Pathogenesis of bone loss after total hip arthroplasty

Bone loss with or without evidence of aseptic loosening is a long term complication after total hip arthroplasty (THA). It occurs with all materials and in all prosthetic systems in use or that have been used to date. Bone loss after THA can be a serious problem in revision surgery because bone deficiencies may limit reconstructive options, increase the difficulty of surgery, and necessitate autogenous or allogenic bone grafting. There are three factors adversely affecting maintenance of bone mass after THA: (1) bone loss secondary to particulate debris; (2) adaptive bone remodeling and stress shielding secondary to size, material properties, and surface characteristics of contemporary prostheses; and (3) bone loss as a consequence of natural aging. This chapter reviews the mechanisms of the primary causes of bone loss after THA.

Early inpatient rehabilitation after elective hip and knee arthroplasty

CONTEXT

Inpatient rehabilitation after elective hip and knee arthroplasty is often necessary for patients who cannot function at home soon after surgery, but how soon after surgery inpatient rehabilitation can be initiated has not been studied.

OBJECTIVE

To test the hypothesis that high-risk patients undergoing elective hip and knee arthroplasty would incur less total cost and experience more rapid functional improvement if inpatient rehabilitation began on postoperative day 3 rather than day 7, without adverse consequences to the patients.

DESIGN

Randomized controlled trial conducted from 1994 to 1996.

SETTING

Tertiary care center.

PARTICIPANTS

A total of 86 patients undergoing elective hip or knee arthroplasty and who met the following criteria for being high risk: 70 years of age or older and living alone, 70 years of age or older with 2 or more comorbid conditions, or any age with 3 or more comorbid conditions. Of the 86 patients, 71 completed the study.

INTERVENTIONS

Random assignment to begin inpatient rehabilitation on postoperative day 3 vs postoperative day 7.

MAIN OUTCOME MEASURES

Total length of stay and cost from orthopedic and rehabilitation hospital admissions, functional performance in hospitals using a subset of the functional independence measure, and 4-month follow-up assessment using the RAND 36-item health survey I and the functional status index.

RESULTS

Patients who completed the study and began inpatient rehabilitation on postoperative day 3 exhibited shorter mean (+/-SD) total length of stay (11.7+/-2.3 days vs 14.5+/-1.9, P<.001), lower mean (+/-SD) total cost ($25891+/-$3648 vs $27762+/-$3626, P<.03), more rapid attainment of short-term functional milestones between days 6 and 10 (36.2+/-14.4 m ambulated vs 21.4+/-13.3 m, P<.001; 4.8+/-0.8 mean transfer functional independence measure score vs 4.3+/-0.7, P<.01), and equivalent functional outcome at 4-month follow-up.

CONCLUSION

These data showed that high-risk individuals were able to tolerate early intensive rehabilitation, and this intervention yielded faster attainment of short-term functional milestones in fewer days using less total cost.

Treatment of pelvic osteolysis associated with a stable acetabular component inserted without cement as part of a total hip replacement

Thirty-five patients who had had a primary total hip replacement with a porous-coated acetabular component inserted without cement had a revision procedure to treat pelvic osteolysis. The mean age at the time of the revision operation was forty-nine years (range, twenty-nine to eighty-five years). Forty-six distinct pelvic osteolytic lesions were noted radiographically around the thirty-five cups. These lesions ranged in size from 0.5 by 0.5 centimeter to 6.3 by 2.7 centimeters (mean, 2.6 by 1.7 centimeters). Fourteen of the thirty-five patients had no or only slight occasional pain at the time of diagnosis of the pelvic osteolysis, fifteen patients had pain attributed to a loose femoral component, one had pain related to a spontaneous fracture of the greater trochanter, and one had pain related to recurrent subluxation. The remaining four patients had pain in the groin despite radiographically stable implants. All of the metal-backed porous-coated acetabular components were stable according to the preoperative radiographs, and the stability was confirmed at the time of the revision. The metal shell was left in place and the acetabular liner was exchanged in all thirty-five patients. The osteolytic lesions were debrided, and thirty-four of the forty-six lesions were filled with allograft bone chips. The patients were evaluated a minimum of two years (range, two to five years; mean, 3.3 years) after the revision operation, and all thirty-five sockets were found to be radiographically stable. The bone grafts appeared to have consolidated, and none of the osteolytic defects had progressed. One-third of the lesions were no longer visible on radiographs, regardless of whether or not they had been filled with bone graft. The remaining lesions had decreased in size. It appears that, in the short-term, exchange of the liner and debridement of the granuloma, with or without use of allograft bone chips in the osteolytic defect, is a reasonable alternative to revision of the socket provided that the metal shell is solidly fixed at the time of the revision operation. If the metal shell has been markedly damaged by the femoral head, the locking mechanism for the polyethylene liner is not intact, or a satisfactory replacement liner is not available, then revision of the porous-coated acetabular component is indicated. These results must be considered preliminary. Since osteolysis may take several years to redevelop after a revision, additional follow-up is required.

Free vascularized fibula grafting for the treatment of osteonecrosis of the femoral head

Sixty-five patients (88 hips) who received free vascularized fibula grafting for treatment of osteonecrosis of the femoral head at the University of Pittsburgh Medical Center, were followed for at least 3 years (average followup, 5.5 years; range, 3-7 years). There were 46 men and 19 women with an average age of 37 years (range, 20-52 years). All patients were evaluated using history, physical examination, Harris Hip Score, anteroposterior and lateral radiographs, and magnetic resonance images. The classification system of Steinberg et al (1995) was used to stage the disease. At final evaluation, 31 hips (35.2%) were rated excellent (Harris Hip Score > 90 points, minimal or no pain), 30 hips (34.1%) were rated good (Harris Hip Score 80-89 points, slight pain), seven hips (8%) were rated fair (Harris Hip Score 70-79, slight or moderate pain), and 20 hips (22.7%) were rated poor (Harris Hip Score < 70, pain). Twenty hips in 17 patients required total hip arthroplasty. In the remaining hips, the disease apparently arrested and the contour of the femoral head was preserved. Kaplan-Meier survivorship analyses showed that the probability of conversion to total hip arthroplasty within an average of 5.5 years after free vascularized fibula grafting was 28% for Stage II hips and 38% for Stages III and IV hips. The hip survival rate for subgroups at 5.5 years was 100% for Stages IC and IIA, 94% for Stage IIB, 50% for Stage IIC, 80% for Stage IIIB, 58% for Stage IIIC, 72% for Stage IVA, and 58% for Stage IVB. Free vascularized fibula grafting is a reliable operation and can preserve hip function and diminish pain successfully.

Effect of a centralizing device on cement mantle deficiencies and initial prosthetic alignment in total hip arthroplasty

Sixty consecutive patients undergoing a primary hybrid total hip arthroplasty were randomized to receive a femoral component either with or without a distal centralizing device. The group with a centralizer had significantly fewer patients with cement mantle deficiencies (excessively thin areas of cement) than the group without a centralizer (P < .001). Furthermore, the centralizer group was, on average, in a neutral alignment, whereas the group without a centralizer was in a varus alignment (P < .001). It was concluded that the distal centralizing device significantly decreases the incidence of cement mantle deficiences and reproducibly aids in achieving a more neutral prosthetic alignment.

Identification of inducible nitric oxide synthase in human macrophages surrounding loosened hip prostheses

Exposure of rodent macrophages to certain cytokines and endotoxin results in the synthesis of inducible nitric oxide synthase (iNOS or NOS-II) leading to the production of large amounts of nitric oxide (NO). Cultures of human macrophages, in contrast, do not produce iNOS after cytokine stimulation, and their ability to act as a physiological source of NO remains questionable. Here we have used immunohistochemistry and in situ hybridization to demonstrate the presence of iNOS within human macrophages present in the interfacial membrane and pseudocapsule that surround failed prosthetic hip joints. Synovial tissue recovered from normal human joints did not express iNOS. Many of the iNOS-positive macrophages within the interfacial membrane had phagocytosed large amounts of polyethylene wear debris, suggesting a role for phagocytic stimuli in inducing iNOS in human macrophages. These findings additionally support a role for NO in modulating the localized bone resorption that accompanies the aseptic loosening of prosthetic joints.

Large amounts of polyethylene debris in the interface tissue surrounding bipolar endoprostheses. Comparison to total hip prostheses

A histologic and biochemical comparison of interface membranes around femoral components of bipolar endoprostheses (n = 17) and total hip prostheses (n = 17) inserted without cement was conducted. The patients' profiles were similar in both groups with respect to age, sex, primary diagnosis, weight, and the interval between primary and revision arthroplasty. Macroscopically, marked circumferential abrasion of the polyethylene insert in the retrieved bipolar cups was noted. Histologic analysis revealed significantly larger amounts of polyethylene debris in the bipolar group. The membranes from the bipolar group also produced significantly greater amounts of prostaglandin E2 (P < .05). The inflammatory membranes associated with large amounts of polyethylene debris may have contributed to aseptic loosening and osteolysis in patients with a bipolar hip prosthesis.

Technique for generating submicrometer ultra high molecular weight polyethylene particles

Ultra high molecular weight polyethylene wear debris is believed to have a major role in aseptic loosening of prosthetic joints. In order to study the cellular and host response to this and other such particulate debris, a source of fine ultra high molecular weight polyethylene debris is needed. We have described a technique to fracture the GUR 4150 primary ultra high molecular weight polyethylene grain, which reproducibly generated particles less than 1 micron in size. Furthermore, the particle morphology was similar to that of ultra high molecular weight polyethylene particles generated in vivo and retrieved from interfacial tissues. The fabricated polyethylene particles ranged from 0.1 to 33 microns in diameter, with a mean of 2.3 +/- 0.2 micron. Sixty percent of the particles were smaller than 1 micron and 90% were smaller than 7 microns. Using filtration and sedimentation, it is possible to acquire finer particle fractions. These particles are currently being used for biological response studies.

Effect of superior and superolateral relocations of the hip center on hip joint forces. An experimental and analytical analysis

With the extensive use of uncemented acetabular components in total hip arthroplasty, relocation of the hip center has become increasingly necessary to avoid bulk grafts and to promote contact between the porous prosthetic surface and bone. Compared with the anatomic hip center, superolateral relocation theoretically results in higher hip joint forces and has been shown in cemented acetabular components to result in an increased clinical failure rate. This study experimentally and analytically compared the hip joint forces in normal, superior, and superolateral hip center locations during both single-leg stance and stairclimbing, performing this comparison over a wide range of hip joint applied flexion moments. An advanced loading fixture was designed to allow any applied moment to be set independently of femoral position, incorporating all three major muscle groups active in stairclimbing position: extensors, abductors, and adductors. For all positions and moments tested, it was found that superolateral relocation caused significant increases in the total hip joint force, but did not affect the nonsagittal force component. Also, superior-only hip center relocation did not significantly affect the total joint force magnitudes or directions. The force increase on hip center lateralization can be attributed to a corresponding increase in the adduction moment. Results from the static analytical model developed supported these findings. The results of this study suggest that superolateral hip center relocation should be avoided and that superior-only relocation may be mechanically acceptable within the confines of the osseous anatomy of the acetabulum.

Revision total hip arthroplasty with a retroperitoneal approach to the iliac vessels

Although vascular complications during revision total hip arthroplasty are rare, the results can be devastating. Reports in the literature describe staged operations, with the first procedure being abdominal to remove cement and/or the acetabular component followed by a second joint reconstruction procedure. A protocol was developed that combines a retroperitoneal approach with revision total hip arthroplasty in one operative procedure in patients at risk for vascular injury. The patient first undergoes a retroperitoneal incision and the iliac artery and vein are dissected free of surrounding tissue. A silicone loop is placed around the iliac artery and vein and brought out through the wound. The wound is temporarily closed using staples. Revision total hip arthroplasty then proceeds in the usual fashion. If hemorrhage is encountered, bleeding can be rapidly controlled by tensioning the abdominal vessel loops and opening the incision for exposure to the vessels. No complications have been encountered in 23 patients when using this approach.

Physical modelling of hip joint forces in stair climbing

A test device has been developed and validated to simulate physiologic loading of the hip during stair climbing. Forces about the hip joint were measured in static simulations of stair climbing using simulated extensor, abductor and adductor muscle groups to support the joint. Femoral flexion angle (to model step length and height) and applied hip flexion moment (to model trunk lean) were varied to examine the effects of different loading conditions on the hip. In stair climbing the maximum total joint force was six times body weight at 34 degrees of femoral flexion and 60 N m of hip flexion moment. Joint forces increased with hip flexion moment and varied little with femoral flexion angle, except for the posteriorly directed force. This component, which twists implants about the femoral shaft, increased with femoral flexion angle but changed little with hip flexion moment.

Aseptic loosening in uncemented total hip arthroplasty in a canine model

An in vivo canine model was developed to investigate the histologic and biochemical parameters associated with aseptic loosening. Thirty-eight canines had cementless total hip arthroplasty. Experimental groups were designed specifically to investigate the relative contributions of implant motion and particulate debris (cobalt chrome alloy, titanium aluminum vanadium, and polyethylene) on the resultant periprosthetic tissues. Tissues from a stable, well-ingrown prosthesis provided a control. Importantly, the histologic and biochemical characteristics of the experimentally induced membranes consistently correlated with previous in vitro reports of tissues retrieved at revision surgery for aseptic loosening. Implant motion and all 3 particulate debris groups resulted in increased numbers of macrophages in the periprosthetic membranes. The histologic findings paralleled the increase in levels of biochemical mediators of bone resorption as measured by collagenase, gelatinase, prostaglandin E2, and interleukin-1 activity. The most striking results were seen in the histology and biochemistry of the particle groups with highly cellular membranes showing increased biochemical activity when compared with controls. The clinical relevance of this work lies in the description of an in vivo model of aseptic loosening that can be used to investigate the effects of numerous variables implicated in aseptic loosening. Ultimately, the model may serve as a basis for developing therapeutic interventions.

Predicting discharge outcome after elective hip and knee arthroplasty

The objective of this prospective study was to determine if differences exist between individuals who require an inpatient rehabilitation program after elective hip and knee arthroplasty from those patients who can be discharged directly home. Multiple variables consisting of baseline demographics, social status, insurance status, medical history, pain level, quantitative strength, range of motion, and functional ability were examined. The primary outcome measure was the discharge destination from the orthopedic service and consisted of either a discharge to home or a discharge to an inpatient rehabilitation unit. Of the 162 patients followed, 65 (40%) were discharged to an inpatient rehabilitation unit, whereas 97 were discharged to home. The patients discharged to inpatient rehabilitation tended to live alone, were significantly older (mean difference = 6.3 yr), and had increased comorbid conditions (p < 0.001 for all variables). Patients discharged to a rehabilitation unit reported significantly greater pain levels than those discharged to home (P < 0.001). The attainment of a supervision level of function demonstrated greater differences between groups than the attainment of independent function for all functional measures. A logistic regression model was developed that predicted 76% of the discharges to rehabilitation by the third physical therapy session postsurgery. In conclusion, predictive markers do exist that differentiate individuals who require further inpatient therapy services after joint replacement surgery.

Comparison of interface membranes obtained from failed cemented and cementless hip and knee prostheses

Biochemical and histologic analyses were performed on interface membranes obtained at revision of aseptically loosened hip implants (n = 36) and knee implants (n = 16). Clinical failure occurred sooner in patients with uncemented total hip implants (Group 1) than in patients with cemented implants (Group 2) (p < 0.02). There was no difference in time to revision between the patients with uncemented implants (Group 3) and patients with cemented total knee implants (Group 4). Histologically, more small (< 5 mu) polyethylene particles were found within macrophages and fibroblasts in membranes from Groups 1 and 2. Polyethylene particles from failed total knees (> 10-100 mu) were larger than those from failed total hips. Large polyethylene fragments and foreign-body giant cells were more common in failed knees than failed total hip membranes. Biochemically, proteinase and cytokine activity in the tissue culture supernatant from all groups was higher than in the control tissue (p < 0.01). The activities of stromelysin, prostaglandin E2, interleukin-1 alpha, interleukin-1 beta, and tumor necrosis factor-alpha were higher in Groups 1 and 2 than in Groups 3 and 4 (p < 0.05). These findings support the hypothesis that interface membranes enveloping femoral (hip) and tibial (knee) components of failed total joint implants may promote bone resorption and aseptic loosening. The reason for slower failure of knee implants as compared with hip prostheses may be the lower level of biochemical activity and macrophage density that correlates closely with larger polyethylene particles.

The characterization of cytokines in the interface tissue obtained from failed cementless total hip arthroplasty with and without femoral osteolysis

The histologic, biochemical, and immunohistologic characteristics of the interface membranes surrounding the femoral component of failed cementless total hip arthroplasty (THA) in patients with (Group I) and without (Group II) radiographic evidence of focal endosteal erosion (osteolysis) were studied. Group I membranes had more macrophages and small particles of polyethylene debris in the membrane, but both groups had similar amounts of metal particles. A greater activity level of interleukin-1 (IL-1), tumor necrosis factor (TNF), and interleukin-6 (IL-6) was seen in the culture supernatant of the membranes from Group I than in that of Group II. Group I membranes also had more cells (macrophages, fibroblasts, and endothelial cells) that stained positively with anti-IL-6 antibody. These results suggest that IL-6, IL-1, and TNF play a role in the focal femoral osteolysis observed in patients with failed cementless hip prostheses.

A biochemical, histologic, and immunohistologic analysis of membranes obtained from failed cemented and cementless total knee arthroplasty

Biochemical, histologic, and immunohistochemical analyses were performed on 34 interface membranes obtained from 33 patients during revision total knee arthroplasty. The membranes had surrounded components of cementless (n = 11) and cemented (n = 23) knee prostheses that were aseptically loose. None of these implant failures was caused by catastrophic polyethylene erosion leading to metal-to-metal contact. The histologic findings were similar in the membranes from cemented and cementless knee components: small polyethylene debris within macrophages and large birefringent polyethylene debris within foreign-body giant cells. Metallic debris was seen in membranes from both groups, but cemented membranes had more polymethylmethacrylate particles and more hyalinization. Intracytoplasmic asteroid bodies were observed in several foreign-body giant cells in both types of membranes. No significant differences were found between the two groups in levels of collagenase, prostaglandin E2 (PGE2), interleukin-1 (IL-1), interleukin-6 (IL-6), or tumor necrosis factor-alpha (TNF-alpha), nor in the population of inflammatory cells stained with IL-1, IL-6, and TNF-alpha antibodies. Membranes that had surrounded components with radiographic evidence of diffuse or localized periprosthetic bone loss released significantly more collagenase, IL-1, IL-6, and TNF than did membranes from components without bone loss. These two groups, however, did not have significantly different PGE2 levels. These findings suggest that polyethylene and metal debris may play a role in macrophage activation and the release of mediators of bone resorption in the membranes surrounding failed cemented and cementless total knee implants.

In vivo and in vitro analysis of membranes from hip prostheses inserted without cement

Biochemical, histological, and immunohistochemical studies of interface membranes surrounding failed hip prostheses that had been inserted without cement were done to examine specific factors involved in the development of aseptic loosening. Membranes from sixty-four femoral components were obtained from sixty-three patients during revision arthroplasty. Fifty-seven membranes were from implants that articulated with polyethylene (thirty-two were made of cobalt-chromium alloy and twenty-five, titanium alloy), and seven were from unipolar endoprostheses made of cobalt-chromium alloy that did not articulate with polyethylene. The membranes from implants with a polyethylene articulation produced significantly higher levels of collagenase and interleukin-1 (p < 0.05). However, there was no significant difference in the levels of prostaglandin E2 between the three groups. Furthermore, membranes from implants with roentgenographic evidence of focal osteolysis (endosteal erosion) released significantly higher levels of interleukin-1 (p < 0.05) than did membranes from implants without focal osteolysis. Although the membranes from the titanium-alloy implants tended to contain more metal debris than those from the cobalt-chromium-alloy implants, the biochemical findings were not significantly different between these two groups. Many macrophages that were filled with polyethylene and metal debris were present in the membranes from both groups with a polyethylene articulation. Few T lymphocytes or B lymphocytes were identified in the three groups.

Primary total hip arthroplasty in the older patient: optimizing the results

Preoperative planning is the first step in adult reconstructive surgery of the hip. When executed properly, it provides a template of the procedure for the whole surgical team. Thorough planning also helps the team anticipate intraoperative problems and avert complications. It reduces surgical trial and error, thus reducing operative time. Planning shortens the learning curve for a new implant system, improves technical skills for performing THA, and ultimately can improve the clinical results.

Total knee arthroplasty in patellectomized patients

Thirty-two total knee arthroplasties (TKAs) in patellectomized patients were evaluated with recent clinical and radiographic examinations. Eighteen patients had a primary TKA with a mean follow-up period of 49 months, and 14 patients had a revision TKA with a mean follow-up period of 36 months. A control group of 13 TKA patients with intact patellae were randomly generated but matched for age, sex, follow-up data, diagnosis, and prosthesis. Posterior cruciate ligament-retaining types of prostheses were used in the primary knees, while the revision knees underwent arthroplasties with the more constrained, posterior cruciate ligament-substituting prostheses. All knees were evaluated based on the Knee Society's clinical and radiographic scoring system. In addition, 18 patients (9 primary, 9 revision) underwent isokinetic dynamometer testing for quantitation of peak quadriceps and hamstring torque. The knee score, indicative of pain relief, averaged 82.5 in the primary group (16 good/excellent, 1 fair, 1 poor), 86.5 in the revision group (12 good/excellent, 1 fair, 1 poor), and 93.9 in the control group (13/13 excellent). The function score averaged 59.7 in the primary group (6 good/excellent, 6 fair, 6 poor), 60.0 in the revision group (5 good/excellent, 2 fair, 7 poor), and 80.9 in the control group (12 good/excellent, 1 fair). The lower function scores predominantly reflected the patients' difficulty in independently climbing or descending stairs. This was also reflected in their higher flexion to extension peak torque ratios. There was one failure in the primary group requiring a revision and one failure in the revision group requiring a knee fusion. There was no radiographic evidence of impending failure in any of the remaining knees. No clinical or radiographic differences were found between the patients with osteoarthritis or rheumatoid arthritis. Although the knee and function scores were lower in the patellectomized patients, the overall results were generally satisfactory without a high incidence of failures. Satisfactory results were obtained in the primary TKAs using the minimally constrained prostheses when the posterior cruciate ligament was intact. Revision TKAs, in which the posterior cruciate ligament was absent, also demonstrated satisfactory results with the more constrained, posterior cruciate ligament-substituting prostheses.