Comparison of High versus Low Volume Ultrasound-Guided Interscalene Plexus Block Utilizing Contrast Magnetic Resonance Imaging: Epidural Spread, Phrenic Nerve Involvement, Tissue Distribution – A Randomized, Controlled Trial.
Introduction: Previous studies have found that higher local anesthetic (LA) injection volumes for interscalene blocks (ISB) are associated with undesired spread to structures surrounding the brachial plexus, including the phrenic nerve, muscle, and epidural space, along with increased risk for complications [1, 2]. This randomized controlled study utilized contrast magnetic resonance imaging (MRI) to directly visualize differential-volume LA distribution, and compared analgesic efficacy and clinical signs of respiratory impairment. Material and Methods: The study was registered with EudraCT (2013-004219-36) and clinicaltrials.gov (NCT02175069). After power analysis [based on 1], 30 patients undergoing shoulder surgery were randomized to receive a standardized ultrasound-guided ISB with catheter placement by a single blinded operator. Either 20ml or 5ml of ropivacaine 0.75% supplemented with MRI contrast dye gadopentetat-dimeglumin were injected, followed by MRI imaging. Surgery was performed under general anesthesia with supplemental fentanyl as required. Data collected at various time points up to 24 hours post block include sensory and motor function, numerical rating scale (NRS) pain scores, time to start of patient-controlled analgesia (PCA) via catheter, analgesic consumption, diaphragmatic ultrasonography and spirometry. MRI sequences were individually analyzed by two experienced blinded radiologists. Results: The 20ml and 5ml groups were well-balanced for age, sex and body mass index. All blocks provided fast onset and adequate intraoperative analgesia with equally low supplemental fentanyl requirement (p=0.620). The incidence of epidural spread (Fig 1a) was equal across groups, but spread reaching the intervertebral foramen (Fig 1b), phrenic nerve involvement (Fig 2a) and copious intramuscular deposition of local anesthetic (Fig 2b) was significantly more frequent in the 20ml group (see Table). Similarly, comparing the 20ml to 5ml group, the incidence of diaphragmatic paralysis was twice as high (n=8 vs 4), and changes from baseline peak respiratory flow rate were larger (Δ = -2.66 [±1.99] vs -1.69 [±2.0] l/min), however, the results were not statistically significant. There were no significant differences in postoperative NRS scores across groups at any time point. While the 20ml group requested PCA start later than the 5ml group (755 [±230] vs 497 [±233] minutes, p=0.003), total PCA consumption was not different (118 [±116] vs 142 [±92] ml ropivacaine 0.2%, p=0.449). Discussion: To the best of our knowledge, this study is the first to visualize LA distribution after ISB in humans utilizing MRI. The data suggest that epidural spread occurs after interscalene nerve block irrespective of injection volume; however, lower volumes cause less spread to the intervertebral foramen, distal phrenic nerve and muscle. The latter may have toxic effects associated with long-term phrenic palsy . With the advent of longer-acting LAs, it is possible that low volumes for interscalene blocks could provide excellent analgesia with less potential for phrenic block and side effects.
 Fritsch et al, Reg Anesth Pain Med 2013  Riazi et al, Br J Anaesth 2008  Hogan et al, Anesthesiology 2013