Serratus anterior plane blocks for median sternotomy anesthesia: back to the drawing board Authors: Rohesh J Fernando, Salvador E Pena, Tomasz Dobrzanski, David Watts Published: 2026 Pain after cardiac surgery can pose a significant problem, and as many as 75% of patients experience moderate or severe levels of pain within the first few days (1). Recently, regional anesthesia has become increasingly utilized to aid in pain control, with a myriad of different blocks being explored (2). Fascial plane blocks are a specific type of regional block in which injection of local anesthetic between muscle layers can provide analgesia by targeting the nerves relevant for the surgical site as they course through the area (3). For pain related to a sternotomy incision, the fascial plane blocks that would typically be used are the superficial parasternal intercostal plane (SPIP) block (formerly known as the pectointercostal fascial plane block), deep parasternal intercostal plane (DPIP) block (formerly the transverse thoracic muscle plane block), and the erector spinae block (3). The SPIP and DPIP blocks target the anterior cutaneous branches of the intercostal nerves, whereas the erector spinae block targets the ventral and dorsal rami. While the serratus anterior plane (SAP) block also targets the intercostal nerve, it primarily affects the lateral cutaneous branches, while not reliably blocking the anterior cutaneous branch needed to cover midline incisions. Thus, although the SAP block has its role in cardiac surgery, it would seem more appropriate for an approach via thoracotomy, where the lateral cutaneous branch of the intercostal nerve provides sensory innervation (3). The SAP block can be performed in the superficial and deep locations. The superficial SAP block is performed between the latissimus dorsi and the serratus muscle and is more likely to block the intercostobrachial, long thoracic and thoracodorsal nerves which course through this plane, with the only sensory nerve being the intercostobrachial (4,5). The deep SAP block is performed between the serratus muscle and the external intercostal muscle to block the lateral cutaneous branch before its division between anterior and posterior branches. Again, this anterior division of the lateral cutaneous branch is distinct from the anterior branch of the intercostal nerve, as demonstrated in Figure 1, and should not be confused (6). Chest wall innervation. Note that the anterior branch (referred to as “division” instead of “branch” on the figure for simplicity) of the lateral cutaneous branch of the intercostal nerve is distinct from the anterior cutaneous branch of the intercostal nerve. Pec, pectoralis. We therefore read with great interest the feasibility trial performed by Bailey et al. in which bilateral SAP catheters were used to treat pain from cardiac surgery via sternotomy (7). In this double-blinded, placebo-controlled, single-center trial, 52 patients were randomized, with 25 patients being analyzed per group after one patient was excluded from each group due to not receiving the intended block. Continuous SAP blocks were performed postoperatively in the intensive care unit. Bailey et al. stated that the block they did was most similar to the deep SAP block (7). The protocol dictated that the patients should initially receive 20 mL of either 0.2% ropivacaine or normal saline (10 mL through the needle and then 10 mL through the catheter), after which 10 mL was given every 2 hours in a programmed intermittent bolus strategy. Since catheters were placed bilaterally, this protocol was utilized for both sides with the intended duration of 72 hours (7). The authors did not find a difference in pain scores, opioid consumption, or recovery metrics, although the study was not specifically designed to determine this; rather, the study was a feasibility study, and the authors concluded that it was not possible for them to perform a randomized controlled trial as a single center due to low enrollment rates and high rates of adverse events, specifically pneumothorax (7). Perhaps the most controversial aspect of the study was the choice of SAP block for midline sternotomy coverage. In the original study by Blanco et al. with 4 healthy volunteers, there was a lack of midline sensory coverage (4). This correlates with the apparent absence of midline staining in cadaveric studies by Biswas et al. and Perez-Herrero et al., suggesting it is very unlikely that the anterior cutaneous branch is anesthetized by a SAP block (8,9). In fact, Perez-Herrero et al. specifically studied the dye staining of SPIP in conjunction with deep SAP blocks in an attempt to demonstrate the addition of midline coverage by the SPIP (9). Their study further confirmed a lack of staining of the anterior cutaneous branch of the intercostal nerve and therefore, lack of midline sensory coverage by the SAP block. Taken together, neither cadaveric nor human studies support the use of the SAP block for coverage of the sternum. Given these questions regarding its utility for sternotomy, we would not have chosen this block for this patient population. Furthermore, we would have suggested performing cold and pinprick sensory testing to assess the block for efficacy over the sternum, as pain scores, opioid consumption, and quality of recovery metrics introduce patient- and provider-specific biases in the perception and assessment of pain, respectively. Although the data do indeed suggest that this trial was not feasible at their institution, there are several additional issues that remain unexplained. First, the authors stated that they used intention-to-treat in their analysis and therefore included all patients regardless of the catheter location (7). The use of intention-to-treat seems reasonable, but this statement introduces the possibility that not all catheters were placed in the desired area. Although the authors stated that the “optimal” location had not been previously elucidated (7), there is no information provided on how much variation there was in catheter location within this study or how often the catheter was placed in the intended location. Additionally, the infusion was stopped early in 5 patients in the treatment group and 6 patients in the placebo group. The authors do not give specific reasons for this other than stating that the majority of instances were due to “nonclinical reasons” (7). Further confounding the picture is the finding that 4 patients in the treatment group and 6 patients in the placebo group experienced kinking of the catheter. This would have presumably affected drug delivery, but no information is given on the duration of the kink and time to resolution. Finally, the authors listed the intervention being discontinued in 9 patients in the treatment group and 10 patients in the placebo group. While the number for the treatment group could be explained by the sum of the infusion being stopped early and catheter kinking, this rationale does not equate for the placebo group. We were not able to find a clear delineation of the 10 terminated catheters in the placebo group and therefore are unsure how exactly this was number was determined. Regardless, these problems reduce the ability to make accurate conclusions about the efficacy of the treatment. Although one may argue that these issues are less of a concern since the primary purpose was not to define efficacy, they also raise questions about the generalizability of these findings as it relates to feasibility. Catheter kinking may not have been preventable, but it is plausible that the early termination of the infusion could have been avoided since it was not due to clinical reasons. Giving the specific reasons would help elucidate this as something like unintentional pulling out the catheter represents a very different issue than stopping the infusion early due to miscommunication. Another perplexing finding was the high rate of pneumothorax at 12% (7). The authors acknowledge this is much higher than the previously reported rates and propose it could be related to catheter placement. Furthermore, if it were related to the catheter, they put forward the suggestion that it could be a result of the catheter stylettes. As an alternative hypothesis, they conject that perhaps their protocol increased their rate of detection due to being more meticulous, although this seems less likely as patients often receive chest radiography routinely postoperatively after cardiac surgery. Regardless, this highlights the previous question of whether the efficacy of the infusion could have been compromised by inappropriate catheter placement. We applaud the effort to place two catheters, however this strategy often limits the amount of local anesthetic that can be injected into the fascial plane (10). Given the need to avoid local anesthetic toxicity with a heart that is potentially irritable or dysfunctional after cardiac surgery, careful planning is required (11). Adequate volume allowing for sufficient local anesthetic spread is of great importance in these fascial plane blocks, as shown by Biswas et al. in their studies investigating optimal injectate spread in the SAP block (8). There are successful studies utilizing the bilateral catheter strategy in the cardiac surgery population, however with differences in both local anesthetic choice and method of infusion. For example, Rubin et al. demonstrated improved pain scores and reduced opioid consumption with either superficial or DPIP catheters compared to no regional anesthetic in a retrospective study (12). This was done for patients that underwent cardiac surgery with median sternotomy with 10 mL 0.1% ropivacaine infused continuously every hour. In contrast, although the surgical incision was similar, the study by Bailey et al. utilized 10 mL every 2 hours using an intermittent bolus technique (7). This comparatively represents less volume over time, which could be important for a fascial plane block. Another notable difference between the two studies mentioned above is the method of infusion. While it has been shown that intermittent bolus programing improves local spread compared to a continuous rate in the epidural space (13), there is disagreement about the optimal dosing strategy regarding dosing for facial plane blocks (14). Cadaveric studies postulate that bolus-dosed strategies may propel local anesthetic deeper along fascial planes by providing high driving hydraulic pressure yielding greater cranio-caudal spread (15,16). Despite improved spread with bolus-based dosing strategies for indwelling catheters in cadaveric studies, clinical studies have yielded conflicting results (17-20). A significant consideration in fascial plane block studies is the rate at which the bolus is administered. While the type of pump was not specified in the present study by Bailey et al., a commonly used pump delivers a local anesthetic bolus at a rate of 250 mL/hour, an equivalent of administering a 10 mL bolus over 2 minutes and 24 seconds, which is significantly slower than a proceduralist would dose that volume (21). Nevertheless, it is possible that the initial manual injections were administered at a faster rate than described above, which could have theoretically manifested as an improvement in early outcomes. However, this was still an unlikely possibility based on the type of block. For the SAP block to achieve midline coverage, this would require local anesthetic to spread a great distance posteriorly to the ventral rami of the spinal nerve or anteriorly to the anterior branch of the intercostal nerve. This degree of local spread likely did not occur in the study, as evidenced by similar pain scores and opioid consumption at all timepoints between the two groups. These findings further emphasize that the SAP block may not be the most effective approach to regional anesthesia in the post-sternotomy patient. In summary, the study by Bailey et al. investigating the benefits of bilateral SAP catheters for median sternotomy in cardiac surgery does not support a larger randomized controlled trial for both mechanistic and technical reasons. While this study appears to have been challenging to carry out, there are several questions that remain about the study and technique, which could limit generalizability. We appreciate the effort to place bilateral catheters with an aim to provide superior analgesia for a surgery with moderate to high pain scores, and we recognize the challenges that remain in determining the best strategy for dosing peripheral nerve catheters. Most importantly, SAP blocks have not been shown to provide sternal coverage in existing literature, and this was only further supported by the lack of analgesic and quality of recovery benefit in this study. Thus, a larger randomized controlled trial of SAP blocks for this patient population would likely not be worthwhile, particularly when considering that alternative blocks have demonstrated efficacy for median sternotomy. Supplementary The article’s supplementary files as Acknowledgments None. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Footnotes Provenance and Peer Review: This article was commissioned by the editorial office, Journal of Thoracic Disease. The article has undergone external peer review. Funding: None. Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0504/coif). R.J.F. serves as an unpaid editorial board member of Journal of Thoracic Disease from September 2025 to August 2027. The other authors have no conflicts of interest to declare. References Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0504/prf Supplementary Materials The article’s supplementary files as