We thank Sethuraman and Suresh [1] for their interest in our recent study on the role of erector spinae plane (ESP) block in acute gastrointestinal injury (AGI) [2]. We appreciate the insights that the mechanism of action of ESP block on local nerve block remains unclear and the craniocaudal spread of the injectate may vary widely due to various patient-specific factors. In the ESPAGI study, it was impractical to determine the levels of spinal nerve block through sensory assessment, because most of the patients were administrated with sedatives. However, we have observed that some patients with paralytic ileus experienced their first flatus within several minutes after first receiving the ESP block. This phenomenon could be explained by the effective blockade of local sympathetic nerves innervating the bowel. In a cadaveric study of anatomical spread of dye with ESP block, the median number of stained spinal nerves was 3.5 (IQR: 2–4) [3]. Thus, the ESP block was performed at thoracic (T) level 8 in our study, and the ropivacaine could diffuse and affect at least half of the targeting sympathetic nerves. Different from thoracic paravertebral block, ESP block cannot rapidly spread ropivacaine into the paravertebral space. The vascular pathway on the intertransverse connective tissue complex and the hyperpermeability of the fascia to macromolecules probably allow a gradual seepage of local anesthetic into the paravertebral space [4]. Through these ways, continuous low-dose administration of ropivacaine after a bolus infusion may permit the spread of ropivacaine into the paravertebral space. Besides, prolonged supine position of ICU patients might also facilitate drug diffusion.

We have corresponding explanations for the queries to some results [1]. First, according to the study protocol, patients who received the intervention for at least 3 days would be included in the analysis. The median duration of intervention in both cohorts was 7 days (IQR: 5–7). The patients with catheter displacement received intervention for more than 3 days and were included in the analysis. Furthermore, the outcomes of the ESPAGI study were already compared between and within groups [2]. The relatively small numbers of patients in each group led to imbalances in some indicators at baseline. The comparison of outcomes between and within groups should be interpreted with caution. There were no significant differences in high-sensitivity C-reactive protein (hs-CRP) with intra-group comparison. Hence, it was incautious to conclude that hs-CRP declined drastically in control group. Besides, abdominal perfusion pressure (APP) was calculated by mean arterial pressure (MAP) and intra-abdominal pressure (IAP) (shown in protocol). Therefore, the MAP was not presented anymore. Finally, the protective effect of ESP block on mortality in AGI patients could partly attributed to the improvement of organ dysfunction and enteral nutrition status [5], as confirmed by the decreased Sequential Organ Failure Assessment (SOFA) score and increased enteral feeding dose in our results. Considering the pathophysiology of AGI is rather complex, the 28-day all-cause mortality should be interpreted as exploratory outcome and needs to be further confirmed in future studies.