Application Of MR Imaging Features Of Primary Frozen Shoulder In Differential Diagnosis And Surgical Treatment

Background:Primary frozen shoulder(PFS)is a common condition describing patients with restricted active and passive movement of the glenohumeral joint.Pain and limitation of shoulder movements are patients’ main presenting symptoms.After the onset,it seriously affect the patient’s sleep and daily activities.Secondary stiff shoulder(SSS)and PFS have similar clinical manifestations,but the treatment options for the two are often different.Most of the PFS can be relieved by conservative treatment.For patients with no obvious improvement in symptoms for more than 6 months of conservative treatment,surgery can be selected,while SSS usually requires active treatment of the primary disease through surgery.Therefore,accurate identification of PFS and SSS has important clinical significance for formulating a reasonable treatment option.Magnetic resonance imaging(MRI)is a good imaging tool of tendons and ligaments.There have been many studies on the MRI manifestations of PFS,but most ly with contrast agents.The contrast agent for enhanced MRI is useful to show abnormal structures,but commonly accompanied by complications such as allergic reactions,and increased economic burden of patients.So,the enhanced MRI is not a routine method in clinical work.Furthermore,there is few research on the differential diagnosis of PFS.Non-enhanced MRI is a routine method,which can show the abnormal signs of frozen shoulder well.Therefore,studying the non-enhanced MRI manifestations of PFS and screening out characteristic manifestations with differential diagnostic value are still of clinical significance and worthy of further discussion.The abnormal signs found in preoperative MRI can provide reference for the treatment option of PFS.In terms of surgical treatment,the degree of manual release and arthroscopic release of the joint capsule is still worth discussing.Objective:1.To retrospectively analyze the non-enhanced MRI imaging findings of the PFS,and screen for MRI signs with diagnostic value.2.To evaluate the diagnostic value of non-contrast MRI features of PFS3.To evaluate midterm clinical results of manipulation under anesthesia combined with arthroscopic capsular release and synovium debridement for PFS.Method:1.This study retrospectively included 154 patients with PFS and 154 patients with SSS diagnosed in our hospital from January 2013 to December 2017,and collected and analyzed the shoulder MRI images of the patients in our hospital.Main indicators such as the thickness and width of the axillary sac,the thickness of the coracohumeral ligament,the subacromial space,and the fluid hyperintensity on proton-density(PD)weighted sequences images were evaluated.Screen the characteristic MRI imaging findings of patients with PFS by comparison.2.This study included 123 patients with PFS and SSS between January 2013 and Ju ne2019.The MRI images in each case were collected and randomly assigned to two specialist.Difference of MRI features between two groups was analyzed.Features of edema signal in the subcoracoid,axillary recess,and long head of the biceps tendon,corac ohumeral ligament thickening more than 2 mm,and axillary recess thickening more than 4 mm were analyzed and the accuracy for diagnosis of PFS was investigated respectively.3.From January 2013 to December 2017,33 patients of PFS were treated with manipulation under anesthesia combined with arthroscopic capsular release and synovium debridement for PFS.MRI and X-ray examination were performed before operation.Visual analog scale(VAS)was used to evaluate the improvement of shoulder pain before and afte r operation.Constant score was used to evaluate the improvement of shoulder function,and the flexion,abduction,external rotation and internal rotation of shoulder joint were recorded.The improvement of clinical indexes was analyzed to evaluate the eff ect of operation.Result:1.The thickness of the axillary recess and the width of the axillary recess in the PFS group were greater than those in the SSS group and the difference was statistically significant(P(27)0.000).There was no statistically significant difference between the two groups in terms of the long diameter of the biceps tendon,the subacromial interval,and the bicipital groove angle(P(29)0.05).In terms of high-intensity edema signals on PD weighted sequences images,the frequency of patients with high-intensity edema signals in the long head of biceps tendon,subcoracoid bursa and axillary recess was 56 cases(36.4%),which was significantly higher than 10 cases(6.5%)in SSS group.In the SSS group,38 cases(25.0%)had high-intensity edema signals at the long head of biceps tendon,subcoracoid bursa,axillary recess and subacromial bursa,which was significantly higher than that of 9cases(5.8%)in PFS group.2.Compared with SSS,combination of high-intensity signals on PD weighted sequences images in the axillary recess capsule,subcoracoid bursa,and synovium of the long head of the biceps had higher proportion in PFS group and the difference was statistically significant(P(27)0.000).In detection of PFS,combination of high-intensity signals in the axillary recess capsule,subcoracoid bursa,and synovium of the long head of the biceps tendon showed a sensitivity,specificity,positive likelihood ratio,negative likelihood ratio,and area under the curve(AUC)were 52.9%,94.3%,9.3,0.5,0.736,respectively.Corresponding results for diagnosis criteria of coracohumeral ligament thickening more than 2 mm were 62.8%,54.3%,1.4,0.7,0.585,respectively.The results for diagnosis criteria of axillary recess thickening more than 4 mm were 80.4%,40.0%,1.3,0.5,0.602,respectively.Reliability testing showed that Kappa value of combination of high-intensity signals in the axillary recess capsule,subcoracoid bursa,and synovium of the long head of the biceps tendon was more than 0.802,ICC of coracohumeral ligament thickening more than 2 mm was more than 0.761,ICC of axillary recess thickening more than 4 mm was more than 0.909.3.None of the patients occurred fractures or labrum tear;all patients’ incision healed by stage I.All the 33 patients were followed up and the mean follow-up was 24.2 months(range:20 to 31).The average thickness of axillary recess measured by MRI before operation was 4.93?0.94 mm,the acromiohumeral interval(AHI)measured by X-ray before operation was 6.86 ± 2.05 mm There were no infection or nerve lesion during the follow-up.At the final follow-up,patients’ range of motion of the affected shoulder,VAS score and Constant score were significantly improved comparing with the results preoperatively(P<0.001).Conclusion:1.The maximum thickness and width of axillary recess in PFS were significantly higher than those in SSS.Combination of edema signals in the subcoracoid bursa,axillary recess capsule,and long head of the biceps tendon is one of the characteristic MRI findings of PFS.2.Combination of edema signals in the subcoracoid bursa,axillary recess capsule,and long head of the biceps tendon on non-contrast MRI showed high specificity and acceptable diagnostic accuracy for detecting PFS,it helps to differentiate it from SSS and reduce the rate of misdiagnosis.3.MRI examination before operation of PFS guided the operation well.Manipulation under anesthesia combined with arthroscopic capsular release and debridement of synovium achieved a good midterm clinical outcome without complication for PFS.