- Email: [email protected]
Definition and epidemiology of osteoarthritis of the hand: A review
Osteoarthritis and Cartilage (2000) 8, Supplement A, $2-$7 © 2000 OsteoArthritis Research Society International doi:l 0.1053/joca.2000.0326, available online at http://www.idealibrary.comon | II | --~[®
Osteoarthritis and Cartilage
1063-4584/00/0A00S2+06 $35.00/0
O A ~ ]
OSTEOARTHRITIS
RfSF~,RCH '~OCIFTY
Journal of the OsteoArthritis Research Society International
Definition and epidemiology of osteoarthritis of the hand: a review D. J. Hart and T. D. Spector Twin Research and Genetic Epidemiology Unit, St Thomas' Hospital, London SE1 7EH, U.K.
Summary The hand is a common site of peripheraljoint involvementin osteoarthritis(OA) and although often underestimatedas a cause of disability, the effecton quality of life from limitationsin performing activitiesof daily living such as dressing and feeding may be considerable. Hand OA may also be an important indicatorof a systemic tendency to OA which may involve weight bearing joints, notably the hips and knees.1The definition of hand OA, particularlyfor epidemiologicalstudies, has undergone reassessmentand revision over the last few years. In this paper we examinethe issues relating to this and consider the epidemiologyof hand OA. © 2000 OsteoArthritis Research Society International
Key words: Hand osteoarthritis,Definitions, Epidemiology.
Defininghandosteoarthritisfor epidemiological studies
osteophyte at all joint sites. Although the original intention of the system was that grades would be defined by standard radiographs, the films finally chosen as the standards did not easily conform to the written descriptions. There has also been variation in the descriptions provided by the authors at various times. 6"~ This has resulted in poor reproducibility between observers and centers. The second potential problem with the system lies in its emphasis on the osteophyte. While the precise chronological sequence of events in OA remains uncertain, the majority of current data point to a primary alteration of articular cartilage, which is then followed by a variable subchondral bony reaction. Two of the four grades in the revised KellgrenLawrence scheme refer only to osteophytes. An individual with joint space narrowing but no visible osteophytes cannot be classified as having OA using the KellgrenLawrence system. The system therefore assumes that joint space loss occurs after osteophytosis. Some joints do not easily fit into this grading system. Hand joints, for example, may be narrowed and sclerotic without the presence of osteophytes. The widespread realization of these deficiencies in the current procedure for the radiographic assessment of OA has led to a reappraisal. A number of alternative approaches to this problem have emerged. Kallman et aL8 developed scales for grading the prevalence and progression of individual radiological features of OA.of the hand. These included osteophytes, joint space narrowing, subchondral sclerosis, lateral deformity, and cortical collapse. The reproducibility of grading with these more explicit standards is greater, both within and between observers, than using the Kellgren and Lawrence scales. Using this scoring system, each individual joint in the hand is given a score; then, depending on the question being investigated, the distal interphalangeal (DIP), the proximal interphalangeal (PIP) and the carpometacarpal (CMC) joints may each be considered as separate joint groups or, alternatively, the hand may be considered a unit and given an aggregate score based on these separate scores. These grading standards have been used successfully to assess both the incidence and rate of progression of
Epidemiological studies of hand osteoarthritis (OA) require explicit diagnostic criteria to classify the disease in the general population. There is no absolute clinical, radiological or pathological standard with which the epidemiology of hand OA can be compared. 2 Definitions of osteoarthritis may therefore be based upon any combination of clinical features, radiographic changes and pathological findings. These definitions can be put to a number of different uses. Among the most important is classification of individuals as being with or without OA for the purposes of etiologic or therapeutic research. While the requirements of any definition in each of these two areas of research are not identical, they share certain common objectives. The method should ideally be accurate, reproducible, noninvasive, convenient and relatively inexpensive. While sophisticated radiological techniques such as MRI, scintigraphy and macroradiographs may have some advantages in the assessment of OA, 3,4 these methods are not always available, are costly in large population-based studies and are of uncertain reproducibility. Therefore, a simple radiographic system using X-rays for assessing osteoarthritis, if appropriately validated, would meet the criteria needed in large epidemiological studies. Most studies to date have used a system of grading radiographic severity that was developed by Kellgren and Lawrence. s'8 This system assigns one of five grades (0-4) to OA at various joint sites, including the hand. Grading is performed by comparing the index radiograph with reproductions in a radiographic atlas,s The criteria for increasing severity of OA relate to the sequential appearance of osteophytes, joint space loss, subchondral sclerosis and cyst formation. There are a number of problems with this system. The main difficulties include inconsistencies in the interpretation of the grading system and the prominence given to the Address correspondence to: Dr Tim D. Spector, Twin Research & Genetic Epidemiology Unit, St Thomas' Hospital, Lambeth Palace Road, London SE1 7EH, U.K. Tel: + 44 171 928 9292 Ext. 2084; Fax: + 44 171 922 8234; E-mail: [email protected] $2
Osteoarthritis and Cartilage Vol. 8 Suppl A hand OA as well as to identify risk factors for both the development and progression of hand OA. Following these changes recent groups have developed standard atlases of radiographs for the assessment of OA in population studies focusing on individual hand joints and grading features of OA separately.9'~° These atlases are well validated against previous grading methods, ~ and universal adoption of these methods will allow for standardization across studies and comparison of different population groups. In 1990, the Diagnostic and Therapeutic Criteria Committee of the American College of Rheumatology suggested that radiography is of less value than clinical examination for classifying symptomatic OA of the hand. ~2 These recommendations were based on the ability of clinical and radiological features to discriminate between patients with hand OA and patients attending a rheumatology clinic with other causes of hand pain, mainly rheumatoid arthritis. However, population based studies dispute these findings. 13-1s In the general population where the prevalence of OA is much higher than that of other forms of arthritis, accurate diagnosis depends on distinguishing OA cases from people with no joint disease at all. For this purpose, it appears that radiology may be more useful than clinical examination. 13-~s In a recent population-based survey of rheumatic diseases in women aged 45-65 years, the association of clinical and radiographic measures of OA of the hand with reported symptoms and with radiographic OA of the knee was examined. ~3 The knee was examined as it is the large joint that is most often affected in association with hand OA. Clinical and radiographic grades correlated with each other and were associated with symptoms to a similar degree. When analyzed in combination, they were only marginally better at predicting symptoms than when tested individually. In contrast, there were marked differences in the relation of clinical and radiological changes to radiographic knee OA. Hand radiology was a better predictor of knee disease than hand examination. Hand radiology was found to provide a better overall assessment of OA than physical examination of the interphalangeal joints or carpometacarpal joints. TM A further population-based study of 6584 men and women in Zoetermeer, The Netherlands resulted in the same conclusions. ~s In this study, the presence of clinical OA was substantially lower than that of radiographic OA and there was a strong relationship between radiographic hand and knee OA. 15 The authors of this study concluded that radiography is the method of choice for the confirmation of hand OA in populations or subgroups within populations. A recent study examining how often Heberden's nodes and radiological DIP OA co-exist in the same digit found Heberden's nodes were not synonymous with DIP osteophytes. The study concluded that radiological DIP osteophytes were a better marker of knee and multiple joint OA than Heberden's nodes, and they may be an imperfect surrogate for hand OA in the absence of radiology. 16 The available evidence would suggest that radiography is better than clinical examination in defining hand OA in population-based epidemiological studies. ~3-~s However, where radiography is not available, clinical examination might be an acceptable, if less preferable, option. TM
The prevalence of hand OA The hand is commonly affected by OA. One study estimated that the prevalence of OA of the hand was 38%
$3 in women over the age of 66 years and 24.5% for men. 17 Similar results were found in a study of 500 subjects with symptomatic limb joint OA referred to rheumatologists. Of 847 affected joints, 30% involved the hands. 18 In the Baltimore longitudinal study of aging, which studied the incidence of hand OA in 177 men, the incidence was highest at the distal interphalangeal joints and increased with age in all hand joints reaching a maximum of 106/1000 person-years in those men aged 60 and above at study entry. 19 The prevalence of radiological OA in individuals aged over 15 years in the different subgroups of hand joints has been estimated for DIP joints (males 19%, females 23%), PIP joints (males 6%, females 11%), metacarpophalangeal joints (males 9.5%, females 10.5%), CMC joints (males 8%, females 12%), wrists (males 6.5%, females 3%). 2° However, of those with radiological OA, symptomatic disease ranged from 20 to 40%.
Risk factors for developing hand OA AGE
The factor most closely associated with the development of hand OA is age. 2° Although OA first appears at the metatarso-phalangeal joints from the age of 25 years, the greatest incidence of OA occurs at 45 years when OA develops in the interphalangeal joints and the first metacarpophalangeal joint. 21 The mechanism for this age-related development of OA is unknown. GENETIC PREDISPOSITION
A strong genetic component to the development of hand OA is thought to be present, particularly in women. In the 1940s, studies by Stecher found Heberden's nodes were three times as common in sisters of affected women as in the general population. 22 In the 1950s and 1960s epidemiological studies2° found that the rates of generalized OA were roughly twice as common in first-degree relatives of probands with OA as in the general population. These data are compatible with either a monogenic autosomal transmission, dominant in women and recessive in men, or a polygenic heredity with a higher threshold of expression in men in whom a significant proportion of the nodes might be of traumatic origin. Limitations of these early studies included case selection methods, poor case definition and lack of age matching within relatives. Many family studies for hand OA and generalized OA have used families selected on the basis of one affected individual. This may result in biased estimates of familial aggregation which can be eliminated by use of twin studies or using large population data sets. Using improved methods two recent studies have confirmed the results of these early studies using large population-based study designs. The Baltimore Longitudinal Study of Aging 23 and the Framingham Offspring Study 24 have both shown significant familial clustering for hand and knee OA. Both studies used unselected population samples, and a particular strength of the Framingham Offspring Study design was that all subjects were assessed at adult ages. Demonstrating aggregation in families does not exclude the possibility that clustering may be accounted for by shared family e0vironment rather than by genetic factors alone. This can be studied separately by using twin studies. Twin studies are also matched for age, which is important in OA where disease expression is age-dependent. Recent twin data from the St Thomas'
$4
D.J. Hart and T. D. Spector: Definitions and epidemiology of hand OA
Hospital U.K. Adult Twin Registry found a greater association of radiological features of hand OA amongst monozygotic (MZ) twins when compared with dizygotic (DZ) twins, confirming a genetic influence with a heritability of 65% for hand OA. 25 In these studies, definitions of generalized OA. and hand OA have differed. In the Baltimore study, OA was defined by three different groups: (1) a single joint, (2) a number of affected joints, and (3) sum of grades of affected joints. Using these methods significant sib-sib correlations were shown for DIP and PIP joints using the second method, and for all IP joints, CMC joints and a p~olyarticular group of hand and knee for the last method. 23 In the Framingham data, family groups were defined as having at least two parents and one affected offspring with hand OA. OA in an individual was defined by counting the number of affected joints from all IP joints, MCP, CMC and knee joints giving a maximum score of 32 affected joints. The mean hand score was 3.4 for parents and 1.4 for offspring. 24 Other genetic markers that have been examined in generalized nodal OA.are the HLA antigens. 26-29 Here the results are conflicting. Although increased frequencies of HLA-AIB82e and HLA-B827 have been reported, some studies have found no such association.2a'29 There are also several reports of families exhibiting premature OA linked to the COL2A1 gene on chromosome 12 encoding type II procollagen. 3°.3~ There is now considerable impetus to search for the genes involved in OA and this is perhaps the most exciting development in the genetics of OA. Two approaches being studied are genome-wide linkage studies and candidate gene studies. A number of recent studies have examined linkage to hand OA with several suggesting linkage to 2q. 32-34 Potential candidate genes mapping to this region include fibronectin, the alpha-2 chain of collagen type V, the IL-1 RA cluster and the interleukin-8 receptor. 32 A number of candidate genes have been implicated in OA such as vitamin D receptor, estrogen receptor and insulin-like growth factor 1.35 37 However, the effect of these candidate genes is modest and likely to explain only a small part of the genetic contribution to OA. In the next decade testing of genes from the human genome will reveal the genetic structure of OA in much greater detail. GENDER AND HORMONAL FACTORS
Gender has been clearly associated with OA, with an effect particularly related to OA at specific sites. Over the age of 55 years, OA is more common in women, usually involving several joints, mostly the interphalangeals, the first carpometacarpal and the knees. 3s Clinical and population-based studies of generalized OA suggest that the onset of OA is related either to the perimenopausal period or to an episode of hormone imbalance.39 The role of sex hormones in hand OA remains to be clarified but is suggested by the increased rate of OA in older women. A case-control study of women with OA showed that these women had double the rate of hysterectomy compared with age-matched controls with rheumatoid arthritis and from the general population. 4° In addition, women with a previous hysterectomy have been shown to have significantly higher rates of OA of the first CMC joint than control women who have not had a hysterectomy. 4~ This increase persists after adjustment for possible confounders including age, obesity, parity and smoking status. 4~ In contrast, however, no clear-cut relationship of OA with oestrogen use or hysterectomy in women was found in the Framingham OA Study. 42
A number of recent epidemiological studies of OA have found a protective effect of oestrogen replacement therapy and OA. 43-4s The Chingford Study was the only one to examine the association in hand OA and found a 60% non-significant protective effect for DIP OA, and 30% effect in CMC OA. 43 The effect was seen only in current users. The mechanism of the protection is unclear but has important implications for etiopathogenesis of OA.
OBESITY
There are conflicting results regarding the association of obesity with hand OA. Secondary analysis of data from the National Health examination survey noted a significant association of body mass index (BMI) with the presence of hand OA in men after adjustment for age, race and skin fold thickening but not after adjustment for waist girth and seat breadth. 4e An association between radiographic hand OA and BMI in men 47 was also observed in the longitudinal, prospective study of 70-year-old people in G6teborg. In contrast, however, no association was observed between indices of obesity and hand OA in men in the Baltimore Longitudinal Study of Aging. 4a For women, no association was observed between BMI and hand OA in the National Health Examination Survey. 46 However, an association was observed between BMI and the grade of hand OA for women in the Tecumseh Community Health Study. 49 Obesity was also modestly associated with DIP and CMC OA in women in the Chingford Study. s° Although the mechanism by which obesity may predispose an individual to hand OA is unknown, it may be that this association is in part due to a metabolic effect.
MECHANICAL FACTORS
The most important, potentially modifiable risk factor for hand OA is repetitive joint usage associated with mechanical stress. Stecher and Karnosh s~ observed that neither radiological OA nor Heberden's nodes form in paralysed hands. This suggested the possibility that for these to develop requires the wear and tear that accompanies regular use of joints. 22 Further data suggesting a role for trauma in the development of hand OA are that the DIP joints of both sexes are the most commonly affected hand joints in rnost individuals;22.sl in particular the DIP joint of the index finger; s2 that OA occurs in more severe form in the DIP than the PIP or CMC joints; 53 and that almost every joint in the right hand of right-handed people of either sex had more frequent and severe disease than their left hands, s2 Studies of occupational groups exposed to 'repetitive use of small hand joints have shown an excess of hand OA in joints that were used repetitively compared to other joints. 54 In Lawrence's classical surveys, coal miners were shown to have more OA than dockers who in turn had more OA than office workers. 38 The right hand was usually involved more than the left. s2 In addition, a higher rate of Heberden's nodes and hand OA was found in British cotton mill workers than aged-matched controls, s4. Also, in a study of three different occupational groups in a Virginia textile mill, women whose jobs required fine pincer grip (which increased the force across the distal interphalangeal joints) had significantly more distal interphalangeal joint OA than those whose jobs required repeated power grip. s5 Recent data examining the association of incident hand OA
Osteoarthritis and Cartilage Vol. 8 Suppl A and grip strength found that maximal grip strength in men were at increased risk of OA in PIP, MCP and CMC joints, and only MCP joints in women. No association was seen with grip strength and DIP joints, s6
Patterns of hand OA GENERALIZEDOA The classification of generalized OA (GOA) is controversial. The term 'generalized OA' was coined by Kellgren and Moore s7 for what they thought was a discrete entity: involvement of at least three joints or a group of joints. In a study of 391 cases of OA they found 120 patients, largely middle-aged women, who had a polyarthritis with a distinct pattern of joint involvement and a characteristic presentation and course. The joints involved were the distal interphalangeal joints (86%), the carpometacarpal joints of the thumb (60%), proximal interphalangeal joints (41%), metatarsophalangeal joints of the big toe (33%), knee (53%, and nearly always bilateral involvement), spine (48%) and hips (30%). The onset of symptoms was often acute and accompanied by signs of joint inflammation. The mean age at presentation was 52 years. Twenty per cent gave a positive family history of disease, suggesting a hereditary influence on its development. Although this view has been challenged by some observers who believe that multiple involvement occurs at random, 58 statistical analysis tends to confirm that individuals with generalized OA form a distinct population, s9 The pattern of radiographic involvement of hand joints was examined in a population-based study of women aged 45-64. s9 There was clear evidence of clustering in joint involvement, with significantly more women having four or more joints affected compared to the expected number. The major determinants of a pattern of joint involvement were symmetry, which was the most important, and clustering by row. EROSIVEOA Erosive OA is believed to be a clinical subset of generalized OA 6° which characteristically involves the hands of middle-aged women. The onset may be acute, and the subsequent course punctuated by episodes of acute pain, swelling, and erythema, affecting both the DIP and PIP joints. The clinical course tends to be aggressive. The diagnosis is essentially radiological and depends upon the presence of articular surface erosions. 61 In a study of 500 consecutive patients attending a rheumatology clinic with symptomatic limb joint osteoarthritis, 24 patients were identified by radiological criteria to have erosive OA. These were age-sex matched with 24 patients from the same series who presented with OA of the hand. Those with erosive OA had nearly twice as many radiographically abnormal joints in the hands as the controls. This was almost entirely due to an increase in distal and proximal interphalangeal joint involvement, 71% of which were erosive. Otherwise, only minor differences were present between the two groups in terms of distribution and incidence of osteoarthritic changes. There were no distinguishing serological or other clinical differences. Natural history Few studies have evaluated the longitudinal radiographic progression of OA of the interphalangeal joints of
$5 the hand. In a longitudinal study, left hand-wrist X-rays of 386 white males, who were participants of the Baltimore Longitudinal Study of Aging and who had been followed for at least 5 years with two or more visits, were examined for prevalence and progression of OA of the distal and proximal interphalangeal joints of the hand. 62 Using the life table method of analysis, the progression of OA was defined by the following criteria: (1) an increase in the severity of radiographic changes of the joints previously affected, and (2) an increase in the number of new joints affected. The results indicated that OA in both the distal and proximal interphalangeal joints progressed at a faster rate in the older population than in individuals less than 60 years of age. Furthermore, OA in the interphalangeal joints progressed at the same rate whether the starting point was a Kellgren grade of 0 (no disease) or 1 (doubtful disease). Incident hand OA data from the Framingham Study has shown that in subjects without baseline OA, rates of incident hand OA was much higher in women, and DIP OA was the most frequently affected joint followed by CMC, PIP and then MCP joints. Prevalent OA in any joint markedly increased risk of OA in other joints in the same row. 63 OA has been shown to progress very slowly in one prospective study of men aged less than 60 years on entry into the study who were followed for 20 years. 19 It took approximately 10-20 years for any of the radiological features to progress one grade. It took even longer for subjects to progress from the intermediate to the late stages of OA, a possible 'burnout' phenomenon. 64 A further study of 59 subjects followed for 10 years showed deterioration in Kellgren-Lawrence grade in 50% and new osteophytes appearing in 48% of DIP joints. There were no obvious risk factors for fast deterioration. 6s
References 1. Hochberg MC. Epidemiology of osteoarthritis: current concepts and new insights. J Rheumatol 1991; 18(Suppl 27):11-16. 2. Spector TD, Cooper C. Radiological assessment of osteoarthritis: Whither Kellgren and Lawrence? Osteoarthritis Cart 1993;1:203-6. 3. Buckland-Wright C, Macfarlane D, Lynch J. Quantitative microfocal radiographic assessment of disease and progression in the hand. J Rheumatol 1991; 27(Suppl):40-1. 4. Backhaus M, Kamradt T, Sandrock D, Loreck D, Fritz J, Wolf KJ, et al. Arthritis of the finger joints: A comprehensive approach comparing conventional radiography, scintigraphy, ultrasound, and MRI. Arthritis Rheum 1999;42:1232-45. 5. Kellgren JH. The Epidemiology of Chronic Rheumatism. Atlas of Standard Radiographs, Vol. 2. Oxford: Blackwell Scientific 1963. 6. Kellgren JH, Lawrence JS. Radiological assessment of osteoarthritis. Ann Rheum Dis 1957;16:494-501. 7. Lawrence JS. Rheumatism in Populations. London: William Heinemann Medical Books Ltd 1977. 8. Kallman DA, Wigley FM, Scott WW Jr, Hochberg MC, Tobin JD. New radiographic grading scales for osteoarthritis of the hand: reliability for determining prevalence and progression. Arthritis Rheum 1989;32:1584-91. 9. Burnett S, Hart DJ, Cooper C, Spector TD. A Radiographic Atlas of Osteoarthritis. London: SpringerVerlag 1994.
S6
D.J. Hart and T. D. Spector: Definitions and epidemiology of hand OA
10. Altman RD, Hochberg M, Murphy WA, Wolfe F, Lequesne M. Atlas of Radiographic features in osteoarthritis. Osteoarthritis Cart 1995;3(Suppl A):3-70. 11. Hart DJ, Harris PA, Chamberlain A. Reliability and reproducibility of grading radiographs for OA of the hand. Br J Rheum 1992;$1:231 (abstract). 12. Altman R, Alarcon G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology Criteria for the classification and reporting of osteoarthritis of the hand. Arthritis / Rheum 1990;33:1601-10. 13. Dieppe P, Cushnaghan J. The natural course and prognosis of osteoarthritis. In: Moskowitz R, Howell DJ, Goldberg VM, Mankin JH, Eds. Osteoarthritis: Diagnosis and Medical Surgical Management, 2nd ed. London: Saunders 1992:399-412. 14. Hart DJ, Spector TD, Egger P, Coggon D, Cooper C. Defining osteoarthritis of the hand for epidemiologic studies. The Chingford study. Ann Rheum Dis 1994;53:220-3.15. Valkenburg HA. Clinical versus radiological osteoarthritis in the general population. In: Peyron JG, Ed. Epidemiology of Osteoarthritis. Paris: Geigy 1981: 53-8. 16. Cicuttini FM, Baker J, Hart DJ, Spector TD. Relation between Heberden's nodes and distal interphalangeal joint osteophytes and their role as markers of generalised disease. Ann Rheum Dis 1998;57: 246-8. 17. Seen JC, Hochberg MC. Osteoarthritis: 1. Epidemiology. Md State Med J 1984;33:712-6. 18. Cushnaghan J, Dieppe P. Study of 500 patients with limb joint osteoarthritis. 1. Analysis by age, sex, and distribution of symptomatic joint sites. Ann Rheum Dis 1991 ;50:8-13. 19. Kallmann DA, Wigley FM, Scott WW Jr, Hochberg MC, Tobin JD. The longitudinal course of hand osteoarthritis in a male population. Arthritis Rheum 1990;33:1323-32. 20. Lawrence JS, Bremner JM, Biers E Osteoarthritis. Prevalence in the population and relationship between symptoms and X-ray changes. Ann Rheum Dis 1966;25:1-5. 21. Allander E. Prevalence, incidence and remission rates of some common rheumatic diseases or symptoms. Scand J Rheumatol 1974;3:145-8. 22. Stecher RM. Heberden's nodes. Heredity in hypertrophic arthritis of the finger joints. Am J Med Science 1941 ;201:801-5. 23. Hirsh R, Lethbridge-Cejku M, Hanson R, Scott WW, Reichle R, Plato CC, et al. Familial aggregation of osteoarthritis; data from the Baltimore Longitudinal Study on aging. Arthritis Rheum 1998;41: 1227-32. 24. Felson DT, Couropmitree N, Chaisson E, Hannan MT, Zhang Y, McAlindon TE, et al. Evidence for a Mendelian gene in segregation analysis of generalized radiographic osteoarthritis. Arthritis Rheum 1998;41:1064-71. 25. Spector TD, Ciccuttini F, Baker J, Loughlin J, Hart DJ. Genetic influences on osteoarthritis in women: a twin study. BMJ 1996;312:940-3. 26. Pattrick M, Manhire A, Ward M, Doherty M. HLA-A, B antigens and u 1-antitrypsin phenotypes in nodal and generalised osteoarthritis and erosive osteoarthritis. Ann Rheum Dis 1989;48:470-5.
27. Brodsky T, Appelboom A, Govaerts Farney JP. Antig~nes HLA et nodosites d'Heberden. Acta Rheumatologica 1979;3:95-103. 28. Ercilla MG, Brancos MA, Breysse G, Alonso G, Vives J, Castillo R, et al. HLA antigens in Forestier's disease, ankylosing spondylitis and polyarthrosis of the hands. J Rheumatol 1977;4(Suppl 3):89-93. 29. Benavides G, Cerantes A, Silva B, Kantona G, Lardizabal J. HLA and Heberden's nodes in Mexican mestizos. Clin Rheumatol 1985;4:97-8. 30. Palotie A, Vaisanen P, Ott J, Ryhanen L, Elima K, Vikkula M, et al. Previous predisposition to familial osteoarthritis linked to type II collagen gene. Lancet 1989;i:924-7. 31. Ala-Kokko L, Baldwin CT, Moskowitz RW, Prokup DJ. Single based mutation in the type II procollagen gene (COL2A1) as a cause of primary osteoarthritis. Proc Natl Acad Sci USA 1990;87:6565-8. 32. Wright GD, Hughes AE, Regan M, Doherty M. Association of two loci on chromosome 2q with nodal osteoarthritis. Ann Rheum Dis 1996;55:317-9. 33. Chapman KE, Mustafa Z, Irven C, Smith A, Carr A, Clipsham K, et al. A comprehensive genome-wide screen for susceptibility loci in osteoarthritis: linkage to chromosome 2q and 11q. Am J Hum Genet 1998;63(suppl):A16 (abstract). 34. Lepp&vuori JK, Kujala UM, Kinnunen J, Kaprio J, Nissila M, Heliovaara M, et al. Genome scan for predisposing loci for distal interphalangeal joint osteoarthritis: evidence for a locus on 2q. Am J Hum Genet 1999;65:1060-7. 35. Keen RW, Hart DJ, Lanchbury JS, Spector TD. Association of early osteoarthritis of the knee with a Taq I polymorphism of the Vitamin D receptor gene. Arthritis Rheum 1997;40:1444-9. 36. Ushiyama T, Ueyama H, Inoue K, Nishioka J, Ohkubo I, Hukuda S. Estrogen receptor gene polymorphism and generalized osteoarthritis. J Rheumatol 1998;25:134-7. 37. Meulenbelt I, Bijkerk C, Miedema HS, Breedveld FC, Hofman A, Valkenburg HA, et al. A genetic association study of the IGF-1 gene and radiological osteoarthritis in a population-based cohort study. Ann Rheum Dis 1998;57:371-4. 38. Lawrence JS. Rheumatism in Populations. London: William Heinemann Medical Books Ltd 1977. 39. Spector TD, Campion GD. Generalized osteoarthritis: a hormonally mediated disease. Ann Rheum Dis 1989;48:523-7. 40. Spector TD, Brown GC, Silman AJ. Increased rates of prior hysterectomy and gynaecological operations in women with osteoarthritis. BMJ 1988;297:899-900. 41. Spector TD, Hart DJ, Brown P, Almeyda J, Dacre JE, Doyle DV, et al. Frequency of osteoarthritis in hysterectomized women. J Rheumatol 1991 ;18:1877-83. 42. Felson DT. The epidemiology of knee osteoarthritis: results from the Framingham osteoarthritis study. Semin Arthritis Rheum 1990;20:42-50. 43. Spector TD, Nandra D, Hart DJ, Doyle DV. Is hormone replacement therapy protective for hand and knee osteoarthritis in women? The Chingford study. Ann Rheum Dis 1997;56:432-4. 44. Oliveria SA, Felson DT, Klein RA, Reed JI, Walker AM. Estrogen replacement therapy and the development of osteoarthritis. Epidemiology 1996;7:415-9.
Osteoarthritis and Cartilage Vol. 8 Suppl A 45. Nevitt MC, Cummings SR, Lane NE, Hochberg MC, Scott JC, Pressman AR, et al. Association of estrogen replacement therapy with the risk of osteoarthritis of the hip in elderly white women. Study of Osteoporotic Fractures Research Group. Arch Intern Med 1996;156:2073-80. 46. Davis MA, Neuhaus JM, Ettinger WH, Muller WH. Body fat distribution and osteoarthritis. Am J Epidemiol 1990;132:701-7. 47. Bagge E, Bjelle A, Eden S, Svanborg A. Factors associated with radiographic osteoarthritis: results from the population study of 70-year-old people in Goteborg. J Rheumatol 1991 ;18:1218-22. 48. Hochberg MC, Lethbridge-Cejku M, Plato CC, Wigley FM, Tobin JD. Factors associated with osteoarthritis of the hand in males: data from the Baltimore longitudinal study of aging. Am J Epidemiol 1991;134: 1121-7. 49. Sowers MF, Zobel D, Weissfeld L, Hawthorne VM, Carman W. Progression of osteoarthritis of the hand and metacarpal bone loss: a twenty-year follow-up of incident cases. Arthritis Rheum 1991 ;34:36-42. 50. Hart DJ, Spector TD. The relationship of obesity, fat distribution and osteoarthritis in women in the general population: the Chingford study. J Rheumatol 1993;20:331-5. 51. Stecher RM, Karnosh LJ. Heberden's nodes: VI the effect of nerve injury upon the formation of degenerative joint disease of the fingers. Am J Med Sci 1947;213:181-3. 52. Acheson RM, Chan YK, Clemett AR. New Haven survey of joint diseases XII: distribution and symptoms of osteoarthritis in the hands with reference to handedness. Ann Rheum Dis 1970;29:275-86. 53. Plato CC, Norris AH. Osteoarthritis of the hand: agespecific joint-digit prevalence rates. Am J Epidemiol 1979;2:169-80. 54. Lawrence JS. Rheumatism in cotton operatives. Br J Int Med 1961;18:270-6. 55. Hadler NM, Gillings DB, Imbus HR, Levitin PM, Makuc D, Utsinger PD, et al. Hand structure and function in
$7
56. 57. 58.
59.
60.
61. 62. 63.
64.
65.
an industrial setting: influence of three patterns of stereotyped repetitive usage. Arthritis Rheum 1978; 21:210-20. Chaisson CE, Zhang Y, Sharma L, Kannel W, Felson DT. Grip strength and risk of developing radiographic hand OA. Arthritis Rheum 1999;42:33-8. Kellgren JH, Moore R. Generalised osteoarthritis and Heberden's nodes. BMJ 1952;1:181-5. O'Brien WM, Clemett AR, Acheson RM. Symptoms and patterns of OA of the hand in the New Haven survey of joint diseases. In: Bennett H, Wood P, Eds. Population Studies of Rheumatic Diseases. Amsterdam: Excerpta Medica 1968:398. Egger P, Cooper C, Hart DJ, Doyle DV, Coggon D, Spector TD. Patterns of joint involvement in osteoarthritis of the hand: The Chingford study. J Rheumatol 1995;22:1509-13. Crain DC. Interphalangeal osteoarthritis characterized by painful, inflammatory episodes, resulting in deformity of the proximal and distal articulations. JAMA 1961 ;175:1949-53. Cobby M, Cushnaghan J, Creamer P, Dieppe P, Watt I. Erosive osteoarthritis: is it a separate disease entity? Clin Radiol 1990;42:258-63. Busby J, Tobin J, Ettinger W, Roadarmel K, Plato CC. A longitudinal study of osteoarthritis of the hand: the effect of age. Ann Hum Biol 1991;18:417-24. Chaisson CE, Zhang Y, McAlindon TE, Hannan MT, Aliabadi P, Naimark A, et al. Radiographic hand osteoarthritis: incidence, patterns and influence of pre-existing disease in a population based sample. J Rheumatol 1997;24:1337-43. Sabiston CP, Adams ME, Li DKB. Magnetic resonance imaging of osteoarthritis: correlation with gross pathology using an experimental model. J Orthop Res 1987;5:164-72. Harris PA, Hart DJ, Dacre JE, Huskisson EC, Spector TD. The progression of radiological hand osteoarthritis over ten years: a clinical follow-up study. Osteoarthritis Cart 1994;2:247-52.
Osteoarthritis and Cartilage
1063-4584/00/0A00S2+06 $35.00/0
O A ~ ]
OSTEOARTHRITIS
RfSF~,RCH '~OCIFTY
Journal of the OsteoArthritis Research Society International
Definition and epidemiology of osteoarthritis of the hand: a review D. J. Hart and T. D. Spector Twin Research and Genetic Epidemiology Unit, St Thomas' Hospital, London SE1 7EH, U.K.
Summary The hand is a common site of peripheraljoint involvementin osteoarthritis(OA) and although often underestimatedas a cause of disability, the effecton quality of life from limitationsin performing activitiesof daily living such as dressing and feeding may be considerable. Hand OA may also be an important indicatorof a systemic tendency to OA which may involve weight bearing joints, notably the hips and knees.1The definition of hand OA, particularlyfor epidemiologicalstudies, has undergone reassessmentand revision over the last few years. In this paper we examinethe issues relating to this and consider the epidemiologyof hand OA. © 2000 OsteoArthritis Research Society International
Key words: Hand osteoarthritis,Definitions, Epidemiology.
Defininghandosteoarthritisfor epidemiological studies
osteophyte at all joint sites. Although the original intention of the system was that grades would be defined by standard radiographs, the films finally chosen as the standards did not easily conform to the written descriptions. There has also been variation in the descriptions provided by the authors at various times. 6"~ This has resulted in poor reproducibility between observers and centers. The second potential problem with the system lies in its emphasis on the osteophyte. While the precise chronological sequence of events in OA remains uncertain, the majority of current data point to a primary alteration of articular cartilage, which is then followed by a variable subchondral bony reaction. Two of the four grades in the revised KellgrenLawrence scheme refer only to osteophytes. An individual with joint space narrowing but no visible osteophytes cannot be classified as having OA using the KellgrenLawrence system. The system therefore assumes that joint space loss occurs after osteophytosis. Some joints do not easily fit into this grading system. Hand joints, for example, may be narrowed and sclerotic without the presence of osteophytes. The widespread realization of these deficiencies in the current procedure for the radiographic assessment of OA has led to a reappraisal. A number of alternative approaches to this problem have emerged. Kallman et aL8 developed scales for grading the prevalence and progression of individual radiological features of OA.of the hand. These included osteophytes, joint space narrowing, subchondral sclerosis, lateral deformity, and cortical collapse. The reproducibility of grading with these more explicit standards is greater, both within and between observers, than using the Kellgren and Lawrence scales. Using this scoring system, each individual joint in the hand is given a score; then, depending on the question being investigated, the distal interphalangeal (DIP), the proximal interphalangeal (PIP) and the carpometacarpal (CMC) joints may each be considered as separate joint groups or, alternatively, the hand may be considered a unit and given an aggregate score based on these separate scores. These grading standards have been used successfully to assess both the incidence and rate of progression of
Epidemiological studies of hand osteoarthritis (OA) require explicit diagnostic criteria to classify the disease in the general population. There is no absolute clinical, radiological or pathological standard with which the epidemiology of hand OA can be compared. 2 Definitions of osteoarthritis may therefore be based upon any combination of clinical features, radiographic changes and pathological findings. These definitions can be put to a number of different uses. Among the most important is classification of individuals as being with or without OA for the purposes of etiologic or therapeutic research. While the requirements of any definition in each of these two areas of research are not identical, they share certain common objectives. The method should ideally be accurate, reproducible, noninvasive, convenient and relatively inexpensive. While sophisticated radiological techniques such as MRI, scintigraphy and macroradiographs may have some advantages in the assessment of OA, 3,4 these methods are not always available, are costly in large population-based studies and are of uncertain reproducibility. Therefore, a simple radiographic system using X-rays for assessing osteoarthritis, if appropriately validated, would meet the criteria needed in large epidemiological studies. Most studies to date have used a system of grading radiographic severity that was developed by Kellgren and Lawrence. s'8 This system assigns one of five grades (0-4) to OA at various joint sites, including the hand. Grading is performed by comparing the index radiograph with reproductions in a radiographic atlas,s The criteria for increasing severity of OA relate to the sequential appearance of osteophytes, joint space loss, subchondral sclerosis and cyst formation. There are a number of problems with this system. The main difficulties include inconsistencies in the interpretation of the grading system and the prominence given to the Address correspondence to: Dr Tim D. Spector, Twin Research & Genetic Epidemiology Unit, St Thomas' Hospital, Lambeth Palace Road, London SE1 7EH, U.K. Tel: + 44 171 928 9292 Ext. 2084; Fax: + 44 171 922 8234; E-mail: [email protected] $2
Osteoarthritis and Cartilage Vol. 8 Suppl A hand OA as well as to identify risk factors for both the development and progression of hand OA. Following these changes recent groups have developed standard atlases of radiographs for the assessment of OA in population studies focusing on individual hand joints and grading features of OA separately.9'~° These atlases are well validated against previous grading methods, ~ and universal adoption of these methods will allow for standardization across studies and comparison of different population groups. In 1990, the Diagnostic and Therapeutic Criteria Committee of the American College of Rheumatology suggested that radiography is of less value than clinical examination for classifying symptomatic OA of the hand. ~2 These recommendations were based on the ability of clinical and radiological features to discriminate between patients with hand OA and patients attending a rheumatology clinic with other causes of hand pain, mainly rheumatoid arthritis. However, population based studies dispute these findings. 13-1s In the general population where the prevalence of OA is much higher than that of other forms of arthritis, accurate diagnosis depends on distinguishing OA cases from people with no joint disease at all. For this purpose, it appears that radiology may be more useful than clinical examination. 13-~s In a recent population-based survey of rheumatic diseases in women aged 45-65 years, the association of clinical and radiographic measures of OA of the hand with reported symptoms and with radiographic OA of the knee was examined. ~3 The knee was examined as it is the large joint that is most often affected in association with hand OA. Clinical and radiographic grades correlated with each other and were associated with symptoms to a similar degree. When analyzed in combination, they were only marginally better at predicting symptoms than when tested individually. In contrast, there were marked differences in the relation of clinical and radiological changes to radiographic knee OA. Hand radiology was a better predictor of knee disease than hand examination. Hand radiology was found to provide a better overall assessment of OA than physical examination of the interphalangeal joints or carpometacarpal joints. TM A further population-based study of 6584 men and women in Zoetermeer, The Netherlands resulted in the same conclusions. ~s In this study, the presence of clinical OA was substantially lower than that of radiographic OA and there was a strong relationship between radiographic hand and knee OA. 15 The authors of this study concluded that radiography is the method of choice for the confirmation of hand OA in populations or subgroups within populations. A recent study examining how often Heberden's nodes and radiological DIP OA co-exist in the same digit found Heberden's nodes were not synonymous with DIP osteophytes. The study concluded that radiological DIP osteophytes were a better marker of knee and multiple joint OA than Heberden's nodes, and they may be an imperfect surrogate for hand OA in the absence of radiology. 16 The available evidence would suggest that radiography is better than clinical examination in defining hand OA in population-based epidemiological studies. ~3-~s However, where radiography is not available, clinical examination might be an acceptable, if less preferable, option. TM
The prevalence of hand OA The hand is commonly affected by OA. One study estimated that the prevalence of OA of the hand was 38%
$3 in women over the age of 66 years and 24.5% for men. 17 Similar results were found in a study of 500 subjects with symptomatic limb joint OA referred to rheumatologists. Of 847 affected joints, 30% involved the hands. 18 In the Baltimore longitudinal study of aging, which studied the incidence of hand OA in 177 men, the incidence was highest at the distal interphalangeal joints and increased with age in all hand joints reaching a maximum of 106/1000 person-years in those men aged 60 and above at study entry. 19 The prevalence of radiological OA in individuals aged over 15 years in the different subgroups of hand joints has been estimated for DIP joints (males 19%, females 23%), PIP joints (males 6%, females 11%), metacarpophalangeal joints (males 9.5%, females 10.5%), CMC joints (males 8%, females 12%), wrists (males 6.5%, females 3%). 2° However, of those with radiological OA, symptomatic disease ranged from 20 to 40%.
Risk factors for developing hand OA AGE
The factor most closely associated with the development of hand OA is age. 2° Although OA first appears at the metatarso-phalangeal joints from the age of 25 years, the greatest incidence of OA occurs at 45 years when OA develops in the interphalangeal joints and the first metacarpophalangeal joint. 21 The mechanism for this age-related development of OA is unknown. GENETIC PREDISPOSITION
A strong genetic component to the development of hand OA is thought to be present, particularly in women. In the 1940s, studies by Stecher found Heberden's nodes were three times as common in sisters of affected women as in the general population. 22 In the 1950s and 1960s epidemiological studies2° found that the rates of generalized OA were roughly twice as common in first-degree relatives of probands with OA as in the general population. These data are compatible with either a monogenic autosomal transmission, dominant in women and recessive in men, or a polygenic heredity with a higher threshold of expression in men in whom a significant proportion of the nodes might be of traumatic origin. Limitations of these early studies included case selection methods, poor case definition and lack of age matching within relatives. Many family studies for hand OA and generalized OA have used families selected on the basis of one affected individual. This may result in biased estimates of familial aggregation which can be eliminated by use of twin studies or using large population data sets. Using improved methods two recent studies have confirmed the results of these early studies using large population-based study designs. The Baltimore Longitudinal Study of Aging 23 and the Framingham Offspring Study 24 have both shown significant familial clustering for hand and knee OA. Both studies used unselected population samples, and a particular strength of the Framingham Offspring Study design was that all subjects were assessed at adult ages. Demonstrating aggregation in families does not exclude the possibility that clustering may be accounted for by shared family e0vironment rather than by genetic factors alone. This can be studied separately by using twin studies. Twin studies are also matched for age, which is important in OA where disease expression is age-dependent. Recent twin data from the St Thomas'
$4
D.J. Hart and T. D. Spector: Definitions and epidemiology of hand OA
Hospital U.K. Adult Twin Registry found a greater association of radiological features of hand OA amongst monozygotic (MZ) twins when compared with dizygotic (DZ) twins, confirming a genetic influence with a heritability of 65% for hand OA. 25 In these studies, definitions of generalized OA. and hand OA have differed. In the Baltimore study, OA was defined by three different groups: (1) a single joint, (2) a number of affected joints, and (3) sum of grades of affected joints. Using these methods significant sib-sib correlations were shown for DIP and PIP joints using the second method, and for all IP joints, CMC joints and a p~olyarticular group of hand and knee for the last method. 23 In the Framingham data, family groups were defined as having at least two parents and one affected offspring with hand OA. OA in an individual was defined by counting the number of affected joints from all IP joints, MCP, CMC and knee joints giving a maximum score of 32 affected joints. The mean hand score was 3.4 for parents and 1.4 for offspring. 24 Other genetic markers that have been examined in generalized nodal OA.are the HLA antigens. 26-29 Here the results are conflicting. Although increased frequencies of HLA-AIB82e and HLA-B827 have been reported, some studies have found no such association.2a'29 There are also several reports of families exhibiting premature OA linked to the COL2A1 gene on chromosome 12 encoding type II procollagen. 3°.3~ There is now considerable impetus to search for the genes involved in OA and this is perhaps the most exciting development in the genetics of OA. Two approaches being studied are genome-wide linkage studies and candidate gene studies. A number of recent studies have examined linkage to hand OA with several suggesting linkage to 2q. 32-34 Potential candidate genes mapping to this region include fibronectin, the alpha-2 chain of collagen type V, the IL-1 RA cluster and the interleukin-8 receptor. 32 A number of candidate genes have been implicated in OA such as vitamin D receptor, estrogen receptor and insulin-like growth factor 1.35 37 However, the effect of these candidate genes is modest and likely to explain only a small part of the genetic contribution to OA. In the next decade testing of genes from the human genome will reveal the genetic structure of OA in much greater detail. GENDER AND HORMONAL FACTORS
Gender has been clearly associated with OA, with an effect particularly related to OA at specific sites. Over the age of 55 years, OA is more common in women, usually involving several joints, mostly the interphalangeals, the first carpometacarpal and the knees. 3s Clinical and population-based studies of generalized OA suggest that the onset of OA is related either to the perimenopausal period or to an episode of hormone imbalance.39 The role of sex hormones in hand OA remains to be clarified but is suggested by the increased rate of OA in older women. A case-control study of women with OA showed that these women had double the rate of hysterectomy compared with age-matched controls with rheumatoid arthritis and from the general population. 4° In addition, women with a previous hysterectomy have been shown to have significantly higher rates of OA of the first CMC joint than control women who have not had a hysterectomy. 4~ This increase persists after adjustment for possible confounders including age, obesity, parity and smoking status. 4~ In contrast, however, no clear-cut relationship of OA with oestrogen use or hysterectomy in women was found in the Framingham OA Study. 42
A number of recent epidemiological studies of OA have found a protective effect of oestrogen replacement therapy and OA. 43-4s The Chingford Study was the only one to examine the association in hand OA and found a 60% non-significant protective effect for DIP OA, and 30% effect in CMC OA. 43 The effect was seen only in current users. The mechanism of the protection is unclear but has important implications for etiopathogenesis of OA.
OBESITY
There are conflicting results regarding the association of obesity with hand OA. Secondary analysis of data from the National Health examination survey noted a significant association of body mass index (BMI) with the presence of hand OA in men after adjustment for age, race and skin fold thickening but not after adjustment for waist girth and seat breadth. 4e An association between radiographic hand OA and BMI in men 47 was also observed in the longitudinal, prospective study of 70-year-old people in G6teborg. In contrast, however, no association was observed between indices of obesity and hand OA in men in the Baltimore Longitudinal Study of Aging. 4a For women, no association was observed between BMI and hand OA in the National Health Examination Survey. 46 However, an association was observed between BMI and the grade of hand OA for women in the Tecumseh Community Health Study. 49 Obesity was also modestly associated with DIP and CMC OA in women in the Chingford Study. s° Although the mechanism by which obesity may predispose an individual to hand OA is unknown, it may be that this association is in part due to a metabolic effect.
MECHANICAL FACTORS
The most important, potentially modifiable risk factor for hand OA is repetitive joint usage associated with mechanical stress. Stecher and Karnosh s~ observed that neither radiological OA nor Heberden's nodes form in paralysed hands. This suggested the possibility that for these to develop requires the wear and tear that accompanies regular use of joints. 22 Further data suggesting a role for trauma in the development of hand OA are that the DIP joints of both sexes are the most commonly affected hand joints in rnost individuals;22.sl in particular the DIP joint of the index finger; s2 that OA occurs in more severe form in the DIP than the PIP or CMC joints; 53 and that almost every joint in the right hand of right-handed people of either sex had more frequent and severe disease than their left hands, s2 Studies of occupational groups exposed to 'repetitive use of small hand joints have shown an excess of hand OA in joints that were used repetitively compared to other joints. 54 In Lawrence's classical surveys, coal miners were shown to have more OA than dockers who in turn had more OA than office workers. 38 The right hand was usually involved more than the left. s2 In addition, a higher rate of Heberden's nodes and hand OA was found in British cotton mill workers than aged-matched controls, s4. Also, in a study of three different occupational groups in a Virginia textile mill, women whose jobs required fine pincer grip (which increased the force across the distal interphalangeal joints) had significantly more distal interphalangeal joint OA than those whose jobs required repeated power grip. s5 Recent data examining the association of incident hand OA
Osteoarthritis and Cartilage Vol. 8 Suppl A and grip strength found that maximal grip strength in men were at increased risk of OA in PIP, MCP and CMC joints, and only MCP joints in women. No association was seen with grip strength and DIP joints, s6
Patterns of hand OA GENERALIZEDOA The classification of generalized OA (GOA) is controversial. The term 'generalized OA' was coined by Kellgren and Moore s7 for what they thought was a discrete entity: involvement of at least three joints or a group of joints. In a study of 391 cases of OA they found 120 patients, largely middle-aged women, who had a polyarthritis with a distinct pattern of joint involvement and a characteristic presentation and course. The joints involved were the distal interphalangeal joints (86%), the carpometacarpal joints of the thumb (60%), proximal interphalangeal joints (41%), metatarsophalangeal joints of the big toe (33%), knee (53%, and nearly always bilateral involvement), spine (48%) and hips (30%). The onset of symptoms was often acute and accompanied by signs of joint inflammation. The mean age at presentation was 52 years. Twenty per cent gave a positive family history of disease, suggesting a hereditary influence on its development. Although this view has been challenged by some observers who believe that multiple involvement occurs at random, 58 statistical analysis tends to confirm that individuals with generalized OA form a distinct population, s9 The pattern of radiographic involvement of hand joints was examined in a population-based study of women aged 45-64. s9 There was clear evidence of clustering in joint involvement, with significantly more women having four or more joints affected compared to the expected number. The major determinants of a pattern of joint involvement were symmetry, which was the most important, and clustering by row. EROSIVEOA Erosive OA is believed to be a clinical subset of generalized OA 6° which characteristically involves the hands of middle-aged women. The onset may be acute, and the subsequent course punctuated by episodes of acute pain, swelling, and erythema, affecting both the DIP and PIP joints. The clinical course tends to be aggressive. The diagnosis is essentially radiological and depends upon the presence of articular surface erosions. 61 In a study of 500 consecutive patients attending a rheumatology clinic with symptomatic limb joint osteoarthritis, 24 patients were identified by radiological criteria to have erosive OA. These were age-sex matched with 24 patients from the same series who presented with OA of the hand. Those with erosive OA had nearly twice as many radiographically abnormal joints in the hands as the controls. This was almost entirely due to an increase in distal and proximal interphalangeal joint involvement, 71% of which were erosive. Otherwise, only minor differences were present between the two groups in terms of distribution and incidence of osteoarthritic changes. There were no distinguishing serological or other clinical differences. Natural history Few studies have evaluated the longitudinal radiographic progression of OA of the interphalangeal joints of
$5 the hand. In a longitudinal study, left hand-wrist X-rays of 386 white males, who were participants of the Baltimore Longitudinal Study of Aging and who had been followed for at least 5 years with two or more visits, were examined for prevalence and progression of OA of the distal and proximal interphalangeal joints of the hand. 62 Using the life table method of analysis, the progression of OA was defined by the following criteria: (1) an increase in the severity of radiographic changes of the joints previously affected, and (2) an increase in the number of new joints affected. The results indicated that OA in both the distal and proximal interphalangeal joints progressed at a faster rate in the older population than in individuals less than 60 years of age. Furthermore, OA in the interphalangeal joints progressed at the same rate whether the starting point was a Kellgren grade of 0 (no disease) or 1 (doubtful disease). Incident hand OA data from the Framingham Study has shown that in subjects without baseline OA, rates of incident hand OA was much higher in women, and DIP OA was the most frequently affected joint followed by CMC, PIP and then MCP joints. Prevalent OA in any joint markedly increased risk of OA in other joints in the same row. 63 OA has been shown to progress very slowly in one prospective study of men aged less than 60 years on entry into the study who were followed for 20 years. 19 It took approximately 10-20 years for any of the radiological features to progress one grade. It took even longer for subjects to progress from the intermediate to the late stages of OA, a possible 'burnout' phenomenon. 64 A further study of 59 subjects followed for 10 years showed deterioration in Kellgren-Lawrence grade in 50% and new osteophytes appearing in 48% of DIP joints. There were no obvious risk factors for fast deterioration. 6s
References 1. Hochberg MC. Epidemiology of osteoarthritis: current concepts and new insights. J Rheumatol 1991; 18(Suppl 27):11-16. 2. Spector TD, Cooper C. Radiological assessment of osteoarthritis: Whither Kellgren and Lawrence? Osteoarthritis Cart 1993;1:203-6. 3. Buckland-Wright C, Macfarlane D, Lynch J. Quantitative microfocal radiographic assessment of disease and progression in the hand. J Rheumatol 1991; 27(Suppl):40-1. 4. Backhaus M, Kamradt T, Sandrock D, Loreck D, Fritz J, Wolf KJ, et al. Arthritis of the finger joints: A comprehensive approach comparing conventional radiography, scintigraphy, ultrasound, and MRI. Arthritis Rheum 1999;42:1232-45. 5. Kellgren JH. The Epidemiology of Chronic Rheumatism. Atlas of Standard Radiographs, Vol. 2. Oxford: Blackwell Scientific 1963. 6. Kellgren JH, Lawrence JS. Radiological assessment of osteoarthritis. Ann Rheum Dis 1957;16:494-501. 7. Lawrence JS. Rheumatism in Populations. London: William Heinemann Medical Books Ltd 1977. 8. Kallman DA, Wigley FM, Scott WW Jr, Hochberg MC, Tobin JD. New radiographic grading scales for osteoarthritis of the hand: reliability for determining prevalence and progression. Arthritis Rheum 1989;32:1584-91. 9. Burnett S, Hart DJ, Cooper C, Spector TD. A Radiographic Atlas of Osteoarthritis. London: SpringerVerlag 1994.
S6
D.J. Hart and T. D. Spector: Definitions and epidemiology of hand OA
10. Altman RD, Hochberg M, Murphy WA, Wolfe F, Lequesne M. Atlas of Radiographic features in osteoarthritis. Osteoarthritis Cart 1995;3(Suppl A):3-70. 11. Hart DJ, Harris PA, Chamberlain A. Reliability and reproducibility of grading radiographs for OA of the hand. Br J Rheum 1992;$1:231 (abstract). 12. Altman R, Alarcon G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology Criteria for the classification and reporting of osteoarthritis of the hand. Arthritis / Rheum 1990;33:1601-10. 13. Dieppe P, Cushnaghan J. The natural course and prognosis of osteoarthritis. In: Moskowitz R, Howell DJ, Goldberg VM, Mankin JH, Eds. Osteoarthritis: Diagnosis and Medical Surgical Management, 2nd ed. London: Saunders 1992:399-412. 14. Hart DJ, Spector TD, Egger P, Coggon D, Cooper C. Defining osteoarthritis of the hand for epidemiologic studies. The Chingford study. Ann Rheum Dis 1994;53:220-3.15. Valkenburg HA. Clinical versus radiological osteoarthritis in the general population. In: Peyron JG, Ed. Epidemiology of Osteoarthritis. Paris: Geigy 1981: 53-8. 16. Cicuttini FM, Baker J, Hart DJ, Spector TD. Relation between Heberden's nodes and distal interphalangeal joint osteophytes and their role as markers of generalised disease. Ann Rheum Dis 1998;57: 246-8. 17. Seen JC, Hochberg MC. Osteoarthritis: 1. Epidemiology. Md State Med J 1984;33:712-6. 18. Cushnaghan J, Dieppe P. Study of 500 patients with limb joint osteoarthritis. 1. Analysis by age, sex, and distribution of symptomatic joint sites. Ann Rheum Dis 1991 ;50:8-13. 19. Kallmann DA, Wigley FM, Scott WW Jr, Hochberg MC, Tobin JD. The longitudinal course of hand osteoarthritis in a male population. Arthritis Rheum 1990;33:1323-32. 20. Lawrence JS, Bremner JM, Biers E Osteoarthritis. Prevalence in the population and relationship between symptoms and X-ray changes. Ann Rheum Dis 1966;25:1-5. 21. Allander E. Prevalence, incidence and remission rates of some common rheumatic diseases or symptoms. Scand J Rheumatol 1974;3:145-8. 22. Stecher RM. Heberden's nodes. Heredity in hypertrophic arthritis of the finger joints. Am J Med Science 1941 ;201:801-5. 23. Hirsh R, Lethbridge-Cejku M, Hanson R, Scott WW, Reichle R, Plato CC, et al. Familial aggregation of osteoarthritis; data from the Baltimore Longitudinal Study on aging. Arthritis Rheum 1998;41: 1227-32. 24. Felson DT, Couropmitree N, Chaisson E, Hannan MT, Zhang Y, McAlindon TE, et al. Evidence for a Mendelian gene in segregation analysis of generalized radiographic osteoarthritis. Arthritis Rheum 1998;41:1064-71. 25. Spector TD, Ciccuttini F, Baker J, Loughlin J, Hart DJ. Genetic influences on osteoarthritis in women: a twin study. BMJ 1996;312:940-3. 26. Pattrick M, Manhire A, Ward M, Doherty M. HLA-A, B antigens and u 1-antitrypsin phenotypes in nodal and generalised osteoarthritis and erosive osteoarthritis. Ann Rheum Dis 1989;48:470-5.
27. Brodsky T, Appelboom A, Govaerts Farney JP. Antig~nes HLA et nodosites d'Heberden. Acta Rheumatologica 1979;3:95-103. 28. Ercilla MG, Brancos MA, Breysse G, Alonso G, Vives J, Castillo R, et al. HLA antigens in Forestier's disease, ankylosing spondylitis and polyarthrosis of the hands. J Rheumatol 1977;4(Suppl 3):89-93. 29. Benavides G, Cerantes A, Silva B, Kantona G, Lardizabal J. HLA and Heberden's nodes in Mexican mestizos. Clin Rheumatol 1985;4:97-8. 30. Palotie A, Vaisanen P, Ott J, Ryhanen L, Elima K, Vikkula M, et al. Previous predisposition to familial osteoarthritis linked to type II collagen gene. Lancet 1989;i:924-7. 31. Ala-Kokko L, Baldwin CT, Moskowitz RW, Prokup DJ. Single based mutation in the type II procollagen gene (COL2A1) as a cause of primary osteoarthritis. Proc Natl Acad Sci USA 1990;87:6565-8. 32. Wright GD, Hughes AE, Regan M, Doherty M. Association of two loci on chromosome 2q with nodal osteoarthritis. Ann Rheum Dis 1996;55:317-9. 33. Chapman KE, Mustafa Z, Irven C, Smith A, Carr A, Clipsham K, et al. A comprehensive genome-wide screen for susceptibility loci in osteoarthritis: linkage to chromosome 2q and 11q. Am J Hum Genet 1998;63(suppl):A16 (abstract). 34. Lepp&vuori JK, Kujala UM, Kinnunen J, Kaprio J, Nissila M, Heliovaara M, et al. Genome scan for predisposing loci for distal interphalangeal joint osteoarthritis: evidence for a locus on 2q. Am J Hum Genet 1999;65:1060-7. 35. Keen RW, Hart DJ, Lanchbury JS, Spector TD. Association of early osteoarthritis of the knee with a Taq I polymorphism of the Vitamin D receptor gene. Arthritis Rheum 1997;40:1444-9. 36. Ushiyama T, Ueyama H, Inoue K, Nishioka J, Ohkubo I, Hukuda S. Estrogen receptor gene polymorphism and generalized osteoarthritis. J Rheumatol 1998;25:134-7. 37. Meulenbelt I, Bijkerk C, Miedema HS, Breedveld FC, Hofman A, Valkenburg HA, et al. A genetic association study of the IGF-1 gene and radiological osteoarthritis in a population-based cohort study. Ann Rheum Dis 1998;57:371-4. 38. Lawrence JS. Rheumatism in Populations. London: William Heinemann Medical Books Ltd 1977. 39. Spector TD, Campion GD. Generalized osteoarthritis: a hormonally mediated disease. Ann Rheum Dis 1989;48:523-7. 40. Spector TD, Brown GC, Silman AJ. Increased rates of prior hysterectomy and gynaecological operations in women with osteoarthritis. BMJ 1988;297:899-900. 41. Spector TD, Hart DJ, Brown P, Almeyda J, Dacre JE, Doyle DV, et al. Frequency of osteoarthritis in hysterectomized women. J Rheumatol 1991 ;18:1877-83. 42. Felson DT. The epidemiology of knee osteoarthritis: results from the Framingham osteoarthritis study. Semin Arthritis Rheum 1990;20:42-50. 43. Spector TD, Nandra D, Hart DJ, Doyle DV. Is hormone replacement therapy protective for hand and knee osteoarthritis in women? The Chingford study. Ann Rheum Dis 1997;56:432-4. 44. Oliveria SA, Felson DT, Klein RA, Reed JI, Walker AM. Estrogen replacement therapy and the development of osteoarthritis. Epidemiology 1996;7:415-9.
Osteoarthritis and Cartilage Vol. 8 Suppl A 45. Nevitt MC, Cummings SR, Lane NE, Hochberg MC, Scott JC, Pressman AR, et al. Association of estrogen replacement therapy with the risk of osteoarthritis of the hip in elderly white women. Study of Osteoporotic Fractures Research Group. Arch Intern Med 1996;156:2073-80. 46. Davis MA, Neuhaus JM, Ettinger WH, Muller WH. Body fat distribution and osteoarthritis. Am J Epidemiol 1990;132:701-7. 47. Bagge E, Bjelle A, Eden S, Svanborg A. Factors associated with radiographic osteoarthritis: results from the population study of 70-year-old people in Goteborg. J Rheumatol 1991 ;18:1218-22. 48. Hochberg MC, Lethbridge-Cejku M, Plato CC, Wigley FM, Tobin JD. Factors associated with osteoarthritis of the hand in males: data from the Baltimore longitudinal study of aging. Am J Epidemiol 1991;134: 1121-7. 49. Sowers MF, Zobel D, Weissfeld L, Hawthorne VM, Carman W. Progression of osteoarthritis of the hand and metacarpal bone loss: a twenty-year follow-up of incident cases. Arthritis Rheum 1991 ;34:36-42. 50. Hart DJ, Spector TD. The relationship of obesity, fat distribution and osteoarthritis in women in the general population: the Chingford study. J Rheumatol 1993;20:331-5. 51. Stecher RM, Karnosh LJ. Heberden's nodes: VI the effect of nerve injury upon the formation of degenerative joint disease of the fingers. Am J Med Sci 1947;213:181-3. 52. Acheson RM, Chan YK, Clemett AR. New Haven survey of joint diseases XII: distribution and symptoms of osteoarthritis in the hands with reference to handedness. Ann Rheum Dis 1970;29:275-86. 53. Plato CC, Norris AH. Osteoarthritis of the hand: agespecific joint-digit prevalence rates. Am J Epidemiol 1979;2:169-80. 54. Lawrence JS. Rheumatism in cotton operatives. Br J Int Med 1961;18:270-6. 55. Hadler NM, Gillings DB, Imbus HR, Levitin PM, Makuc D, Utsinger PD, et al. Hand structure and function in
$7
56. 57. 58.
59.
60.
61. 62. 63.
64.
65.
an industrial setting: influence of three patterns of stereotyped repetitive usage. Arthritis Rheum 1978; 21:210-20. Chaisson CE, Zhang Y, Sharma L, Kannel W, Felson DT. Grip strength and risk of developing radiographic hand OA. Arthritis Rheum 1999;42:33-8. Kellgren JH, Moore R. Generalised osteoarthritis and Heberden's nodes. BMJ 1952;1:181-5. O'Brien WM, Clemett AR, Acheson RM. Symptoms and patterns of OA of the hand in the New Haven survey of joint diseases. In: Bennett H, Wood P, Eds. Population Studies of Rheumatic Diseases. Amsterdam: Excerpta Medica 1968:398. Egger P, Cooper C, Hart DJ, Doyle DV, Coggon D, Spector TD. Patterns of joint involvement in osteoarthritis of the hand: The Chingford study. J Rheumatol 1995;22:1509-13. Crain DC. Interphalangeal osteoarthritis characterized by painful, inflammatory episodes, resulting in deformity of the proximal and distal articulations. JAMA 1961 ;175:1949-53. Cobby M, Cushnaghan J, Creamer P, Dieppe P, Watt I. Erosive osteoarthritis: is it a separate disease entity? Clin Radiol 1990;42:258-63. Busby J, Tobin J, Ettinger W, Roadarmel K, Plato CC. A longitudinal study of osteoarthritis of the hand: the effect of age. Ann Hum Biol 1991;18:417-24. Chaisson CE, Zhang Y, McAlindon TE, Hannan MT, Aliabadi P, Naimark A, et al. Radiographic hand osteoarthritis: incidence, patterns and influence of pre-existing disease in a population based sample. J Rheumatol 1997;24:1337-43. Sabiston CP, Adams ME, Li DKB. Magnetic resonance imaging of osteoarthritis: correlation with gross pathology using an experimental model. J Orthop Res 1987;5:164-72. Harris PA, Hart DJ, Dacre JE, Huskisson EC, Spector TD. The progression of radiological hand osteoarthritis over ten years: a clinical follow-up study. Osteoarthritis Cart 1994;2:247-52.