PMC

(A) Overview of the in situ loading set-up at TOMCAT beamline, SLS, PSI, Switzerland. (B) Custom made loading device used to pull the screw out of the tibia (sample placement is illustrated without surrounding chamber). The blue arrow indicates the pull-out direction.

Maximum force and stiffness (A,B) presented as box-and-whisker plots showing the range of the data, the quartiles, and the median; ^*represents p < 0.05. Significant (p < 0.05) correlations obtained between stiffness (C,E) and maximum force (D,F) with BV/TV and screw insertion are shown with Spearman correlation coefficient ρ, p-value and R-squared value of linear regression curve.

(A) Screw insertion parameters showing distance (d) between the screw and the tibial plateau as well as tilt (t) measured on a 2D radiograph. (B) Trabecular ROI started 1 mm proximal to the screw edge and ended 1 mm distal to the screw edge (i). ROI included trabecular bone and excluded the cortex as drawn in yellow in original (ii) and segmented (iii), as exemplified in three transversal cuts from the unloaded scan.

Typical crack patterns from 2D slice cuts (A) from two steps before rupture (left) to the step at or immediately after rupture (right). Yellow arrows indicate the main cracks. Data in (A) is presented as GIF in Supplementary Data Sheets (– for respectively Crack Type 1, Type 2, and Type 3 samples). (B) Force vs. displacement curves during in situ pullout of all samples in each crack group, where the dark blue curves are control samples and the light orange curves are BMP + Za samples. Crack Type 1 showed rupture close to screw mainly inside trabecular bone. Crack Type 2 failed through large cortical cracks. Crack Type 3 indicates that failure started presumably away from the screw and propagated toward the interface.

Contact between screw threads and cortex (A) and bone volume fraction in ROI defined in Figures (B) for each treatment group presented with box-and-whisker plots showing the range of the data, the quartiles, and the median; ^*represents p < 0.05. In insert are illustrated the cutting planes used in (C,D): transversal (i, full line) and longitudinal (ii, dotted line). (C) Partial cuts of cortical bone (beige) segmentation to visualize the contact with the screw (light red: whole contact, dark red: with threads) for a typical sample from each treatment group. (D) Cuts of raw scans to visualize trabecular bone in the same samples. Data in (C) is presented in 3D as (Supplementary Data Sheets – for respectively control, BMP and BMP + Za samples).