Primary osteoarthritis (OA) is a common debilitating disease involving focal cartilage loss and formation of periarticular bone. The knee and the hip commonly are affected. In the UK, it is estimated that 34% of the population older than 45 years has OA of the knee,8 and 19% of the population older than 55 years has OA of the hip.5 Most cases of OA of the hip and knee are idiopathic. Risk factors have been established in relation to age, obesity, race, occupation, injury, and joint deformity.4,5 Genetic factors have been implicated with increasing frequency in epidemiologic and genetic studies. Candidate gene studies and genome-wide scanning have provided important initial results.5,10
Segregation of primary OA as a Mendelian trait is rare. This prompted our investigation of families with an affected sibling pair. Investigation of families containing at least two siblings who have had total joint replacement for end-stage primary idiopathic OA (identified from clinical, radiologic, and histologic evidence) or have advanced radiographic changes of OA has shown evidence for linkage to six loci on chromosomes 2q, 4q, 6, 11q, 16p, and 16q.9 Particularly strong linkages have been found in females with OA of the hip, especially on chromosomes 6 and 11q.2,10 These studies showed that this particular group of patients may harbor genetic traits that increase the risk of hip OA, and therefore warrant additional investigation.
To determine if this increased risk of symptomatic OA of the hip is present in the next generation, we investigated 49 families with at least one affected female sibling pair who had total hip arthroplasty (THA) for end-stage primary OA. We aimed to determine the relative risk of symptomatic and radiographic OA in these 145 children.
MATERIALS AND METHODS
Patients were recruited from a cohort of 49 families with at least one affected female sibling pair (189 individual members). Local (OxREC CO.185) and National (MREC/02/2/109) ethical approval was obtained. Members of the families with an affected sibling pair were contacted and details of their children were requested. The children (hereafter referred to as offspring) then were invited to participate in the study. One hundred forty-five offspring from affected families and 119 spouses, acting as controls, volunteered. Table 1 shows the number of participants, the method of recruitment, and the reasons for refusal in nonparticipants.
Exclusion criteria included patients with a history of significant trauma or other predisposing factors, such as developmental dysplasia of the hip; however, no patients were excluded on these grounds.
Patients were examined in a research clinic, and the presence of OA was determined by clinical and radiologic examinations. The clinical enquiry focused on symptoms and distribution of joint pain and stiffness. Symptoms elicited were those of joint pain, including groin, buttock, thigh, and knee pain. The distribution of symptoms and the presence of joint-line tenderness, crepitus, deformity, range of motion (ROM), and Heberden’s nodes were recorded.
Patients who had any symptoms or signs suggestive of OA had plain radiographic examination of the affected joints. Radiographs were graded for the presence of OA1 independently without individual knowledge of patients by the two main researchers (JS and AC). Grade 1 OA was determined to be the minimum grade required to diagnose radiologic OA, which correlates with a Kellgren and Lawrence Grade 2.6 Patients were considered affected if they had clinical symptoms and signs and a minimum of radiologic Grade 1 OA.
In patients with symptoms suggestive of OA who declined radiologic examination, we used the presence of Heberden’s nodes as a definitive sign of primary OA. Six patients declined radiologic examination, and of these, four patients were determined to be affected.
Differences in the frequency of symptomatic radiologic idiopathic OA between offspring and controls were compared using chi square testing and contingency table analysis. The p values were calculated using Fisher’s exact test. Relative risk ratios were calculated with 95% confidence intervals.
In the 145 offspring and 119 control patients studied, 51 (35.2%) and 12 (10.2%), respectively, were affected clinically and radiologically with primary OA (Table 2). The frequency of OA of the hip was greater in the offspring compared with in the control patients (p = 0.05). Osteoarthritis of the knee was more common (p = 0.03) in the control patients (Table 3). As the affected parents of the offspring had severe end-stage OA of the hip, these data imply that predisposing genetic factors show some degree of joint specificity.
The male to female ratio was 1:1.2 in the offspring group and 1.2:1 in the control group. We did not find a greater rate of affected female patients as might have been suggested by previous studies, with a ratio of affected male to female patients of 1.13:1 (Table 4).
Six offspring were identified as having severe end-stage OA, three already had total joint replacements [two had THAs, one had total knee arthroplasty (TKA)], and one offspring was waiting for a TKA. Four control patients were found to have had severe end-stage knee disease.
There was a difference in the frequency of OA in the offspring compared with the control patients (35.2% versus 10.2%; p = 0.000001), which was reflected by an increased relative risk to the offspring of 3.5 (Table 5).
Evidence that there is a significant genetic component in the development of primary OA comes from epidemiologic, twin pair, sibling risk, and more recently, linkage analysis studies.
In 1941 Stecher13 reported that Heberden’s nodes were twice as common in mothers and three times as common in the sisters of affected white females than would be expected by chance alone. Kellgren et al7 found that polyarticular OA in first degree relatives of patients with radiologic OA was more than twice as common as would be expected by chance alone. Twin studies have been used effectively to highlight the contribution of genetics in the causation of OA. In 1996, Spector et al12 showed a heritability of 39–65% with a concordance rate of 0.64 in monozygous twins as opposed to 0.38 in dizygous twins. MacGregor et al11 did an additional twin study focusing on radiographic OA of the hip in female twins and found heritability of 50%. In 1997, a sibling risk study by Chitnavis et al3 showed that siblings of patients with end-stage primary OA had twice the risk of THA and five times the risk of TKA compared with control patients. Most epidemiologic studies involve only one generation with no evidence of transmission of OA to the next generation.
The current study clearly shows vertical transmission of OA from families with female sibling pairs with end-stage OA of the hip to their offspring. We found a 3.5 (95% confidence interval, 2.0–6.2) times (p = 0.000001) risk to the offspring in comparison to the control patients. The pattern of OA is more likely to involve the hip (p = 0.05), suggesting that genetic factors predispose to a particular form of OA.
Some families had multiply affected offspring whereas others had none affected, indicating that different genetic and environmental factors also are involved.
The offspring group was younger with an average age of 47 years. This cohort was generally of a preretirement age and may have found it difficult to take the time to visit our institution. This may have resulted in a selection bias with more symptomatic than asymptomatic offspring volunteering to participate. However, the offspring and control patients were closely matched for age, making this unlikely to have been a significant factor.
The clinical spectrum of OA is wide, and the significance of isolated radiologic signs is not certain. One-hundred forty-nine of 264 patients were not evaluated radiographically because they did not have any clinical signs or symptoms; however, a percentage of these patients will have radiologic signs of OA. We set standards for the definition of our affected patients based on symptoms and signs in combination with Grade 1 radiologic changes of OA to minimize these confounding factors as much as possible. Radiographic grading of OA in our patients by the main researchers (JS and AC) may be subject to intraobserver or interobserver error.
Factors such as acetabular dysplasia or slipped upper femoral epiphysis play a part in the development of OA of the hip. We assessed all radiographs for features of these conditions and found no apparent differences between the offspring and control patients.
Our study reinforces the evidence that genetic factors play a significant role in the development of OA. We investigated a selected group of families with an affected female sibling pair with end-stage primary idiopathic OA requiring THA. Our results showed that these offspring have a 3.5 times greater risk of having primary OA develop as compared with age-matched control patients, with most having OA of the hip. Linkage analysis, which is being done at this time, will determine whether known loci are involved in the vertical transmission.
The findings of this study add significant importance to the role of genetics in the cause of primary OA of the hip.
1. Burnett S, Hart D, Cooper C, Spector T: A Radiographic Atlas of Osteoarthritis. London, Springer 1994.
2. Chapman K, Mustafa Z, Dowling B, et al: Finer linkage mapping of primary hip osteoarthritis susceptibility on chromosome 11q in a cohort of affected female sibling pairs. Arthritis Rheum 46:1780–1783, 2002.
3. Chitnavis J, Sinsheimer J, Clipsham K, et al: Genetic influences in end-stage osteoarthritis: Sibling risks of hip and knee replacement for idiopathic osteoarthritis. J Bone Joint Surg 79B:660–664, 1997.
4. Cooper C, McAlindon T, Snow S, et al: Mechanical and constitutional risk factors for symptomatic knee osteoarthritis: Differences between medial tibiofemoral and patellofemoral disease. J Rheumatol 21:307–313, 1994.
5. Felson DT: Epidemiology of hip and knee osteoarthritis. Epidemiol Rev 10:1–28, 1988.
6. Kellgren JH, Lawrence JS: Radiological assessment of osteo-arthrosis. Ann Rheum Dis 16:494–502, 1957.
7. Kellgren J, Lawrence J, Bier F: Genetic factors in generalised osteo-arthritis. Ann Rheum Dis 22:237–255, 1963.
8. Lawrence JS: Osteoarthritis. In Lawrence JS(ed). Rheumatism in Populations. London, Heinemann Medical 1–31, 1977.
9. Loughlin J: Genetic epidemiology of primary osteoarthritis. Curr Opin Rheumatol 13:111–116, 2001.
10. Loughlin J: Genome studies and linkage in primary osteoarthritis. Rheum Dis Clin North Am 28:95–109, 2002.
11. MacGregor AJ, Antoniades L, Matson M, Andrew T, Spector TD: The genetic contribution to radiographic hip osteoarthritis in women: Results of a classic twin study. Arthritis Rheum 43:2410–2416, 2000.
12. Spector TD, Cicuttini F, Baker J, Loughlin J, Hart D: Genetic influences on osteoarthritis in women: A twin study. BMJ 312:940–943, 1996.
13. Stecher R: Heberden’s nodes: The incidence of hypertrophic arthritis of the fingers. N Engl J Med 222:300–308, 1941.