The fact that THA is a remarkably effective, reliable, and cost-effective surgical procedure cannot be disputed. 19,20,22,26 However, the outcome of THA in a given group of patients can differ significantly depending on the type of outcome measurement device used. 5 Furthermore, various patient-specific factors such as the socioeconomic status, functional ability, and level of patient expectations may alter the outcome of joint replacement. 26 Charnley 6 recognized the importance of these patient-related factors and suggested that the outcome of THA in two groups of patients only should be compared if the groups are similar in other respects.
The goal of evaluating the outcome of THA more reproducibly stimulated the development of hip scoring systems. The first evaluation system dates to 1931. 11 Subsequently hip scoring systems such as those of Merle d’Aubigne and Postel 23 or the Harris Hip score (HHS) 14 were developed, and continue to be used widely as clinical outcome instruments. The high level of variability among traditional hip scores has resulted in a diverse group of instruments using different languages and definitions, scaling factors, and weighting of the various domains. This has resulted in an array of disparate descriptive terms and numerical scores that cannot be compared with each other and have not all been validated scientifically. 16 Therefore, studies on long-term hip outcome may suffer serious methodologic deficiencies.
In response to a call from the orthopaedic community for a uniform method of evaluation and reporting of outcome of hip arthroplasty, the Hip Society, Society International Congress of Orthopaedics and Traumatology (SICOT), and the American Academy of Orthopaedic Surgeons (AAOS) convened a task force that established a system of Clinical and Radiographic Terminology (CART). 18
Using CART as the basis, Professor M. E. Müller established the International Documentation and Evaluation System (IDES). 24 Using that system and its precursors, data on more than 58,000 primary and revision hip replacements have been collected and recorded from 45 participating European centers since 1965 and stored in a centralized computerized database.
We analyzed the functional outcome of primary THA in a large cohort of patients during a 15-year period and reported these data in accordance with the principles of CART. The major purpose of the study was to evaluate the influence of demographic-related and patient-related factors on functional outcome of THA by subanalyzing and reporting these data for patients of different gender, age, diagnostic groups, body mass index, and Charnley classes.
MATERIALS AND METHODS
The current study was restricted to patients older than 20 years having primary THA and the respective followup examinations of the original implants. The description of outcome of patients with revision arthroplasty was not a study goal and therefore all revision-intervention and revision-followup records were excluded. There were 21,997 patients having 25,990 hip replacements. Bilateral hip replacements had been done in 3993 patients (18.2%). There were 11,754 (53.4%) women and 10,243 (46.6%) men with a median age of 66 years (range, 20–96 years). Patients were divided into three age groups of 20 to 50 years (8.3%), 51 to 70 years (54.6%), and older than 70 years (37.1%). Distribution of main diagnoses was osteoarthritis in 75.6% of patients, developmental dysplasia in 8.7% of patients, inflammatory arthritis in 3.2% of patients, fracture of the femoral neck in 6.3% of patients, and other diagnoses in 6.2% of patients. At the time of surgery, 68.7% of patients were in Charnley Class A, 28.2% of patients were in Charnley Class B, and 3.1% of patients were in Charnley Class C (Table 1). Because of the long duration of the study, patients had coexistent conditions develop or received treatment for arthritis of other joints. This had resulted in a change in the distribution of patients in various Charnley groups. At the latest followup, 57.6% of patients were in Charnley Class A, 18.7% of patients were in Charnley Class B, and 23.7% of patients were in Charnley Class C. There was hybrid fixation of components (cemented stem, uncemented cup) in 44.4% of cases, reverse hybrid fixation (uncemented stem, cemented cup) in 13.7% of cases, fully uncemented fixation in 14.4% of cases, and fully cemented fixation in 27.5% of cases. The THAs were done using various surgical approaches, by surgeons of variable experience, and in one of the 45 central European institutions.
The data were derived from the hip database of our institution. Preoperative and postoperative clinical, radiographic and implant data of consecutive cases were documented with optically readable code sheets according to the standards of the International Documentation and Evaluation System (IDES). 24 Clinical data derived from the physical examination and the patient interview were captured by collating the outcome of the respective parameter to the best matching answer category for this parameter on the IDES sheet, based on the CART system. The institute uses three different types of comprehensive and extensive data sheets. The IDES Sheet-A records the preoperative, perioperative, and postoperative information of the patient until discharge from hospital. The IDES Sheet-C serves for documentation of clinical and radiographic followups including a section for patient satisfaction. The IDES Sheet-B is completed for patients having revision surgery. The circumstances leading to and the details of revision surgery are documented in this sheet.
Patients were categorized according to gender, age, main diagnosis, body mass index, and Charnley class 6 at the time of the primary THA. The categorizations were based on the detailed information extracted from the IDES sheets relating to patient demographics, underlying disorder, and functional limitations caused by unilateral (Charnley A) or bilateral hip disease (Charnley B), or other coexistent conditions affecting walking capacity such as angina pectoris or arthropathy of other joints (Charnley C). The distribution of patients based on Charnley classification, gender, age, and diagnosis is shown in Table 1.
Functional Hip Scores
The information contained in the IDES Sheet-C also was used to calculate the HHS. All domains for HHS, except for sitting and the use of public transport, are recorded on the IDES sheets. Functional daily activities such as getting up from chair, stair climbing, and walking distance were used to determine the latter domains. Where more answer options are given on the IDES sheets than is needed to calculate the score, domains were bundled to synchronize the respective parameters.
Serial AP and lateral radiographs of the operated joint were reviewed using a standardized measurement tool for evaluation of hip components. All changes around the uncemented and cemented component were documented. 2,10,13,15 Detailed radiographic data, including component positioning or migration, wear, osteolysis, and other parameters were obtained and recorded in the IDES C-Sheets. Acetabular and femoral loosening was defined by comparing the postoperative and followup radiographs. The following was deemed to represent radiographic loosening of the components: continuous radiolucencies around the socket in Zones 1 to 3 according to DeLee and Charnley 8; a superior migration equal to or greater than 5 mm, protrusion or a progressive tilt of the socket 10; a fracture of the socket, stem, or the cement mantle; stem subsidence equal to or greater than 3 mm; radiolucencies measuring greater than 2 mm at the bone-cement or bone-stem interface; and circumferential radiolucencies at the bone-cement or bone-stem interface or multiple small areas of focal osteolysis, or any large osteolytic defects around the stem.
Statistical analysis was done according to a cross sectional study design separately for each followup year until postoperative Year 15. All available followups of patients included in the study were grouped according to years after primary intervention. In case of multiple examinations within 1 year, the followup nearest the middle of the year was used. The outcomes of the various variables were calculated as prevalence for the respective followup year and represent 16 (preoperative status and 15 followup years) cross sectional observations of independent patient groups where each followup group forms a subgroup of the primary patient sample. Ninety-five percent confidence intervals were calculated for each parameter and graphically displayed as shaded areas around the curves. Differences of outcome measurements between patients groups were considered as significant where respective confidence bands were not overlapping. The statistical analysis corresponds therefore to a univariate analysis of the data. All graphic and statistical evaluations were done using SAS 8.2 (SAS Institute Inc, Cary, NC).
The IDES-C followup sheets contain a section that is completed by, or for the patient evaluating the outcome of surgery as perceived by the patient. The patient satisfaction is categorized into excellent, good, fair, and poor depending on patient’s opinion regarding the success of arthroplasty in addressing symptoms, limitations, and need for medication. The outcome of arthroplasty was perceived to be excellent or good by more than 93% of the patients in the short-term to mid-term followup. Patient satisfaction declined gradually, but slightly, with time such that in the later followup periods between 85% and 90% of patients were satisfied with the outcome of the intervention.
Figure 1 shows the percentage of patients with excellent or good satisfaction in the Charnley classes. A satisfactory outcome was most durable for patients in Charnley Class C. None of the other demographic factors had a significant influence on patient satisfaction with outcome.
Harris hip scores were evaluated for the entire patient population and for each demographic, diagnostic, or functional subgroup. The early (Years 1–5) functional outcome was excellent or good for 80% to 87% of the entire patient population, expressed by a score result greater than or equal to 80. Furthermore, this favorable outcome, despite a slight decline, remained very high throughout the study with more than 70% of patients having excellent to good outcomes at 15 years.
The proportion of male patients with an excellent HHS (≥90 points) was significantly higher than in the female patient group during the greatest part of the followup (Fig 2A). There was no significant difference between age groups 20 to 50 years and 51 to 70 years. However, patients older than 70 years had constantly lower functional scores than the younger patients throughout all followups (Fig 2B). Although the number of patients with an excellent functional outcome expressed by HHS constantly decreased with increasing body mass index, the differences were not always statistically significant (Fig 2C). The improvement in HHS was most dramatic for patients in Charnley Class A (Fig 2D). However, patients in Charnley Classes B and C despite having a relatively lower HHS similarly were satisfied with the outcome of THA (Fig 1).
Because of large variations of outcome within the diagnostic groups no significant difference, as measured by HHS, was detected for patients based on their underlying disorders.
The intensity of pain before THA and at regular intervals thereafter was evaluated. More than 95% of patients had disabling pain before THA. The surgical intervention was successful in alleviating pain for almost all patients such that at 2 years after arthroplasty, more than 95% of patients were completely pain-free or only had mild pain. Despite a decline in pain relief, more than 86% of the patients had none or only mild pain at 15 years. The frequency of patients with complete pain relief in different Charnley classes also was evaluated and was similar in all the three groups. Patients in Charnley Class B had a slightly poorer prognosis with respect to the longevity of pain relief (Fig 3A). Of all other factors assessed, only the preoperative diagnosis had a significant influence on the proportion of patients who were pain free. The patient group with a fracture of the femoral neck had the poorest outcome with respect to complete pain relief (Fig 3B).
Range of Motion
Patients in this study group had examinations and recordings of the hip ROM preoperatively and at their followups. Hip flexion, divided into different ranges, as the most important representative of hip motion was analyzed. Before THA, only 21% of patients were able to achieve a hip flexion in excess of 90° whereas 1 year after the surgery, hip flexion greater than 90° was possible in 73% of the hips. The hip flexion range did not peak until the fifth year after surgery when 79% of hips achieved hip flexion in excess of 90°. With time, a decline in ROM was observed. Subanalysis according to the Charnley classes revealed that patients in Charnley Class B had the poorest preoperative and postoperative hip flexion (Fig 4A). Also, the preoperative diagnosis had a significant influence on hip flexion where a smaller number of patients with developmental dysplasia achieved a hip flexion greater than 90° (Fig 4B). None of the other demographic factors significantly influenced the range of hip flexion.
Overall, ambulation was the parameter most sensitive to the influence of the demographic factors being studied. Before THA, more than 48% of patients required walking aids for ambulation. One year postoperatively the percentage had declined to 32.1% and reached the lowest level in postoperative Year 5 when only 12.8% of patients required walking aids. Preoperatively, only 8.5% of patients were able to walk unaided for more than 60 minutes compared with 46% of patients in the first year after the joint replacement. During the following 2 years, there was additional improvement in ambulation and the maximum proportion of patients achieving an excellent walking status was not reached until 3 years postoperatively (59.3%). Preoperative differences between the two genders were even more pronounced postoperatively and the male patient group had a significantly larger proportion of individuals able to walk at least 60 minutes during the first 10 years after the intervention (Fig 5A). Minimal preoperative differences in ambulatory status among the three age groups became significant after arthroplasty between patients younger than and older than 70 years, and to a lesser extent between patients younger than 50 years and those 51 to 70 years (Fig 5B). The wide confidence intervals in the diagnostic groups did not allow precise conclusions regarding differences in ambulatory capacities. However, compared with other diagnoses, twice as many individuals with fractures of the femoral neck were ambulating distances more than 60 minutes preoperatively. Postoperatively, this patient group with the highest median age (70 years) did not regain the walking capacity as fast and as extensively as other patients. Individuals with inflammatory arthritis were the poorest performers preoperatively and postoperatively (Fig 5C). Despite minimal preoperative differences, there were significantly fewer patients with excellent postoperative ambulatory status in the obese group compared with the normal weight group (Fig 5D). Charnley class had the most profound influence on ambulatory capacity. The gain in walking capacities was most impressive for patients in Charnley Class A and least for patients in Charnley Class C. There was, however, a progressive decline in ability to ambulate unaided beginning 5 years postoperatively across all Charnley classes. This deterioration in walking capacities was most pronounced for patients in Charnley Class B (Fig 5E). Stair climbing activities were influenced similarly by these demographic factors.
The frequency of acetabular and femoral component loosening was determined. There was a gradual and constant increase in the incidence of radiographic loosening for femoral and acetabular components during the study. The number of radiographically loose stems increased constantly from 0.87% in postoperative Year 1 to 5.8% in Year 5, 12.5% in Year 10 to 14.4% in Year 14 after surgery. The number of radiographically loose cups increased from initial 0.8% in postoperative Year 1 to 1.9% in Year 5 and 9.5% in Year 10 to 16.7% in postoperative Year 14. The prevalence of femoral stem loosening was higher than the prevalence of acetabular cup loosening in the first postoperative decade. This relationship was reversed for the second postoperative decade.
Although the medical community has set forth uniform data elements collected before and after each treatment, usually no well-defined spectrum of validated and risk-adjusted outcomes for medical and surgical interventions exists. Outcome measures used in evaluating and reporting the results of THA suffer the same limitations. 21
To standardize outcome measures, CART was developed in the early 1990s to be used for building blocks of information and to form a common language for reporting the outcome of THA. This development helped define a core minimal data set of items to be used in the evaluation of patients being treated for hip disease. The IDES, which has been used at our institution for more than 20 years, is based on the principles and concepts of CART. The current study used this uniform outcome assessment to determine the long-term functional results of THA in a very large group of patients for 15 years.
This study corroborates many previous studies showing that the outcome of THA is excellent and durable as measured by various outcome parameters and patient reported criteria. The study did not intend to evaluate the survivorship of THA components that has been studied previously and reported extensively. 2,4,7,9,27,28 Rather, it focuses on the various parameters of outcome evaluation.
The strong time dependency of functional outcome after THA was shown clearly. Many of the functional gains and maximum improvement for pain relief are achieved during the first 2 years after THA. However, in distinction to previous information, this study showed that some functional parameters, most notably walking capacity, continue to improve for at least 3 and as many as 5 years after THA. The study also showed a time dependent gradual decline in function, as measured by all outcome parameters that begins on average, approximately 5 years after THA and continues thereafter.
Demographic factors have a stronger effect on the functional outcome of THA than on pain relief or satisfaction with the treatment result.
Differences in ambulation and stair climbing capacities were observed between the two gender groups preoperatively and were more marked after the intervention so that postoperatively male patients had a higher likelihood of gaining the highest function than the female patients. The course of gain and loss of ambulatory capacities is approximately the same for the two gender groups with a maximum function 3 years postoperatively and an approximate parallel decline in the proportion of patients with excellent walking capacities thereafter. These differences in functional outcome also are reflected in the HHS where the proportion of male patients with an excellent result is significantly higher in the short-term and mid-term followup. Greenfield et al 12 observed similar differences between the gender groups. In their study, male patients had a significantly higher Activity of Daily Living Score at 1 year postoperatively. Gender as the sole influential factor may not entirely explain the functional differences. In our patient cohort, women had a median age of 67 years at THA compared with 65 years in men. Also, diagnoses with a less favorable functional outcome, such as inflammatory arthritis and femoral neck fracture were more prevalent in the female cohort. Aging is one of the most relevant factors causing deterioration of overall functional status. In our study, differences in the proportion of patients with excellent walking and stair climbing activities and an excellent HHS were less pronounced between age groups 20 to 50 years and 51 to 70 years but remarkable between age groups older than and younger than 70 years. Similarly, the literature reports that increasing age is associated with poorer functional outcome. 17,25
The influence of the main diagnosis had a strong effect on the postoperative functional parameters. Although sample size caused confidence intervals to be wider in the inflammatory arthritis group and the group with fracture of the femoral neck it was apparent that these patients had poorer walking and stair climbing capacities than patients with osteoarthritis or dysplasia. The relative improvement in walking capacities is smallest for patients with fracture of the femoral neck. This group had the highest proportion of patients with excellent walking capacities preoperatively (17.6%). Postoperatively however, the gain in walking capacities lagged behind that of patients with osteoarthritis and dysplasia and did not reach a comparably high maximum. Important factors contributing to the poor performance of patients with fracture of the femoral neck were the highest median age of all diagnostic groups (70 years) and a higher proportion of female patients (62%). Patients with inflammatory arthritis had the poorest preoperative and postoperative ambulatory capacities. Many patients suffering from multijoint involvement are known to have reduced ambulatory capacities. Accordingly, the percentage of those patients belonging to Charnley Class C was 19% compared with 1% in the osteoarthritis group, 2% in the dysplasia group, and 9% in the group with femoral neck fractures. Hip motion, as expressed by flexion range is affected differently by the main diagnosis. Hip dysplasia was the only diagnostic group with a statistically lower percentage of patients having at least 90° flexion. This group also had the lowest ROM of all diagnostic groups preoperatively, which suggests stiffness of the soft tissue structures around the hip was not reversed completely by the intervention. Although Young et al 29 concluded in their review that the effect of diagnosis on outcome was uncertain, our results with this large group of patients clearly showed the influences of specific disease categories on functional outcome after THA.
Body mass index had a clear association with walking capacity and overall outcome as expressed by HHS. Increasing body mass index was associated with a decreasing number of patients with excellent walking capacities or an excellent HHS. Similar to the relationship between male and female patients the preoperatively minimal differences are amplified after arthroplasty.
Because the Charnley classification reflects the influence of the patient’s locomotor system and overall health status on ambulation, it also was included in the analysis as a more general demographic factor. The impact of Charnley class was profound and affected all parameters assessed. For many functional parameters, Charnley classification was the one factor that had the greatest influence on overall outcome. The current study showed that the degree of change in function with time is markedly dependent on the Charnley class of the patient. Although the patients in Charnley Class B had the lowest median age (64 years), this cohort had the most rapid deterioration of function after THA, starting approximately 6 to 7 years after arthroplasty. Explanations therefore could be an aggravation of symptoms on the diseased, not yet operatively-treated contralateral side or a deterioration of the mechanical status of one or both THAs.
According to the HHS system, more of the patients in Charnley Classes B and C had suboptimal outcomes and compromised scores. The latter was, however, not entirely true when individual functional outcome measures were evaluated. The HHS declined and became distorted for these patients because of the influence of their coexistent conditions. 3–5 Based on our findings, and in agreement with Callaghan et al, 5 we think there is value in reporting the results of THA separately for the Charnley classes; likewise, there is value in reporting individual parameters, instead of composite scores.
The current study suffered some shortfalls. It is a report on the outcome of THA using different design prostheses, various fixation methods, with the surgery being done in multiple centers and by surgeons of various experiences. The aforementioned factors clearly influence the performance of THA, and in particular may have influenced the incidence of pain, radiographic lucency, and the ROM. Also, the current study was limited to a cohort of patients with documented followups from 1 to 15 years after primary THA whereby none of the patients had a complete record of 15 followups. Possibly, the outcome of THA, using the presented methodology was altered positively by inclusion of patients who were lost to followup before the end of the study and by exclusion of patients with revision surgery within 15 years after primary THA.
Despite the aforementioned limitations, the current study was unique in many aspects and benefits from considerable strength. It reported outcome measures using uniform and stringent method of data collection in a very large patient population. The inclusion of patients from various orthopaedic practices, academic and nonacademic, may maximize the ability to generalize the conclusions of this study.
Total hip arthroplasty provides considerable improvement in function with respect to mobility and ROM and permanently relieves pain in a majority of patients. The optimal outcome of the intervention is not detected until 3 to 5 years after arthroplasty and despite deterioration with time, the results of THA are durable and satisfactory. Various demographic factors influence the functional outcome with the Charnley classification having the most profound effect on postoperative functional parameters.
Berry DJ, Harmsen WS, Cabanela ME, Morrey BF: Twenty-five-year survivorship of two thousand consecutive primary Charnley total hip replacements: Factors affecting survivorship of acetabular and femoral components. J Bone Joint Surg 84A:171–177, 2002.
2. Brand RA, Pedersen DR, Yoder SA: How definition of “loosening” affects the incidence of loose total hip reconstructions. Clin Orthop 210:185–191, 1986.
3. Bryant MJ, Kernohan WG, Nixon JR, Mollan RA: A statistical analysis of hip scores. J Bone Joint Surg 75B:705–709, 1993.
4. Callaghan JJ, Albright JC, Goetz DD, Olejniczak JP, Johnston RC: Charnley total hip arthroplasty with cement: Minimum twenty-five-year follow-up. J Bone Joint Surg 82A:487–497, 2000.
5. Callaghan JJ, Dysart SH, Savory CF, Hopkinson WJ: Assessing the results of hip replacement: A comparison of five different rating systems. J Bone Joint Surg 72B:1008–1009, 1990.
6. Charnley J: The long-term results of low-friction arthroplasty of the hip performed as a primary intervention. J Bone Joint Surg 54B:61–76, 1972.
7. Dawson J, Jameson-Shortall E, Emerton M, et al: Issues relating to long-term follow-up in hip arthroplasty surgery: A review of 598 cases at 7 years comparing 2 prostheses using revision rates, survival analysis, and patient-based measures. J Arthroplasty 15:710–717, 2000.
8. DeLee JG, Charnley J: Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop 121:20–32, 1976.
9. Emery D, Britton A, Clarke H, Grover M: The Stanmore total hip arthroplasty: A 15- to 20-year follow-up study. J Arthroplasty 12:728–735, 1997.
10. Engh CA, Massin P, Suthers KE: Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components. Clin Orthop 257:107–128, 1990.
11. Ferguson AB, Howorth MB: Slipping of the upper femoral epiphysis. JAMA 97:1867–1872, 1931.
12. Greenfield S, Apolone G, McNeil BJ, Cleary PD: The importance of co-existent disease in the occurrence of postoperative complications and one-year recovery in patients undergoing total hip replacement: Comorbidity and outcomes after hip replacement. Med Care 31:141–154, 1993.
13. Gruen TA, McNeice GM, Amstutz HC: “Modes of failure” of cemented stem-type femoral components: A radiographic analysis of loosening. Clin Orthop 141:17–27, 1979.
14. Harris WH: Traumatic arthritis of the hip after dislocation and acetabular fractures: Treatment by mold arthroplasty. J Bone Joint Surg 51A:737–755, 1969.
15. Hodgkinson JP, Shelley P, Wroblewski BM: The correlation between the roentgenographic appearance and operative findings at the bone-cement junction of the socket in Charnley low friction arthroplasties. Clin Orthop 228:105–109, 1988.
16. Johanson NA: Outcomes Assessment. In Callaghan JJ, Rosenberg AG, Rubash HE, (eds). The Adult Hip. Philadelphia, Lippincot Raven Publishers 853–863, 1998.
17. Johnsson R, Thorngren KG: Function after total hip replacement for primary osteoarthritis. Int Orthop 13:221–225, 1989.
18. Johnston RC, Fitzgerald RH, Harris WH, et al: Clinical and radiographic evaluation of total hip replacement: A standard system of terminology for reporting results. J Bone Joint Surg 72A:161–168, 1990.
19. Jones CA, Voaklander DC, Johnston DW, Suarez-Almazor ME: Health related quality of life outcomes after total hip and knee arthroplasties in a community based population. J Rheumatol 27:1745–1752, 2000.
20. Liang MH, Cullen KE, Larson MG, et al: Cost-effectiveness of total joint arthroplasty in osteoarthritis. Arthritis Rheum 29:937–943, 1986.
21. Lingard E, Hashimoto H, Sledge C: Development of outcome research for total joint arthroplasty. J Orthop Sci 5:175–177, 2000.
22. March LM, Cross MJ, Lapsley H, et al: Outcomes after hip or knee replacement surgery for osteoarthritis: A prospective cohort study comparing patients’ quality of life before and after surgery with age-related population norms. Med J Aust 171:235–238, 1999.
23. Merle d’Aubigné R, Postel M: Functional results of hip arthroplasty with acrylic prosthesis. J Bone Joint Surg 36A:451–475, 1954.
24. Paterson D: The International Documentation and Evaluation System (IDES). Orthopedics 16:11–14, 1993.
25. Pettine KA, Aamlid BC, Cabanela ME: Elective total hip arthroplasty in patients older than 80 years of age. Clin Orthop 266:127–132, 1991.
26. Wiklund I, Romanus B: A comparison of quality of life before and after arthroplasty in patients who had arthrosis of the hip joint. J Bone Joint Surg 73A:765–769, 1991.
27. Wroblewski BM, Fleming PA, Siney PD: Charnley low-frictional torque arthroplasty of the hip: 20-to-30 year results. J Bone Joint Surg 81B:427–430, 1999.
28. Wroblewski BM, Siney PD: Charnley low-friction arthroplasty of the hip: Long-term results. Clin Orthop 292:191–201, 1993.
© 2003 Lippincott Williams & Wilkins, Inc.
29. Young NL, Cheah D, Waddell JP, Wright JG: Patient characteristics that affect the outcome of total hip arthroplasty: A review. Can J Surg 41:188–195, 1998.