Classification of structural properties of human bones is essential for understanding differential response to loading and fracture risk. Ribs, in particular, are frequently fractured during motor vehicle crashes and are linked to high mortality rates, especially in elderly individuals. While many studies describe variation of bone properties with respect to age and sex differences, these parameters explain only a small amount of variability in rib properties. The focus of this study was to investigate the ability of geometry to predict the response of ribs to dynamic loading.

A total of 122 complete mid-level ribs from 76 fresh post-mortem human subjects (15-99 years of age, 20 females, 56 males) were excised and their span length (Sp.Le) and curve length (Cv.Le) were measured from head to costochondral junction. The ribs were tested in a custom-built pendulum fixture simulating a dynamic frontal impact to the thorax. Peak force (FPEAK) was defined as the maximum force in the primary loading direction prior to failure. Linear structural stiffness (K) was calculated as the slope of 20-80% of the elastic portion of the force-displacement curve. Sections were removed from mid-shaft of the rib and thin-sections were prepared. Cross-sectional microscopic images were obtained at 40x magnification with an Olympus VS120 slide scanner. Measurements were manually made in cellSens Dimension® imaging software (Olympus Corporation) to obtain total subperiosteal area (Tt.Ar) and cortical area (Ct.Ar). Section modulus was calculated for the pleural and cutaneous cortices independently (ZPLE and ZCUT, respectively) in ImageJ software using a customized macro.

Tt.Ar, Ct.Ar, and Z all have positive relationships with FPEAK and K. When considering only cross-sectional geometric values, F was best predicted by Ct.Ar (p-R²=0.673) and ZPLE (p-R²=0.668) and K was best predicted by Tt.Ar (p-R²= 0.437) and ZPLE (p-R²=0.494). Robusticity, an index used to establish the relationship between longitudinal growth and transverse expansion, allows for the combination of gross and cross-sectional geometric parameters. This index was calculated in four ways:​ Tt.Ar/Sp.Le, Tt.Ar/Cv.Le, Ct.Ar/Sp.Le, and Ct.Ar/Cv.Le, and these variations were individually evaluated in terms of their ability to predict FPEAK and K. Preliminary findings indicate that FPEAK is best predicted by Ct.Ar/Sp.Le (p-R²=0.709), but K is best predicted by Tt.Ar/Cv.Le (p-R²=0.784).

Identifying accurate predictors of structural properties of ribs may improve the ability to assess fracture risk. Additionally, detailed cross-sectional properties can contribute to improved physical and computational models, injury criteria, and clinical assessments of bone fragility