Gait-simulating fatigue loading analysis and sagittal alignment failure of spinal pelvic reconstruction after total sacrectomy: comparison of 3 techniques

Object: Reconstruction after total sacrectomy is a critical component of malignant sacral tumor resection, permitting early mobilization and maintenance of spinal pelvic alignment. However, implant loosening, graft migration, and instrumentation breakage remain major problems. Traditional techniques have used interiliac femoral allograft, but more modern methods have used fibular or cage struts from the ilium to the L-5 endplate or sacral body replacement with transiliac bars anchored to cages to the L-5 endplate. This study compares the biomechanical stability under gait-simulating fatigue loading of the 3 current methods.

Methods: Total sacrectomy was performed and reconstruction was completed using 3 different constructs in conjunction with posterior spinal screw rod instrumentation from L-3 to pelvis: interiliac femur strut allograft (FSA); L5-iliac cage struts (CSs); and S-1 body replacement expandable cage (EC). Intact lumbar specimens (L3-sacrum) were tested for flexion-extension range of motion (FE-ROM), axial rotation ROM (AX-ROM), and lateral bending ROM (LB-ROM). Each instrumented specimen was compared with its matched intact specimen to generate an ROM ratio. Fatigue testing in compression and flexion was performed using a custom-designed long fusion gait model.

Results: Compared with intact specimen, the FSA FE-ROM ratio was 1.22 ± 0.60, the CS FE-ROM ratio was significantly lower (0.37 ± 0.12, p < 0.001), and EC was lower still (0.29 ± 0.14, p < 0.001; values are expressed as the mean ± SD). The difference between CS and EC in FE-ROM ratio was not significant (p = 0.83). There were no differences in AX-ROM or LB-ROM ratios (p = 0.77 and 0.44, respectively). No failures were noted on fatigue testing of any EC construct (250,000 cycles). This was significantly improved compared with FSA (856 cycles, p < 0.001) and CS (794 cycles, p < 0.001).

Conclusions: The CS and EC appear to be significantly more stable constructs compared with FSA with FE-ROM. The 3 constructs appear to be equal with AX-ROM and LB-ROM. Most importantly, EC appears to be significantly more resistant to fatigue compared with FSA and CS. Reconstruction of the load transfer mechanism to the pelvis via the L-5 endplate appears to be important in maintenance of alignment after total sacrectomy reconstruction.