Effect of decompression range on decompression limit of cervical laminoplasty

BACKGROUNDS: Cervical posterior decompression surgery is used to relieve ventral compression indirectly by incorporating a backward shift of the spinal cord, and this indirect decompression is bound to be limited. This study aimed to determine the decompression limit of posterior surgery and the effect of the decompression range. METHODS: We retrospectively reviewed the data of 129 patients who underwent cervical open-door laminoplasty through 2008 to 2012 and were grouped as follows: C4–C7 (n = 11), C3–C6 (n = 61), C3–C7 (n = 32), and C2–C7 (n = 25). According to the relative location of spinal levels within a decompression range, the type of decompression at a given level was categorized as external decompression (ED; achieved at the levels located immediately external to the decompression range margin), internal decompression (ID; achieved at the levels located immediately internal to the decompression range margin), and central decompression (CD; achieved at the levels located in the center, far from the decompression range margin). The vertebral–cord distance (VCD) was used to evaluate the decompression limit. The C2–C7 angle and VCD on post-operative magnetic resonance images were analyzed and compared between groups. The relationship between VCD and decompression type was analyzed. Moreover, the relationship between the magnitude of the ventral compressive factor and the probability of post-operative residual compression at each level for different decompression ranges was studied. RESULTS: There was no significant kyphosis in cervical curvature (> −5°), and there was no significant difference among the groups (F = 2.091, P = 0.105). The VCD of a specific level depended on the decompression type of the level and followed this pattern: ED < ID < CD (P < 0.05). The decompression type of a level was sometimes affected by the decompression range. For a given magnitude of the ventral compressive factor, the probability of residual compression was lower for the group with the larger VCD at this level. CONCLUSIONS: Our study suggests that the decompression range affected the decompression limit by changing the decompression type of a particular level. For a given cervical spinal level, the decompression limit significantly varied with decompression type as follows: ED < ID < CD. CD provided maximal decompression limit for a given level. A reasonable range of decompression could be determined based on the relationship between the magnitude of the ventral compressive factor and the decompression limits achieved by different decompression ranges.