Objective:Both lumbar disc degeneration(LDD)and nerve root compression(NRC)may lead to the morphological changes of lumbar multifidus(LM)characterized by myofibers atrophy or fatty infiltration(FI).Hides has found that healthy populations have asymmetry degree of cross-sectional area(ADCSA)between 2.9%and 5.8%,and a hypothesis has been proposed that the LM ADCSA greater than 10%suggests an underlying pathological state of the lumbar vertebra.In this study,LM morphological indicators at corresponding levels were analyzed in patients with L5/S1 LDD or with combined NRC at the same level to investigate the correlation between LDD and NRC degree and LM morphological changes,and to verify the hypothesis.Methods:This was a cross-sectional study.In this study,960inpatients who had undergone plain MRI of the lumbar vertebra were selected by 2 radiologists with more than 10 years of working experience following preset inclusion and exclusion rules,and 64 were finally included.LDD and NRC degree were respectively evaluated using the corresponding Pfirrmann rating system.LM Cross-sectional area(CSA),fatty infiltration rate(FIR)and ADCSA were measured by Image J software.All manually evaluated indexes were completed independently by the 2 surgeons and the mean value was taken after the consistency was verified.The gender,age,height,BMI,grades of L4/5 disc degeneration according to Pfirrmann system,L4/5(D),grades of L5/S1 disc degeneration according to Pfirrmann system,L5/S1(D),grades of nerve root compromise at L5/S1 level according to Pfirrmann system,L5/S1(N),LM CSA,LM FIR and LM ADCSA of the samples included were collected.SPSS software was used for statistical calculations:ICC test was used to verify the consistency of the measurement results of the 2surgeons;linear regression combined with multiple linear regression was used to analyze the correlation between LDD and NRC degree and LM CSA,FIR and ADCSA(for more accurate research results,dummy variables were set for L4/5(D),L5/S1(D)and L5/S1(N)with the lowest level included as the reference variable);Fisher’s exact test was used to verify the hypothesis that"the LM ADCSA greater than 10%suggests an underlying pathological state of the lumbar vertebra".Results:1.Both surgeons had good reliability of LM morphological indicators(ICC≥0.93,p<0.001);2.Patients with L5/S1(N)level II,III,and IV had an average decrease of 1.983 cm~2,2.663 cm~2,and 4.384 cm~2in LM CSA compared with those with level I(p<0.01);3.Patients with L5/S1(N)level III had an average increase of 12.3%in LM FIR compared with those with level I(p<0.01);4.No significant correlation was observed between L5/S1(D)and LM morphological changes(p>0.05);5.No significant correlation was observed between L4/5(D),L5/S1(D)and L5/S1(N)and LM ADCSA(p>0.05);6.No statistical difference was observed in the distribution of L4/5(D),L5/S1(D)and L5/S1(N)levels when LM ADCSA was divided into 2 groups with 10%as the threshold(p>0.05).Conclusions:1.NRC is significantly correlated with the decrease of LM CSA and the increase of FIR at corresponding levels;2.There is no significant correlation between LDD and LM morphological changes;3.LM ADCSA changes cannot reflect LDD or NRC. |