfusion requirements, pain
medication usage, or serum levels of interleukin-6. Most of the arthroplasties
were performed with hybrid fixation. There was no difference in cement
grading. The rate of limp at six weeks was higher in the standard-incision
group than in the minimally invasive surgery group (46.8% compared with
21.4%). Other rehabilitation milestones were similar. There was no significant
difference between the groups with regard to the rate of complications, but
there were two dislocations in the minimally invasive surgery group as
compared with none in the standard-incision group. The only clinically
relevant advantages of single-incision posterior minimally invasive total hip
arthroplasty were a reduction in blood loss and a lower rate of limp at six
weeks. Longer-term follow-up is necessary to determine if minimally invasive
total hip arthroplasty is superior in terms of daily function and the
durability of fixation.
Lai KA, Shen WJ, Yang CY, Shao CJ, Hsu JT, Lin RM. The use of
alendronate to prevent early collapse of the femoral head in patients with
nontraumatic osteonecrosis. A randomized clinical study. J Bone Joint Surg
Am. 2005;87:2155-9.
Forty patients with Steinberg grade-II or III osteonecrosis of the femoral
head were randomized to receive 70 mg of alendronate orally once weekly for
twenty-five weeks. The control group did not receive any medication. The
patients were followed for a minimum of two years with use of the Harris hip
score, serial radiographs, and magnetic resonance imaging scans. Only two of
twenty-nine femoral heads in the treatment group collapsed, whereas nineteen
of twenty-five in the control group collapsed (p < 0.001). In addition, one
hip in the treatment group went on to total hip arthroplasty, compared with
sixteen hips in the control group. Many risk factors are associated with
osteonecrosis. This particular protocol may be effective only for selected
subgroups of patients. Larger patient populations and longer follow-up periods
are necessary in order to fully assess the clinical effectiveness of using
alendronate to prevent collapse and progression to total hip arthroplasty.
Kalisvaart KJ, de Jonghe JF, Bogaards MJ, Vreeswijk R, Egberts TC,
Burger BJ, Eikelenboom P, van Gool WA. Haloperidol prophylaxis for elderly
hip-surgery patients at risk for delirium: a randomized placebo-controlled
study. JAm Geriatr Soc. 2005;53:1658-66.
This randomized, double-blind, placebo-controlled trial was an attempt to
define the role of routine administration of haloperidol to high-risk
geriatric patients to prevent delirium following hip surgery. The study
included 430 patients who were seventy years old or more and were at risk for
postoperative delirium. Haloperidol (1.5 mg per day, divided into three doses)
was given at the time of admission and was continued for seventy-two hours
after surgery. The investigators recorded the prevalence of delirium as well
as its severity and duration. The duration of hospital stay was also
determined. Delirium was reported in 15.8% of the patients, and the
prophylactic administration of haloperidol had no effect on its prevalence.
However, it did significantly decrease the severity and duration of the
dementia (5.4 compared with 11.8 days; p < 0.01). In addition, the mean
length of the hospital stay was reduced by more than five days. The medication
had no side effects. This appears to be a safe and cost-effective treatment
for this particular patient population.
Tak E, Staats P, Van Hespen A, Hopman-Rock M. The effects of an
exercise program for older adults with osteoarthritis of the hip. J
Rheumatol. 2005;32:1106-13.
The investigators randomized 109 patients who had osteoarthritis of the hip
to an eight-week exercise program and compared them with controls who followed
no formal exercise program. All of the patients were at least fifty-five years
old and lived independently. Outcome was measured according to pain, the
Harris hip score, a quality of life assessment, body mass index, and
disability. Patients were followed prospectively for three months. The
exercise program, consisting of observed strength-training, home exercises,
and dietary advice, was effective for decreasing pain and improving hip
function. It did not affect the quality of life or the body mass index.
However, there was insufficient follow-up to determine if these changes would
continue. Exercise and lifestyle alterations can reduce pain and improve
functional status in the short term. This is especially important for medical
delivery systems in which there is a long waiting time for total hip
arthroplasty.
Bal BS, Haltom D, Aleto T, Barrett M.
Early complications of primary total hip replacement performed with a
two-incision minimally invasive technique. J Bone Joint Surg
Am. 2005;87:
2432-8.872432
2005
[PubMed][CrossRef]
Moore MS, McAuley JP, Young AM, Engh CA
Sr. Radiographic signs of osseointegration in porous-coated acetabular
components. Clin Orthop Relat Res.
2006;444:
176-83.444176
2006
[PubMed][CrossRef]
Daniel J, Pynsent PB, McMinn DJ.
Metal-on-metal resurfacing of the hip in patients under the age of 55 years
with osteoarthritis. J Bone Joint Surg Br.
2004;86:
177-84.86177
2004
[PubMed][CrossRef]
Treacy RB, McBryde CW, Pynsent PB.
Birmingham hip resurfacing arthroplasty. A minimum follow-up of five years.
J Bone Joint Surg Br.
2005;87:
167-70.87167
2005
[PubMed][CrossRef]
Schmalzried TP, Silva M, de la Rosa MA,
Choi ES, Fowble VA. Optimizing patient selection and outcomes with total hip
resurfacing. Clin Orthop Relat Res.
2005;441:
200-4.441200
2005
[PubMed][CrossRef]
Kurtz S, Mowat F, Ong K, Chan N, Lau E,
Halpern M. Prevalence of primary and revision total hip and knee arthroplasty
in the United States from 1990 through 2002. J Bone Joint Surg
Am. 2005;87:
1487-97.871487
2005
[CrossRef]
Friesecke C, Plutat J, Block A. Revision
arthroplasty with use of a total femur prosthesis. J Bone Joint Surg
Am. 2005;87:
2693-701.872693
2005
[CrossRef]
Unger AS, Lewis RJ, Gruen T. Evaluation
of a porous tantalum uncemented acetabular cup in revision total hip
arthroplasty: clinical and radiological results of 60 hips. J
Arthroplasty. 2005;20:
1002-9.201002
2005
[CrossRef]
Kurtz SM, Hozack W, Turner J, Purtill J,
MacDonald D, Sharkey P, Parvizi J, Manley M, Rothman R. Mechanical properties
of retrieved highly cross-linked crossfire liners after short-term
implantation. J Arthroplasty.
2005;20:
840-9.20840
2005
[PubMed][CrossRef]
Silva M, McClung CD, Dela Rosa MA, Dorey
FJ, Schmalzried TP. Activity sampling in the assessment of patients with total
joint arthroplasty. J Arthroplasty.
2005;20:
487-91.20487
2005
[PubMed][CrossRef]
Park YS, Moon YW, Lim SJ, Yang JM, Ahn
G, Choi YL. Early osteolysis following second-generation metal-on-metal hip
replacement. J Bone Joint Surg Am.
2005;87:
1515-21.871515
2005
[PubMed][CrossRef]
Clarke MT, Darrah C, Stewart T, Ingham
E, Fisher J, Nolan JF. Long-term clinical, radiological and histopathological
follow-up of a well-fixed McKee-Farrar metal-on-metal total hip arthroplasty.
J Arthroplasty. 2005;20:
542-6.20542
2005
[PubMed][CrossRef]
Lindahl H, Malchau H, Herberts P,
Garellick G. Periprosthetic femoral fractures classification and demographics
of 1049 periprosthetic femoral fractures from the Swedish National Hip
Arthroplasty Register. J Arthroplasty.
2005;20:
857-65.20857
2005
[PubMed][CrossRef]
Ricci WM, Bolhofner BR, Loftus T, Cox C,
Mitchell S, Borelli J Jr. Indirect reduction and plate fixation, without
grafting, for periprosthetic femoral shaft fractures about a stable
intramedullary implant. J Bone Joint Surg Am.
2005;87:
2240-5.872240
2005
[PubMed][CrossRef]
Farrell CM, Springer BD, Haidukewych GJ,
Morrey BF. Motor nerve palsy following primary total hip arthroplasty.
J Bone Joint Surg Am.
2005;87:
2619-25.872619
2005
[PubMed][CrossRef]
Keating JF, Grant A, Masson M, Scott NW,
Forbes JF. Randomized comparison of reduction and fixation, bipolar
hemiarthroplasty, and total hip arthroplasty. Treatment of displaced
intracapsular hip fractures in healthy older patients. J Bone Joint
Surg Am. 2006;88:
249-60.88249
2006
[CrossRef]
Blomfeldt R, Tornkvist H, Ponzer S,
Soderqvist A, Tidermark J. Comparison of internal fixation with total hip
replacement for displaced femoral neck fractures. Randomized, controlled trial
performed at four years. J Bone Joint Surg Am.
2005;87:
1680-8.871680
2005
[PubMed][CrossRef]