Li Peiyu, Li Jie, Kiram Abdukahar, Tian Zhen, Sun Xing, Qin Xiaodong, Shi Benlong, Qiu Yong, Liu Zhen, Zhu Zezhang
Division of Spine Surgery, Department of Orthopedic Surgery, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China.
Division of Spine Surgery, Department of Orthopedic Surgery, Nanjing Drum Tower Hospital, Clinical College of Nanjing University of Chinese Medicine, Nanjing, China.
Spine J. 2025 Feb;25(2):347-358. doi: 10.1016/j.spinee.2024.09.029. Epub 2024 Sep 28.
The global alignment and proportion (GAP) score was developed to predict mechanical complications (MCs) after adult spinal deformity surgery but showed limited sensitivity in the Asian population. Considering variations in sagittal parameters among different ethnic groups, our team developed the ethnicity-adjusted GAP score according to the spinopelvic parameters of 566 asymptomatic Chinese volunteers (C-GAP score). Notably, degenerative scoliosis (DS) patients with MCs following corrective surgery have more severe paraspinal muscle degeneration. For DS patients with various sagittal alignments, the unevenly distributed degeneration of paraspinal muscle may exert different influences on MC occurrence and largely affect the accuracy of the C-GAP score in clinical assessment. Therefore, incorporating paraspinal muscle degeneration indices within the C-GAP score may improve its accuracy in predicting MC occurrence.
We aimed to clarify the influence of paraspinal muscle degeneration on the C-GAP score predicting MC occurrence following DS surgery and modify the C-GAP score with paraspinal muscle degeneration parameters.
A retrospective case-control study.
A total of 107 adult degenerative scoliosis patients.
Demographic information, postoperative sagittal spinopelvic parameters, the GAP score, the C-GAP score, and paraspinal muscle degeneration parameters.
A total of 107 DS patients undergoing posterior spinal fusion surgery (≥4 vertebrae) with a minimum of 2 years follow-up (or experiencing MCs within 2 years) were retrospectively reviewed. Their C-GAP score was calculated based on our previous study and patients were divided into 3 C-GAP categories, "proportioned" (P), "moderately disproportioned" (MD), and "severely disproportioned" (SD). Relative cross-sectional area (cross-sectional area of muscle-disc ratio×100, rCSA) and fat infiltration rate, FI% at L1/2, L2/3, L3/4, and L4/5 discs were quantitatively evaluated using magnetic resonance imaging (MRI). In each C-GAP category, patients were additionally divided into the MC group and the non-MC group to analyze their paraspinal muscle degeneration. A multivariable logistic regression model consisting of the CSA-weighted average FI% (total FI%) and the C-GAP score, C-GAPM was constructed. The area under the curve (AUC) of the receiver operating characteristic (ROC) curves was used to evaluate the predictability of the GAP score, the C-GAP score, FI%, and C-GAPM. This project was supported by the National Natural Science Foundation of China (No.82272545) and Special Fund of Science and Technology Plan of Jiangsu Province (No.BE2023658).
For all 107 patients, FI% at L1/2, L2/3, L3/4, and L4/5 discs and the total FI% of the MC group (n=32) were significantly higher than those of the non-MC group (n=75). The MC rates of 3 original GAP categories, P, MD, and SD categories were 25.00% (6/24), 27.03%(10/37), and 34.78% (16/46) (χ2=0.944, p=.624). Based on the C-GAP score, the MC rates of the P, MD, and SD categories were 11.90% (5/42), 34.69% (17/49), and 62.50% (10/16), showing significant differences (χ2=15.137, p=.001). In the C-GAP MD category, compared with the non-MC group (n=32), the MC group (n=17) has a higher total FI% (26.16(22.95, 34.00) vs 22.67(16.39, 27.37)), p=.029). A similar trend was identified in the C-GAP SD category (34.79±11.56 vs 19.00±5.17, p=.007), but not in the C-GAP P category (25.09(22.82, 32.66) vs 24.66(17.36, 28.63), p=.361). The AUC of the GAP score, the C-GAP score, the total FI%, and C-GAPM were respectively 0.601, 0.722, 0.716, and 0.772.
Paraspinal muscle degeneration exerts a significant effect on the occurrence of MC in the C-GAP MD, SD instead of P category. The integration of paraspinal muscle FI% with the C-GAP score (C-GAPM) enables a more accurate prediction of MCs following DS surgery. Surgeons should pay adequate attention to paraspinal muscle degeneration during surgical planning and postoperative management for patients in the C-GAP MD and SD categories.
全球对线与比例(GAP)评分旨在预测成人脊柱畸形手术后的机械并发症(MCs),但在亚洲人群中敏感性有限。考虑到不同种族矢状位参数的差异,我们的团队根据566名无症状中国志愿者的脊柱骨盆参数制定了种族调整后的GAP评分(C-GAP评分)。值得注意的是,矫正手术后发生MCs的退变性脊柱侧凸(DS)患者的椎旁肌退变更为严重。对于矢状位对线不同的DS患者,椎旁肌不均匀分布的退变可能对MC的发生产生不同影响,并在很大程度上影响C-GAP评分在临床评估中的准确性。因此,将椎旁肌退变指标纳入C-GAP评分可能会提高其预测MC发生的准确性。
我们旨在阐明椎旁肌退变对预测DS手术后MC发生的C-GAP评分的影响,并使用椎旁肌退变参数对C-GAP评分进行修正。
一项回顾性病例对照研究。
共107例成年退变性脊柱侧凸患者。
人口统计学信息、术后矢状位脊柱骨盆参数、GAP评分、C-GAP评分和椎旁肌退变参数。
回顾性分析107例行后路脊柱融合手术(≥4个椎体)且至少随访2年(或在2年内发生MCs)的DS患者。根据我们之前的研究计算他们的C-GAP评分,并将患者分为3个C-GAP类别,即“比例合适”(P)、“中度比例失调”(MD)和“严重比例失调”(SD)。使用磁共振成像(MRI)定量评估L1/2、L2/3、L3/4和L4/5椎间盘水平的相对横截面积(肌肉-椎间盘横截面积比值×100,rCSA)和脂肪浸润率FI%。在每个C-GAP类别中,患者进一步分为MC组和非MC组,以分析其椎旁肌退变情况。构建了一个由CSA加权平均FI%(总FI%)和C-GAP评分组成的多变量逻辑回归模型C-GAPM。采用受试者操作特征(ROC)曲线下面积(AUC)评估GAP评分、C-GAP评分、FI%和C-GAPM的预测能力。本项目得到了国家自然科学基金(No.82272545)和江苏省科技计划专项基金(No.BE2023658)的支持。
对于所有107例患者,MC组(n = 32)在L1/2、L2/3、L3/4和L4/5椎间盘水平的FI%以及总FI%均显著高于非MC组(n = 75)。原始GAP分类中P、MD和SD类别的MC发生率分别为25.00%(6/24)、27.03%(10/37)和34.78%(16/46)(χ2 = 0.944,p = 0.624)。基于C-GAP评分,P、MD和SD类别的MC发生率分别为11.90%(5/42)、34.69%(17/49)和62.50%(10/16),差异有统计学意义(χ2 = 15.137,p = 0.001)。在C-GAP MD类别中,与非MC组(n = 32)相比,MC组(n = 17)的总FI%更高(26.16(22.95, 34.00) 对 22.67(16.39, 27.37)),p = 0.029)。在C-GAP SD类别中也发现了类似趋势(34.79±11.56对19.00±5.17,p = 0.007),但在C-GAP P类别中未发现(25.09(22.82, 32.66)对24.66(17.36, 28.63),p = 0.361)。GAP评分、C-GAP评分、总FI%和C-GAPM的AUC分别为0.601、0.722、0.716和0.772。
椎旁肌退变对C-GAP MD、SD类别而非P类别中MC的发生有显著影响。将椎旁肌FI%与C-GAP评分(C-GAPM)相结合能够更准确地预测DS手术后的MCs。对于C-GAP MD和SD类别的患者,外科医生在手术规划和术后管理过程中应充分关注椎旁肌退变情况。