The main finding of this meta-analysis is that ultrasound-guided ESPB significantly reduced opioid consumption 24 hours after surgery. It further reduced pain scores and patients who need rescue analgesia, besides, it prolonged the time to first request of rescue analgesia. Despite of the high heterogeneity, the main finding was consistent in subgroup analyses.
Possible mechanisms for findings
Ultrasound-guided ESPB is a peri-paravertebral regional anesthesia technique which is supposed to block the dorsal and ventral rami of the thoracic and abdominal spinal nerves , and thereby to block the anterior, posterior, and lateral thoracic and abdominal walls. However, the mechanisms of action and spread of LA are not fully elucidated. Several potential mechanisms have been posited. one of the suggested mechanisms of ultrasound-guided ESPB is paravertebral spread of LA , LA infiltration was observed from injection site to three vertebral levels cranially and four levels caudally . Based on this mechanism, Coşarcan SK et al. reported a modification ESPB and got good pain relief in various surgeries. However, the mechanism of paravertebral spread of LA remained debated in several cadaveric studies[28-30]. Another potential mechanism is epidural spread of LA. Schwartzmann A et al , Tulgar S, et al and Altıparmak B, et alfound unilateral erector spinae plane block result in bilateral sensory blockade in some patients, epidural spread of the LA during ESPB may explain this result. Moreover, some evidence indicated that penetration of LA acted on dorsal and ventral rami through the connective tissues and branch communication leaded to visceral analgesia [34, 35]
Implications for clinical researches
Our findings demonstrated that ultrasound-guided ESPB was associated with a reduction of opioid consumption, which further proved the effectiveness of ESPB. However, ultrasound-guided ESPB has only been utilized in clinical setting for about three years, several important issues have not been resolved yet. First, the optimal concentration, volume and type of LA in ESPB is not well established. Although 20 and 30ml of 0.25% bupivacaine or 0.5% ropivacaine were recommended, concentrations of 0.25-0.5% bupivacaine 10-20ml were used in ultrasound-guided ESPB among all 12 RCTs included in this meta-analysis. Is bupivacaine more preferred than ropivacaine? why? We tried to make a judgment but stop by the insufficient evidence. More researches of ultrasound-guided ESPB on concentration, volume, type of LA are necessary. Next, although no complications of ultrasound-guided ESPB have been reported in all included RCTs, risks such as LA toxicity, vascular puncture and pneumothorax still need our attention. Two studies have reported pneumothorax associated with ESPB [37, 38], and Selvi O et al.  reported unintended motor block linked to ESPB. More complications may appear as the increased use of ultrasound-guided ESPB in population. Last, compared to other regional block techniques such as transversus abdominis plane block (TAPB), serratus plane block (SPB), and Quadratus Lumborum Block (QLB), is the erector spine block more effective in some operations where the block areas overlap? Several RCTs on these topics published recently but far from achieving convincing conclusions[40-42]
Strengths and limitations
Our meta-analysis has several strengths. As far as we know, this is the first meta-analysis to evaluate the efficacy of ultrasound-guided ESPB in adults undergoing GA surgery. Besides, we performed this meta-analysis in compliance with the Cochrane Handbook and the PRISMA statement. Several notable limitations should be considered when interpreting the results. Firstly, the trials included have a modest sample size which could magnify the treatment effect. Secondly, the substantial heterogeneity was observed, one major factor result in heterogeneity is the diversity of surgery types (breast, lumbar spine, hip, abdominal etc). Parietal pain is more prominent in breast and lumbar spine, while visceral pain is the main component of postoperative pain following abdominal surgeries. The use of different types of opioid and supplementary analgesics such as paracetamol [23,24] may also add an extra heterogeneity. Furthermore, owing to all patients were under GA surgeries, sensory blocking could not be evaluated adequately to exclude potential block failures of ESPB. Last, although we conducted a comprehensive literature search, it is hard to rule out the possibility of missing studies.