Mode of inheritance of hand osteoarthritis in ethnically homogeneous pedigrees

Human Biology, Dec 2002 by Livshits, Gregory, Kalichman, Leonid, Cohen, Zvi, Kobyliansky, Eugene

Abstract The aim of the present study was to investigate the extent and mode of inheritance of hand osteoarthritis (OA) using a large sample of ethnically homogeneous pedigrees. Two types of segregation analysis (SA) models were examined. Type I models used the data adjusted for potential significant covariates, particularly age and sex, prior to genetic analysis. Type II models incorporated effects of the potential covariates into major gene penetrance functions, permitting an account of the genotype covariate-specific effect on study variables. The results of this study strongly supported the hypothesis of a major gene effect and additional multifactorial component. The best-fitting model was the Mendelian one with an additive type of inheritance. The estimates obtained using the standard three-factor variance decomposition analysis suggest that age (72.8%) and major gene (14.5%) are the main sources of interindividual differences in the development of hand OA. The contribution of the putative major gene on age- and sex-adjusted OA phenotype variation was 55% in the present study.

Osteoarthritis (OA) is a process of joint disintegration that is characterized by changes in the synovium, cartilage loss, and reactive new bone formation at joint margins, the latter primarily by proliferation of osteoarticular tissue in the capsular attachments and subchondral bone marrow (Sokoloff 1987; Macfarlane et al. 1991; Buckwalter et al. 2000). Overall, 60% of the population over 35 years of age report arthritic symptoms varying from occasional joint stiffness to severe restriction of motion with persistent deep pain (Buckwalter and Martin 1995). Recent studies have demonstrated a clear genetic effect on interindividual variation of age-adjusted radiographic OA of the hands, knees (Spector et al. 1996; Chitnavis et al. 1997; Hirsch et al. 1998; Felson et al. 1998; Bijkerk et al. 1999), hip joints (Chitnavis et al. 1997), and the so-called "primary generalized OA" (Williams and Jimenez 1995). The estimated genetic effects ranged from 27% (Chitnavis et al. 1997) to 65% (Spector et al. 1996), depending upon the type of relatives chosen for use in the study and the skeletal area. The great majority of the studies used linear regression techniques (mostly familial correlations) to determine the extent of a possible genetic determination of OA. We are aware of only one publication (Felson et al. 1998) in which a complex segregation analysis (SA) was undertaken to search for a pattern of OA inheritance. This study found discernible genetic contributions to OA, which confirmed the results of the other above-mentioned studies. Furthermore, those investigators provided evidence for a major recessive gene effect as well as a residual multifactorial component to be acting as a mode of intergenerational transmission of OA. Spector et al. (1999), however, pointed out some problematic aspects in the research of Felson et al.: (1) the lines of evidence of genetic heterogeneity by sex were vague; (2) aggregated OA scores for both hands and knees were used (the development of OA can be caused by different factors in these two sites); (3) heavily skewed data that can mimic the effect of a major gene were used in the SA.

The aim of our study was to investigate the extent and mode of inheritance of hand OA by using a large sample of ethnically homogeneous pedigrees. Two types of genetic models were employed. The type I models examined the data adjusted for age and sex variation prior to SA. This type can be directly compared with the analyses of Felson et al. (1998). The type II models incorporated the effect of covariates into an SA (Elston 1981; Ginsburg and Livshits 1999) and allow one to address the concerns raised by Spector et al. (1999).

Materials and Methods

The present research involved radiographic assessment of hand bones, anthropological measurements, and interviewing of subjects. All examinations, measurements, and interviews of the participants in the study were carried out only according to their agreement. All the study subjects signed informed-consent documents before entering the respective projects, and the latter were approved by the Helsinki Ethics Committee of Tel-Aviv University.

Sample. The subjects of the present study come from Chuvasha and live in many small villages in the Chuvasha and Bashkirostan Autonomies, The Russian Federation. The study cohort included 629 males aged 18-89 years (mean 46.3) and 561 females aged 18-90 years (mean 48.2). The data were retrieved from 1190 family members belonging to 349 nuclear families ranging from 3 to 7 individuals per family. Some of the nuclear families were combined into 3 four-generation complex pedigrees and 21 into three-generation complex pedigrees, a feature that bears significant importance for an SA. Table 1 presents a breakdown of the research sample, including the data on number of pairs of first- and second-degree relatives. All studied families were recruited and enrolled randomly, i.e., regardless of the outcome of any of the measured variables.