SGEM#381: Put Your Hand on My Shoulder and Reduce It

Date: October 27th, 2022 Reference: Hayashi et al. Comparative efficacy of sedation or analgesia methods for reduction of anterior shoulder dislocation: A systematic review and network meta-analysis. AEM October 2022 Guest Skeptic: Dr. Justin Morgenstern is an emergency physician and the creator of the #FOAMed project called First10EM.com   Case: A 19-year-old man presents to the emergency department (ED) with his first time anterior should dislocation after trying to recreate one of his favourite scenes in the movie Lethal Weapon. He is in significant pain, but your charge nurse informs us that, like most days since the pandemic started, the department is completely full of admitted patients, and there is nowhere safe to perform a procedural sedation, let alone have the staff to do it. The patient asks, through clenched teeth, whether there are any other options to get his shoulder back in. Background: We have covered shoulder issues a few times on the SGEM. There was an episode looking at diagnosing rotator cuff injuries (SGEM#74), the best position post-dislocation immobilization (SGEM#121) and using point of care ultrasound (POCUS) to diagnose shoulder dislocations (SGEM#288). The shoulder joint has the widest range of motion of any joint in the human body. This makes it very useful and very susceptible being dislocated. The vast majority of shoulder dislocations are anterior. Young active men are at greatest risk for dislocating their shoulder. There is also a wide range of options to diagnose shoulder dislocations (clinically, POCUS, x-ray) and dozens of reduction techniques. Some clinicians perform reductions without any analgesics at all, while others choose from a variety of options, including peripheral nerve blocks, intra-articular anesthesia, and full procedural sedation. Procedural sedation might represent one of the greatest advancements for the practice of emergency medicine, allowing us to perform a large variety of necessary but painful procedures without causing our patients pain. Although minor adverse events, such as brief apnea or hypoxia, are common, significant adverse events are very rare, and the benefits are clear (Bellolio 2016). However, for most departments, procedural sedation represents a logistical challenge that can increase a patient’s length of stay. Peripheral nerve blocks (PNBs) can be very effective at controlling pain, but require a degree of practitioner skill. The use of ultrasound to guide these procedures has increased their popularity in recent years. There have been a few randomized control trials (RCTs) of peripheral nerve blocks for shoulder dislocation, but without definitive results (Raeyat Doost 2017; Blaivas 2011). Intra-articular anesthetic (IAA) injections are another option, and seem like they should be incredibly easy, considering that the humeral head is not sitting in the glenoid fossa, and so the joint is wide open and supposedly easy to access. Intra-articular injection has been compared with procedural sedation (PS) for shoulder dislocation, with some potential benefits (Wakai 2011).  However, in one study, emergency physicians missed the joint space almost half the time when performing landmark-based shoulder injections (Omer 2021). Therefore, uncertainty remains about the ideal technique to provide analgesia and/or sedation for the reduction of anterior shoulder dislocations. Clinical Question: What is the safety and efficacy of intravenous sedation, intra-articular injection, and peripheral nerve block for the reduction of anterior shoulder dislocations. Reference: Hayashi et al. Comparative efficacy of sedation or analgesia methods for reduction of anterior shoulder dislocation: A systematic review and network meta-analysis. AEM October 2022 Population: RCTs that assessed sedation of analgesia methods for the reduction of anterior shoulder dislocations diagnosed on either physical exam or x-ray in patients older than 15 years of age. Exclusions: Allergies to study medications, multiple traumas, fractures (except Hill-Sachs and Bankart lesions), hemodynamic instability, or respiratory distress. Intervention: Intravenous (IV) sedation, intra-articular anesthetic (IAA) injection, and peripheral nerve blocks (PNB). Comparison: Patients who received either a placebo or no sedation. Outcome: Primary Outcome: There were three primary outcomes - Immediate success rate, patient satisfaction, and ED length of stay (LOS) Secondary Outcomes: Adverse events, pain score, time required for reduction, number of reduction attempts, and total success rate of the reduction. This is an SGEMHOP episode. Normally we have one of the authors on the show. This time we have the corresponding author who is an orthopedic trauma surgeon in Japan. Dr. Yamamoto was kind enough to give a shout out to his co-investigators and send responses to our ten nerdy questions. I can understand how hard it would be to talk nerdy in another language. Dr. Yamamoto's co-authors incliuded Minoru Hayashi, Kenichi Kano, Naoto Kuroda, Akihiro Shiroshita and Yuki Kataoka who are member of Scientific Research WorkS Peer Support Group (SRWS-PSG). SRWS-PSG is a scientific research group mainly conducting systematic reviews Authors’ Conclusions: “The results of our NMA indicated that three sedation or analgesia methods (IVS, IAA, and PNB) might result in little to no difference in the success rate of reduction and patient satisfaction. IAA and PNB had no adverse respiratory events.” Quality Checklist for Therapeutic Systematic Reviews: The clinical question is sensible and answerable. Yes The search for studies was detailed and exhaustive. Yes The primary studies were of high methodological quality. No The assessment of studies were reproducible. Yes The outcomes were clinically relevant. Yes There was low statistical heterogeneity for the primary outcomes. Unsure The treatment effect was large enough and precise enough to be clinically significant. No Results: After full-text review, they identified 16 RCTs that fulfilled their inclusion and exclusion criteria. These trials encompass a total of 957 patients. Of the 16 studies, 11 compared IV sedation to intra-articular injection, four compared nerve blocks to sedation, and one compared intra-articular injection to nothing. Key Result: There were no statistical differences in immediate success rate between techniques, uncertainty regarding patient satisfaction and intra-articular anesthetic had the shortest length-of-stay. Primary Outcomes: Immediate Success: There were no statistical differences IAA vs IVS: RR 0.93, 95% CI 0.84 to 1.02 PNB vs IVS: RR 1.13, 95% CI 0.84 to 1.52 Patient Satisfaction: The evidence was uncertain, with no statistical differences IAA vs IVS: SMD -0.47, 95% CI -1.41 to 0.48 PNB vs IVS: SMD -0.60, 95% CI -1.43 to 0.23 ED Length of Stay: The evidence was classified as very uncertain, but IAA had statistically shorter length of stays than IVS, whereas there was not a statistical difference between PNB and IVS IAA vs IVS: MD -107 minutes, 95% CI -203 to -13 PNB vs IVS: SMD -26 minutes, 95% CI -149 to 96 Secondary Outcomes: Adverse events: Two of the studies reported no adverse events. Respiratory events were the most common in the IVS group. Psychological agitation and drowsiness were reported in the IAA group, and mild local anesthetic systemic toxicity was reported in the PNB group. Pain score: IAA might be lower than PNB (SMD -1.8), but there were not differences noted between either IAA or PNB and IVS. Time for reduction: Both IAA and PNB might take longer than IVS (by 5 and 15 minutes respectively). Number of reduction attempts: Very uncertain, but no clear differences between the groups. Total success rate of reduction: No clear differences. We asked Dr. Yamamoto and his team of co-authors ten nerdy questions. They sent their written responses which are listed below: 1) Uncertainty: Rather than just focusing on statistical significance, you use the language of uncertainty throughout your results, with almost all the results being very uncertain. This language is not used in all meta-analyses. Can you comment on why you phrased your results this way? It is a very important point. We used “uncertain” as a result of evaluating the confidence of the evidence using the CINeMA tool. The confidence is the credibility of results from NMA and covers six domains: (i) within-study bias, (ii) reporting bias, (iii) indirectness, (iv) imprecision, (v) heterogeneity, and (vi) incoherence [Nikolakopoulou et al PLoS Med 2020]. The concept of imprecision includes statistical significance. We believe that our evaluation using confidence is more appropriate in assessing the credibility of the evidence. Dr. Yamamoto Considering both certainty of evidence and effect size are important for readers. Therefore, clinicians should not judge the effectiveness of interventions based on effect size alone, but should also consider the certainty of the evidence. 2) Accounting for Bias in the Meta-Analyses: I always find it difficult to appropriately account for the potential bias of individual studies when reading a meta-analysis. If I wouldn’t trust the results of a single RCT, it doesn’t help to mix it in with a bunch of other trials with similar methodologic issues. This is the classic GIGO – garbage in garbage out – problem. You perform a secondary analysis that focuses only on studies with the lowest risk of bias, and in that analysis IV sedation actually was statistically better than intra-articular injections. How do you account for bias in a meta-analysis, and which of these outcomes do you trust? A meta-analysis including only RCTs with low risk of bias is more reliable than the others. However,