Journal of the American Academy of Orthopaedic Surgeons 320 Osteoarthritis (OA) of the hip is manifested as degeneration of the tissues of the hip joint, including hyaline cartilage, fibrocartilage, bone, and synovium. Hip arthritis can result from several different patterns of joint failure. Underlying pathologic changes due to condi- tions such as osteonecrosis, trauma, sepsis, Paget’s disease, and rheuma- toid arthritis can produce degenera- tion of the joint. Conditions such as developmental dysplasia of the hip (DDH) and slipped capital femoral epiphysis (SCFE) leave the patient with predisposing anatomic abnor- malities that can later result in os- teoarthritic changes. When any of these conditions can be identified (Table 1), the degenerative process is termed “secondary OA.” When neither an anatomic abnormality nor any specific disease process can be identified, the condition is called “primary OA,” which is, therefore, a diagnosis made by exclusion. Clinical and epidemiologic stud- ies indicate that OA of the hip is a distinct entity that behaves different- ly from OA in other synovial joints. 1 Patients who have undergone total hip replacement (THR) because of a Dr. Hoaglund is Professor (Emeritus) of Ortho- paedic Surgery, University of California at San Francisco Medical Center, San Francisco. Dr. Steinbach is Professor of Radiology and Ortho- paedic Surgery, University of California at San Francisco Medical Center. Reprint requests: Dr. Hoaglund, Department of Orthopaedic Surgery, UCSF Medical Center, 500 Parnassus Avenue (MU 320-W), San Francisco, CA 94143-0728. Copyright 2001 by the American Academy of Orthopaedic Surgeons. Abstract Primary osteoarthritis (OA) of the hip has a distinct etiology and epidemiology compared with other types of arthropathy in the hip joint. Arthritis of the hip can be secondary to conditions such as osteonecrosis, trauma, sepsis, or rheumatoid arthritis. Certain conditions, such as congenital hip disease and slipped capital femoral epiphysis, involve predisposing anatomic abnormalities; in such cases, the term “secondary OA” is used. When either an anatomic abnormality cannot be determined or other specific causative entities are not identified, primary OA is the diagnosis of exclusion. The prevalence of hip OA is about 3% to 6% in the Caucasian population and has not changed in the past four decades. In contrast, studies in Asian, black, and East Indian populations indicate a very low preva- lence of hip OA. Statistics on patients who underwent total hip replacement for primary OA in San Francisco and Hawaii demonstrate a virtual absence of the condition in Asians and low rates in the black and Hispanic populations. Family studies from Sweden, Britain, and the United States show increased rates of hip OA in first-degree relatives of the index patient when compared with the normal population. Occupations requiring heavy lifting, farming, and elite sports activi- ty are associated with increased rates of hip OA. The low prevalence of hip OA in Asian and black populations in their native countries; the low incidence of total joint replacement for primary OA in Asian, black, and Hispanic populations in North America; and the familial association of hip OA in Caucasians all suggest that genetic factors may be involved in the occurrence of this disease. J Am Acad Orthop Surg 2001;9:320-327 Primary Osteoarthritis of the Hip: Etiology and Epidemiology Franklin T. Hoaglund, MD, and Lynne S. Steinbach, MD diagnosis of primary OA rarely undergo total knee replacement, and vice versa. 2 In two other studies, Japanese and Hong Kong Chinese populations had virtually no pri- mary OA of the hip but a consider- able incidence of knee OA. 3,4 Epidemiology The modern search for the causes of OA started with the work of British scientists Kellgren and Lawrence, who carried out epidemiologic stud- ies encompassing several ethnic groups in a number of geographic areas, including western European and specifically British Caucasians, African and Jamaican blacks, and Native Americans. 5-7 Responses to questionnaires and the clinical findings in randomly selected pop- ulations were evaluated, and radio- graphs of various joints were ex- Franklin T. Hoaglund, MD, and Lynne S. Steinbach, MD Vol 9, No 5, September/October 2001 321 amined with the use of standard- ized grading. 5-7 Patients were con- sidered to have generalized OA if three or more joint groups were involved. Evaluation of these data was the basis for an atlas that is still used today to characterize the various degrees of OA. 5 These researchers found that the incidence of OA was not related to latitude or longitude. The highest prevalence of OA occurred in the Caucasians. It was noted that OA pref- erentially affected the distal inter- phalangeal joints and the first carpo- metacarpal joints of the hands, the knees, and the first metatarsopha- langeal joints. Comparisons between ethnic groups were not particularly helpful in delineating the etiology of OA; however, the marked ethnic and racial differences in the incidence of hip OA were important observa- tions. 6 Population Studies Lawrence and Sebo 6 studied the comparative incidence of radio- graphic OA of the hip by evaluating pelvic radiographs of European Caucasians, blacks, and Native Americans from nine geographic areas. The term “radiographic OA” was used because there was no intent to distinguish primary or sec- ondary OA from other specific causes of hip disease. Kellgren’s grading system was used to characterize the degree of hip arthritis as absent (grade 0), doubtful (grade 1), mini- mal (grade 2), moderate (grade 3), or severe (grade 4). The incidences of grade 2-4 and grade 3-4 OA were higher in the Caucasian males than in the females (Table 2). The rate of moderate and severe OA in Cau- casians was three to four times that in blacks and Native Americans. Hoaglund et al 4 also evaluated conventional radiographs of the hands, knees, and hips of 500 Hong Kong Chinese subjects. An ex- tremely low rate of OA of the hip (1%) was found, but the incidence of hand OA was similar to that in European Caucasians. A second study was done 20 years after the original study. Pelvic radiographs of 999 Hong Kong Chinese men ob- tained during intravenous pyelog- raphy demonstrated no change in the incidence of hip OA. 8 In another study, a group of 51,777 East Indians who visited an orthopaedic clinic in India were examined. 9 The incidence of OA was found to be less than 0.1%. Pelvic radiographs of American Caucasians demonstrated that the prevalence of hip OA in patients over 54 years of age was 2.7% to 3.5%, 10 which is lower than the rate seen in European Caucasians. 6-11 In a Swedish study, 12 pelvic radio- graphs obtained during double- contrast colon examinations were evaluated. The overall incidence of hip OA was 3.4% in 1964. There has been no change in this rate over the past four decades, as demonstrated in follow-up studies performed in 1984 and 1993. 13 In summary, population studies show that the rate of moderate to severe primary or secondary OA of the hip in Caucasians is 3% to 6%, compared with 1% or less in East Indians, 9 blacks, Hong Kong Chinese, and Native Americans. The prevalence has not changed in four decades, suggesting that genetic and/or environmental factors remain constant. Hip Disease in Hip Surgery Patients Specific causes of hip disease and OA in patients who underwent hip surgery in the 1940s and 1950s have been evaluated. 14-16 The rela- tive incidence of various hip dis- eases described in these series is quite different from the current incidence. During that era, 20% to 50% of surgical cases were due to Legg-Calvé-Perthes disease, DDH, or SCFE. One might have expected a higher incidence of adults pre- senting with a painful hip due to arthritic changes from these causes, because there was no prophylactic treatment for these conditions or good early surgical reconstruction at that time. There has been an ongoing de- bate about whether the high rate of Table 1 Causes of Secondary OA and Their Radiographic Appearance Condition Radiographic Features * Osteonecrosis Sclerosis, lucency, flattening of femoral head Paget’s disease Osseous enlargement, trabecular coarsening, cortical thickening Inflammatory arthropathy Joint-space narrowing, osteopenia, erosion Traumatic remodeling Distortion of osseous contour with sclerosis and remodeling Degenerative dysplasia Shallow acetabulum with increased acetabular of the hip index, subluxation, or dislocation Slipped capital femoral Medial and/or posterior displacement of epiphysis femoral head, convexity at head-neck junction, short femoral head, coxa vara, short broad femoral neck * In addition to joint-space narrowing and osteophytes. Primary Osteoarthritis of the Hip Journal of the American Academy of Orthopaedic Surgeons 322 secondary OA in that period con- tinues today. Studies of THR pa- tients in Europe and the United States have provided information about the frequency of all causes of hip arthritis in various populations. The prevalence data were compa- rable among Caucasian population groups in the United States, Scan- dinavia, and western Europe. The rate of THR varied from approxi- mately 60 per 100,000 persons in the Mayo Clinic data to 140 per 100,000 in Norway 17 (Table 3). When other specific causes of hip disease were excluded, it was noted that primary OA was the underlying disorder in nearly 90% of osteoar- thritic hips. Similar published statis- tics for Africa, South America, Hong Kong, or Japan are not available. The rate of THR for all residents of San Francisco has been mea- sured for the years 1984 through 1988 using data from the 17 hospi- tals in or near the city (Table 4). 22 Preoperative pelvic radiographs were evaluated for arthritis with- out the examiner’s knowledge of the gender or race of the patient. Total hip replacement statistics for Caucasians were compared with those for the sizable non-Caucasian populations (i.e., Asian, Hispanic, and black). Caucasians had a rate of THR of 75 per 100,000 persons, blacks had half the incidence, and Asians had only one tenth. There were markedly lower rates of THR performed because of primary OA in non-Caucasians; for example, the rate was 1.3 per 100,000 for the Chinese population, compared with 43 per 100,000 for Caucasians. The rates of THR for the other causes of hip disease, including secondary OA, were not signifi- cantly different among the various ethnic groups. This study was repeated in Ha- waii, 23 with its large Asian and Pacific Islander population (750,000 persons). In the total Hawaiian population, medical insurance is available to more than 90% of the patients, thus minimizing the effect of access to care as a factor. The re- sults of this study confirmed the markedly higher rates of THR in Caucasians, who predominantly had primary OA (Table 4). The rate of THR for secondary OA in the Caucasian population was not sig- nificantly different from the rates for the Chinese, Filipino, Hawaiian, and Japanese populations. Incidence of Predisposing Anatomic Abnormalities In 1965, Murray 24 suggested that secondary OA was more common than primary OA. In that study, ra- diographs were evaluated looking for evidence of DDH or femoral head tilt. Previous DDH was found in 37% of British women with OA, and a femoral-head tilt deformity was noted in 74% of British men with OA (Table 5). His conclusion that OA was in fact secondary 90% of the time has been supported by subsequent analysis of the data in uncontrolled series of hip surgery patients in the United States 25,26 and South Africa. 27 However, there is now considerable information from more recent surgical series that argues against such a high inci- dence of either DDH or SCFE as the cause of OA. Measurements of the center-edge (CE) angle in blacks (who are rela- tively unaffected by DDH and hip OA) are the same as those in Cau- casians. 28 (The CE angle is a mea- sure of hip socket coverage. A low angle indicates a shallow socket, which can predispose to secondary OA). In one study, acetabular depth measurements and CE angles in British patients were the same in normal and osteoarthritic hips. 29 The incidence of DDH based on CE angle measurements in Hong Kong Chinese men was not different from that in British men, 8 yet osteoar- thritic hips are much less frequent in the Chinese. Although Japanese persons have significantly smaller Table 2 Incidence of Hip OA in Male and Female Subjects in Population Studies 6 Incidence (M/F), % Grades Grades Subjects (No.) 2-4 3-4 Caucasian (1,451) 20/12 6.5/5.75 Black (503) 2.3/3 1/1.6 Native American (545) 9/8.0 2.7/1.6 Table 3 Relation Between THR and Primary OA in Various Nations 17-21 Sweden Norway Denmark Finland USA United Kingdom France Belgium Rate of THR per 100,000 persons 130 140 82 58 60 54 108 116 Incidence of primary OA, % NA 68 86 56 63 NA NA NA * NA = data not available Franklin T. Hoaglund, MD, and Lynne S. Steinbach, MD Vol 9, No 5, September/October 2001 323 CE angles than British persons, they have much less hip OA. 30 Lane et al 31 found no difference in the incidence of an abnormal CE angle between American Caucasian women with OA of the hip and those without it (4% to 5% in each group). In this population, DDH did not account for more than a few osteoarthritic hips. If primary OA of the hip is due to a subtle dysplasia, it should be seen frequently in Japanese persons, with their extremely high rate of dysplasia; however, primary OA is rare in that population. 3,32 The incidence of DDH in Caucasians is 0.5%. If OA devel- oped in all children with DDH, this would account for only one eighth of the cases of OA in the Caucasian pop- ulation. Hawaiian Japanese women have the same rate of THR for dys- plastic OA as Caucasian women, but their rate of THR performed for pri- mary OA is only one tenth of that for Caucasian women. 23 In summary, when controlled studies of acetabular measurements are done and the rates of DDH and OA for each racial subgroup are considered, DDH accounts for only a small percentage (5% to 10%) of hip OA in Caucasians. Since the original publications of Harris, 26 Murray, 24 Stulberg and Harris, 25 and Solomon, 27 which sug- gested that OA was often caused by SCFE (Table 5), there has been new information that SCFE or subclinical SCFE does not account for the high rate of primary OA. The tilt defor- mity seen in the adult with OA has been shown to be due to a remod- eling process that causes progres- sion of osteophytes, rather than to an old epiphyseal slip. 33 Goodman et al 34 identified a subtle or subclini- cal slip deformity in 8% of cadaver bone specimens from a large collec- tion of disarticulated femora and pelves. There are some similarities of this subclinical SCFE to SCFE, such as a lower incidence in women, increased left-side prevalence, and increased incidence in specimens from black subjects. However, clini- cal SCFE in the adolescent occurs in only 1 of every 800 boys (an inci- dence of only 0.125%). Whether subclinical SCFE is a forme fruste of SCFE remains to be proved. Even if it is, anatomically normal femora accounted for more than 80% of the OA in their series. Furthermore, although there is little primary OA in the black and Polynesian popula- tions, as well as a low incidence of THR performed because of OA, 22,23 these populations have two to four times the rate of SCFE in Cauca- sians. Table 4 Rates of THR for Secondary and Primary OA by Ethnic/Racial Background in Two Studies 22,23*† Chinese Japanese Filipino Caucasian Hispanic Black Hawaiian San Francisco study 22 Women 1.8/1.4 2.9/2.9 0.9/3.1 6.7/52.6 0/6.2 1.1/19.2 … Men 0.4/1.1 0/0 0/0 1.8/33.9 0.8/6.3 0/10.1 … Hawaii study 23 Women 0.31/0.84 0.17/0.24 0.06/0.25 0.29/2.35 … … 0/0.30 Men 0/0.52 0.06/0.08 0/0.25 0.11/2.32 … … 0.06/0.16 * Adapted with permission from Hoaglund FT, Oishi CS, Gialamas GG: Extreme variations in racial rates of total hip arthroplasty for primary coxarthrosis: A population-based study in San Francisco. Ann Rheum Dis 1995;54:107-110. Also adapted with permission from Oishi CS, Hoaglund FT, Gordon L, Ross PD: Total hip replacement rates are higher among Caucasians than Asians in Hawaii. Clin Orthop 1998;353:166-174. † Values are expressed as rates for secondary OA/primary OA per 100,000 population/yr. Table 5 Incidence of OA Due to DDH and Femoral-Head Tilt in Men and Women in Four Studies 24-27 Stulberg Murray 24 Solomon 27 Harris 26 and Harris 25 Women DDH 37% 69% 68% 65% Tilt 11% 4.8% 10% … Men DDH 11% 8% 15% 25% Tilt 74% 45% 66% … Primary Osteoarthritis of the Hip Journal of the American Academy of Orthopaedic Surgeons 324 Etiology Body Weight Bioengineering studies have shown that forces across the hip joint are multiplied by a leverage system involving muscles pulling against the weight of the body. For example, an individual standing on one foot has a force across the hip joint of three to four times his or her body weight. Therefore, obesi- ty could be contributory in the eti- ology of primary OA of the hip. Although OA of the knee has been shown to be consistently related to obesity, there have been conflict- ing results regarding the relation- ship of body weight to the develop- ment of OA of the hip. Saville and Dickson 35 found no difference in the average body weight of patients with primary or secondary OA of the hip compared with control sub- jects. Spector 36 and Tepper and Hochberg 37 found no relationship between body weight or body-mass index and the occurrence of hip OA. However, other studies have shown an increased risk of OA of the hip in obese individuals. 38,39 The con- flicting findings concerning body weight and hip OA may be related in part to the fact that patients gain weight as a result of the immobility imposed by a painful hip. Occupational Factors The physical demands of work have been estimated to account for 40% of cases of hip OA, 40 and carry- ing heavy loads has been associated with an increased rate of hip OA. Axmacher and Lindberg 41 found that farmers were at increased risk for hip OA in studies conducted in England, Sweden, and France—all countries that are known to have high rates of hip OA. However, the researchers pointed out that farmers are usually members of families that have been involved in farming for many generations; therefore, heredi- tary factors cannot be ruled out. Studies conducted in Sweden in- dicated an increased risk of OA in patients who had been involved in track, field, and racket sports as well as among soccer players (especially those who played professionally). 42 The findings from these studies support the hypothesis that sports may exacerbate hereditary predis- position to OA of the hip. Other Anatomic Considerations There is a known association be- tween arthritis and previous DDH in patients with the mechanical fac- tors that predispose to secondary OA, such as an acetabular abnor- mality or femoral anteversion. Therefore, investigations have also been done to determine the role of femoral anteversion alone in pa- tients with primary OA of the hip. However, small studies involving cadaver specimens and patients with hip arthropathy have shown variable results. 43-46 In cadaver specimens without OA, there was no correlation between femoral anteversion and acetabular angles. A study of Hong Kong Chinese cadaver bones showed a high inci- dence of femoral anteversion com- pared with specimens obtained from Caucasian subjects. 47 Inas- much as hip OA is rare in the Hong Kong Chinese population, this ob- servation is inconsistent with ante- version being a factor in primary OA of the hip. Femoral anteversion in osteo- arthritic and normal subjects has been measured with the use of biplanar and/or conventional (AP and cross-table lateral) radio- graphs; a mild increase in antever- sion was found in some series but not in others. 43 This increase in femoral anteversion could be due to measurement error. Using ac- curate CT techniques, Reikerås et al 46 found no difference in femoral anteversion between control sub- jects and a series of Caucasian pa- tients with OA. Family Studies In 1984, Lindberg 48 studied 289 siblings of 184 Swedish patients who had undergone THR for primary coxarthrosis. Coxarthrosis occurred in 8% of siblings. This represents a statistically significant (P<0.05) increase over the rate of 3.8% in con- trol subjects. In an unpublished study, 72 pa- tients with primary OA who un- derwent THR responded to a writ- ten family questionnaire. Analysis of the responses confirmed Lind- berg’s findings. Of the 238 first- degree relatives over the age of 50, 13% reported a THR or symptoms of hip pain consistent with OA of the hip. This is significantly higher than the 3% to 6% prevalence of OA of the hip diagnosed radiographi- cally. In black patients in San Francisco, the incidence of THR for primary OA was only 30% of that for Cau- casian patients. 22 This is the same as the incidence of Duffy or GM blood groups reported for black persons in the western part of the United States. 49 (All Europeans have Duffy blood groups, but black Africans do not have them.) Therefore, the dif- ference between the rates of THR for primary OA in Caucasians and in blacks is consistent with a genetic etiology. In a prospective cross-sectional survey of 264 British patients who underwent THR for idiopathic OA of the hip, the familial prevalence of THR was ascertained with a questionnaire distributed to their siblings. 50 By using spouses as a control group, the researchers cal- culated the relative risk of THR in siblings as 1.86 (95% CI, 0.93 to 3.60). With the use of the threshold liabil- ity model, the heritability of end- stage OA of the hip was estimated to be 27%. Twin studies in women have shown that the heritability rates of OA of the knee and hand are 49% and 65%, respectively. 51 In more Franklin T. Hoaglund, MD, and Lynne S. Steinbach, MD Vol 9, No 5, September/October 2001 325 recent twin studies of OA of the hip, women showed a 50% heritability (95% CI, 20% to 74%). 52 In summary, two different factors appear to be involved in the etiology of OA. First, the lower rates of pri- mary OA in population studies of non-Caucasians and the difference in rates of THR for primary OA in patients with various racial back- grounds living in a common envi- ronment suggest that it is a disease primarily of Caucasians. Second, the relatively high incidence of the disease in siblings of Swedish and English patients who underwent THR because of primary OA sug- gests that genetic factors are opera- tional as well. Radiographic Criteria In the studies performed by Hoaglund and co-workers in San Francisco 22 and Hawaii, 23 the radio- graphs were read without knowl- edge of the race of the patient. The rate of secondary OA in the Cau- casian population was not signifi- cantly different from the rates for Chinese, Filipino, Hawaiian, and Japanese subjects. The use of radiographic criteria has been criticized as being generally inadequate to establish that a case classified as primary OA is actually secondary OA. For example, DDH is thought to be underrepresented in women with primary OA, and SCFE or a variation of it is thought to be underrepresented in men. However, evaluation of the radio- graphs of equivalent large Japanese and Caucasian populations have demonstrated that these criteria can be used to distinguish between pri- mary and secondary OA. The cumulative rates of THR for secondary OA (measured in cases per 100,000 person-years) are similar for Japanese women (0.17) and Caucasian women (0.29), but the rate for primary OA in Japanese women was only one tenth (0.24) of that for the Caucasian women (2.35). If large numbers of cases of primary OA are in fact secondary OA, there should be sizable numbers in the Japanese women as well as in the Caucasian women. The statistics for men show the same disparity. The rate of sec- ondary OA in Japanese men is 0.6, in contrast to the rate of 0.11 for Cau- casian men. If no variant of SCFE were distinguished on the basis of radiographic criteria, comparable numbers of primary OA would be seen in both Caucasian and Japanese men. However, the rate of THR for Caucasians with primary OA is 29 times that for the Japanese men, but the rate of SCFE in Japanese men is only one half to one fifth of that in Caucasians. It seems clear, therefore, that use of radiographic criteria did not lead to underrepresentation of DDH in the female subjects. Similarly, use of those criteria did not lead to under- representation of SCFE or subclinical SCFE in male subjects. Summary Contributions to our understanding of the etiology of OA of the hip have come from international popu- lation studies, joint replacement reg- istries, epidemiologic investigations, and, more recently, analysis of fam- ily pedigrees to investigate genetic factors in this disease. Attempts to elucidate the etiology of OA of the hip depend on our ability to recog- nize and separate this condition from other causes of hip disease, as well as our ability to distinguish secondary OA from primary OA. Secondary OA of the hip occurs when a condition such as DDH, SCFE, or Legg-Calvé-Perthes disease results in a predisposing anatomic abnormality that leads to hip degen- eration due to mechanical factors. A specific diagnosis of secondary OA is dependent on radiographic diag- nosis and is more sensitive in the early stages of degeneration, before it progresses and remodeling occurs. Radiographic studies of both Cau- casian and Asian patients with and without OA suggest that a specific diagnosis can be made. Disagree- ment may arise in specific cases, but in a large series such cases should not affect the overall conclusions. It is noteworthy that studies of the Asian populations in San Fran- cisco and Hawaii showed similar prevalences of all types of hip ar- thritis in those who underwent THR but virtually no underlying primary OA. As the same radiographic cri- teria were applied to all races, the high incidence in Caucasians indi- cates that the conditions are separa- ble at least 90% of the time. The contribution of subclinical SCFE to the overall incidence of pri- mary OA is small. 34 In a study of cadaver femora and pelves, the un- derlying anatomy was normal in 83% of the specimens, and only 8% had subclinical SCFE. Interestingly, grade 2 OA developed in only 37% of the femora with subclinical SCFE; therefore, it appears that something other than the anatomic deformity is the main etiologic factor underly- ing OA. The results of comparison studies of Caucasian, black, Native Ameri- can, and Hong Kong Chinese popu- lations; the joint replacement statis- tics; and the comparative rates of THR in American non-Caucasian populations indicate that primary OA is a disease primarily of Cauca- sians of European ancestry. Inci- dence rates do not appear to have changed in the past four decades, suggesting that the etiologic factors have not changed. Although some studies suggest that factors such as obesity, high-performance athletics, strenuous occupations, and femoral neck anteversion may be contribu- tory to OA, there is also evidence to the contrary. The minimal inci- dence of THR for primary OA in American Asians compared with Primary Osteoarthritis of the Hip Journal of the American Academy of Orthopaedic Surgeons 326 Caucasians living in a similar envi- ronment suggests a genetic etiology. Apparently normal hip anatomy is present in more than 80% of pa- tients in whom OA develops, and only a small percentage of patients with subclinical SCFE are subse- quently found to have OA. There- fore, it is reasonable to postulate an underlying defect in articular carti- lage or bone that leads to the even- tual development of OA. References 1. Cushnaghan J, Dieppe P: Study of 500 patients with limb joint osteoarthritis: I. Analysis by age, sex, and distribu- tion of symptomatic joint sites. Ann Rheum Dis 1991;50:8-13. 2. Bachmeier CJ, March LM, Cross MJ, et al: A comparison of outcomes in osteo- arthritis patients undergoing total hip and knee replacement surgery. Osteo- arthritis Cartilage 2001;9:137-146. 3. Hoaglund FT, Shiba R, Newberg AH, Leung KYK: Diseases of the hip: A comparative study of Japanese Oriental and American white patients. J Bone Joint Surg Am 1985;67:1376-1383. 4. Hoaglund FT, Yau ACMC, Wong WL: Osteoarthritis of the hip and other joints in southern Chinese in Hong Kong. J Bone Joint Surg Am 1973;55: 545-557. 5. Kellgren JH, Lawrence JS: Radio- logical assessment of osteo-arthrosis. Ann Rheum Dis 1957;16:494-502. 6. Lawrence JS, Sebo M: The geography of osteoarthrosis, in Nuki G (ed): The Aetiopathogenesis of Osteoarthrosis. Kent, Ohio: Pitman Medical, 1980, pp 155-183. 7. Solomon L, Beighton P, Lawrence JS: Rheumatic disorders in South African Negro: Part II. Osteo-arthrosis. S Afr Med J 1975;49:1737-1740. 8. Lau EMC, Lin F, Lam D, Silman A, Croft P: Hip osteoarthritis and dyspla- sia in Chinese men. Ann Rheum Dis 1995;54:965-969. 9. Mukhopadhaya B, Barooah B: Osteo- arthritis of hip in Indians: An anatomi- cal and clinical study. Indian J Orthop 1967;1:55-62. 10. Lawrence RC, Hochberg MC, Kelsey JL, et al: Estimates of the prevalence of selected arthritic and musculoskele- tal diseases in the United States. J Rheumatol 1989;16:427-441. 11. van Saase JLCM, van Romunde LKJ, Cats A, Vandenbroucke JP, Valkenburg HA: Epidemiology of osteoarthritis: Zoetermeer survey—Comparison of radiological osteoarthritis in a Dutch population with that in 10 other popula- tions. Ann Rheum Dis 1989;48:271-280. 12. Danielsson LG: Incidence and progno- sis of coxarthrosis: 1964. Clin Orthop 1993;287:13-18. 13. Danielsson L, Lindberg H: Prevalence of coxarthrosis in an urban population during four decades. Clin Orthop 1997;342:106-110. 14. Adam A, Spence AJ: Intertrochanteric osteotomy for osteoarthritis of the hip: A review of fifty-eight operations. J Bone Joint Surg Br 1958;40:219-226. 15. Lloyd-Roberts GC: Osteoarthritis of the hip: A study of the clinical pathol- ogy. J Bone Joint Surg Br 1955;37:8-47. 16. Gade H: A contribution to the surgical treatment of osteoarthritis of the hip- joint. Acta Chir Scand Suppl 1947; 120:95. 17. Havelin LI, Espehaug B, Vollset SE, Engesaeter LB, Langeland N: The Norwegian arthroplasty register: A survey of 17,444 hip replacements 1987–1990. Acta Orthop Scand 1993; 64:245-251. 18. Malchau H, Herberts P, Ahnfelt L: Prognosis of total hip replacement in Sweden: Follow-up of 92,675 opera- tions performed 1978-1990. Acta Orthop Scand 1993;64:497-506. 19. Overgaard S, Knudsen HM, Hansen LN, Mossing N: Hip arthroplasty in Jut- land, Denmark: Age and sex-specific incidences of primary operations. Acta Orthop Scand 1992;63:536-538. 20. Paavolainen P, Hamalainen M, Musto- nen H, Slatis P: Registration of arthro- plasties in Finland: A nationwide prospective project. Acta Orthop Scand Suppl 1991;241:27-30. 21. Madhok R, Lewallen DG, Wallrichs SL, Ilstrup DM, Kurland RL, Melton LJ III: Trends in the utilization of prima- ry total hip arthroplasty, 1969 through 1990: A population-based study in Olmsted County, Minnesota. Mayo Clin Proc 1993;68:11-18. 22. Hoaglund FT, Oishi CS, Gialamas GG: Extreme variations in racial rates of total hip arthroplasty for primary coxarthrosis: A population-based study in San Francisco. Ann Rheum Dis 1995;54:107-110. 23. Oishi CS, Hoaglund FT, Gordon L, Ross PD: Total hip replacement rates are higher among Caucasians than Asians in Hawaii. Clin Orthop 1998; 353:166-174. 24. Murray RO: The aetiology of primary osteoarthritis of the hip. Br J Radiol 1965;38:810-824. 25. Stulberg SD, Harris WH: Acetabular dysplasia and development of osteo- arthritis of the hip, in Harris WH (ed): The Hip: Proceedings of the Second Open Scientific Meeting of the Hip Society, 1974. St Louis: CV Mosby, 1974, pp 82-93. 26. Harris WH: Etiology of osteoarthritis of the hip. Clin Orthop 1986;213:20-33. 27. Solomon L: Patterns of osteoarthritis of the hip. J Bone Joint Surg Br 1976; 58:176-183. 28. Skirving AP: The centre-edge angle of Wiberg in adult Africans and Cauca- sians: A radiographic comparison. J Bone Joint Surg Br 1981;63:567-568. 29. Croft P, Cooper C, Wickham C, Coggon D: Osteoarthritis of the hip and acetab- ular dysplasia. Ann Rheum Dis 1991; 50:308-310. 30. Yoshimura N, Campbell L, Hashimoto T, et al: Acetabular dysplasia and hip osteoarthritis in Britain and Japan. Br J Rheumatol 1998;37:1193-1197. 31. Lane NE, Nevitt MC, Cooper C, Press- man A, Gore R, Hochberg M: Acetab- ular dysplasia and osteoarthritis of the hip in elderly white women. Ann Rheum Dis 1997;56:627-630. 32. Nakamura S, Ninomiya S, Nakamura T: Primary osteoarthritis of the hip joint in Japan. Clin Orthop 1989;241: 190-196. 33. Resnick D: The ‘tilt deformity’ of the femoral head in osteoarthritis of the hip: A poor indicator of previous epiphysi- olysis. Clin Radiol 1976;27:355-363. 34. Goodman DA, Feighan JE, Smith AD, Latimer B, Buly RL, Cooperman DR: Subclinical slipped capital femoral epiphysis: Relationship to osteoarthro- sis of the hip. J Bone Joint Surg Am 1997;79:1489-1497. 35. Saville PD, Dickson J: Age and weight in osteoarthritis of the hip. Arthritis Rheum 1968;11:635-644. 36. Spector TD: The fat on the joint: Osteo- arthritis and obesity. J Rheumatol 1990; 17:283-284. 37. Tepper S, Hochberg MC: Factors asso- ciated with hip osteoarthritis (OA): Data from the National Health and Nutrition Franklin T. Hoaglund, MD, and Lynne S. Steinbach, MD Vol 9, No 5, September/October 2001 327 Examination Survey (NHANES-1) [ab- stract]. Arthritis Rheum (Suppl) 1991; 34:S33. 38. Heliövaara M, Mäkelä M, Impivaara O, Knekt P, Aromaa A, Sievers K: Asso- ciation of overweight, trauma and workload with coxarthrosis: A health survey of 7,217 persons. Acta Orthop Scand 1993;64:513-518. 39. Felson DT: Epidemiology of hip and knee osteoarthritis. Epidemiol Rev 1988;10:1-28. 40. Hochberg MC, Lethbridge-Cejku M: Epidemiologic considerations in the pri- mary prevention of osteoarthritis, in Hamerman D (ed): Osteoarthritis: Pub- lic Health Implications for an Aging Population. Baltimore: Johns Hopkins University Press, 1997, pp 169-186. 41. Axmacher B, Lindberg H: Coxarthrosis in farmers. Clin Orthop 1993;287:82-86. 42. Marti B, Knobloch M, Tschopp A, Jucker A, Howald H: Is excessive run- ning predictive of degenerative hip disease? Controlled study of former elite athletes. BMJ 1989;299:91-93. 43. Halpern AA, Tanner J, Rinsky L: Does persistent fetal femoral anteversion con- tribute to osteoarthritis? A preliminary report. Clin Orthop 1979;145:213-216. 44. Terjesen T, Benum P, Anda S, Svenning- sen S: Increased femoral anteversion and osteoarthritis of the hip joint. Acta Orthop Scand 1982;53:571-575. 45. Reikerås O, Høiseth A: Femoral neck angles in osteoarthritis of the hip. Acta Orthop Scand 1982;53:781-784. 46. Reikerås O, Bjerkreim I, Kolbenstvedt A: Anteversion of the acetabulum and femoral neck in normals and in pa- tients with osteoarthritis of the hip. Acta Orthop Scand 1983;54:18-23. 47. Hoaglund FT, Low WD: Anatomy of the femoral neck and head, with com- parative data from Caucasians and Hong Kong Chinese. Clin Orthop 1980;152:10-16. 48. Lindberg H: Prevalence of primary coxarthrosis in siblings of patients with primary coxarthrosis. Acta Orthop Scand 1986;203:273-275. 49. Reed TE: Distributions and tests of independence of seven blood group systems in a large multiracial sample from California. Am J Hum Genet 1968;20:142-150. 50. Chitnavis J, Sinsheimer JS, Clipsham K, et al: Genetic influences in end- stage osteoarthritis: Sibling risks of hip and knee replacement for idiopathic osteoarthritis. J Bone Joint Surg Br 1997;79:660-664. 51. Spector TD, Cicuttini F, Baker J, Loughlin J, Hart D: Genetic influences on osteoarthritis in women: A twin study. BMJ 1996;312:940-943. 52. MacGregor AJ, Antoniades L, Matson M, Andrew T, Spector TD: The genet- ic contribution to radiographic hip osteoarthritis in women: Results of a classic twin study. Arthritis Rheum 2000;43:2410-2416. . (especially those who played professionally). 42 The findings from these studies support the hypothesis that sports may exacerbate hereditary predis- position to OA of the hip. Other Anatomic