Predictive value of postoperative hypolordosis in the occurrence of adjacent segment disease after lumbar fusion surgery / 中华骨科杂志

ABSTRACT

Objective:To investigate the value of relative lumbar lordosis (RLL) and lumbar distribution index (LDI) in predicting the occurrence of adjacent segment disease (ASDis) after lumbar fusion surgery.Methods:This study retrospectively reviewed 163 consecutive patients (58 males and 105 females) who had undergone lumbar fusion and had been followed over 2 years,with an average age of 58.7 years; among them, 74, 71, and 18 patients had undergone fusion of one-level, two-level, and three-level, respectively. They were divided into the non-ASDis group and ASDis group based on the presence of ASDis or not. Pre- and post-operative spinopelvic parameters were measured on the upright lateral radiographs. RLL was calculated as measured lumbar lordosis (LL) minus ideal LL, and LDI was calculated as the ratio of postoperative low lumbar lordosis (LLL) to LL. Each parameter was stratified into 1 "aligned" subgroup and 3 "disproportioned" subgroups in accordance with values. Cochran-Armitage test of trend andlogistic analysis were performed to investigate the association between these two parameters and the occurrence of ASDis.Results:The average follow-up duration after initial surgery was 46±14 months (range, 25 to 134 months). Twenty-four (14.7%) patients were diagnosed as ASDis. The age ( t=3.13, P=0.002) and the proportion of 2-level and 3-level fusion (χ 2=10.27, P=0.006) in the ASDis group were significantly higher than those in the non-ASDis group ( P<0.05). There were no statistically significant differences between groups with respect to other general data. The ratios of moderate and severe hypolordosis of RLL were significantly higher in the ASDis group than that in the non-ASDis group (χ 2=16.92, P<0.001). There was also a significant linear trend with higher degree of hypolordosis being associated with higher rates of ASDis. However, distribution of four statuses of LDI did not differ statistically between groups. After controlling the confounders, the logistic regression analysis revealed that age, odd ratio ( OR)=1.07, 95% CI (1.01, 1.13), P=0.018), moderate[ OR=4.34, 95% CI (1.03, 18.41), P=0.046] and severe hypolordosis [ OR=11.64, 95% CI (1.30, 104.49), P=0.028] were significantly associated with the occurrence of ASDis. Conclusion:A significant association between postoperative RLL and occurrence of ASDis after lumbar fusion surgery were detected. Setting surgical goals according to RLL may help reduce the ASDis rate. However, LDI is not identified to be predictive factors of the occurrence of ASDis.