| Objective:To summarize the clinical features of thoracic spinal stenosis(TSS),analysis the mid-long term efficacy of posterior surgical treatment for TSS and discuss the clinical prognostic factors and risk factors related to surgical complications.Method:Retrospective cohort study.Cases of TSS treated with posterior decompression and fusion surgery in the Department of Orthopedics of our hospital from January 2007 to December 2020 were collected.The EMRS system was used to collect the baseline data,surgical information and complications of all cases.The cross-section area and the canal diameter occupied ratio were measured by electronic imaging system.The modified Japanese Orthopaedic Association score(m JOA)was used to evaluate the clinical efficacy preoperatively,5 days postoperatively and the last follow-up.Describe the population characteristics of TSS and follow up the mid-long term outcomes of posterior decompression and fusion surgery for TSS.The 22 factors that may affect the curative effect were compared between groups and univariate analysis.Binary logistic regression was applied to reanalyze the statistically significant factors in the univariate analysis results,and the independent risk factors affecting the mid-long term efficacy of TSS patients were obtained.The cases were divided into complication group and non-complication group according to whether complications occurred.Clinical,surgical and imaging datas were compared between the two groups.The indicators with statistical difference in univariate analysis and clinical significance were included into multivariate binary logistic regression and the risk factors of complications were obtained.Result:Part I:Clinical features1.General data: Based on the inclusion and exclusion criteria,123 patients with TSS were included in this study,including 70 males(56.91%)and 53 females(43.09%),with a mean age of 54.17±9.65 years.The median duration from initial onset to clinical diagnosis was 12 months.2.Main etiologies of TSS: According to the classification of pathogenic factors,there were 92 cases(74.80%)of the ossification of ligamentum flavum(OLF),9 cases(7.32%)of the ossification of posterior longitudinal ligament(OPLL),13 cases(10.57%)of the thoracic disc herniation(TDH),7 cases(5.69%)of the OLF combined with TDH,2 cases(1.63%)of the OLF combined with OPLL.3.Symptoms: There were 79 cases(64.23%)complained of "weakness of lower limbs and instability of walking",21 cases(17.07%)complained of "numbness and/or pain in lower limbs",9 cases(7.32%)complained of " Intermittent claudication ",10 cases(8.13%)complained of " radiative pain in chest and abdomen ",4 cases(3.25%)complained of "thoracoabdominal band sensation".4.TSS imaging manifestations: There were 190 ossified segments in 92 OLF patients.The distribution of ossification segments were in sequence of lower thoracic vertebrae77.37%(147/190),middle thoracic vertebrae 17.37%(33/190)and upper thoracic vertebrae 5.26%(10/190).According to Sato classification,enlarged type accounted for31%(59/190),tuberous type accounted for 23%(44/190),extened type accounted for19%(36/190),lateral type accounted for 17%(32/190)and fused type accounted for10%(19/190).According to Kuh classification,round type accounted for 63%,beak type accounted for 37%.There were 24 ossified segments in 9 OPLL patients.The distribution of ossification segments were in sequence of upper thoracic vertebrae 66.67%(16/24),lower thoracic vertebrae 20.83%(5/24)and middle thoracic vertebrae 12.5%(3/24).According to sagittal CT ossification,beak type accounted for 45%(4/9),linear type accounted for 33%(3/9),continuous waveform type accounted for 11%(1/9),continuous cylindrical type accounted for 11%(1/9)and mixed type accounted for 0%(0/9).There were 29 protruding segments in 13 patients with TDH.The distribution of protruding segments were in sequence of lower thoracic vertebrae 79.31%(23/29),middle thoracic vertebrae 20.69%(6/29),upper thoracic vertebrae 0%(0/29).Part II: Analysis of factors associated with prognosis1.A total of 76 patients were followed up,with a median follow-up time of 46(14-84)months.The average recovery rate calculated from m JOA score was 58.95%±0.28%.The excellent rate of operation was 61.84%.2.Logistic regression model was applied for univariate analysis,and the results showed that the poor prognosis factors were age,preoperative m JOA score,surgical segments,operation time,high-signal changes on preoperative MRI,residual rate of cross-section area,residual rate of axial anteroposterior diameter,residual rate of axial lateral median diameter,residual rate of sagittal anteroposterior diameter.The data with statistically significant differences were included in the multivariate binary Logistic regression model.The results of the multifactorial binary logistic regression model showed that the independent risk factors of poor prognosis were the number of surgical segments(OR=3.181,P=0.001)and the residual rate of spinal canal area(OR=0.711,P=0.031).Part III: Analysis of factors associated with complications1.In this study,31 of 123 TSS patients had postoperative complications,with a total complication incidence of 25.20%.There were 18 cases(14.63%)with cerebrospinal fluid leakage(CSF),3 cases(2.44%)with spinal cord or nerve root injury,3 cases(2.44%)with transient neurological deterioration,3 cases(2.44%)with poor wound healing,2 cases(1.63%)with urinary tract infection,2 cases(1.63%)with electrolyte imbalance.2.Logistic regression model was applied for univariate analysis,and the results showed that the risk factors associated with complications were age,duration of disease,diabetes mellitus,preoperative m JOA,operative segments,operative time,blood loss,high-signal changes on preoperative MRI.The results of the multifactorial binary logistic regression model showed that the independent risk factors for complications were preoperative m JOA(OR=0.667,P=0.011)and number of surgical segments(OR=1.530,P=0.016).3.Subgroup analysis: Univariate analysis of CSF related factors was performed.The results showed that the risk factors included surgical segments(OR=1.499,P=0.025),high-signal changes on preoperative MRI(OR=1.241,P=0.001),duration of disease(OR=1.029,P=0.014),operation time(OR=1.010,P=0.001),and blood loss(OR=1.002,P=0.001).Univariate analysis of spinal cord or nerve root injury related factors was performed.The results showed that the risk factors included preoperative m JOA(OR=0.219,P=0.010)and residual rate of cross-section area(OR=0.937,P=0.039).Subgroup analysis was performed on patients with OLF,and univariate Logistic regression equation was constructed between each imaging classification and encroachment index and the occurrence of complications.The results showed that in Sato classification,compared with the lateral type,the extended type and enlarged type did not significantly increase the risk of complications,and the difference was not statistically significant.However,fused type(OR=13.333,P=0.030)and tuberous type(OR=16.000,P=0.016)increased the incidence of complications,and the difference was statistically significant.In Kuh classification,compared with round type,beak type increased the risk of complications,and the difference was statistically significant(OR=2.812,P=0.029).Conclusion:1.TSS patients can be caused by OLF,OPLL and TDH,OLF was the main cause of TSS,mostly happened in males.Most TSS patients were admitted to the hospital due to unilateral or bilateral lower limb weakness and walking instability.OLF and TDH often occur in the lower thoracic vertebra,and OPLL often occurs in the upper thoracic vertebra.2.The mid-long term clinical efficacy of posterior decompression and fusion surgery for TSS is satisfactory.The risk of poor prognosis increased 2.185 times for each additional surgical segment and 0.43 times for each 10% reduction in residual spinal canal area.3.The total incidence of complications after TSS is high,and CFL and spinal cord injury are the most common complications.The risk of complicatons decreased 0.333 times for every 1 point increase in preoperative m JOA.The risk of complications increased 0.53 times for each additional number of surgical segments.The risk factors of spinal cord or nerve root injury were preoperative m JOA and axial residual rate of cross-section area.Postoperative complications are more likely to occur in OLF patients with fused type,tuberous type and beak type. |