Preservation of short-term cervical posterior muscle volume and cervical alignment: a comparison between three types of modified muscle-sparing laminoplasty and conventional open-door laminoplasty for multilevel degenerative cervical myelopathy.

BACKGROUNDS

Several modified muscle-sparing laminoplasty (LP) techniques have been developed to prevent axial symptoms in the treatment of multilevel degenerative cervical myelopathy (MDCM). However, the postoperative changes in posterior muscle volume (PMV) and cervical alignment following these procedures remain controversial. This study aimed to compare the short-term postoperative changes in PMV and cervical alignment between conventional LP and three types of muscle-sparing LPs: unilateral muscle-preservation laminoplasty (UL), double-door laminoplasty (DL), and intermuscular "raising roof" laminoplasty (RL).

METHODS

Consecutive MDCM patients who underwent LP, DL, UL, or RL between February 2022 and May 2022 at a same ward were enrolled. Baseline data and surgical characteristics were collected. Preoperative and postoperative PMV were semiautomatically segmented and evaluated, and cervical alignment was measured. The PMV loss ratio was calculated as the change in PMV divided by the preoperative PMV.

RESULTS

A total of 79 MDCM patients were included in this study (LP: 20, DL: 20, UL:23, RL:16). No significant differences were observed in the preoperative demographic data, surgical characteristics and radiological variables. The preoperative C2-7 cobb angles of four groups were 11.59 ± 8.80, 9.54 ± 11.18, 11.08 ± 11.32 and 10.75 ± 10.90, respectively(p = 0.962). The preoperative PMV were 2.726 ± 0.79, 2.607 ± 0.752, 2.808 ± 0.724, 2.686 ± 0.674*10mm, respectively(p = 0.802). At 1-year follow-up, all four groups showed favorable and comparable JOARR(p = 0.443). The postoperative PMV of four groups were 2.430 ± 0.68, 2.355 ± 0.621, 2.416 ± 0.667, and 2.602 ± 0.666 *10mm(p = 0.606), and the PMV loss ratio were 10.0%±12.9%, 8.7%±9.6%, 13.6%±10.4%, and 2.8%±11.0%, respectively (p = 0.033). The modified LP groups did not present significant difference in PMV loss or cervical alignment parameters compared with LP. After adjustment, the RL group presented significant lower PMV loss ratio (p = 0.035) compared with LP. Multivariate regression revealed that the loss in cervical alignment was correlated to the PMV loss ratio.

CONCLUSIONS

The DL group has significant lower postoperative CL-E loss than LP. After adjustment, the RL procedure demonstrated better efficacy in PMV preservation compared to conventional LP. The loss of cervical lordosis at neutral and extension positions was associated with the PMV loss ratio.