S. Jacobsen and S. Sonne-Holm

Transcription

1 Rheumatology 2005;44: Advance Access publication 12 October 2004 Hip dysplasia: a significant risk factor for the development of hip osteoarthritis. A cross-sectional survey S. Jacobsen and S. Sonne-Holm doi: /rheumtology/keh436 Objectives. The aim of this cross-sectional survey of 2232 women and 1336 men (age range yr) was to investigate individual risk factors for hip joint osteoarthritis (OA). Methods. Standardized, weight-bearing pelvic radiographs were evaluated. Radiological hip joint OA was defined as minimum joint space width (JSW) 2.0 mm. Hip dysplasia was evaluated according to common radiographic indices. Radiographic findings were correlated to general health and lifestyle information obtained at baseline examinations and questionnaires. The study focused on age; self-reported hip pain, occupational exposure to repeated daily lifting, body mass index, smoking and hip dysplasia. Results. Hip dysplasia (HD) prevalence ranged from % depending on the radiographic index applied. Hip OA prevalence was % in subjects <60 yr of age and % in subjects 60 yr of age. Of factors entered into logistic regression analyses, only age (P<0.001 for right hips and P<0.001 for left hips) and hip dysplasia (P<0.001 for right hips and P ¼ for left hips) were significantly associated with hip OA prevalence in women. In men, only hip dysplasia was associated with hip OA prevalence, P<0.001 in right hips and P ¼ in left hips. Conclusions. Of the individual risk factors investigated in this study, only age and hip dysplasia were associated with the development of hip osteoarthritis. KEY WORDS: Hip, Hip dysplasia, Osteoarthritis, Epidemiology of osteoarthritis. Systemic and biomechanical factors interact in the heterogeneous pathogenesis of osteoarthritis (OA) [1]. It is generally acknowledged that reduced load-transferring areas in dysplastic hips and accompanying soft tissue injuries may lead to premature degeneration [2 4]. However, the importance of hip dysplasia has not, to our knowledge, been investigated in cross-sectional surveys relative to other potential risk factors for hip OA. We have investigated the influence of individual risk factors on the prevalence of hip OA in this cross-sectional survey of 4151 participants of the Third Copenhagen City Heart Study: The Osteoarthritis Substudy Cohort. The survey focused on hip dysplasia, body mass index, smoking and occupational exposure to repeated daily lifting. Material and methods The Copenhagen City Heart Study: The Osteoarthritis Substudy (CCHS III) The CCHS is a longitudinal survey of an adult, Caucasian cohort selected from the county of O sterbro in Copenhagen using a random social security number algorithm. The survey has registered lifestyle factors and general health since the beginning of 1976 [5]. From 1991 to 1994 antero-posterior (AP) pelvic and lateral lumbar spine radiographs were recorded in 1533 men with an average age of 65 yr (range yr) and 2618 women with an average age of 65 yr (range yr). Radiographs were obtained standing: with the feet pointed straight forward, the lower extremities were positioned in neutral abduction adduction along the functional axis of the lower extremity. In AP pelvic radiographs the X-ray beam was centred two finger breadths over the symphysis pubis in the vertical midline. The X-ray beam in lateral lumbar spine radiographs was centred at the apical midpoint of the iliac crest. The tube to film distance was 120 cm in all cases. Two radiography technicians obtained all the radiographs. Radiographic parameters of hip OA Minimum joint space width (JSW) was measured at three locations: (1) at the lateral margin of the subchondral sclerotic line ( the sourcil ), (2) at the apical transection of the weightbearing surface by a vertical line through the centre of the femoral head and (3) at the medial margin of the weight-bearing surface bordering on the fovea, or as a fourth measurement if minimum JSW was found outside the three standard locations of measurements (Fig. 1). Minimum JSW was selected as the smallest of these three measurements. One observer (SJ) performed all measurements using a 0.1 mm graded magnifying glass (Peak, Japan). Radiographic definition of hip OA The authors used the minimum JSW value of 2.0 mm as their sole radiographic OA discriminator, regardless of formations of osteophytes, subchondral sclerosis or subchondral cysts, according to the recommendations of Lanyon et al. [6]. This cut-off value equalized average minimum JSW 2 standard deviations (S.D.). Department of Orthopaedic Surgery, Hvidovre University Hospital of Copenhagen, Denmark. Submitted 6 April 2004; revised version accepted 10 September Correspondence to: S. Jacobsen, Baneledet 17, DK-2830 Virum, Denmark. sjac@dadlnet.dk 211 Rheumatology Vol. 44 No. 2 ß British Society for Rheumatology 2004; all rights reserved

2 212 S. Jacobsen and S. Sonne-Holm FIG. 1. Measurements of hip joint space widths. Radiographic parameters of hip dysplasia Acetabular morphology was assessed by the acetabular depth ratio (ADR) as defined by Stulberg and Harris [3] and Cooperman et al. [7] (cut-off value for definite dysplasia set at 250ø) [3, 7]. Containment of the femoral head was assessed by Wiberg s [2] centre-edge (CE) angle (pathological cut-off value set at 20 ) and Heyman and Herndon s [8] femoral head extrusion index (FHEI) (cut-off value set at 25%) (Fig. 2a c). The cut-off values were defined in accordance with the authors recommendations. The lateral margin of the sourcil was used to designate the lateral acetabular rim in relevant measurements. Shenton s line was evaluated. Pelvic rotation during X-ray recording was assessed using To nnis foramen obturator index (FOI), in which maximum horizontal width of the right obturator foramen was divided by left obturator foramen width [9]. All measurements of the radiological parameters of hip dysplasia were performed by a single reader (SJ). Physical and occupational parameters At the CCHS III examinations of the weight and height of all participants was measured, and body mass index (BMI ¼ weight/height 2 in kg/m 2 ) was calculated. The CCHS III questionnaire recorded the nature and duration of occupations since leaving school. For each occupation reported, the CCHS III have registered different levels of repeated lifting during a typical working day. The questions concerning workloads were

3 Hip dysplasia and hip OA 213 (a) (b) (c) FIG. 2. (a) The CE angle of Wiberg [2] designates the angle between a line perpendicular to the horizontal teardrop line through the femoral head centre and a line from the femoral head centre to the lateral edge of the acetabular weight-bearing surface. (b) The acetabular depth ratio (ADR) [2, 3]. A line is drawn from the teardrop shadow designating the medial and inferior margin of the acetabular cavity to the lateral edge of the acetabular weight-bearing surface (acetabular width). A second line perpendicular to the first is drawn to intersect the medial corner of the superior weight-bearing surface of acetabulum. The length of this line is the acetabular depth. (c) The femoral head extrusion index (FHEI) [8], expresses the percentage of the femoral head diameter uncovered by the acetabulum, or the percentage of the femoral head lateral for Perkin s line. formulated along the guidelines of The Danish National Board of Industrial Injuries, using the following categories: (1) primarily seated occupation, (2) standing, walking occupation, no repeated lifting, (3) daily repeated lifting equivalent to kg or kg, (4) repeated daily lifting equivalent to (50 100) 20 kg, or (20 50) 50 kg, (5) repeated daily lifting equivalent to ( ) 20 kg or (50 100) 50 kg, and (6) repeated daily lifting equivalent to ( ) 20 kg or ( ) 50 kg. Assessment of hip pain At the baseline examination subjects were asked the following questions: (1) Have you experienced recurrent hip pain during the last 12 months? and (2) Have you experienced frequent and recurrent deep pain in the groin during the last 12 months? In a prior study of the relationship between self-reported pain in or around the hip joint and joint space narrowing, the authors found good correlation between hip pain and groin pain and JSW narrowing. In an earlier study we found a significant relationship between self-reported hip pain and radiological hip OA [10]. No statement on laterality or bilaterality was obtained. Smoking The CCHS III questionnaire recorded the smoking habits of the participants. We have chosen the following questions from the CCHS III questionnaire regarding smoking: (1) How many years of smoking? and (2) How many cigarettes a day? In this study we decided to omit cigar or pipe smokers because they constituted an insignificant proportion of smokers. As the numerical unit we have used package-years, i.e. the number of years smoking one pack of cigarettes (20 cigarettes) a day. Exclusion criteria for radiographs In a cadaver study we have found that extreme pelvic rotation and inclination/reclination had significant influence on the radiographic morphometry of hip dysplasia and measurements of minimum JSW [11]. To stay inside 3 of error of measurements of the CE angle, FOI inclusion limits of were applied to radiographs. Furthermore, radiographs with pelvic inclinations outside 2 S.D. of the mean were omitted from the study. Median pelvic inclination was 38 (0 82 ), and 1 S.D. was 9.4, according to lateral lumbar spine radiographs. Inclusion limits of pelvic inclination thus ranged from 19 to 56. Furthermore, radiographs in which measurements were inaccurate due to obesity were omitted. Exclusion criteria for subjects The following exclusion criteria were applied: (1) former hip surgery of any hip, (2) former fractures of any hip, (3) former treatment of childhood hip disorders and (4) a history of inflammatory arthritis of any joint. Reproducibility Intra-observer reproducibility of measurements of JSW and parameters of dysplasia was assessed by blinded rereading of a subset of 50 radiographs 4 weeks after the first reading (SJ), using intra-class coefficients. Statistical analysis The chosen critical limit of minimum JSW at 2.0 mm designating definite hip OA was tested for its ability to explain self-reported hip and groin pain by 2 analyses. The association between hip dysplasia and self-reported pain was assessed by 2 analyses. Continuous variables such as age, BMI, package-years of smoking and years of exposure to different levels of repeated daily lifting were tested against significantly reduced minimum JSW (2.0 mm) by multiple regression analyses. In simple calculations a significance level of P<0.05 was employed. In multiple regression analyses, the significance level was adjusted to P<0.005 (Bonferroni s adjustment). Sex-related differences in radiological or physical parameters were investigated by independent samples

4 214 S. Jacobsen and S. Sonne-Holm t-tests and 2 tests with odds ratios. All statistical analyses were performed with the SPSS 12.0 statistical software (SPSS Inc., Chicago, IL). At the initialization of the survey general acceptance was made by the Ethical Committee of Copenhagen for the conduction of the work. Informed patient consent was obtained from each participant in the study. Results Inclusion Applying FOI limits of and limits of pelvic inclination inside 2 S.D. of the mean meant that 314 of the CCHS III radiographs had to be excluded from further analysis (7.6%). Applying further individual exclusion criteria to subjects resulted in the final inclusion of 1336 men and 2232 women (age range yr). Background data and sex-related differences for the included subjects are presented in Table 1. Reproducibility of radiographic measurements Intra-reader repeatability of measurements of minimum JSW was acceptable: intra-class correlation coefficients of minimum JSW were r ¼ 0.91 for right hips and r ¼ 0.87 for left hips. Intra-class coefficients of right hip CE angle measurements were r ¼ 0.85 for right hips and r ¼ 0.82 for left hips. Intra-class coefficients for right hip ADR were r ¼ 0.82 and r ¼ 0.80 for left hip ADR. Intra-class coefficients for right hip FHEI were r ¼ 0.89 and r ¼ 0.90 for left hip FHEI. Prevalence of hip OA and hip dysplasia Prevalence of hip OA and hip dysplasia according to pre-defined criteria are summarized in Table 2. Mean minimum JSW was 3.65 mm (S.D. ¼ 0.91 mm) in men and 3.56 mm (S.D. ¼ 0.81 mm) in women. The difference in minimum JSW was statistically significant (P ¼ 0.004). The mean CE angle was 34.6 (S.D. ¼ 7.7 ) in both sexes. There were no cases of hip subluxation in the material (broken Shenton s line >5 mm). TABLE 1. Background data (1336 men/2232 women) Number Mean Percentage Sex-related differences Parameter M W M W M W P OR (95% CI) Age BMI (kg/m 2 ) <0.001 BMI<25 kg/m < ( ) BMI 25 and <30 kg/m < ( ) BMI 30 kg/m ( ) Past or actual smoking < ( ) Package-years a <0.001 Seated occupation < ( ) Years of seated occupation Standing, walking occupation < ( ) Years of standing occupation Occupation with some lifting b < ( ) Years with some lifting Occupation with moderate lifting < ( ) Years with moderate lifting Occupation with heavy lifting < ( ) Years with heavy lifting 1.9 More than one category of occupation ( ) Complaining of hip pain < ( ) Complaining of groin pain ( ) Either hip or groin pain < ( ) Hip and groin pain ( ) a Package-years ¼ years of smoking 1 package or more of cigarettes a day (20 cigarettes). b Occupations according to the guidelines of The Danish National Board of Industrial Injuries: (1) primarily seated occupation, no repeated lifting; (2) standing, walking occupation, no repeated lifting; (3) daily repeated lifting equivalent to kg or kg; (4) daily repeated lifting equivalent to (50 100) 20 kg or (20 50) 50 kg. TABLE 2. Prevalence of hip osteoarthritis and hip dysplasia (for 1430 subjects <60 yr of age: 574 men/856 women, and 2138 subjects 60 yr of age: 779 men/1359 women) Right hip (M/W) Left hip (M/W) Either hip (M/W) Bilateral (M/W) Parameter No % No % No % No % JSW 2.0 mm (<60 yr) 14/9 2.5/1.0 12/11 2.1/1.3 22/17 3.9/2.0 4/3 0.7/0.3 JSW 2.0 mm (60 yr) 34/69 4.4/5.0 41/68 5.3/5.0 58/ /7.9 17/29 2.2/2.1 CE angle 20 48/77 3.6/3.4 42/75 3.1/3.4 73/ /5.4 17/32 1.5/1.4 ADR 250ø 146/ / / / / / /74 5.2/3.3 FHEI 25% 50/ /5.2 35/97 2.6/4.3 71/ /7.7 14/42 1.0/1.9 JSW, minimum joint space width; CE, Wiberg s centre edge (angle); ADR, Murray and Stulberg s acetabular depth ratio; FHEI, Heyman and Herndon s femoral head extrusion index; M, men; W, women.

5 Hip dysplasia and hip OA 215 TABLE 3. The relationship between radiological hip OA (minimum JSW 2.0 mm) and self-reported hip and groin pain (men n ¼ 1336/women n ¼ 2232) Men Women Hip pain Groin pain Hip pain Groin pain Radiological OA P OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) Right hip JSW 2.0 mm/left hip JSW>2.0 mm ( ) ( ) ( ) ( ) Right hip JSW mm/left hip JSW<2.0 mm < ( ) ( ) ( ) ( ) Right hip JSW 2.0 mm/left hip JSW 2.0 mm ( ) ( ) ( ) ( ) Right hip JSW>2.0 mm/left hip JSW>2.0 mm ( ) ( ) ( ) ( ) JSW, minimum joint space width; OR, odds ratio; 95% CI, 95% confidence interval. TABLE 4. The relationship between hip dysplasia (CE angle 20 ) and self-reported hip and groin pain (men n ¼ 1336/women n ¼ 2232) Men Women Hip pain Groin pain Hip pain Groin pain Hip dysplasia P OR (95%CI) P OR (95% CI) P OR (95%CI) P OR (95% CI) Right hip CE 20 /left hip CE> ( ) ( ) ( ) ( ) Right hip CE>20 /left hip CE ( ) ( ) ( ) ( ) Right hip CE 20 /left hip CE ( ) ( ) ( ) ( ) Right hip CE>20 /left hip CE> ( ) ( ) ( ) ( ) CE, centre edge angle; OR, odds ratio; 95% CI, 95% confidence interval. Pain and OA Of the study subjects 670 women (30%) and 308 men (23.1%) complained of recurrent and deep hip pain during the last 12 months prior to index examination. The sex-related difference was statistically significant (P<0.001). Three hundred and thirty-seven women (14.2%) and 176 men (13.2%) complained of deep and recurrent groin pain. This difference was not significant. The associations between self-reported pain and hip OA is shown in Table 3. Since information on uni- or bilaterality of self-reported pain was unfortunately not registered at the index examination we have investigated the relationship between pain and JSW assuming a minimum JSW 2.0 mm, and JSW 2.0 mm separately for each side and in combination in both sexes. We found that pathologically reduced JSW (2.0 mm) was associated with self-reported hip and groin pain. Pain and dysplasia The association between self-reported pain and hip dysplasia according to Wiberg s CE angle is presented in Table 4. Since no information of uni- or bilaterality of self-reported pain was obtained at the index examination, we have investigated the relationship between pain and hip dysplasia (CE angle 2.0 mm) for the right and left sides separately and in combination in both sexes. Self-reported hip and groin pain were not associated significantly with hip dysplasia. Age and minimum JSW The relationship between age and mean minimum JSW is represented in Fig. 3. A steady decrease in mean minimum JSW was observed after the fifth decade of life in women, while minimum JSW was relatively unaltered in men throughout life. Logistic regression analysis The continuous variables of age, BMI, parameters of dysplasia, different levels of occupational exposure to daily repeated lifting and package-years of smoking were entered into regression analyses. Ten different variables were entered, and the significance level was adjusted accordingly to P<0.005 (Bonferroni s adjustment: 0.05/10). Hip OA was significantly influenced by age in women (P<0.0001). OR and 95% CI for right-side hips were 1.09 ( ). OR and 95% CI for left-side hips were 1.08 ( ). Hip OA was significantly influenced by Wiberg s CE angle in women: P ¼ in right hips and P ¼ in left hips. OR and 95% CI for right-side hips were 1.09 ( ) and 1.07 ( ) for left-side hips. With regard to men, of the entered potential risk factors only the CE angle was statistically associated with hip OA defined as minimum JSW 2.0 mm: P<0.001 [OR ¼ 1.1 (95% CI )] in right hips and P ¼ [OR ¼ 1.6 (95% CI )] in left hips. Age was not associated with hip OA in men after Bonferroni s adjustment: P ¼ in right hips and P ¼ 0.01 in left hips. In this cross-sectional analysis we found neither positive nor adverse effects of occupational exposure to repeated lifting, BMI or smoking on the prevalence of hip OA in either sex. Discussion In this survey of 1336 men and 2232 women of all age groups we focused upon age, hip dysplasia, BMI, smoking and occupational exposure to repeated daily lifting as potential risk factors for radiological hip OA. A minimum JSW 2.0 mm was our single radiographic OA discriminator. Recorded variables were entered into logistic regression analyses. We found that age and hip dysplasia were significantly associated with radiological hip OA in women, and only hip dysplasia was associated with hip OA in men. Epidemiological studies have documented that ageing is the cardinal risk factor for incident OA. However, while the incidence of OA before the age of 50 is lower among women than among men, it increases progressively in women after the menopause, and remains relatively unaltered among men after the fifth decade of life [12]. This suggests that a decrease in gonadal steroids in postmenopausal women plays a role in the development of OA [13]. In asymptomatic subjects Lanyon et al. [6] found that minimum hip JSW progressively decreased in post-menopausal women, while it remained relatively unaltered in men throughout life.

6 216 S. Jacobsen and S. Sonne-Holm 4,2 4,1 Minimum Joint Space Width (mm) 4,0 3,9 3,8 3,7 3,6 3,5 Male right hip minimum JSW Male left hip minimum JSW Female right hip minimum JSW Female left hip minimum JSW 3, FIG. 3. Relationship between age and minimum joint space width. Age groups We also found a progressive decrease in mean minimum hip JSW after the fifth decade of life in female study subjects as opposed to male study objects where minimum JSW remained relatively unaltered throughout life. In this survey we had no opportunity to investigate the influence of oral contraception or post-menopausal hormone replacement therapy on the prevalence of radiological hip OA among women, which is a noteworthy limitation of the study and the results presented. Since Wiberg s 1939 thesis on the subject [2], hip dysplasia (HD) has been assumed to be a pre-osteoarthritic condition leading to premature radiological hip OA. In vivo and in vitro studies have documented that the reduction in load-transferring areas in dysplastic hips may lead to degeneration. While premature hip osteoarthritis (OA) invariably seems to develop in subluxated hips, the extent and rate at which degeneration develops in moderately dysplastic hips is not defined. Only few longitudinal, and no casecontrolled, studies exist [7, 14 16]. The cross-sectional epidemiology on the subject is contradictory; lacking consensus definitions of dysplastic morphology and radiological OA, and qualified estimates of the background prevalence of HD. Several studies are based on pelvic radiographs obtained for other purposes, i.e. urograms or colon radiographs with little or no information on recording techniques [17 20]. Hip OA grading has mostly been performed by using the composite radiological classifications of Kellgren and Lawrence or the revision by Croft [21, 22]. Both classifications seem to have limited application to the hip joint. Kellgren and Lawrence s classification has the implicit notion of a chronological sequence of degeneration attached: joint space narrowing leads to osteophytes, which lead to subchondral sclerosis, and so forth. This sequence has not been evidenced in the literature, to our knowledge. The OA classification of Croft does not have a certain chronology attached, which is a distinct advantage. However, Croft s classification is based solely on male urograms, thereby emphasizing formations of osteophytes and subchondral sclerosis, which is more frequently encountered in male hips, rather than joint space narrowing, which is progressively more marked in female hips after the fifth decade of life [23]. Applying Croft s radiological discriminator, one runs the risk of inflating male hip OA prevalence, and underestimating female hip OA prevalence. The terminology regarding joint space narrowing of both classifications is inaccurate, using terms such as marked or gross and pre-supposes a knowledge of the normal distribution of hip joint space width in asymptomatic subjects. This knowledge has just recently been forthcoming [6]. Finally, inter- and intrarater reproducibility of composite readings seems to be inferior to repeated readings of individual features of OA, notably minimum JSW [24 27]. In this study we found that the prevalence of mild to moderate hip dysplasia was higher than expected, ranging from 5.4 to 12.6% depending on the radiographic index applied, and that the presence of hip dysplasia did in fact influence hip OA prevalence significantly. In a recent study, Birrell et al. [28] reported a prevalence of acetabular dysplasia of 30% in 195 patients presenting with new episodes of hip pain at their GP, but with no significant relationship between dysplasia and radiological OA. The authors termed the condition: the syndrome of symptomatic adult acetabular dysplasia or SAAD. In the present study we found a significant association between self-reported hip and groin pain and radiological hip OA (as defined by minimum JSW 2.0 mm) in both sexes. Although unilaterality or bilaterality of self-reported hip pain was unfortunately not registered at the baseline examination, which would have been much preferred, we have investigated the association between pain and reduced JSW in various combinations of unilaterality and bilaterality of healthy and degenerated hips. Unlike Birrell et al. we did not find consistent associations between hip dysplasia and self-reported hip or groin pain in either sex. It is notoriously difficult to correlate self-reported sensations of pain in or around the hip joint with actual radiological evidence of degeneration, no matter how advanced the questionnaires might be. In this study no fewer than 30% of women and 23.1% of men complained of hip pain, and 14% of women and 13% of men complained of groin pain, while actual radiological hip OA prevalence is 5 7% in this and other epidemiological studies. In a recent longitudinal case-controlled study we followed 81 subjects (27 men/54 women) with mild to moderate hip dysplasia but without radiological hip OA at admission (CE 6 20 ) and 136 control subjects with normal hips for 10 yr to monitor rates of JSW narrowing. We found no differences in self-reported hip, gluteal and groin pain between controls and subjects with dysplasia [29].

7 Hip dysplasia and hip OA 217 An association between being overweight and hip OA has not been thought to be as important as in knee OA. However, some studies suggest a significant relationship between overweight and symptomatic hip OA. Vingård [30] found significantly increased odds ratios for development of end-stage hip OA among 239 men, if BMI>mean BMI þ 1 S.D. Adjusted odds ratios varied between 1.67 (95% CI ) and 2.49 (95% CI ). Oliveira et al. [31] found significant correlations between incident, symptomatic hip OA and overweight in 134 matched case control pairs of women aged yr. The authors calculated an odds ratio of 3.4 (95% CI ) for women having a BMI between and 27.8 kg/m 2. Marks and Allegrante [32] found BMIs in the overweight and obese range in 70% of 586 females and 435 males requiring total hip replacements for end-stage hip OA. The authors found that the percentage of overweight or obese subjects with end-stage hip OA was higher than the values reported in the adult population generally, but do not state whether the difference was statistically significant. The study design did not permit conclusions about whether overweight antedated hip OA [32]. Cooper et al. [33] reported a definitive positive relationship of increasing BMI to hip OA in a case control study of 611 patients listed for total hip replacement compared with an equal number of age- and sexmatched subjects. The correlation was positive regardless of gender. Odds ratios were adjusted for individual risk covariates and were 1.9 (95% CI ) for men with BMI 28 kg/m 2 and 1.7 (95% CI ) for women with BMI 28 kg/m 2 [33]. In this study, we found no significant influence on hip OA prevalence by increasing BMI. The possible causal relationship between heavy physical workloads over prolonged periods of time and the development of hip OA has been the focus of many studies, but women are virtually absent in the studies. In one of the best-executed studies of occupational lifting and hip OA, Coggon et al. [34] found no association among women, but an association in men. In this study we did not find that occupational exposure to varying levels of repeated daily lifting influenced radiological hip OA prevalence in either sex. The Framingham Osteoarthritis Study has demonstrated a modest but significant inverse relationship of cigarette smoking to knee OA [35]. The model has not been evaluated in regard to hip OA in any major surveys, to our knowledge. We found no positive or adverse effects of cigarette smoking on the prevalence of hip OA in women. Of the well-known risk factors for the development of hip OA, this study only documented significant relationships between age and hip dysplasia and radiological hip OA in 1336 men and 2232 women of the Copenhagen City Osteoarthritis Substudy. Rheumatology Key messages Of the individual risk factors associated with the development of OA in general, only age and residual hip dysplasia were associated with the prevalence of hip osteoarthritis. The study was based on epidemiological information and standardized, weightbearing pelvic radiographs of 1332 men and 2232 women. Acknowledgements This study received financial support from the Research Board of the University Hospital of Rigshospitalet, the Danish Medical Research Council, the Danish Rheumatism Association, the SAHVA Foundation and Sygekassernes Helsefond. The authors have declared no conflicts of interest. References 1. Jordan JM, Kington RS, Lane NE, Nevitt MC, Zhang Y, Sowers M. Systemic risk factors for osteoarthritis. Ann Intern Med 2000;133: Wiberg G. Studies on dysplastic acetabulae and congenital subluxation of the hip joint. Acta Orthop Scand (Suppl) 1939;58: Stulberg SD, Harris WH. Acetabular dysplasia and development of osteoarthritis of hip. In: Harris WH (ed.). The Hip. Proceedings of the Second Open Scientific Meeting of the Hip Society. St Louis, MO: Mosby, 1974: Stulberg SD. Unrecognized childhood hip disease: a major cause of idiopathic osteoarthritis of the hip. In: Cordell LD, Harris WH, Ramsey PL, MacEwen GD (eds). Proceedings of the Third Open Scientific Meeting of the Hip Society. St Louis, MO: Mosby, 1975: Schnohr P, Jensen G, Lange P, Scharling H, Appleyard M. The Copenhagen city heart study Österbroundersøgelsen tables with data from the third examination Eur Heart J 2001(Suppl. H): Lanyon P, Muir K, Doherty S, Doherty M. Age and sex differences in hip joint space among asymptomatic subjects without structural change: implications for epidemiologic studies. Arthritis Rheum 2003;48: Cooperman DR, Wallensten R, Stulberg SD. Acetabular dysplasia in the adult. Clin Orthop 1983;175: Heyman CH, Herndon CH. Legg Perthes disease. A method for the measurement of the roentgenographic result. J Bone Joint Surg (Am) 1950;32-A: Tonnis D. Normal values of the hip joint for the evaluation of X-rays in children and adults. Clin Orthop 1976;119: Jacobsen S, Sonne-Holm S, Søballe K, Gebuhr P, Lund B. The relationship of hip joint space to self reported hip pain. Osteoarthritis Cartilage 2004;12: Jacobsen S, Sonne-Holm S, Lund B et al. Pelvic orientation and assessment of hip dysplasia in adults. Acta Orthop Scand 2004; in press. 12. Oliveria SA, Felson DT, Reed JI, Cirillo PA, Walker AM. Incidence of symptomatic hand, hip, and knee osteoarthritis among patients in a health maintenance organization. Arthritis Rheum 1995;38: Oliveria SA, Felson DT, Klein RA, Reed JI, Walker AM. Estrogen replacement therapy and the development of osteoarthritis. Epidemiology 1996;7: Afoke NYP, Byers PD, Hutton WC. Contact pressures in the human hip joint. J Bone Joint Surg (Br) 1987;69B: Bergmann G, Graichen F, Rohlmann A. Hip joint loading during walking and running, measured in two patients. J Biomechanics 1993;26: Hasegawa Y, Iwata H, Mizuno M, Genda E, Sato S, Miura T. The natural course of osteoarthritis of the hip due to subluxation or acetabular dysplasia. Arch Orthop Trauma Surg 1992;111: Ali-Gombe A, Croft PR, Silman AJ. Osteoarthritis of the hip and acetabular dysplasia in Nigerian men. J Rheumatol 1996;23: Croft P, Cooper C, Wickham C, Coggon D. Osteoarthritis of the hip and acetabular dysplasia. Ann Rheum Dis 1991;50: Lau EM, Lin F, Lam D, Silman A, Croft P. Hip osteoarthritis and dysplasia in Chinese men. Ann Rheum Dis 1995;54: Smith RW, Egger P, Coggon D, Cawley MI, Cooper C. Osteoarthritis of the hip joint and acetabular dysplasia in women. Ann Rheum Dis 1995;54: Kellgren J, Lawrence, JS Radiological assessment of osteo-arthritis. Ann Rheum Dis 1957;16: Croft P, Cooper C, Wickham C, Coggon D. Defining osteoarthritis of the hip for epidemiologic studies. Am J Epidemiol 1990;132:

8 218 S. Jacobsen and S. Sonne-Holm 23. Jacobsen S, Sonne-Holm S, Søballe K, Gebuhr P, Lund B. Radiologic case definitions and prevalences of osteoarthritis of the hip. Acta Orthop Scand 2004; in press. 24. Sun Y, Günther KP, Brenner H. Reliability of radiographic grading of osteoarthritis of the hip and knee. Scand J Rheumatol 1997;26: Auleley GR, Giraudeau B, Dougados M, Ravaud P. Radiographic assessment of hip osteoarthritis progression: impact of reading procedures for longitudinal studies. Ann Rheum Dis 2000;59: Dougados M, Gueguen A, Nguyen M et al. Radiological progression of hip osteoarthritis: definition, risk factors and correlations with clinical status. Ann Rheum Dis 1996;55: Dougados M, Gueguen A, Nguyen M et al. Radiographic features predictive of radiographic progression of hip osteoarthritis. Rev Rheum 1997;64: Birrell F, Silman A, Croft P, Cooper C, Hosie G, Macfarlane G. Syndrome of symptomatic adult acetabular dysplasia (SAAD) syndrome. Ann Rheum Dis 2003;62: Jacobsen S, Sonne-Holm S, Søballe K, Gebuhr P, Lund B. Joint space width in hip dysplasia. J Bone Joint Surg (Br) 2004; in press. 30. Vingård E. Overweight predisposes to coxarthrosis. Acta Orthop Scand 1991;62: Oliveria SA, Felson DT, Cirillo PA, Reed JI, Walker AM. Body weight, body mass index, and incident symptomatic osteoarthritis of the hand, hip, and knee. Epidemiology 1999;10: Marks R, Allegrante JP. Body mass indices in patients with disabling hip osteoarthritis. Arthritis Res 2002;4: Cooper C, Inskip H, Croft P et al. Individual risk factors for hip osteoarthritis: obesity, hip injury, and physical activity. Am J Epidemiol 1998;147: Coggon D, Kellingray S, Inskip H, Croft P, Campbell L, Cooper C. Osteoarthritis of the hip and occupational lifting. Am J Epidemiol 1998;147: Felson DT, Anderson JJ, Naimark A, Hannan MT, Kannel WB, Meenan RF. Does smoking protect against osteoarthritis? Arthritis Rheum 1989;32: Clinical Vignette doi: /rheumatology/keh336 Anti-Jo-1 antibody-positive arthropathy presenting as severe osteolysis of the distal interphalangeal joints A 66-yr-old housewife developed arthralgia of the shoulders, hands and fingers with Raynaud s phenomenon from She recognized swelling of the finger joints, dry cough and dyspnoea on exertion in Arthralgia subsequently disappeared but deformities of the DIPs remained. The patient attended a consultation at our hospital for breathlessness and joint deformities in Severe interstitial pneumonia was identified on chest radiography and CT. However, 67 Ga scintigraphy showed no uptake in the lungs. Radiography of the hands revealed joint destruction in the interphalangeal joints and DIPs. Severe marginal erosion of the sides of the middle phalanges and proximal ends of the distal phalanges was recognized. Dislocations of DIPs had resulted from these bone changes (the figure shows radiographs of the patients hands). No abnormalities were apparent in other joints. Signal uptake on 99m Tc bone scintigraphy was recognized only in the DIPs. Fat-suppression gadolinium-enhanced MRI of the hands revealed no proliferation of synovium or accumulation of joint fluid. Laboratory data on admission revealed normal biochemistry. Negative results were obtained for CRP, RF and ANA. However, anti-jo-1 antibodies were detected in serum. No treatment was prescribed, as active arthritis and interstitial pneumonia were not present. Two years later, joint symptoms and lung function have not deteriorated. We, the authors, declare that there are no conflicts of interest. M. YAMAMOTO, S.HARADA, M.OHARA, C.SUZUKI Y. NAISHIRO, H.YAMAMOTO, H.TAKAHASHI H. TAKAHASHI, K.IMAI Correspondence to: M. Yamamoto. mocha@cocoa. plala.or.jp