Livshits G, Yakovenko K, Kobyliansky E
Department of Anatomy and Anthropology, Sackler Faculty of Medicine, Tel Aviv University, Ramat Aviv, Tel Aviv, 69978, Israel.
Bone. 2003 Feb;32(2):191-8. doi: 10.1016/s8756-3282(02)00954-7.
Despite the obvious epidemiological significance of bone size (BS) and geometry (BG) traits as risk factors for osteoporotic fracture, very little is still known concerning the extent of their genetic determination. In the present paper we report the results of quantitative genetic analysis of a number of BG and BS indices, as well as of BMD measurements, obtained on a large pedigree-based sample (296 nuclear families, 1208 individuals) of plain hand radiographs. The families studied were all ethnically Caucasians (Chuvasha) living in small villages along the Volga River (Russia). The sample consisted of 636 men and 572 women, aged 18-91 years. To assess hand bone size we used the outcome of principal component analysis conducted on 48 measurements of metacarpal bones and proximal phalanges (PC-BS). Two BG indices, average metacarpal cortical index and breaking bending resistance index (BBRI), also measured on metacarpal and proximal phalanges were used. Again the outcome of the principal component PC-BBRI was examined in the genetic analysis. PC-BS measurements strongly correlated with body length (r = 0.75, P < 0.001) and weight (r = 0.39, P < 0.001), suggesting that they indeed reflected hand skeleton size. Familial correlations for all studied traits, adjusted for covariates (sex, age, etc.), were all highly significant statistically. For example, parent/offspring correlations ranged between 0.248 (P < 0.001) for phalangeal BMD and 0.385 (P < 0.001) for PC-BBRI. Maximum likelihood estimates of the variance component analysis confirmed these results, indicating that approximately 58 to 66% of the residual variance of the studied traits was attributable to genetic effects. Bivariate analysis clearly revealed that while genetic variation of the phalangeal BMD was independent of the genetic effects influencing hand BS and BG, the latter two were strongly interrelated. A substantial proportion of PC-BS and PC-BBRI variation was due to shared genetic (r(G) = 0.468 +/- 0.063) and environmental (r(E) = 0.704 +/- 0.052) factors.
尽管骨大小（BS）和几何形状（BG）特征作为骨质疏松性骨折的危险因素具有明显的流行病学意义，但关于其遗传决定程度仍知之甚少。在本文中，我们报告了对大量基于家系的样本（296个核心家庭，1208人）的手部X线平片进行的多项BG和BS指数以及骨密度测量的数量遗传学分析结果。所研究的家庭均为居住在伏尔加河沿岸（俄罗斯）小村庄的白种人（楚瓦什人）。样本包括636名男性和572名女性，年龄在18 - 91岁之间。为了评估手部骨大小，我们使用了对掌骨和近端指骨进行48次测量后进行主成分分析的结果（PC - BS）。还使用了在掌骨和近端指骨上测量的两个BG指数，即平均掌骨皮质指数和抗弯强度指数（BBRI）。在遗传分析中再次检查了主成分PC - BBRI的结果。PC - BS测量值与身高（r = 0.75，P < 0.001）和体重（r = 0.39，P < 0.001）密切相关，表明它们确实反映了手部骨骼大小。对所有研究性状进行协变量（性别、年龄等）调整后的家族相关性在统计学上均高度显著。例如，指骨骨密度的亲子相关性在0.248（P < 0.001）之间，PC - BBRI的亲子相关性在0.385（P < 0.001）之间。方差成分分析的最大似然估计证实了这些结果，表明所研究性状的剩余方差中约58%至66%可归因于遗传效应。双变量分析清楚地表明，虽然指骨骨密度的遗传变异独立于影响手部BS和BG的遗传效应，但后两者密切相关。PC - BS和PC - BBRI变异的很大一部分是由于共享的遗传（r(G) = 0.468 +/- 0.063）和环境（r(E) = 0.704 +/- 0.052）因素。
