Article révisé par les pairs
Résumé : We tested the hypothesis that instrumentation applied across two levels of the lumbar spine inhibits rotation at these levels and that this inhibition is accompanied by more rotation at the levels above and below. The rotation in the transverse plane of each of the five lumbar vertebrae of 10 cadaveric lumbar spines was produced by dead weights applied to a torsion bar attached at T12. The lumbar spines were first measured without instrumentation being applied. Three different types of instrumentation were then applied in turn across two levels of each lumbar spine: AO internal fixation from L2 to L4, lumbosacral Louis plates from L4 to S1, and Cotrel-Dubousset (CD) pedicular screws and rods from L2 to L4 and then from L4 to SI. For each of these four, a set of measurements of rotation in the transverse plane was taken. The Louis plates allowed less rotation across L4-S1 than CD, although the result was not significant. The AO fixator applied across L2-L4 allowed significantly less rotation across L2-L4 than CD. The AO fixator was compared with the spine with no instrumentation to determine if less rotation across L2-L4 was associated with more rotation across LI-L2 and L4-L5; the differences were significant (p < 0.05). © 1995 Raven Press, Ltd., New York.