The results of treatment of recurrent dislocation following total hip arthroplasty have been discouraging. At the Mayo Clinic, 116 (1 per cent) of 10,500 patients who had a total hip arthroplasty needed a revision because of instability, and operative treatment was successful in only sixty (61 per cent) of ninety-eight12,14. Numerous alternatives, including trochanteric reattachment or advancement, reorientation of one component or both, lengthening of the femoral neck, elimination of impingement, long-term bracing, capsulorrhaphy, supplementation of the wall of the acetabular component with a polyethylene wedge, placement of an elevated inner acetabular liner, conversion to a bipolar replacement with removal or retention of the acetabular component, and even resection arthroplasty, have been described for the treatment of this problem. However, other than resection arthroplasty, none of these procedures have been reported to be successful in more than 80 per cent of patients8,12-14. Anderson et al., in a recent study, reported that six (29 per cent) of twenty-one patients had repeat dislocation after insertion of an S-ROM constrained acetabular component (Joint Medical Products, Stamford, Connecticut). The purpose of the current paper is to describe our clinical experience with a different type of constrained acetabular component and to make recommendations with regard to the appropriate use of this implant.
*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund or foundation, educational institution, or other non-profit organization with which one or more of the authors is associated. In addition, funds were received in total or partial support of the research or clinical study presented in this article. The funding source was National Institutes of Health Grant AR43314.
†Des Moines Orthopaedic Surgeons, 6001 Westown Parkway, West Des Moines, Iowa 50266.
‡Department of Orthopaedic Surgery, Indiana University School of Medicine, 541 Clinical Drive, Room 600, Indianapolis, Indiana 46202.
§Department of Orthopaedic Surgery, University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, Iowa 52242. E-mail address for Dr. Callaghan: john-callaghan@uiowa.edu. Please address requests for reprints to Dr. Callaghan.
Fifty-six constrained acetabular components (Omnifit constrained acetabular bearing insert; Osteonics, Allendale, New Jersey) were implanted, between April 1988 and February 1993, in fifty-five patients at two centers (Iowa Methodist Medical Center, Des Moines, Iowa, and Indiana University Medical Center, Indianapolis, Indiana) for the treatment of recurrent dislocation following total hip arthroplasty. There were thirty-six women and nineteen men, and the average age at the time of the operation was seventy-one years (range, thirty-one to ninety-two years). One woman had a bilateral procedure. Thirty-one procedures were performed on the right side and twenty-five were performed on the left side. All procedures were performed or supervised by one of the two senior ones of us (W. N. C. or R. C. J.).
The primary diagnosis was osteoarthrosis in thirty-four hips, post-traumatic osteoarthrosis in ten hips, inflammatory arthritis in eight hips, residua of congenital dysplasia in two hips, infection in one hip, and osteonecrosis in one hip. Each of the fifty-six hips had had an average of six documented dislocations (range, two to twenty dislocations).
The constrained acetabular component was used as a last resort, as a mode of salvage, only when it was believed that other modalities would be ineffective. Other patients who had had dislocation following total hip arthroplasty were managed with more conventional methods. Patients who had had only one dislocation were managed with education. Use of a brace was also considered when the risk of recurrence was believed to be high. All patients were evaluated for evidence of malposition of the components, soft-tissue laxity, and lack of compliance with the advice of the physician. The treatment for most of these patients also was non-operative (education and bracing). If non-operative treatment was unsuccessful, operative exploration was considered.
At the time of the operative exploration, the hip was examined for evidence of malposition of the components, soft-tissue laxity, and impingement. These problems were treated when possible, and the range of motion and stability of the hip then were reevaluated. If the hip was stable, the constrained acetabular component was not inserted and the revision was completed with use of a non-constrained implant, with or without an elevated acetabular liner. An exception was made when the surgeon believed the risks of an additional, future operation to be greater than those associated with use of a constrained acetabular implant because of the patient's age or health status. The constrained implant was chosen for such patients regardless of the intraoperative findings pertaining to stability. Another exception was a patient who had a well fixed femoral or acetabular component in association with extremely poor-quality host bone. It was believed that use of the constrained acetabular component was preferable to revision of the well fixed implant in such patients. In five patients, the indications for use of the constrained component included both poor health and a well fixed implant in association with poor bone quality. Thus, the constrained component was believed to be indicated only for the treatment of complex instability and for patients who were likely to have failure of all other methods of treatment.
An average of three previous procedures (range, one to twelve procedures) had been performed on these fifty-six hips. Seven hips (all with a history of infection) had had a Girdlestone arthroplasty; six, a bulk femoral or acetabular allograft; three, a protrusio cage; two, an arthrodesis; two, a periprosthetic fracture; and one, a proximal femoral replacement prosthesis. In addition, two patients were mentally retarded and another eighteen had severe confusion or Alzheimer disease.
The operative approach for the index operation was transtrochanteric in thirty-two hips, posterolateral in twenty-one, anterolateral in two, and a combination of anterior and posterior in one.
The constrained prosthesis consisted of a polyethylene socket with an inner diameter of twenty-two or twenty-eight millimeters and a locking ring identical to that used in a bipolar prosthesis (the inner bearing). The socket was covered by a polished cobalt-chromium shell, which articulated with another polyethylene socket (the outer bearing). The shell was constrained in the outer polyethylene socket during the manufacturing process (Fig. 1). The entire device was then snapped into a standard-profile metal acetabular shell (Osteonics) with an outer diameter of fifty-two millimeters or more or it was cemented into an acetabular shell made by another manufacturer or into the acetabulum.
Ten acetabular components were inserted with cement and forty-six were inserted without cement. Eight of the components that were inserted without cement were placed into a metal backing that had been inserted without cement one to seven months earlier, in exchange for a standard polyethylene liner. Four of the components that were inserted with cement were placed into a well fixed metal backing that had been inserted previously without cement. The well fixed metal backings, which were made by a different manufacturer, were all in acetabula with bone of such poor quality that the surgeon believed that it would be dangerous to remove them.
A well fixed femoral component that had been inserted one to twenty-one years (average, eight years) earlier was left in place in forty hips. Thirty-three of these components had been inserted with cement, and seven had been inserted without cement. In the other sixteen hips, a new femoral component was implanted with use of cement at the time of the index operation.
At the time of the latest follow-up, the surviving patients were asked to return for clinical and radiographic evaluation. Patients who were unable to return had radiographs made by their local physician and subsequently mailed to us for evaluation. All living patients were either evaluated clinically or interviewed by telephone by us, with use of the standard system of terminology for reporting of results that was described by Johnston et al. Members of the families of the patients who had died were interviewed to determine the function of the hip immediately before the time of death.
At the time of the latest follow-up, thirty-eight patients (thirty-nine hips) were alive, sixteen patients had died, and one patient had been lost to follow-up. Hence, clinical follow-up information was obtained for fifty-five hips (98 per cent). The living patients were followed clinically for an average of sixty-four months (range, thirty-seven to ninety-seven months), and the deceased patients had been followed for an average of twenty-seven months (range, one to eighty-one months). Thirty-four (87 per cent) of the thirty-nine hips in the living patients were evaluated clinically by one of us, and the others were evaluated on the basis of the telephone interview and the radiographs that had been sent by their local physician.
Radiographic Evaluation
Of the thirty-eight living patients (thirty-nine hips), thirty-three were followed radiographically for at least two years (average, fifty-three months; range, twenty-four to ninety-seven months) and only they were included in the radiographic evaluation portion of this study. Anteroposterior radiographs of the pelvis, including the entire femoral component, were made for all thirty-three patients. Additional radiographs, including lateral and oblique radiographs, were available for thirty-one patients (82 per cent). Of the sixteen patients (sixteen hips) who had died, five had been followed radiographically for at least two years (average, forty-two months; range, twenty-four to eighty-one months).
Loosening of femoral components that had been inserted without cement was classified according to the criteria of Engh et al. Fixation by bone ingrowth was defined as no subsidence or migration and little or no formation of a radiopaque line around the stem; stable fibrous ingrowth, as no progressive migration and the formation of non-divergent radiopaque lines around the porous surface of the component; and instability, as either progressive subsidence within the canal or the formation of divergent radiopaque lines that were more widely separated from the stem at its ends.
The criteria for loosening of acetabular components that had been inserted without cement were similar. Fixation by bone ingrowth was defined as no migration and no formation of a circumferential radiolucent line; stable fibrous ingrowth, as formation of a circumferential radiolucent line but no progressive migration; and instability, as definite migration or definite progression in the width of a previously noted circumferential radiolucent line.
Loosening of femoral components that had been inserted with cement was classified according to the criteria of Harris et al.5,6. Subsidence was measured with use of the method of Loudon and Charnley. Definite loosening was defined as subsidence of the femoral component, fracture of the cement or stem, or the presence of a radiolucent line at the interface between the stem of the prosthesis and the cement that had not been seen on the early postoperative radiograph; probable loosening, as the presence of a continuous radiolucent line along the entire bone-cement interface; and possible loosening, as the presence of a radiolucent line that encompassed more than 50 but less than 100 per cent of the circumference of the stem at the bone-cement interface on at least one radiograph.
Migration of acetabular components was evaluated with use of the criteria of Massin et al. Definite loosening was defined as migration of the component or the presence of any new fracture in the cement mantle (in the four hips in which the constrained liner was cemented into a previously placed metal backing, evaluation of the cement mantle was difficult because of the radiopaque metal); probable loosening, as the presence of a circumferential radiolucent line around the entire component at the bone-cement interface; and possible loosening, as the presence of a radiolucent line around 50 to 99 per cent of the component at the bone-cement interface.
Radiolucent lines between the cement and bone or between the prosthesis and bone, as seen on the anteroposterior radiograph, were recorded on the basis of the seven femoral zones described by Gruen et al. and the three acetabular zones described by DeLee and Charnley. In hips in which a previously placed femoral or acetabular component had been retained at the time of the index operation, radiolucent lines that were present at the time of the index operation were recorded as old, whereas those that were progressive and those that were first noted at the time of the latest follow-up evaluation were recorded as new.
New or progressive osteolysis was recorded in a similar fashion and was considered to be present when there was any localized loss of the endosteal cortex of the femur or any osseous loss in the periacetabular region that appeared cystic (at least ten millimeters in length).
Redislocation or Instability following the Index Arthroplasty
One of the sixteen hips in the patients who died had redislocated repeatedly during the follow-up period. In this patient, the constrained component had been cemented into a well fixed metal-backed acetabular component that had previously been inserted without cement (Figs. 2-A and 2-B).
In addition, one (3 per cent) of the thirty-nine hips in the patients who were alive at the latest follow-up evaluation had redislocated repeatedly. In this patient, the acetabular component had been inserted without cement (Figs. 3-A, 3-B, 3-C and 3-D).
Other Reoperations
An additional seven hips (13 per cent) had or needed a reoperation during the follow-up period. Two of these hips were in patients who died, and five were in patients who were alive.
Of the sixteen hips in the patients who died, two had needed a reoperation subsequent to the index arthroplasty for a reason other than dislocation or instability. Both reoperations consisted of débridement (with retention of the components) because of infection. Before the index arthroplasty, one of these patients had a history of infection, which had necessitated several operations (including a resection arthroplasty) on two separate occasions.
Of the thirty-nine hips in the thirty-eight living patients, five (13 per cent) needed a reoperation subsequent to the index arthroplasty for a reason other than dislocation or instability. Two of these hips had a resection arthroplasty because of infection, one hip had revision of the femoral component because of failure of the allograft and loosening of the component, one had internal fixation because of a periprosthetic fracture, and one had revision of the acetabular component because of aseptic loosening.
The patient who was lost to follow-up was last seen eight months postoperatively and had had no subsequent reoperations or dislocations at that time.
Clinical Results
Of the thirty-nine hips in the thirty-eight living patients, twenty-eight (72 per cent) caused no pain at the time of the latest follow-up; seven (18 per cent), mild pain; three (8 per cent), moderate pain; and one (3 per cent), severe pain. Nineteen patients had no limp, eleven (twelve hips) had a mild limp, six had a moderate limp, and two were unable to walk. Twelve patients needed no support to walk, fourteen used a cane, ten (eleven hips) used crutches or a walker, and two were confined to a wheelchair. All patients who used crutches, a walker, or a wheelchair had factors unrelated to the hip that contributed to the disability.
Complications
Aside from the previously noted complications that led to a reoperation, the most common complication was trochanteric non-union, which occurred in ten (31 per cent) of the thirty-two hips that had had a trochanteric osteotomy. However, in twenty of these hips, the greater trochanter or its osseous bed was either completely or almost completely absent along the proximal-lateral aspect of the femur. Only four of the ten hips were associated with a limp, and none dislocated after placement of the constrained component.
Although routine screening for deep venous thrombosis was not performed, one patient had a symptomatic deep venous thrombosis that necessitated anticoagulation. One patient sustained an intraoperative fracture of the femur and perforation, and another had an intraoperative fracture of the acetabulum. One patient sustained an incomplete palsy of the sciatic nerve. Severe heterotopic ossification developed in one patient; it was not treated operatively. One patient died secondary to cardiac complications and another died as a result of lower gastrointestinal complications in the early postoperative period.
Radiographic Results
All postoperative radiographs were evaluated for radiolucent lines, evidence of loosening, and osteolysis. However, only the thirty-eight hips that had been followed radiographically for at least two years will be reported on in detail. (There was no evidence of definite loosening or progressive osteolysis in the hips for which the most recent radiographs had been made less than two years postoperatively.)
The radiographic results were classified according to eight distinct categories: four for the acetabular components and four for the femoral components. The categories for the acetabular components included a new acetabular shell and a new constrained liner, both inserted without cement; an old acetabular shell and a new constrained liner, both inserted without cement; a new constrained liner inserted directly into the acetabulum with cement; and an old acetabular shell inserted without cement and a new constrained liner inserted with cement. The categories for the femoral components included a new femoral component inserted with cement, an old femoral component inserted with cement, a new femoral component inserted without cement, and an old femoral component inserted without cement. A component was considered new if it had been implanted at the time of the index operation (that is, at the same time as the constrained acetabular liner), and it was considered old if it had been placed at the time of a previous operative procedure.
Acetabular Components
A new acetabular shell and a new constrained liner, both inserted without cement: Twenty-seven hips were treated with a new acetabular shell combined with a new constrained acetabular liner, both inserted without cement, at the time of the index operation. At the time of the latest radiographic follow-up, twenty-two (81 per cent) of these components were fixed by bone ingrowth, five (19 per cent) had stable fibrous fixation, and none were loose. No component had been revised. Two hips (7 per cent) had evidence of acetabular osteolysis.
An old acetabular shell and a new constrained liner, both inserted without cement: In seven hips, a new constrained acetabular liner was placed without cement into an old metal shell that had been inserted without cement one to seven months earlier. The old standard-type polyethylene liner was simply replaced with the new constrained liner. At the time of the latest radiographic follow-up, two components were fixed by bone ingrowth, three had stable fibrous fixation, and two were loose. One of the latter two components was revised because of aseptic loosening. None of these hips had evidence of acetabular osteolysis.
A new constrained liner inserted into the acetabulum with cement: A constrained liner was cemented directly into two acetabula. At the time of the latest radiographic follow-up, one component was believed to be stable and the other was classified as possibly loose. Neither hip had evidence of acetabular osteolysis or needed a revision of the component because of aseptic loosening.
An old acetabular shell inserted without cement and a new constrained liner inserted with cement: Two hips were treated with a constrained liner that was cemented into a well fixed metal shell that had been previously inserted without cement. As the metal backing had been made by a different manufacturer, simple exchange of the liner was not possible and, because of the rigid fixation of the acetabular component and the poor quality of the acetabular bone, the surgeon believed that it was dangerous to attempt to remove the metal backing. Thus, the constrained liner was cemented into the metal shell. At the time of the latest radiographic follow-up, one component was possibly loose and the other was definitely loose (at the liner-cement interface). Neither component had been revised because of aseptic loosening, and neither hip had acetabular osteolysis.
Femoral Components
A new femoral component inserted with cement: Nine new femoral components were inserted with cement at the time of the index operation. At the time of the latest radiographic follow-up, all nine appeared stable. None had been revised because of aseptic loosening, and no hip had evidence of femoral osteolysis.
An old femoral component inserted with cement: Twenty-four hips had a well fixed femoral component that had been inserted with cement one to twenty-one years earlier and that was left in place at the time of the index operation. At the time of the latest radiographic follow-up, twenty components (83 per cent) were stable, two (8 per cent) were possibly loose, and two (8 per cent) were definitely loose. There were no revisions because of aseptic loosening. Two hips had evidence of femoral osteolysis.
A new femoral component inserted without cement: No femoral component was inserted without cement at the time of the index operation.
An old femoral component inserted without cement: Five hips had a well fixed femoral component that had previously been inserted without cement and that was left in place at the time of the index operation. At the time of the latest radiographic follow-up, all five components had evidence of fixation by bone ingrowth. No component had been revised because of aseptic loosening. One hip had evidence of femoral osteolysis.
In the present series, the new constrained acetabular component successfully prevented instability in thirty-eight (97 per cent) of the thirty-nine hips in the living patients after an average duration of follow-up of five years. This short-term rate of clinical success appears to be far superior to the previously reported rates of success of any other type of treatment for recurrent instability after total hip arthroplasty1.
Neither of the recurrent dislocations in the current series was related solely to the component itself; both were due at least in part to the operative technique. In one hip the acetabular component had been cemented into a previously placed acetabular shell that had been made by a different manufacturer, and in the other the component had been placed onto a bulk allograft. Four of the other seven reoperations were related to a preexisting condition, including a history of infection (three) and loosening of the femoral component secondary to resorption of the allograft (one). It should also be noted that the constrained acetabular component was used by the senior ones of us only to treat the most severe instability. Routine instability was treated with other techniques, including use of a conventional extended-lip liner.
Use of a constrained component has two major theoretical disadvantages: increased polyethylene wear and increased interfacial stresses. Impingement with large increases in interfacial stresses occurs occasionally, as seen in our patient who had failure at the bone-prosthesis interface and our patient who had failure at the cement-prosthesis interface. If increased polyethylene wear and interfacial stresses do occur, then both the femoral and the acetabular component will loosen. Although we remain concerned about this possibility, we did not find an increased prevalence of loosening of the component in this study. Analysis of the thirty-eight hips that had been followed radiographically for at least two years revealed that only two (6 per cent) of the thirty-four acetabular components that had been inserted without cement and only two (6 per cent) of the thirty-three femoral components that had been inserted with cement appeared definitely loose. (The number of acetabular components that had been inserted with cement and the number of femoral components that had been inserted without cement were too small for meaningful analysis.) These rates are comparable with those that have been reported, after a similar duration of follow-up, after complex revision total hip arthroplasty9. An additional important concern is that a dislocation of the constrained component cannot be treated non-operatively.
It is important to realize that the primary goal was a stable hip with no additional dislocations. Thus, the patients and the surgeons were willing to accept the increased risk of polyethylene wear, osteolysis, and loosening. Although a detailed radiographic analysis was performed, it was limited by the number of hips that had been followed radiographically for at least two years (thirty-eight [68 per cent] of fifty-six) and by the variety of reconstructive procedures that were included in this series. In addition, the clinical results in terms of pain, limp, and walking ability were difficult to interpret. Twenty-seven (49 per cent) of our fifty-five patients were more than seventy-five years old, twenty (36 per cent) had severe mental impairment, and many others had severe physical disability related to factors other than the hip. All of these limitations reflect the complexity of this population of patients. Given the limited indications for use of this implant, it is probable that a homogeneous population of patients will never become available for study. The radiographic results after such a short-term follow-up interval are of limited utility.
In summary, we are encouraged by the early results of use of the constrained acetabular component for patients who had complex instability. However, the surgeons who performed the operations were experienced in the techniques of revision total hip arthroplasty. This component should not be considered as a first-line treatment option for patients who have routine instability after total hip arthroplasty. Because of the potential long-term problems that have been described, we recommend its use only as a salvage measure for the treatment of severe instability.
NOTE: The authors thank Nancy Keesling, R.N., Patricia Kratz, R.N., Renee Shirer, M.A.H.P., and Linda Bigley, R.N., for assistance in the collection of the data and in contacting the patients, as well as David Heck, M.D., Edward Hellman, M.D., and Patrick Sullivan, M.D., for assistance in the clinical and radiographic analyses.
Anderson, M. J.; Murray, W. R.; and Skinner, H. B.: Constrained acetabular components. J. Arthroplasty,9: 17-23, 1994.917
1994
[PubMed]
DeLee, J. G., and Charnley, J.: Radiological demarcation of cemented sockets in total hip replacement. Clin. Orthop.,121: 20-32, 1976.12120
1976
[PubMed]
Engh, C. A.; Bobyn, J. D.; and Glassman, A. H.: Porous-coated hip replacement. The factors governing bone ingrowth, stress shielding, and clinical results. J. Bone and Joint Surg.,69-B(1): 45-55, 1987.69-B(1)45
1987
Gruen, T. A.; McNeice, G. M.; and Amstutz, H. C.: "Modes of failure" of cemented stem-type femoral components. A radiographic analysis of loosening. Clin. Orthop.,141: 17-27, 1979.14117
1979
[PubMed]
Harris, W. H., and McGann, W. A.: Loosening of the femoral component after use of the medullary-plug cementing technique. Follow-up note with a minimum five-year follow-up. J. Bone and Joint Surg.,68-A: 1064-1066, Sept. 1986.68-A1064
1986
Harris, W. H.; McCarthy, J. C., Jr.; and O'Neill, D. A.: Femoral component loosening using contemporary techniques of femoral cement fixation. J. Bone and Joint Surg.,64-A: 1063-1067, Sept. 1982.64-A1063
1982
Johnston, R. C.; Fitzgerald, R. H., Jr.; Harris, W. H.; Poss, R.; Müller, M. E.; and Sledge, C. B.: Clinical and radiographic evaluation of total hip replacement. A standard system of terminology for reporting results. J. Bone and Joint Surg.,72-A: 161-168, Feb. 1990.72-A161
1990
Kaplan, S. J.; Thomas, W. H.; and Poss, R.: Trochanteric advancement for recurrent dislocation after total hip arthroplasty. J. Arthroplasty,2: 119-124, 1987.2119
1987
[PubMed]
Katz, R. P.; Callaghan, J. J.; Sullivan, P. M.; and Johnston, R. C.: Long-term results of revision total hip arthroplasty with improved cementing technique. J. Bone and Joint Surg.,79-B(2): 322-326, 1997.79-B(2)322
1997
Loudon, J. R., and Charnley, J.: Subsidence of the femoral prosthesis in total hip replacement in relation to the design of the stem. J. Bone and Joint Surg.,62-B(4): 450-453, 1980.62-B(4)450
1980
Massin, P.; Schmidt, L.; and Engh, C. A.: Evaluation of cementless acetabular component migration. An experimental study. J. Arthroplasty,4: 245-251, 1989.4245
1989
[PubMed]
Morrey, B. F.: Joint Replacement Arthroplasty, p. 861. New York, Churchill Livingstone, 1991.
Olerud, S., and Karlström, G.: Recurrent dislocation after total hip replacement. Treatment by fixing an additional sector to the acetabular component. J. Bone and Joint Surg.,67-B(3): 402-405, 1985.67-B(3)402
1985
Woo, R. Y. G., and Morrey, B. F.: Dislocations after total hip arthroplasty. J. Bone and Joint Surg.,64-A: 1295-1306, Dec. 1982.64-A1295
1982