Predicting the outcome of total hip replacement before surgery would facilitate in determining the relative indications for this procedure. Such predictors are difficult to validate because subjectivity weakens the reliability. The indication for total hip replacement is based primarily on a patient's quality of life, pain, and function, and not simply on the pre-operative radiograph, which usually is used to confirm the underlying diagnosis.3,4,13 Pain and function may be used to evaluate quality of life and for determining a patient's need for total hip replacement.12,14 However, because individuals may exaggerate or minimize pain and loss of function, predicting the outcome of a total hip replacement, based on these qualities, may be difficult. Decreased mental status and subsequent poor ability to report pain has been observed after total hip replacement.2,9,11 However, an individual with debilitating hip pain before surgery but a paucity of osteoarthritis as shown on the radiograph may present a legitimate dilemma for the surgeon. The current study was performed to determine whether a relationship exists between the type and extent of osteoarthritis observed on the preoperative radiograph and the postoperative results of a total hip replacement in patients with primary osteoarthritis.
MATERIALS AND METHODS
Between January 1989 and April 1996, 1171 consecutive primary total hip replacements were performed for primary osteoarthritis. The diagnosis of osteoarthritis was made radiographically and confirmed by pathologic specimen. Standard anteroposterior (AP) radiographs were taken using the same technique and were performed by the same technician. No attempt was made to distinguish inflammatory arthropathy from osteoarthritis based solely on the preoperative radiograph. Rather, all cases of inflammatory arthropathy as diagnosed from a pathologic specimen were excluded from the study.
At the authors' institution, the histopathologic criteria for differentiating inflammatory from noninflammatory arthritis (osteoarthritis) include the presence or absence of acute or chronic inflammatory cells (greater than 2-5 cells per high power field). These cells may include polymorphonuclear leukocytes, lymphocytes, histiocytes, plasma cells, or lymphoid follicles. The presence of a fibrinous pannus also negates the diagnosis of noninflammatory arthropathy. Of these cases 1163 hips (1015 patients) were followed up for an average of 32 months (range, 6-93 months). The followup schedule included visits at 2 months, 6 months, 1 year, 3 years, 5 years, and 7 years after surgery. Patients with less than 2 years of followup were included, rather than including only patients with a minimum 2 years of followup because early and long-term pain relief were studied. The average age of patients in the group was 67.2 years, with a standard deviation of 10.5 years. Fifty-five percent of the patients were women (562 hips). Eight patients (eight hips) were lost to followup or not included because of insufficient medical records. The remainder of the demographic data are included in Table 1.
One author (MAR) used an anterolateral approach on 441 (37.9%) of the hips. The remainder of the total hip replacements were performed using a posterior approach. Cemented (62.9%, 731 hips) and uncemented (37.1%, 432 hips) acetabular components were implanted. Cemented stems were implanted in 75.9% (883 hips) of cases.
An osteoarthritis grading system was used that applies generally to osteoarthritis and that separately grades three ascertained radiographic components of hip osteoarthritis: cartilage space narrowing or loss, osteophyte formation, and the direction of cartilage space loss (femoral head migration). Although AP and frog lateral radiographs were obtained before surgery, only the AP radiographs were used for assessment.
Previous reports provide no complete grading system because the methods either neglect one component of osteoarthritis or group the components together.1,7,15 Thus, cartilage space was graded as normal (no narrowing), narrowing, bone on bone, or bony erosion. Osteophytosis was graded similarly as none, mild, or severe. In general, if osteophyte formation definitely was present, but minimal in size, 1 to 2 mm for example, the grade was mild. Larger or more numerous osteophytes were graded as severe. Finally, the direction of cartilage space loss (femoral head migration) was recorded as global (concentric), superior, or medial (protrusio, for example). Thus, the direction, rather than the actual amount, of femoral head migration, was recorded. Thus, variability attributable to femoral head or acetabular degeneration was minimized.
A grading system similar to the one used in this review has been described previously by Kellgren and Lawrence.7 Their method, which grades osteoarthritis as simply none, questionable, minimal, moderate, or severe, has been shown to have a low intraobserver error rate (r = 0.75).7 Unlike the grading system used in the current report, the classification of Kellgren and Lawrence7 groups together all radiographic components of osteoarthritis. In the current study, each preoperative radiograph was reviewed retrospectively and graded by one author (ARA) to limit intraobserver error. These data were reviewed but not changed by one of the other authors (JBM). Thus, although radiographs were reviewed by an experienced orthopaedic surgeon, the radiographic data used were obtained by one observer (ARA).
All other data, including the Harris hip score5 and the associated pain score, were collected prospectively by the surgeon or surgical assistant. These scores were obtained at each followup and were collected uniformly for all patients. Statistical analysis, including Wilcoxon and t tests (categorical and continuous data, respectively), was performed with the aid of the SAS software package (SAS Institute, Cary, NC). A multivariate linear model at each followup was used to assess the relationship of the preoperative radiograph evaluation to the hip and pain score after controlling variables of age and gender. Thus, age and gender were stratified to minimize any potential differences in one type of arthritis that may be more common in one demographic group than another.
Radiographic analysis of all hips, including degree of cartilage space loss, osteophyte formation, and direction of cartilage space loss, is included in Table 2. All hips showed some degree of cartilage space loss, and osteophyte formation was common. In addition, 70% of hips observed had mainly superior cartilage space loss, as opposed to concentric or medial cartilage space loss. Table 3 lists the preoperative and postoperative hip and pain scores at each followup as a function of cartilage space loss, osteophyte formation, and direction of cartilage space loss (migration). A statistically significant correlation was observed between the degree of cartilage space loss and lower preoperative Harris hip score (p = 0.0001). No such association was observed between the degree of cartilage space loss and preoperative pain score. In addition, preoperative hip and pain scores were independent of the degree of osteophyte formation and direction of femoral head cartilage space loss as observed on the preoperative radiograph.
Patients with a greater degree of cartilage space loss were observed to have statistically less hip pain at 6 months (p = 0.0016) and 1 year (p = 0.0028) after surgery. However, at 3 years or greater, no such association was observed (Wilcoxon test). In addition, patients who had mainly superior cartilage space loss (femoral head migration) before surgery had statistically less pain at 6 months than did patients with global or medial hip cartilage space. However, at 1 year or greater the direction of cartilage space loss was independent of pain scores. In addition, no association between postoperative pain and the degree of osteophyte formation was observed. Finally, throughout the entire postoperative followup (as long as 7 years) no association (t test) was observed between Harris hip scores and the degree of osteoarthritis (the amount of cartilage space loss, the direction of cartilage space loss, or osteophyte formation).
There is little agreement among authors regarding the grading of osteoarthritis based on plain radiographs. Some authors have graded osteoarthritis but did not separate any of its radiographic components.4,7,15 Other methods have included some but not all of these radiographic characteristics.8,10
Interobserver and intraobserver error has been shown to be crucial in data compilation, specifically radiographic reading.6,7 Intraobserver error was not specifically evaluated in the current study. In addition, although using one observer certainly limits interobserver error, it also may introduce systematic bias. Kellgren and Lawrence7 categorized radiographic osteoarthritis as none, doubtful, minimal, moderate, or severe. Although their radiographic assessment was somewhat simplistic, they observed an interobserver correlation coefficient of r = 0.40 when applied to hip osteoarthritis. A lower intraobserver error rate was observed (r = 0.75). However, the design of the study did not specifically address the question of accuracy. Although one observer was used in the current study, multiple observers are more likely to provide validity.
In the current study, AP radiographs were used for assessment and not lateral profiles. Complete assessment of the joint before surgery should include AP and lateral radiographs because cartilage space narrowing on occasion may be seen on the lateral view, with relatively little narrowing observed on the AP view. Similarly, weightbearing may influence apparent cartilage space narrowing. Antero-posterior hip radiographs of patients in the weightbearing position are not obtained routinely before surgery at the authors' institution.
Patients with greater degrees of cartilage space loss had lower preoperative hip scores. In addition, more severe degrees of hip cartilage space loss correlated with more pain relief at 6 months and 1 year. However, no correlation was observed at 3 years or greater. Although, this finding is consistent with the thought that most orthopaedic surgeons may have concerning the reliability of pain relief in patients undergoing total hip replacement with less severe degree of osteoarthritis, this is the first report, to the authors' knowledge, to document this observation statistically. However, unlike the degree of pain relief, no association was observed between the degree of osteoarthritis and postoperative hip scores throughout the followup intervals included the current study.
1. Bissacotti JF, Ritter MA, Faris PM, Keating EM, Cates HE: A new radiographic evaluation of primary osteoarthritis. Orthopedics 17:927-930, 1994.
2. Duggleby W, Lander J: Cognitive status and postoperative pain: Older adults. J Pain Symptom Manage 9:19-27, 1994.
3. Felson DT: Epidemiology of hip and knee osteoarthritis. Epidemiol Rev 10:1-28, 1994.
4. Fox KM, Hochberg MC, Resnik CS, et al: Severity of radiographic findings in hip osteoarthritis associated with total hip arthroplasty. J Rheumatol 23:693-696, 1996.
5. Harris WH: Traumatic arthritis of the hip dislocation and acetabular fractures: Treatment by mold arthroplasty. J Bone Joint Surg 51A:737-755, 1969.
6. Johanson NA, Charlson ME, Szatrowski TP, Ranawat CS: A self-administered hip-rating questionnaire for the assessment of outcome after total hip replacement. J Bone Joint Surg 74A:587-597, 1992.
7. Kellgren JH, Lawrence JS: Radiological assessment of osteo-arthrosis. Ann Rheum Dis 16:494-501, 1957.
8. Kobayashi S, Takoaka K, Saito N, Hisa K: Factors affecting aseptic failure of fixation after primary Charnley total hip arthroplasty. J Bone Joint Surg 79A:1618-1627, 1997.
9. Kwentus JA, Harkins SW, Lignon N, Silverman JJ: Current concepts of geriatric pain and its treatment. Geriatrics 40:48-57, 1985.
10. Lane NE, Nevitt MC, Genant HK, Hochberg MC: Reliability of new indices of radiographic osteoarthritis of the hand and hip and lumbar disc degeneration. J Rheumatol 20:1911-1918, 1993.
11. Levy RN, Levy CM, Snyder J, Digiovanni J: Outcome and long-term results following total hip replacement in elderly patients. Clin Orthop 316:25-30, 1995.
12. Mancuso CA, Ranawat CS, Esdaile JM, Johanson NA, Charlson ME: Indication for total hip and total knee arthroplasties. J Arthroplasty 11:34-46, 1996.
13. Wilson MG, Michet CJ, Ilstrup DM, Melton LJ: Idiopathic symptomatic osteoarthritis of the hip and knee: A population-based incidence study. Mayo Clin Proc 65:1214-1221, 1990.
14. Wright JG, Rudicel S, Feinstein AR: Ask patients what they want. J Bone Joint Surg 76B:229-234, 1993.
© 2000 Lippincott Williams & Wilkins, Inc.
15. Wroblewski BM, Charnley J: Radiographic morphology of the osteoarthritic hip. J Bone Joint Surg 64B:568-570, 1982.