 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1998 The Journal of Bone and Joint Surgery, Inc. The Harris Design-2 Total Hip Replacement Fixed with So-Called Second-Generation Cementing Techniques. A Ten to Fifteen-Year Follow-up* , LONDON, ONTARIO, CANADAInvestigation performed at London Health Sciences Centre, The University of Western Ontario, London
We analyzed the clinical results of 195 Harris Design-2 total hip replacements performed with so-called second-generation cementing techniques in 166 consecutive patients who had osteoarthrosis. The mean age of the patients at the time of the replacement was sixty-seven years and nine months (range, thirty-one to eighty-nine years). Forty-eight patients (fifty-four hips) died before the time of the latest follow-up, but the implants were apparently functioning well at the time of death. Three patients (four hips) were lost to follow-up. Five patients (five hips; 3 percent) had a revision because of aseptic loosening of the acetabular or femoral component, or both, that was related to wear-induced osteolysis. The mean Harris hip score for the 131 hips that were available at the latest follow-up examination at a mean of twelve years (range, ten to fifteen years) after the operation was 89 ± 10 points. On the basis of the Harris hip score, seventy-six hips had an excellent result, thirty-four had a good result, fifteen had a fair result, and six had a poor result at the latest follow-up examination. Radiographically, twelve (9 percent) of the 131 acetabular components and three (2 percent) of the 131 femoral components were probably or definitely loose. At a mean of twelve years, 186 (97 percent) of 191 Harris Design-2 implants were in situ or had been in situ at the time of the patient's death.
Total hip replacement with the use of cement revolutionized the treatment of osteoarthrotic conditions of the hip3,5,9,10,13,15,17,19,21,22,26,27. Since Charnley's breakthrough in the 1960s, more than six million low-friction total hip replacements have been performed. From the beginning, Charnley was concerned about the durability of the implants in terms of wear and aseptic loosening5. Reported rates of revision have varied with the type of implant: they have been as low as 3 to 9 percent for Charnley implants and as high as 25 percent for other implants after ten or even twenty years of follow-up3,12,18,22,26,27 (Table I). Concern was expressed regarding the high prevalence of radiographic loosening after use of so-called first-generation implants and finger-packing cementing techniques; these rates have ranged between 9 and 30 percent for femoral loosening and between 8 and 41 percent for acetabular loosening3,7,12,18,22,26,27 (Table I). To address this problem, techniques for fixation without cement, hybrid replacement (insertion of the acetabular component without cement and insertion of the femoral component with cement), and improved techniques for cementing of total hip replacements have emerged1-3,6,16,17,20,21,24.
We report here the results of insertion of a cobalt-chromium femoral component and an all-polyethylene eccentric acetabular socket2,3,9,10,16,17 (the Harris Design-2 total hip replacement) with so-called second-generation cementing techniques (use of an intramedullary plug, lavage, drying, a cement gun, and multiple acetabular fixation holes) in patients who had osteoarthrosis. The mean duration of follow-up was twelve years (range, ten to fifteen years).
One hundred and ninety-five consecutive Harris Design-2 total hip replacements (Howmedica, East Rutherford, New Jersey) were performed, between July 1978 and June 1985, by the senior ones of us (R. B. B. and C. H. R.) in 166 consecutive patients who had osteoarthrosis (Fig. 1). The mean age of the eighty-two men and eighty-four women at the time of the operation was sixty-seven years and nine months (range, thirty-one to eighty-nine years). One hundred and six right hips and eighty-nine left hips were involved. The acetabular component was all-polyethylene with an eccentric articulation that maximized the thickness of the polyethylene in the superior and posterior quadrants. The femoral component was available in three sizes (small, medium, and large) but did not have proportional dimensions since the length of the stem remained constant for all stem sizes. The femoral component had a collar, a matte finish, and a rounded rectangular cross section.
All of the total hip replacements were inserted through a posterolateral operative approach with which the short external rotators and the femoral attachment of the gluteus maximus were incised during exposure of the joint and then were reattached at the end of the procedure. A posterior capsulectomy of the hip joint was followed by an anterior capsulectomy to aid anterior displacement of the femur17. The Harris Design-2 implants were fixed with so-called second-generation cementing techniques. The acetabulum was medialized to the floor of the foveal notch and reamed until bleeding subchondral bone was encountered. Every effort was made to preserve acetabular subchondral bone. A cebetome burr (Zimmer, Warsaw, Indiana) was used to make ten to fifteen five-millimeter-deep cement-fixation holes in the periphery of the acetabular socket. The osseous bed was meticulously cleaned with a saline-solution lavage system and then was dried. Thrombin-soaked Gelfoam pads were packed into the acetabulum while the cement was being mixed. Simplex-P bone cement (Howmedica) in the doughy stage was placed in the acetabulum and was pressurized with the forty-two-millimeter-diameter circular handle of the acetabular holder. An attempt was made to insert the acetabular component at 40 degrees from the horizontal with approximately 10 degrees of anteversion. The femoral component was selected after the proximal end of the femoral canal had been opened with a box osteotome and an awl and then rasped to a size that corresponded with the size that had been determined with preoperative templating or until resistance was encountered. The femoral neck was prepared with a calcar reamer to improve contact between the collar and the calcar. The femoral canal was cleaned with a Miller cement brush (Zimmer) and then was lavaged with saline solution. A cement intramedullary plug was inserted two centimeters distal to the tip of the femoral component. A thrombin-soaked sponge was packed into the femoral canal while the cement was being mixed. The sponge was removed, and the femoral canal was meticulously dried with gauze sponges. The canal was then filled with cement in a retrograde fashion from the intramedullary plug proximally with use of a cement gun. The proximal cement was pressurized by the surgeon with a clean-gloved thumb by occluding the femoral neck and pressing distally two or three times. The femoral component was inserted carefully; contamination of the cement with blood as well as any rotatory movements were avoided. Every attempt was made to keep the femoral component in a neutral or valgus orientation. The femoral component was anteverted in agreement with the orientation of the retained femoral neck. The patients were prospectively evaluated preoperatively; at six weeks, three months, six months, and one year; and yearly thereafter with use of a modified Harris hip score that incorporated radiographic parameters8. Any abnormal intraoperative findings were recorded on an intraoperative data form. All of the prospective data were gathered by three independent clinical research assistants. The quality of the cementing technique was evaluated radiographically, with use of the zones of Gruen et al.7 on the femoral side and those of DeLee and Charnley5 on the acetabular side, by medical personnel who were not associated with the original operation (M. S., S. M., and M. W.). Radiographic assessment included categorization of the original cementing technique for both the acetabular and the femoral component1. Type A indicated a so-called whiteout between the component and the surrounding bone, type B indicated very good cementing overall but some osseous trabeculae that were incompletely filled with cement, type C indicated an incomplete cement mantle either from abutment of the implant or from voids, and type D indicated an incomplete cement mantle10. The clinical results were examined according to these criteria and were also categorized according to the age of the patient at the time of the replacement.
Forty-eight patients (fifty-four hips) died before the most recent follow-up examination, which was performed ten to fifteen years (mean, twelve years) post-operatively. All of the implants had been in place at the time of death, and the function had been clinically and radiographically satisfactory. Five (3 percent) of the 195 hips had been revised. One patient (one hip) who was seventy-two years old and had residual hemiplegia after a stroke was excluded from the latest follow-up evaluation because mobility was limited to bed-to-chair transfers; however, the result was radiographically satisfactory. Three patients (four hips) were lost to follow-up. Early complications (those occurring less than four weeks postoperatively) included dislocation (three patients), pulmonary embolus (one), and death (one). Late complications included dislocation (two patients), aseptic loosening (five), Brooker et al.4 grade-III or IV heterotopic bone formation (fourteen), and periprosthetic fracture (one). One hundred and thirty-one hips were available for the latest follow-up examination. The results according to the Harris hip score8 were categorized as excellent (90 to 100 points), good (80 to 89 points), fair (70 to 79 points), and poor (less than 70 points). The result was excellent for seventy-six hips, good for thirty-four, fair for fifteen, and poor for six. The clinical results were also categorized according to the age of the patient (by decade) at the time of the total hip replacement (Table II). The mean score was 100 ± 0 points for the patients between thirty and forty years old (one hip), 93 ± 8 points for those between forty-one and fifty (six hips), 93 ± 8 points for those between fifty-one and sixty (twenty-two hips), 90 ± 8 points for those between sixty-one and seventy (seventy-one hips), 83 ± 9 points for those between seventy-one and eighty (twenty-nine hips), and 62 ± 16 point for those between eighty-one and ninety (two hips). The mean score for all 131 hips was 89 ± 10 points.
Five patients (five hips) had a revision (Table III). One of them, a seventy-eight-year-old woman who had unilateral osteoarthrosis, was seen fourteen years post-operatively for a painful left hip. Radiographs revealed evidence of osteolysis on both the acetabular and the femoral side of the joint with definite loosening of both implants. The replacement was successfully revised with use of an extensively coated femoral component and a metal-backed acetabular component inserted without cement. Another patient, a sixty-eight-year-old woman, had a revision eleven years postoperatively because of aseptic loosening of the femoral component and wear of the well fixed acetabular component (Figs. 2-A and 2-B). Radiographs revealed radiolucent scalloping of the femur that was typical of osteolysis with definite loosening (Fig. 2-A). This patient did well following a revision with use of a femoral component inserted with impaction-grafting techniques and an acetabular socket inserted without cement (Fig. 2-B). The third patient who had a revision, an eighty-one-year-old man, had the procedure thirteen years postoperatively because of aseptic loosening of the femoral component; both components were replaced with implants inserted without cement. The fourth patient, a sixty-two-year-old man who had had a bilateral hip replacement, had revision of the left hip six years postoperatively, after radiographs revealed evidence of loosening of the femoral component. Both components were revised because of aseptic loosening of the femoral component and wear of the acetabular socket; the revision components were inserted without cement and with bone-grafting in the trochanteric region. Finally, a revision was performed in an eighty-three-year-old man who had had a bilateral hip replacement and increasing pain in the left hip nine years postoperatively. Radiographically, there was a complete radiolucent line around the acetabular and femoral components. Both components were successfully revised with a hybrid replacement.
The radiographs of the 131 hips that were seen at the time of the latest follow-up were assessed by three orthopaedic surgeons who had not been involved with the original operations. There was a radiolucent line in zone I around thirteen femoral components (10 percent), in zone II around ten (8 percent), in zone III around twelve (9 percent), in zone IV around four (3 percent), in zone V around nine (7 percent), in zone VI around seventeen (13 percent), and in zone VII around seventeen. Radiolucent lines were observed in zone I around fifty acetabular components (38 percent), in zone II around thirty (23 percent), and in zone III around fifty-four (41 percent). The cementing technique for the 111 hips for which the immediate postoperative radiographs were available was examined with respect to the most recent Harris hip score8 (Table IV). The mean score was 90 ± 12 points for the thirty-three femoral components for which the cementing technique had been type A, 89 ± 10 points for the sixty-two for which it had been type B, and 86 ± 9 points for the fifteen for which it had been type C. The femoral component for which the cementing technique had been type D had been revised before the most recent follow-up examination. The mean score was 88 ± 10 points for the thirty acetabular components for which the cementing technique had been type A, 89 ± 11 points for the sixty-nine for which it had been type B, and 92 ± 5 points for the twelve for which it had been type C.
The components were assessed radiographically at the latest examination and categorized as possibly, probably, or definitely loose according to the criteria of Harris et al.10,11,14,23. A radiolucent line at least two millimeters wide surrounding 50 to 99 percent of the implant was considered evidence of possible loosening, and a radiolucent line at least two millimeters wide surrounding the entire implant indicated probable loosening. A component was considered definitely loose if it had migrated. Five femoral components were possibly loose, and two were probably loose. Five femoral components were definitely loose and were revised. The degree of loosening of the femoral components was examined with respect to the initial cementing technique (Table V). One possibly loose and one probably loose component had been cemented with type-A technique; three possibly loose, one probably loose, and one definitely loose component, with type-B; one possibly loose and three definitely loose components, with type-C; and one definitely loose component, with type-D. Ten acetabular components were probably loose, and two were definitely loose.
In the present study, in which the duration of follow-up was ten to fifteen years (mean, twelve years), the clinical and radiographic results of Harris Design-2 total hip replacements performed with so-called second-generation cementing techniques were better than those reported for implants inserted with first-generation techniques3,7,12,18,22,26,27. So-called second-generation cementing techniques, which include the use of an intramedullary cement plug, meticulous cleaning of the femoral canal, and the use of one of three sizes of femoral stem to fit a given femur, seem to be worthwhile3. Whether the use of a proportionally longer femoral stem, a proximally textured or precoated femoral component, a centralizer, or porosity-reduced cement would have improved the results of the present study is not known6. Although the rate of revision was low (3 percent; five of 195 hips), we are concerned by the rates of probable and definite loosening (9 percent [twelve of 131] for the acetabular components and 2 percent [three of 131] for the femoral components) that were noted in the present study as well as in other studies12,18,22,26,27. No metal-backed acetabular components inserted with cement were used in the present study19. The results of the present study are similar to those of Mulroy et al.17 in terms of the rates of revision and radiographic loosening of the femoral component, but they are remarkably different with regard to the rates for the acetabular component. The rate of acetabular revision in the present study was 3 percent (five of 195) compared with 10 percent (eight of eighty-one) in the study of Mulroy et al., and the rate of radiographic loosening of the acetabular component in the present study was 9 percent (twelve of 131) compared with 42 percent (twenty-eight of sixty-seven) in the study by Mulroy et al. Our results are more similar to those reported by others and are perhaps related to factors such as the inclusion of only patients who had osteoarthrosis as well as the use of only one type of implant13,15,25,28,29. In addition, depending on the size of the acetabulum, the acetabular operative technique in our study involved ten to fifteen five-millimeter peripheral cement-fixation holes, whereas Mulroy et al. used three 12.5-millimeter fixation holes in the pubis, ilium, and ischium. The implication is that fixation involving multiple peripheral acetabular holes is more durable than that involving three 12.5-millimeter holes. It is interesting compare the results of the present study and those of other ten or even twenty-year follow-up studies of total hip replacement with cement with the findings reported by Charnley3,12,18,22,26,27 and with those of other studies3 in which the results of replacement with the Charnley implant were studied after the same follow-up intervals. We recognize the limitations of using historic controls instead of a randomized clinical trial. Both the Harris Design-2 and the flat-backed Charnley femoral component have a rounded rectangular cross section and come in a single length. The rate of revision in the present study was slightly lower (3 percent; five of 195 hips) than those (3 to 9 percent) in studies involving so-called first-generation cementing techniques and the Charnley polished flat-backed prosthesis12,18,22,26,27 (Table I). The introduction of so-called second-generation cementing techniques also greatly enhanced radiographic fixation.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |