Abstract
Data on 240 primary Charnley total hip arthroplasties, performed in 211 patients from October 1968 to July 1974, were recorded prospectively. Fifty-two prostheses were implanted in forty-six patients who were thirty-four to fifty-five years old, and 188 prostheses were implanted in 165 patients who were more than fifty-five years old. The results for the younger patients were compared with those for the older patients. Twenty hips were revised over-all, and all five of the infections and four of the five fractures of the femoral stem that occurred in this group were in the older patients. However, the rate of aseptic loosening was similar for the two age-groups (three of the hips in the younger patients and five in the older patients). The probability of revision after twenty years for the two groups, as determined with Kaplan-Meier analysis, was 11.7 per cent for the younger patients and 10.7 per cent for the older patients. There was no significant difference between the two groups with regard to either the number of revisions or the percentage of patients who had radiographic signs of loosening. The median Charnley hip scores during the course of the study were also similar for the two groups.We recommend that Charnley low-friction arthroplasty be used even in younger patients, as the long-term results were excellent and were comparable with those in the elderly age-group.
Total hip arthroplasty with cement has been proved to be a good choice for the treatment of osteoarthrosis of the hip in older patients, but its use in young patients, whose functional demands are higher and whose longer life expectancy makes revision more likely, is still debatable.
The use of total joint implants without cement in young patients has increased, bringing with it new problems, such as severe resorption of bone, which makes subsequent revision operations difficult to perform and causes persistent pain in the thigh.
During the first decade after the introduction of hip arthroplasty, most surgeons were reluctant to perform a joint replacement in patients who were less than sixty years old because the long-term results were unknown. Consequently, there have been few reported studies with long-term follow-up (more than fifteen years) of hip replacements in young patients, and the number of patients in these studies has usually been small6,10,14,18. In 1979, Charnley stated that "between 45 and 65 years of age the surgeon must exercise sensitive and informed judgement in selecting and rejecting patients.... Below 45 years of age we are still only at the beginning of our experience."5 This was the view in our clinic when the Charnley low-friction arthroplasty was introduced here in 1968. However, because of pressure from several patients with severe osteoarthrosis of the hip who were less than forty-five years old and who wished to have an active and pain-free life before they grew old, a number of hip replacements were performed in young patients.
Data from the first 240 primary total hip replacements done by our department were prospectively recorded, and radiographs were kept on file. In Denmark, all patients can be traced at any time through the People's Register. Therefore, a high rate of follow-up is possible, allowing the calculation of highly reliable survival data.
The purpose of the present study was to compare the results of Charnley total hip replacements performed with cement in a series of young patients with the results in a series of older patients who were operated on during the same time-period with the same standardized procedures.
*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.
†Harlow Wood Orthopaedic Hospital, Near Mansfield, Notts NG18 4TH, United Kingdom.
‡Department of Orthopaedics, Odense University Hospital, DK-5100 Odense, Denmark.
The senior one of us (K. H. S.) performed 240 primary Charnley low-friction arthroplasties in 211 patients from October 1968 to July 1974. Fifty-two total hip prostheses were implanted in forty-six patients who were thirty-four to fifty-five years old, and 188 were implanted in 165 patients who were fifty-six years old or more.
During this time, 395 intertrochanteric osteotomies were performed in our department, as this was still the primary operative procedure for the treatment of osteoarthrosis. Total joint replacement was considered as an option only for patients in whom the osteoarthrosis was too far advanced to be treated effectively with an osteotomy. Thus, the hips in this series were all severely osteoarthrotic. Also, significantly more younger patients (twenty hips; 38 per cent) than older patients (thirty-one hips; 16 per cent) had had a previous operation, most frequently an osteotomy, on the hip with the replacement (p = 0.001, chi-square test with Yates correction) (Table I). The two groups differed with regard to the indications for the operation. One hundred and fifty-nine (85 per cent) of the hips in the older patients were treated for osteoarthrosis, compared with thirty-one (60 per cent) of the hips in the younger patients, whereas twelve (6 per cent) of the hips in the older patients were treated for sequelae of congenital hip disease, compared with fifteen (29 per cent) of the hips in the younger patients. Only three hips (all in older patients) were treated for rheumatoid arthritis.
The arthroplasties were performed in a conventional operating room, and the surgeons wore ordinary gowns. No antibiotic or antithrombotic drugs were used. The operative technique was performed according to the recommendations of Charnley, with a lateral approach and a trochanteric osteotomy. No medullary plug was used in the femoral canal, and a metal grid was used to cover the centering hole for the reamer in the acetabulum beginning with the seventy-fifth hip treated. CMW cement (CMW Laboratories, Dentsply, England) was introduced with use of thumb pressure only, about five minutes after the initiation of mixing—when it had achieved a rather firm consistency. Care was taken to maintain a very firm push and a steady position until complete polymerization of the cement had occurred. As was recommended at the time that these operations were performed, a neutral or valgus position of the stem was intended, with the tip of the stem touching the medial wall of the medullary canal of the femur. Because of this contact, a uniform mantle of cement around the stem was not achieved. One hundred and twenty-four of the stems were positioned in neutral; fifty-six, in varus (less than 10 degrees for fifty stems); and forty-three, in valgus (less than 10 degrees for all). The immediate postoperative radiographs of seventeen hips were lost and therefore the position of the stem could not be assessed. In most hips, the cement extended one to three centimeters beyond the tip of the prosthesis (Fig. 1).
A standard femoral prosthesis was inserted in 219 hips (91 per cent), and the remainder of the hips received a straight, narrow stem, which was specially designed for use in small, dysplastic hips. The femoral stem was polished and was of the now-obsolete so-called first-generation, flat-back design (Fig. 1), which has been reported to have a high rate of fracture19. A standard cup with a narrow, indented rim was inserted in 219 hips (91 per cent), and an extra-small cup, without a rim, was used in the remaining twenty-one hips. Most of the small components were inserted in young patients.
Revisions were done on the basis of clinical and radiographic findings, as we do not believe that radiographic failure alone is synonymous with clinical failure unless there is severe resorption of bone.
The patients were seen six months and two, five, and ten years after the operation, and any surviving patients were evaluated at the time of the most recent follow-up in 1990.
Data were prospectively recorded on standardized forms, and the radiographs were kept on file. Clinically, the hips were assessed with the scale of Merle d'Aubigé and Postel, as modified by Charnley4. Radiographically, the femoral components were evaluated with use of the principles described by Harris et al. Similar criteria were used for the cup. Definite loosening was defined as migration of the cup; probable loosening, as a complete radiolucent zone around the cement mantle; and possible loosening, as a radiolucent line along 50 to 99 per cent of the cement-bone interface.
The patients were divided into two groups at the most recent follow-up examination. One group comprised patients who were fifty-five years old or less at the time of the operation and the other group, those who were at least fifty-six years old. Fifty-five years was chosen as the maximum age for the younger patients as, at the time of the operations, total hip arthroplasty was not recommended for patients who were less than fifty-five years old. Furthermore, other authors have used a similar dividing point3,6,14,17. Ahnfelt et al. also used fifty-five years as the upper age-limit in their youngest group of patients.
The preoperative level of activity of the two groups differed. Twenty-seven of the younger patients (twenty-eight hips; 54 per cent) were employed in an occupation that involved heavy manual labor (for example, farmer, laborer, or craftsperson), compared with forty-seven of the older patients (fifty-four hips; 29 per cent). Eleven of the younger patients (twelve hips; 23 per cent) were employed in an occupation that involved light, sedentary labor (for example, domestic worker, clerk, or homemaker), compared with thirty-one of the older patients (thirty-five hips; 19 per cent). Only eight of the younger patients (twelve hips; 23 per cent) were retired or received a pension, compared with eighty-seven of the older patients (ninety-nine hips; 53 per cent).
Ninety-six older patients (111 hips; 59 per cent) and fifteen younger patients (fifteen hips; 29 per cent) died before the time of the most recent follow-up. The median time from the operation to the time of death was similar between the two age-groups (10.8 years for the younger patients and 12.5 years for the older patients) (Table II).
The rate of follow-up of the surviving hips was very high (Table II). Only three of the surviving patients (four hips), all of whom were in the older group, were not able to return for the most recent follow-up examination; all three did not return because of severe, incapacitating disease. However, it was confirmed that none of these patients had had a revision operation. The median duration of follow-up was 17.0 years (range, 15.0 to 20.6 years) for the younger patients and 17.7 years (range, 15.1 to 20.4 years) for the older patients.
Twenty hips (8 per cent) were revised: five (10 per cent) of the fifty-two hips in the younger patients and fifteen (8 per cent) of the 188 hips in the older patients. All of the infections (five hips) and four of the five fractures of the femoral stem occurred in the older patients. The rate of aseptic loosening was higher in the younger patients (three hips; 6 per cent) than in the older patients (five hips; 3 per cent), but this difference was not significant (p = 0.37, Fisher exact test) (Table III). At the time of the most recent follow-up, no additional patients met our previously described criteria for revision. Five patients (2 per cent) had a Charnley score for pain4 of 3 points or less, but none of them had any radiographic signs of loosening. When the two age-groups were compared, no significant differences were found concerning the rate of radiographic loosening of the components or the rate of fracture of the cement around the femoral component (Table III). Of the twelve hips that had fracture of the cement, ten had definite radiographic loosening of the stem, according to the criteria of Harris et al., but no clinical signs of loosening. The fractures of the cement all occurred around the tip of the prosthesis. Only one of the five cups that were loose was associated with definite loosening of the femoral stem.
The median Charnley scores for pain4 preoperatively; at six months and two, five, and ten years after the operation; and at the most recent follow-up examination were identical for the two groups (Fig. 2-A). The older patients had a lower median score for function (4 points) than the younger patients (5 points) at the latest follow-up examination, but this was due to general deterioration of health rather than to problems related to the hip (Fig. 2-B). It should also be noted that the median score for function peaked at five years after the operation and then leveled off. This indicates that the final clinical result of a hip replacement should not be assessed sooner than five years after the operation. There were no substantial differences with regard to the median Charnley scores for motion between the two groups, even at the latest follow-up examination (Fig. 2-C).
Thirty-four of the younger patients (thirty-six hips; 69 per cent) were still able to perform their job after the operation, and at the most recent follow-up examination five still had a physically demanding job.
The data for the two groups were compared with use of a Kaplan-Meier analysis. The main course of the graphs of the probability of revision in the two groups was almost identical. However, the five revisions in the younger patients were done six to sixteen years after the operation, whereas the revisions in the older patients were distributed throughout the entire follow-up period, with three revisions performed within the first postoperative year (Figs. 3-A and 3-B). The rate of revision seemed to be steady over time, and no increase was seen at the time of the longest follow-up, which confirms the findings of Older and Butorac.
The probability of survival of the Charnley low-friction arthroplasty at twenty years in the present study was 88.3 per cent in the younger patients (95 per cent confidence interval, ±9.8 per cent) and 89.3 per cent in the older patients (95 per cent confidence interval, ±5.8 per cent). There was no significant difference between the rates of survival (p = 0.82). The high rate of follow-up makes these figures very reliable.
Total hip replacement in young patients is a controversial issue. Because these patients have a longer life expectancy and because they are very often active professionally and recreationally, higher loads are placed on the prosthesis for a longer time than on those implanted in older, less active patients. As a revision operation yields results that are less favorable than those of a primary operation, great effort has been put into the development of better implants. Recently, the insertion of implants without cement has been recommended, especially in young patients, with the expectation that these implants are associated with fewer long-term complications, such as loosening, and are easier to revise than prostheses inserted with cement. In the current literature, however, there is a trend back toward the insertion of implants with cement, particularly on the femoral side of the arthroplasty, as several studies showing excellent long-term results regarding these implants are now available1,2,13,14,20. Havelin et al., with use of the Norwegian Arthroplasty Register, reported the worst results for implants inserted without cement in young male patients, whereas there was no difference between the results of implants inserted with cement and those inserted without cement in older patients. Still, insertion of implants without cement is often recommended specifically for young patients.
The purpose of the present study was to provide highly reliable long-term survival data on Charnley total hip replacements performed with cement in young patients and to compare the results with those in a group of older patients who were managed in the same manner. Use of the People's Register of Denmark made a high rate of follow-up possible.
We were not able to confirm the findings of Gudmundsson et al. that men who were less than seventy years old had a significantly higher rate of loosening than women who were less than seventy. Stauffer did not find a higher frequency of loosening of the femoral component in patients who were less than sixty years old at the time of the operation, but he did find a higher frequency of loosening of the acetabular cup. Sarmiento et al. found no difference in the rate of revision for patients with osteoarthrosis who were less than fifty years old at the time of the operation, compared with the rate for those who were more than fifty.
In the present study, eight of the younger patients (twelve hips; 23 per cent) were receiving a pension or were retired at the time of the operation, compared with eighty-seven of the older patients (ninety-nine hips; 53 per cent). At the time of the operation, twenty-seven of the younger patients (twenty-eight hips; 54 per cent) were employed in a job that required heavy manual labor, compared with forty-seven of the older patients (fifty-four hips; 29 per cent). The younger patients had a much higher preoperative level of activity than the older patients, despite the different underlying hip conditions, and the demands that the younger patients placed on the hip replacement were expected to cause a higher rate of revision. However, the difference in the level of activity between the two groups did not influence the results at the time of the most recent follow-up. The younger patients did not have a higher frequency of long-term complications than the older patients.
We have previously demonstrated13 that there is little correlation between radiographic evidence of loosening, according to the criteria of Harris et al., and loosening that causes clinical symptoms. In a study by Nicholson, only eighteen of thirty-three radiographically loose stems and sixteen of thirty-one radiographically loose cups were revised. In a study by Older and Butorac, only eleven of thirty radiographically loose components were revised. Collis recommended revision on the basis of clinical findings and not on the basis of radiographic findings alone. We recommend a similarly conservative approach—that is, revision only for symptomatic patients who have radiographic evidence of loosening and close observation of asymptomatic patients who have radiographic evidence of loosening, in order to ensure intervention before eventual severe loss of bone stock renders a revision difficult or impossible. As a consequence, we now make routine radiographs only at the six-month follow-up examination or if a patient is symptomatic.
We believe that a Charnley hip replacement with cement should be a standard choice even for patients who are thirty-four to fifty-five years old, especially as there has not yet been sufficient follow-up of hip replacements performed without cement and the long-term results of reconstruction with these implants are thus not known. This view is consistent with that of Older and Butorac, who, on the basis of similar results, found it difficult to justify the widespread use of hip replacements without cement except in high-risk patients. However, we do not believe that clear guidelines have yet been defined for hip replacement without cement.
In other studies10,16, the uneven distribution of patients who had rheumatoid arthritis in the groups being compared may have influenced the outcomes. However, none of the younger patients in the present study had rheumatoid arthritis, and therefore we had no experience with the use of this type of prosthesis in such patients. Joshi et al. found that a cemented prosthesis in a young patient who has rheumatoid arthritis has a higher possibility of survival than one in a young patient who has osteoarthrosis.
We have seen no proof so far that prostheses inserted without cement are better suited for young patients, as no long-term results are yet available, and we do not plan to use them before such results are published. In the present series, in which the rate of follow-up of the younger patients was 100 per cent and the median duration of follow-up was 17.6 years (range, 15.0 to 20.6 years), the results were very good, even for the younger patients. Similarly, the results in the study by Collis were so favorable that he recommended insertion of prostheses with cement for patients who were less than fifty years old. Even if young patients have a slightly higher risk of long-term complications, the results are still very good.
The number of revisions in the present series was low compared with that in other series with the same or shorter follow-up3,6,10,14,18. The results of most of those studies were so good that the authors stated their intention to use the Charnley low-friction arthroplasty or a similar implant in young patients in the future. We support this view and anticipate similar or better results in the future, with the use of laminar airflow operating rooms, prophylactic antibiotics, and modern techniques of cement application. Barrack et al. reported no radiographic loosening of implants after fifty total hip replacements performed with the use of more advanced cementing techniques in patients who were forty-four years old or less.
Ahnfelt, L.; Herberts, P.; Malchau, H.; and |and |Andersson, G. B.: Prognosis of total hip replacement. A Swedish multicenter study of 4,664 revisions. Acta Orthop. Scandinavica,Supplementum 238: 1990.Supplementum 238
1990
Barrack, R. L.; Mulroy, R. D., Jr.; and |and |Harris, W. H.: Improved cementing techniques and femoral component loosening in young patients with hip arthroplasty. A 12-year radiographic review. J. Bone and Joint Surg.,74-B(3): 385-389, 1992.74-B(3)385
1992
Carter, S. R.; Pynsent, P. B.; and |and |McMinn, D. W. J.: Greater than ten year survivorship of Charnley low friction arthroplasty. J. Bone and Joint Surg.,73-B (Supplement I): 71, 1991.73-B (Supplement I)71
1991
Charnley, J.: The long-term results of low-friction arthroplasty of the hip performed as a primary intervention. J. Bone and Joint Surg.,54-B(1): 61-76, 1972.54-B(1)61
1972
Charnley, J.: Low Friction Arthroplasty of the Hip. Theory and Practice. New York, Springer, 1979.
Collis, D. K.: Long-term (twelve to eighteen-year) follow-up of cemented total hip replacements in patients who were less than fifty years old. A follow-up note. J. Bone and Joint Surg.,73-A: 593-597, April 1991.73-A593
1991
Gudmundsson, G. H.; Hedeboe, J.; and |and |Kjær, J.: Mechanical loosening after hip replacement. Incidence after 10 years in 125 patients. Acta Orthop. Scandinavica,56: 314-317, 1985.56314
1985
[CrossRef]
Harris, W. H.; McCarthy, J. C., Jr.; and |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
Havelin, L. I.; Espehaug, B.; Vollset, S. E.; and |and |Engesæter, L. B.: Early failures among 14,009 cemented and 1,326 uncemented prostheses for primary coxarthrosis. The Norwegian Arthroplasty Register, 1987-1992. Acta Orthop. Scandinavica,65: 1-6, 1994.651
1994
[CrossRef]
Joshi, A. B.; Porter, M. L.; Trail, I. A.; Murphy, J. C. M.; and |and |Hardinge, K.: The survival analysis of Charnley low friction arthroplasty in patients forty years of age or younger. J. Bone and Joint Surg.,74-B (Supplement II): 139, 1992.74-B (Supplement II)139
1992
Kaplan, E. L., and |and |Meier, P.: Nonparametric estimation from incomplete observations. J. Am. Statist. Assn.,53: 457-481, 1958.53457
1958
[CrossRef]
Merle d'Aubigné, R., and |and |Postel, M.: Functional results of hip arthroplasty with acrylic prosthesis. J. Bone and Joint Surg.,36-A: 451-475, June 1954.36-A451
1954
Neumann, L.; Freund, K. G.; and |and |Sørenson, K. H.: Long-term results of Charnley total hip replacement. J. Bone and Joint Surg.,76-B(2): 245-251, 1994.76-B(2)245
1994
Nicholson, O. R.: The Charnley hip replacement—a longish-term review. J. Bone and Joint Surg.,74-B (Supplement I): 37, 1992.74-B (Supplement I)37
1992
Older, J., and |and |Butorac, R.: Charnley low friction arthroplasty (LFA). A 17-21-year follow-up study. J. Bone and Joint Surg.,74-B (Supplement III): 251, 1992.74-B (Supplement III)251
1992
Sarmiento, A.; Ebramzadeh, E.; Gogan, W. J.; and |and |McKellop, H. A.: Total arthroplasty with cement. A long-term radiographic analysis in patients who are older than fifty and younger than fifty years. J. Bone and Joint Surg.,72-A: 1470-1476, Dec. 1990.72-A1470
1990
Stauffer, R. N.: Ten-year follow-up study of total hip replacement. With particular reference to roentgenographic loosening of the components. J. Bone and Joint Surg.,64-A: 983-990, Sept. 1982.64-A983
1982
Sullivan, P. M.; MacKenzie, J. R.; Callaghan, J. J.; and |and |Johnston, R. C.: Total hip arthroplasty with cement in patients who are less than fifty years old. A sixteen to twenty-two-year follow-up study. J. Bone and Joint Surg.,76-A: 863-869, June 1994.76-A863
1994
Wroblewski, B. M.: Fractured stem in total hip replacement. A clinical review of 120 cases. Acta Orthop. Scandinavica,53: 279-284, 1982.53279
1982
[CrossRef]
Wroblewski, B. M.; Siney, P. D.; Raut, V. V.; and |and |Taylor, G.: Charnley low-friction arthroplasty—18-26 year results. In Proceedings of the British Orthopaedic Association. J. Bone and Joint Surg.,74-B (Supplement II): 140-141, 1992.74-B (Supplement II)140
1992