The Value Of High Frequency Ultrasound And MRI In The Diagnosis Of Partial And Full-Thickness Rotator Cuff Tears

ObjectiveThe results of shoulder arthroscopy were used as the gold standard to analyze and compare the diagnostic value of high frequency ultrasound and MRI in rotator cuff partial and full-thickness tear,so as to provide a reasonable and effective examination plan for clinical practice.MethodA retrospective analysis was performed on 127 patients who underwent shoulder arthroscopic surgery in affiliated hospital of Chengde Medical College from February 2015 to November 2019 due to shoulder joint disease,including 59 males and 68 females,aged 29 ?74 years,with an average age of55.8±7.8 years.All the patients underwent MRI and high-frequency US examination before surgery.Preoperative diagnosis of rotator cuff were respectively counted according to high frequency US,MRI and intraoperative:rotator cuff without tear,rotator cuff partial tear and rotator cuff full-thickness tear.According to the intraoperative diagnosis,the patients were divided into two groups: first,the rotator cuff tear group and the rotator cuff non-tear group,while the rotator cuff tear group was divided into the rotator cuff full-thickness tear group and the partial tear group.Secondly,the rotator cuff full-thickness tear group and the rotator cuff non-full-thickness tear group,while the rotator cuff non-full-thickness tear group were divided into the rotator cuff partial tear group and the rotator cuff non-tear group.SPSS 19.0 statistical software was applied,rotator cuff diagnosis in arthroscopy was used as the gold standard,and the sensitivity,specificity,accuracy,positive predictive value and negative predictive value of high frequency US and MRI in the diagnosis of rotator cuff tear,rotator cuff full-thickness tear and rotator cuff partial tear in each groupwere compared,and the chi-square test was performed.If P<0.05,the difference was considered statistically significant.ResultAmong the 127 patients,102 were diagnosed with rotator cuff tear by high-frequency US,25 were diagnosed without tear,including 54 with full-thickness tear and 48 with partial tear.MRI diagnosed rotator cuff tear in105 cases,22 cases without tear,including 58 cases with full-thickness tear and 47 cases with partial tear.Through shoulder arthroscopic surgery,there were 105 cases of rotator cuff tear,22 cases without tear,including 60 cases with full-thickness tear and 45 cases with partial tear.The sensitivity,specificity,accuracy,positive predictive value(PPV)and negative predictive value(NPV)of high-frequency US in diagnosing rotator cuff tear existence were 94.3%,86.4%,92.9%,97.1% and 76.0%,respectively,while MRI diagnosis was 97.1%,86.4%,95.3%,97.1% and 86.4%,respectively.The above results of chi-square test for high-frequency US and MRI: sensitivity P=0.495,specificity P=1,accuracy P=0.552,PPV P=1,and NPV P=0.071,with no significant difference(P>,0.05).The sensitivity,specificity,accuracy,PPV and NPV of high-frequency US in diagnosing full-thickness tear and partial tear were 83.0%,87.5%,84.8%,90.7% and 77.8%,respectively.MRI diagnoses were 91.5%,92.5%,91.9%,94.7% and 88.1%,respectively.The above results of chi-square test and MRI: sensitivity P=0.088,specificity P=0.249,accuracy P=0.121,PPV P=0.268,NPV P=0.06,and there was no significant difference between them(P>,0.05).The sensitivity,specificity,accuracy,PPV and NPV of high-frequency US in diagnosing full-thickness tear and non-full-thickness tear were 81.7%,92.5%,87.4%,90.7% and 84.9%,respectively.The MRI diagnoses were90.0%,94.0%,92.1%,93.1% and 91.3%,respectively.The above results of chi-square test and MRI: sensitivity P=0.103,specificity P=0.593,accuracy P=0.169,PPV P=0.602,and NPV P=0.192,and there was no significant difference between them(P>,0.05).The sensitivity,specificity,accuracy,PPV and NPV of high-frequency US in diagnosing rotator cuff partial tear and non-tear were 87.5%,86.4%,87.1%,94.5% and 79.2%,respectively.The MRI diagnoses were 95.2%,90.5%,93.4%,95.0% and 90.5%,respectively.The above results of chi-square test and MRI: sensitivity P=0.051,specificity P=1,accuracy P=0.237,PPV P=0.769,NPV P=0.029.It was suggested that both high-frequency US and MRI had higher accuracy,sensitivity and specificity in the diagnosis of partial rotator cuff tear,and there was no significant difference in sensitivity,specificity,accuracy and PPV between the two(P>,0.05).However,the NPV of MRI was significantly higher than that of US,with a significant difference(P<0.05).Conclusion1 High frequency US and MRI have high accuracy,specificity and sensitivity in diagnosing rotator cuff tear,especially full-thickness tear,and there is no significant difference between them.2 For the diagnosis of partial rotator cuff tear,MRI has higher diagnostic value than high-frequency US,especially in the negative prediction value.3 High-frequency US can be used for the screening of rotator cuff tear,but MRI should be combined with suspected rotator cuff tear to improve the diagnostic accuracy.