Optimization and standardization of technique for fluoroscopically guided suprascapular nerve blocks.

Abstract

The purpose of this study is to standardize the fluoroscopically guided suprascapular nerve block technique by optimizing patient positioning and imaging features that predict suprascapular notch visibility. One hundred fifty-five consecutive patients underwent suprascapular nerve block from January 2010 through August 2012. The effect of arm position on suprascapular notch visibility and procedure parameters such as fluoroscopy time were evaluated using a chi-square test and a nonparametric Mann-Whitney U test, respectively. The coracoid process "in profile," a clear space between the coracoid process tip and the top of the glenoid, a visible glenohumeral joint space, and bisection of the glenoid by the projection of the scapular spine were identified as possible predictors for suprascapular notch visualization. Their ability to predict suprascapular notch visibility was assessed using positive predictive value (PPV), sensitivity, logistic regression, and receiver operating characteristic (ROC) curve analysis. Procedures performed with the affected arm positioned above the shoulder ("arm up") yielded increased suprascapular notch visualization (91% vs 47%; p<0.0001) and decreased fluoroscopy time (1.3 vs 2.0 minutes; p=0.002) compared with those performed with the affected arm positioned at the patient's side ("arm down"). The four imaging features had 91-95% PPVs for suprascapular notch visibility. Concurrent visualization of all four features yielded the highest discriminative accuracy for suprascapular notch visualization (area under the ROC curve [AUC], 0.870). Discriminative accuracy was good with visualization of only two features (AUC, 0.767-0.844) and fair with visualization of only one feature (AUC, 0.644-0.769). Positioning patients arm up and optimizing several key imaging features allows fluoroscopically guided suprascapular nerve blocks to be performed reliably and confidently.