关节镜下肩关节松解术治疗冻结肩的术后疗效与患者术前病程的相关性分析

Correlation analysis between the preoperative clinical course and postoperative outcomes following treatment of frozen shoulder by arthroscopic shoulder release

目的探讨关节镜下肩关节松解术治疗冻结肩的临床疗效,并分析术后疗效与患者术前病程的相关性。方法回顾分析2015年9月至2017年11月,在我院经关节镜肩关节松解术治疗的34例冻结肩病例。男12例,女22例;左肩20例,右肩14例;平均年龄54岁;平均病程5个月。比较患者术前及末次随访的疼痛视觉模拟评分(visual analogue scale,VAS)、主动肩关节活动度(range of motion,ROM)、肩关节功能评分(constant-murley,C-M)、美国肩肘外科医师协会肩关节功能评分(American shoulder and elbow surgeons,ASES)、美国加州大学洛杉矶分校肩关节功能评分(University of califonia,los angeles,UCLA),对患者末次随访肩关节功能评分与术前病程作相关性分析和线性回归分析。结果34例随访12~37个月,平均25个月。33例(97%)获得良好临床疗效,1例存在持续轻度患侧肩关节疼痛伴活动受限。VAS评分:术前9.00(7.75~9.25)分,末次随访0(0~1)分,差异有统计学意义(P<0.05)。前屈:术前(80.15±19.87)°,末次随访(160.44±9.24)°;外展:术前(73.82±19.23)°,末次随访(158.06±8.92)°;外旋:术前(16.03±11.53)°,末次随访(57.06±10.08)°;体侧内旋活动范围也由手术前的臀部水平恢复到T12水平,差异均有统计学意义(P<0.05)。UCLA肩关节功能评分:术前12.00(10.00~15.00)分,末次随访35.00(30.75~35.00)分,差异有统计学意义(P<0.05)。C-M肩关节功能评分:术前29.50(25.00~36.00)分,末次随访96(84.75~98.00)分,差异有统计学意义(P<0.05)。ASES肩关节功能评分:术前18.33(16.25~33.33)分,末次随访95.00(88.33~98.33)分,差异有统计学意义(P<0.05)。患者术前UCLA、C-M及ASES肩关节功能评分与病程均无显著相关。末次随访UCLA、C-M及ASES肩关节功能评分与术前病程均呈线性负相关(UCLA:r=-0.443,P=0.0086;R2=0.357,P=0.0002。C-M:r=-0.502,P=0.0025;R2=0.393,P<0.0001。ASES:r=-0.538,P=0.0010;R2=0.388,P<0.0001)。结论关节镜下肩关节松解术是治疗冻结肩有效方法,并且术后疗效与患者术前病程间存在明显的线性负相关,即患者病程越长术后效果越差。因此对冻结肩患者建议保守治疗效果不显著时,应尽早行关节镜下肩关节松解术,以获得更好的疗效。

Objective To investigate the clinical efficacy of arthroscopic shoulder release in the treatment of frozen shoulder and its correlation with preoperative clinical course.Methods A retrospective self-controlled study was conducted to analyzed 34 patients(12 males,22 females;20 left shoulders,14 right shoulders)after arthroscopic shoulder release from September 2015 to November 2017.The mean age was 54 years and the mean course was 5 months.VAS,active shoulder ROM,Constant-Murley(C-M)shoulder function scores,American Shoulder and Elbow Surgeon Association(ASES)shoulder function scores,University of Califonia,Los Angeles(UCLA)shoulder function scores were compared by correlation and linear regression analysis preoperatively and at the final follow-up.Results The average follow-up was 25 months(range:12-37 months).Good clinical outcomes were achieved in 33 patients(97%).Persistent mild shoulder pain and limited mobility were observed in 1 patient.VAS decreased from 9(7.75-9.25)preoperatively to 0.01 at the last follow-up with significant differences(P<0.05).The medial internal rotation recovered from the preoperative hip level to the T12 level with statistically significant differences(158.06±8.92)°;(57.06±10.08)°(P<0.05).UCLA shoulder function scores increased from 12.00 preoperatively(10.00-15.00)to 35.00(30.75-35.00)at the final follow-up with significant differences(P<0.05);C-M shoulder function scores increased from 29.50(25.00-36.00)to 96(84.75-98.00)with significant differences(P<0.05);ASES shoulder function scores increased from 18.33(16.25-33.33)to 95.00(88.33-98.33)with significant differences(P<0.05);There were no significant correlations between preoperative UCLA,C-M and ASES shoulder function scores and the clinical course.UCLA,C-M,and ASES shoulder function scores at the final follow-up were linearly negatively correlated with the preoperative course.(UCLA:r=-0.443,P=0.0086;R2=0.357,P=0.0002.C-M:r=-0.502,P=0.0025;R2=0.393,P<0.0001.ASES:r=-0.538,P=0.0010;R2=0.388,P<0.0001).Conclusions Arthroscopic shoulder release is effective in the treatment of frozen shoulders.Significant linear negative correlation between the postoperative outcomes and preoperative clinical course is noted.Longer course indicates worse effects,so that early surgery is recommended after failed conservative treatment.