P12. Lumbosacral residual rod strains in long PSR constructs often exceed in vitro rod bending strains

Abstract

BACKGROUND CONTEXT Surface strain on instrumentation spanning the lumbosacral junction is often recorded during in vitro studies evaluating the biomechanical effects of different pedicle screw and rod (PSR) configurations. Rod strains recorded while releasing the rods from the screws after testing, much like strains recorded while securing the rods to screws prior to testing, are telling of the residual (additive) pre-strains that the rods are subjected to during multi-directional loading. PURPOSE The objectives of this study were to analyze lumbosacral rod strains from multiple rod locations (L5-S and S1-S2, posterior and lateral sides) during PSR L2-S1 vs L2-S2AI rod release, and to compare these to strains observed during multi-directional bending of the same constructs immediately before release. STUDY DESIGN/SETTING In vitro biomechanical study using human cadaveric specimens. PATIENT SAMPLE A total of 13 (L1-Pelvis) cadaveric spines (11F/2M, 58.7 ±6.1 yrs, DEXA 0.734 (±0.110) g/cm2) were studied. OUTCOME MEASURES Primary rod strains at L5-S and S1-2 on posterior and lateral sides. METHODS Thirteen cadaveric specimens were instrumented with ALIF at L5-S1 and PSR instrumentation spanning first L2-S1 and then L2-S2AI. Titanium rods were manually contoured with a French bender by a trained neurosurgeon to fit inside all poly-axial screw heads without significant reduction required, and then instrumented with 4 stacked rosette strain gauges (C2A-06-G1350-350/SP70) facing posteriorly (post) and laterally (lat) at both L5-S and S1-S2 (right side rod). Multi-location rod strains were recorded for both conditions (L2-S1 and then L2-S2AI, with full bilateral rod release and reattachment between conditions) while tested in a 6DOF robot under continuous dynamic loading to 7.5 Nm in flexion, extension, lateral bending (right and left), and axial rotation (right and left), followed by 400 N compression. In both cases after testing (L2-S1 and then L2-S2AI), rod strains were recorded during rod-from-screw release, performed sequentially level by level, caudal to cranial. Differences between rod release strains (from beginning to end) for the different conditions (including all gage locations: L5-S, S1-S2, lateral and posterior) were compared using 1-way ANOVA (p RESULTS The test max rod strains during in vitro multi-directional bending and compression were: 785 uE (L5-Spost - compression), 851 uE (S1-2post - compression), 429 uE (L5-Slat - right axial rotation), and 250 uE (S1-2lat - left axial rotation). The mean (±1 SD) rod release strains immediately following testing were: 426±374 uE [L5-Spost (S1)], 690±375 uE [L5-Spost (S2AI)], 256±147 uE [L5-Slat (S1)], 744±709 uE [L5-Slat (S2AI)], 921±822 uE [S1-2post (S2AI)], and 666±655 uE [S1-2lat (S2AI)] (Fig. 1, p=0.08). The number of cases when release strains after testing were greater than test max strains were: 17% [L5-Spost(S1)]; 31% [L5-Spost(S2AI)]; 46% [S1-2post(S2AI)]; 18% [L5-Slat(S1)]; 69% [L5-Slat(S2AI); and 77% [S1-2lat(S2AI)]. CONCLUSIONS Securing contoured rods to pedicle screws while instrumenting cadaveric spines with L2-S1 and L2-S2AI PSR instrumentation result in considerable lumbosacral rod strains on both the posterior and lateral sides of the rods, with magnitudes often (up to 77% of cases on the lateral side with S2AI) exceeding those recorded during multi-directional loading of the same constructs. Strains experienced during activities of daily living are additive to pre-strains observed during attachment, with high overall strains increasing the risk of 1) rod fracture through fatigue mechanisms and 2) pedicle screw loosening. Thus, these results have clinical implications with regard to rod fit and contouring. FDA DEVICE/DRUG STATUS PSR (Approved for this indication)