Risk factors of implant-related complications in adult degenerative scoliosis with posterior long segment internal fixation / 中华骨科杂志

ABSTRACT

Objective@#To analyze the risk factors of implant-related complications in adult patients with adult degenerative scoliosis (ADS) who underwent long-level internal fixation.@*Methods@#This was a retrospective study that analyzed 99 cases of adult degeneration scoliosis patients who underwent long level posterior pedicle screw fixation in our hospital between June 2013 and January 2016. The internal fixation and related complications were evaluated by measuring and analyzing the radiographic data of the postoperative X-ray films. The timepoints of measurement were pre-operation, before discharge, half a year, 1 year, when complications occur and the final follow-up after operation. Implant-related complications included proximal junctional kyphosis (PJK), proximal junctional failure (PJF), distal junctional kyphosis/failure (DJK/DJF), rod breakage in addition to other radiographic implant-related complications (that were not related to PJF) such as screw loosening, breakage or pullout, or interbody graft and hook or set screw dislodgements seen on follow-up radiographs. The incidence of postoperative implant-related complications was counted. All the cases included in the study were divided into the complication group and the non-complication group. Statistical difference between groups at different follow-up time point was analyzed. Potential risk factors were identified using univariate testing. Multivariate Logistics regression was used to analyze the independent risk factors for implant-related complications. The postoperative functional scores were assessed using the Oswestry Disability Index (ODI), Japanese Orthopaedic Association Scores (JOA), Visual Analogue Scale (VAS), and Lumbar Stiffness Disability Index (LSDI). Functional scores were tested using group t tests. Patients were divided into groups according to PI-LL <10 °, 10°-20 ° and > 20 °. The preoperative and postoperative radiographical parameters and clinical function score among each groupwere compared.The best PI-LL matching value was verified by analyzing the effect of long-segment fusion orthopedics on ADS.@*Results@#Ninety-nine ADS patients who underwent long level posterior fixation were included. The incidence of patients with mechanical complications was 30.3%. Univariate analysis showed that chronic risk factors of postoperative implant-related complications after surgery of ASD included diabetes (OR=3.52, P=0.001) and blood transfusion (OR=2.61, P=0.030); surgical risk factor isosteotomy (OR=4.33, P=0.000); preoperative imaging risk factor was preoperative SVA (OR=1.03, P=0.000); the risk factors for increased risk of the implant-related complications included anemia (OR=1.17, P=0.810), cardiac complications (OR=1.80, P=0.290) and hospital stay (OR=1.11, P=0.110). Independent predictors identified on multivariate Logistics regression modeling included osteotomy (OR=3.05, P=0.032), and preoperative SVA (OR=1.03, P=0.007). The radiographical parameters and clinical function scores of the PI-LL 10°-20° group were better than or partially superior to those of the PI-LL<10° group and the PI-LL>20° group. The postoperative SVA of the PI-LL 10°-20° group was significantly lower than that of the PI-LL<10° group (t=2.399, P=0.020) and the PI-LL>20° group (t=-3.074, P=0.005). The incidence of implant-related complications in the PI-LL 10°-20° group was significantly lower than that in the PI-LL<10° group (t=1.584, P=0.003). Survival analysis showed that the PI-LL 10°-20° group was significantly better than the PI-LL<10° group (χ2=7.782, P=0.005), while the PI-LL 10°-20° group had better survival than PI-LL>20° group, althoughthatwas not statistically significant (χ2=2.542, P=0.111).@*Conclusion@#Risk factors of postoperative implant-related complications after surgery of ASD included osteotomy and preoperative SVA. Patients with one or more of these risk factors should be informed of the risk increase with informed consent. Patients with PI-LL between 10 ° and 20 ° had better postoperative radiographical parameters and clinical functional scores. They should be optimized preoperatively and followed up closely during the postoperative period.