Optimizing the superior trunk block for arthroscopic shoulder surgery: a randomized, double-blind comparison of low- vs. conventional-dose ropivacaine on anesthesia, analgesia, and diaphragmatic function

Abstract

Background

The superior trunk block (STB) has been proposed as an alternative to interscalene block for reducing the risk of hemidiaphragmatic paralysis (HDP) while maintaining adequate analgesia. However, the optimal local anesthetic dosage for STB has not been established. This study aimed to evaluate whether a low dose of ropivacaine could effectively provide surgical anesthesia, reduce the incidence of complete and partial HDP, and improve anesthetic safety and patient comfort during shoulder arthroscopy.

Methods

Eighty-eight patients scheduled for shoulder arthroscopy were enrolled and randomly allocated in a 1:1 ratio to either a low-dose group (LD group; 10 mL of 0.25% ropivacaine) or a conventional-dose group (CD group; 15 mL of 0.5% ropivacaine). The primary outcome was the incidence of HDP assessed at 3 h post-block. Secondary outcomes included rates of successful surgical anesthesia, pain scores, opioid consumption, duration and onset time of the block, handgrip strength, and patient satisfaction.

Results

The incidence of complete and partial HDP at 3 h post-STB was significantly lower in the LD group than in the CD group (31.8% vs. 84.1%, p < 0.001), with an absolute risk reduction of 52.3% (95% CI: 34.8% to 69.8%). All patients in both groups successfully underwent surgery without conversion to general anesthesia with tracheal intubation or requiring supplemental intraoperative analgesia. The overall patient satisfaction score was significantly higher in the LD group compared to the CD group (p = 0.003). However, patients in the LD group reported higher pain scores at 12 h post-block (p = 0.044), exhibited a shorter duration of sensory block (p = 0.001) compared to the CD group.

Conclusions

These findings indicate that a low-dose ropivacaine regimen (10 mL of 0.25%) for STB significantly reduces the incidence of HDP while still providing sufficient surgical anesthesia. However, this approach may be associated with reduced efficacy in postoperative analgesia.

Trial registration

This trial was registered at the Chinese Clinical Trial Registry (No. ChiCTR2000036608).

Background

Shoulder arthroscopy is a common surgical intervention for a variety of shoulder pathologies; however, it is frequently associated with considerable postoperative pain, with intensity comparable to that of major procedures such as gastrectomy or thoracotomy [1]. Interscalene brachial plexus block (ISB) remains the most effective regional technique for postoperative analgesia following shoulder arthroscopy [2]. Nonetheless, ISB carries a substantial risk of hemidiaphragmatic paralysis (HDP), which poses particular challenges for patients with compromised respiratory function [3]. In recent years, considerable effort has been directed toward identifying alternative nerve blocks that mitigate HDP while maintaining analgesic efficacy. The superior trunk block (STB), first introduced in 2014, has emerged as one such alternative [4]. STB is increasingly favored due to its capacity to deliver analgesia comparable to ISB, alongside a potentially reduced risk of HDP [5, 6].

Since 2019, multiple clinical trials have evaluated the impact of STB on the incidence of HDP and its efficacy in providing anesthesia and analgesia in patients undergoing shoulder arthroscopy [5–10]. Nevertheless, reported rates of HDP remain inconsistent across studies. For instance, Kim et al. [6] and Kang et al. [5] reported incidences of complete HDP—defined as a reduction in diaphragmatic excursion exceeding 75% from baseline—as low as 4.8% and 5.3%, respectively. In contrast, Robles et al. [9] observed complete HDP in 33.3% of cases. Similarly, the incidence of partial HDP (25–75% reduction in diaphragmatic movement) varies considerably, ranging from 0% [6] and 5.7% [7] to as high as 71% [5] in different reports. Furthermore, Lee et al. [8] concluded that STB did not significantly reduce the risk of HDP compared with ISB.

The dosage of local anesthetic (LA) administered in STB is hypothesized to be a major factor influencing the incidence of HDP. Our team recently performed the world’s first bilateral shoulder arthroscopy under STB anesthesia, using 10 mL of 0.25% ropivacaine on each side. Notably, neither side exhibited complete HDP and the patient experienced no dyspnea [11]. Based on these findings, we hypothesized that the incidence of HDP would differ significantly between patients receiving a low-dose regimen (10 mL of 0.25% ropivacaine) and those receiving a conventional dose (15 mL of 0.5% ropivacaine) for unilateral STB.

Methods

Study design

This study was a prospective, randomized controlled trial that received approval from the Ethics Committee of Shanghai Changzheng Hospital of Naval Medical University (Approval No. CZEC2020-10). It was also registered at the Chinese Clinical Trial Registry (No. ChiCTR2000036608). Written informed consent was obtained from all participants.

Eligible participants were adults aged 45 to 80 with American Society of Anesthesiologists physical status (ASA) I to III. We excluded patients with chronic obstructive pulmonary disease, transient ischemic attack, sleep apnea syndrome, or contraindications to peripheral nerve block. Additionally, patients who had a score of less than 1 (0 = no sensory block; 1 = partial sensory block; 2 = complete sensory block) on the pinprick test in the dominant area of the C5 and C6 nerve roots (lateral and triangular area of the upper arm) after 30 min of the block were excluded. Patients who were lost to follow-up or required conversion to endotracheal intubation were also excluded. Furthermore, diaphragmatic movement may not be visualized under ultrasound during clinical practice, even with the modified protocol introduced by Kim et al. [6] Therefore, during the preoperative visit, the investigators pre-probed the patient’s diaphragm using portable ultrasound, and patients whose diaphragm could not be visualized under ultrasound were also excluded.

Randomization and blinding

Patients were randomly allocated to the low-dose (10 ml of 0.25% ropivacaine) group (LD group) or conventional-dose (15 ml of 0.5% ropivacaine) group (CD group) in a 1:1 ratio using random numbers generated by SPSS (version 26.0, SPSS Inc., Chicago, USA). The randomization was balanced by randomly permuted blocks with block sizes of 6 and 8. A statistician who was independent of the study team prepared the randomization list. All STB were performed by a dedicated regional anesthesiologist (YYZ) with an assistant who did not play any other role in the study. Before administering the block to the patient, they opened the sealed envelope to learn about the grouping. Patients, follow-up investigators, and surgeons were blinded to the subgroups. To prevent patients from guessing the grouping through different volumes of LA, ropivacaine diluted to 0.25% and 0.5% were drawn up to 20 ml in 20 ml syringes. Furthermore, to ensure that the patient was unaware of the grouping, the regional anesthesiologist performed the block while the patient kept the ultrasound screen out of their eyesight.

Baseline measurements

After enrolment, demographic and morphometric characteristics of each patient were recorded, including age, sex, body mass index, ASA score, surgical side, and comorbid conditions. Handgrip strength was also assessed using a dynamometer (Jamar Plus, USA), and the maximum voluntary isometric contraction measurements were recorded. Standard monitoring was routinely performed once the patient entered the room. Patients inhaled low-flow pure oxygen (2 L min⁻¹) through nasal oxygen tubes, and ECG, non-invasive blood pressure, SpO₂, and heart rate were monitored routinely.

Before the block, following a reported procedure, the diaphragmatic excursion was assessed as a baseline measure using M-mode sonography (Wisonic Labat SP, China) [6, 12, 13]. The probe was positioned below the costal margin, between the midclavicular and anterior axillary lines, and directed medially, cephalad, and dorsally to view the posterior third of the hemidiaphragm. The liver and spleen were used as acoustic windows. The craniocaudal movement of the diaphragm was measured in centimeters during a standardized coached maximal inspiratory effort. Diaphragmatic excursion was assessed three times, and the results were averaged.

Superior trunk block technique

We used the methods described by Kang et al. [5] and Kim et al. [6] and modified the procedure according to our clinical practice. Precise positioning of the superior trunk of the brachial plexus is crucial to ensure the success of the block. Therefore, we developed a stepwise approach to superior trunk localization based on literature review [14, 15]. Ultrasonography was performed with a high-frequency linear ultrasound probe (5–15 MHz), and patients lay supine with their necks exposed. First, the transducer was placed in the supraclavicular fossa to obtain a short-axis view of the brachial plexus and the subclavian artery above the first rib (Fig. 1D). Then, the transducer was manipulated from the supraclavicular fossa to the interscalene groove, and the C7, C6, and C5 nerve roots were identified to better define the contiguous anatomy of the brachial plexus (Fig. 1A-C). Finally, once the C5 nerve root was identified, the transducer moved in the reverse direction, and the superior trunk formed by the C5 and C6 nerve roots was identified. The targeted level of insertion for the injection was proximal to the take off of the suprascapular nerve (Fig. 1E).

Fig. 1.

Serial ultrasonographic images of the brachial plexus nerve above the clavicle and the modified superior trunk block. SCM, sternocleidomastoid muscle; SA, scalenus anterior; SM, scalenus medius; IJV, internal jugular vein; CA, carotid artery; ScA, subclavian artery; ScV, subclavian vein; TP, transverse process; SSN, suprascapular nerve; ST, superior trunk; MT, middle trunk; LA, local anesthetic. C5, C6, and C7 emerge at the level of the corresponding transverse process (A, B, C). The brachial plexus nerve is located lateral to the subclavian artery in the supraclavicular fossa (D). The target injection site for the superior trunk block is where the suprascapular nerve emanates (E). The sandwich technique highlights that the blocking needle is injected with LA on both the superior and inferior sides of the superior trunk (F, G). White asterisks correspond to the needle tip. The white triangle shows the needle path. The grey area shows LA diffusion

After skin sterilization and infiltration with 1 ml of 1% lidocaine, the block needle (Pajunk Sonoplex, Germany) was inserted from lateral to medial using the in-plane technique. A key feature of our approach is the adoption of the sandwich injection technique. Once the needle tip was visualized below the superior trunk, 5 ml of 0.25% ropivacaine was administered in the LD group, while 10 ml of 0.5% ropivacaine was injected in the CD group. The needle tip was then carefully repositioned above the trunk, and the remaining local anesthetic was delivered. Throughout the injection, the needle tip was advanced slowly to promote circumferential spread of local anesthetic around the superior trunk. Under continuous ultrasound guidance, the regional anesthesiologist maintained clear visualization of the entire bevelled needle tip, ensuring it remained oriented toward the trunk at all times.

After the block was completed, each patient received a sensory block assessment by a blind investigator every 5 min until 30 min. Sensory block success was defined as the loss of cold sensation in three specific areas: the lateral side of the mid-humerus (C5), the lateral side of the forearm (C5, C6), and the dorsal surface of the proximal phalanx of the thumb (C6). Patients who achieved satisfactory sensory block were considered to have met the criteria for surgical anesthesia, which was defined as the ability to proceed with surgery without the need for general anesthesia, supplemental block, or local infiltration [16, 17]. The onset time was also recorded. For patients with incomplete or failed blocks, remedial blocks or general anesthesia with tracheal intubation were administered.

Perioperative management

During the procedure, patients were sedated with dexmedetomidine at a loading dose of 0.6 µg/kg and a maintenance dose of 0.4 µg/kg/h. They received oxygen at a rate of 2 L per minute via a nasal oxygen tube. If surgical anesthesia was not achieved, the patient would be transferred to receive general anesthesia. Intraoperatively, patients in both groups received azasetron (10 mg) and dexamethasone (5 mg). Any other intraoperative medications were administered based on the patient’s vital signs and the anesthesiologist’s clinical experience and were recorded accordingly.

After the procedure, patients were transferred to the post-anesthesia care unit (PACU) and monitored until they met the discharge criteria [18]. Subsequently, patients were transferred to the ward for continued care. At the surgical center, all patients received multimodal analgesia, which consisted of intravenous ketorolac 60 mg (or 30 mg if the patient was older than 65 years) every 8 h, oral loxoprofen 60 mg every 8 h, and intra-articular 2% lidocaine 0.1 g on the first postoperative day (for rotator cuff repair). Opioid remedial analgesia was administered if the pain score exceeded 4 on the 11-point numerical rating scale (NRS), where 0 indicates no pain and 10 indicates the worst pain imaginable. Opioid remedial analgesia included intramuscular morphine, oral oxycodone, and topical buprenorphine transdermal patch. All opioid consumption was converted and recorded as intravenous morphine equivalents [19].

Outcomes

The primary outcome was the incidence of HDP, which was defined as the rate of complete and partial HDP in cases after 3 h of the block. The secondary outcomes included the rate of successful surgical anesthesia (defined as the completion of surgery without the need for remedial block or general anesthesia), postoperative pain assessment (measured at PACU, 6-hour intervals during the first 24 h after the block), opioid consumption (including during hospitalization and at discharge), block duration (defined as the time from the block until the patient first felt pain at the surgical site), onset time, degree of motor block (measured using the change in handgrip strength in the PACU compared with the baseline), patients’ satisfaction (measured using a verbal rating scale (VRS), where 0 represents total dissatisfaction and 10 represents the most satisfactory subjective answer). All assessments and data collection were performed by investigators who were unaware of the group allocation.

Statistical analysis

The sample size was calculated using PASS (version 2019, NCSS, Kaysville, Utah, USA) based on the primary outcome. According to a previous study, the overall incidence of complete and partial HDP was estimated to be 76.3% in the CD group [5]. The modified STB was expected to reduce the overall incidence of complete and partial HDP by 50%, so it was projected to be 38.15% in the LD group. A total of 78 patients were required to achieve 90% power with a type-1 error of 0.05. Considering a 10% rate of missed visits and refusals, the final minimum number of subjects needed for the intervention and control groups was 44 each, for a total of at least 88 subjects included in the study.

The primary outcome was analyzed according to the intention-to-treat principle, and worst-case imputation was used for missing data. A per-protocol analysis was also performed to enhance the results. Binary and categorical variables were analyzed using Fisher’s exact test. To account for information loss during data transformation, we also performed a statistical comparison of diaphragm excursion before and after the STB. Quantitative data were tested for normality using the Shapiro-Wilk test. Diaphragmatic excursion before and after STB in the same group was compared using the paired t-test. The intergroup differences were analyzed using the student t-test or the Mann-Whitney U test, depending on the normality of data distribution. A two-sided P value less than 0.05 was considered statistically significant. The statistical analysis was performed using SPSS (version 26.0, SPSS Inc., Chicago, USA).

Results

This text describes a study involving 116 patients who underwent shoulder arthroscopic surgery between March 1, 2021, and December 31, 2022. Out of these patients, 88 consented to participate and were randomly divided into two groups. All patients completed the procedure without needing to switch to general anesthesia with tracheal intubation or a rescue block. Every patient received the assigned block and completed the planned follow-up assessment (Fig. 2). There were no significant differences between the two groups in terms of age, sex, ASA physical status, BMI, duration, or type of surgery (Table 1).

Fig. 2.

Consolidated Standards of Reporting Trials (CONSORT) flow diagram

Table 1.

Patient characteristics

	CD Group (n = 44)	LD Group (n = 44)	P value
Age (years)	58.3 ± 7.2	56.6 ± 7.4	0.390
BMI (kg/m2)	24 ± 2.4	23.5 ± 3.2	0.439
Duration of surgery (min)	68.3 ± 17.4	65.2 ± 16.7	0.387
Sex			1.000
Male	20 (47.7%)	22 (50.0%)
Female	24 (52.3%)	22 (50.0%)
ASA status			0.433
I	10 (22.7%)	14 (31.8%)
II	32 (72.7%)	26 (59.1%)
III	2 (4.5%)	4 (9.1%)
Surgical side			1.000
Right	31 (70.5%)	30 (68.2%)
Left	13 (29.5%)	14 (31.8%)
Surgical procedure			0.549
RCR	39 (88.6%)	36 (81.8%)
Non-RCR	5 (11.4%)	8 (18.2%)

BMI Body mass index (calculated as weight in kilograms divided by height in meters squared), ASA American Society of Anesthesiologists, RCR rotator cuff repair. Values are presented as Mean ± SDs, numbers (%) or medians (interquartile range). The P value for the t test, Mann‒Whitney U test, and Fisher exact test is set at 0.05

The study found that the incidence of HDP was significantly lower in the LD (low dose) group compared to the CD (conventional dose) group after 3 h of the blocks (31.8% vs. 84.1%, p < 0.001). The absolute rate difference was 52.3% (95% CI 34.8% to 69.8%). In the CD group, 3 patients (6.8%) experienced complete HDP, and 34 (77.3%) experienced partial HDP. In contrast, only 14 patients (31.8%) in the LD group experienced partial HDP (Table 2). There was no statistically significant difference between the two groups at baseline during a deep breath. However, after 3 h of the block, the LD group showed significantly less reduction in the mean diaphragmatic excursions than the CD group (3.0 ± 0.8 cm vs. 2.1 ± 1.1 cm, with a mean difference of 0.88 cm and a 95% CI of 0.49 to 1.28 cm, p < 0.001) (Fig. 3). There was no significant difference in SpO₂ between the two groups at baseline and in the PACU.

Table 2.

Incidence of hemidiaphragmatic paralysis

	CD Group (n = 44)	LD Group (n = 44)	P value
Any hemidiaphragmatic paralysis (complete or partial)	37 (84.1%)	14 (31.8%)	<0.001
Hemidiaphragmatic paralysis			<0.001
No	7 (15.9%)	30 (68.2%)
Partial	34 (77.3%)	14 (31.8%)
Complete	3 (6.8%)	0 (0%)

Data are presented as numbers (%). The P value for the Fisher exact test is set at 0.05

Fig. 3.

Changes in diaphragmatic excursion at baseline and 3 h after blocks. Differential comparison of diaphragmatic excursion at baseline and 3 h after blocks in the CD and LD groups (A, B). Comparison of intergroup variability in diaphragmatic excursion (C). ns, not statistically significant; **, P < 0.001

No statistically significant difference in NRS pain scores was observed between the two groups at any of the assessed time points: before block (baseline), in the PACU, or at 6 h, 18 h, and 24 h after the block. NRS scores at 12 h after the block in the LD group were slightly higher than those in the CD group (2 [0, 2] vs. 0 [0, 2], p = 0.044). Although the absolute difference in NRS scores at 12 h was modest, it reached statistical significance and may reflect the shorter duration of sensory block in the LD group. The mean duration of the block was shorter for the LD group (11.8 ± 3.2 h) than the CD group (14.5 ± 4.3 h, p = 0.001). Opioid consumption at the hospital was also higher in the LD group (10 [0, 10] vs. 10 [0, 20] mg morphine equivalents, p = 0.041). Patients in the CD group experienced a greater reduction in grip strength (80.8%) compared to the LD group (68.5%), corresponding to approximately 2.6 kg greater loss of grip strength in the CD group than the LD group (Table 3). There was no statistical difference in opioid prescriptions post-discharge. At the postoperative follow-up, the satisfaction score of the patients in the LD group was higher than in the CD group (8 [8, 9] vs. 7 [7, 8], p = 0.003).

Table 3.

Pain Scores, opioid Consumption, block Duration, adverse Effects, and satisfaction outcomes

		CD Group		LD Group		P value
Sample size		44		44
Rate of successful surgical anesthesia		44 (100%)		44 (100%)		1.000
Pain scores (NRS 0–10)
Baseline		0 (0–0)		0 (0–0)		0.323
In the PACU		0 (0)		0 (0)		1.000
At 6 h		0 (0)		0 (0)		1.000
At 12 h		0 (0–2)		2 (0–2)		0.044
At 18 h		2 (0.5–2.5)		2 (1–2)		0.622
At 24 h		2 (2–3)		2 (2–3.5.5)		0.626
Worst pain		3 (2–3)		3 (2–4)		0.153
Opioid consumption (mg)
Hospitalization		10 (0–10)		10(0–20)		0.041
Discharge prescriptions		50 (15–60)		50 (30–70)		0.293
Block duration (h)		14.5 (4.3)		11.9 (3.2)		0.001
Onset time (min)		15.6 (5.5)		16.1 (5.3)		0.623
Decrease in handgrip strength at PACU		80.8 (14.9)		68.5 (17.7)		0.001
Satisfaction scores (VRS 0–10)		7 (7–8)		8 (8–9)		0.003
Adverse effects
PONV		5 (11.4%)		4 (9.1%)		1.000
Rebound pain		3 (6.8%)		6 (13.6)		0.484

NRS  numerical rating scale, VRS verbal rating scale, PACU postanesthesia care unit, PONV postoperative nausea and vomiting. Values are presented as Mean ± SDs, numbers (%) or medians (interquartile range). The P value for the t test, Mann‒Whitney U test, and Fisher exact test is set at 0.05

Discussion

In our present study, we investigated the effectiveness of using low dose (10 ml of 0.25% ropivacaine) ropivacaine to implement an STB for patients undergoing shoulder arthroscopy to reduce the incidence of HDP. Our data demonstrated that the LD group further reduced the incidence of HDP and provided adequate surgical anesthesia. While the LD group was slightly inferior to the CD group in providing postoperative analgesia.

HDP is a common complication of ISB and poses an unnecessary risk to patients with respiratory compromise [3]. Despite the compensatory effect of the contralateral diaphragm, diaphragm dysfunction has a potential impact on survival and is associated with disorders such as sleep disturbances, exercise intolerance, and drowsiness [12, 20, 21]. The phrenic nerve originates from the C3 to C5 segments and descends obliquely across the surface of the anterior scalene muscle behind the prevertebral fascia, and there are many anatomical variants of the phrenic nerve [14, 22]. Despite the use of extrasheath local anesthetic injection or a reduction in local anesthetic dosage, phrenic nerve palsy cannot be completely avoided during ISB [23–25]. Moreover, ISB may damage the long thoracic nerve, and the dorsal scapular nerve located in the middle scalene muscle with the laterodorsally in-plane approach and even cause additional epidural spreading [14, 26].

Researchers have recently proposed many alternatives to ISB, including costoclavicular blocks. The suprascapular nerve block, despite having an anterior approach that provides a more comprehensive block, does not cover all the nerves innervating the shoulder and may not provide adequate surgical anesthesia [26, 27]. The posterior approach can avoid the incidence of HDP but may not provide sufficient surgical anesthesia, and anatomical variations make visualizing the suprascapular nerve by ultrasonography challenging. Anatomical evidence suggests that the main reason for ensuring adequate block at other alternatives is the diffusion of local anesthetic (LA) to the level of the superior trunk of the brachial plexus [28, 29]. While costoclavicular blocks, supraclavicular blocks, and infraclavicular blocks can achieve sufficient analgesia, they require a large dose of LA, which increases the risk of HDP. Direct injection of 5 mL of radiolabeled dye at the level of the superior trunk of the brachial plexus results in staining of the upper trunk, suprascapular nerve, lateral pectoral nerve, and the C5 and C6 nerve roots [30]. This pattern of dissemination demonstrates the potential for effective analgesia. STB has been reported to provide superior postoperative analgesia compared with costoclavicular blocks using the same dose of LA [31]. In conclusion, STB appears to be one of the most promising alternatives to ISB for regional anesthesia for shoulder surgery.

The brachial plexus nerve and phrenic nerve are both covered by the prevertebral fascia, and a large volume of LA could easily diffuse ventrally or cephalad to the phrenic nerve [14]. Prior studies have found that the volume of 10 ml of ropivacaine had a ceiling effect on block duration [32], and reducing the volume of LA was attempted. However, using 10 ml of 0.5% ropivacaine for STB still resulted in a 67.2% rate of HDP after 3 h of the block, which may be related to the penetration by the concentration gradient. A study by Kim et al. [10] found that the incidence of HDP was 45.2% in the treatment group using 7 mL of 0.5% ropivacaine for STB. Low concentrations of ropivacaine could provide similar analgesic effects as high concentrations while reducing the incidence of side effects associated with LA and minimizing motor block [33, 34]. Therefore, the final choice of using 10 ml of 0.25% ropivacaine for STB resulted in a lower incidence of HDP.

The technique for injection merits further investigation since a lower volume and concentration of LA are applied. Some trials showed high success rates, short onset times, and long block duration when administered LA surrounds the nerve [35, 36]. Benefiting from the advancement in ultrasound technology, the sandwich-injection technique was developed through our clinical practice. We subtly adjusted the needle tip and kept the entire beveled needle tip oriented towards the superior trunk with a slower and lower pressure injection of LA. The effectiveness of the modified technique was also verified by the results that the block onset time was not statistically different between the groups. Moreover, we successfully performed modified STB in a patient undergoing bilateral shoulder arthroscopy, which may be the first report of successful bilateral shoulder arthroscopy under bilateral STB [11].

In previous studies, only Kim et al. [6] used MAC combined with intravenous propofol sedation, however, propofol can cause respiratory depression, which may increase the risk in patients who have developed HDP [37]. Dexmedetomidine is known to produce sedation with a significantly reduced risk of airway obstruction and respiratory depression [38]. It was administered intravenously both prior to and during surgery to reduce intraoperative anxiety and tension, thereby improving patient comfort. Patients under dexmedetomidine sedation remain readily arousable during surgery, enhancing anesthetic safety. Based on previous studies [37, 39] and institutional experience [40, 41], we have innovated the use of MAC under dexmedetomidine sedation with STB in patients undergoing shoulder arthroscopy. Dexmedetomidine exerts hypotensive effects, which help minimize bleeding in the surgical field during shoulder arthroscopy. This results in improved visualization and increased surgeon satisfaction. Notably, dexmedetomidine induces effective sedation while also exhibiting inherent analgesic properties. Under the same conditions, this dual effect may have contributed more substantially to the outcomes in the LD group, potentially narrowing the observed difference in analgesic efficacy compared to what might have been observed without its use.

In terms of postoperative analgesia, the NRS score of the LD group was higher than that of the CD group at the 12 h after the block, which may be due to the shorter duration of the block in the LD group. Although the absolute difference in NRS scores at 12 h was modest, it reached statistical significance and may reflect the shorter duration of sensory block in the LD group.Notably, 6 patients (LD group) and 3 (CD group) in the two groups experienced postoperative rebound pain respectively. However, there is no evidence to suggest that the incidence of rebound pain is related to the volume and concentration of LA [42]. Meanwhile, in our institution, intra-articular infiltration analgesia was routinely given on the first day after surgery to patients who underwent rotator cuff injury repair, and the incidence of rebound pain was much lower than that reported in the literature [42], further studies are needed to confirm its effectiveness in the future. Interestingly, the preserved motor function—reflected by significantly better grip strength in the LD group (68.5% of baseline vs. 80.8% reduction in the CD group)—likely contributed markedly to patient satisfaction, as it facilitates early mobilization and functional recovery. In addition, the avoidance of respiratory symptoms such as mild dyspnea may have further enhanced the overall experience, outweighing the transient difference in pain at a single time point.

This study has several limitations that should be considered. First, although the incidence of complete HDP in LD group was zero—the lowest reported in current literature on STB—the relatively modest sample size may limit the generalizability of this finding. Future studies with larger cohorts are needed to confirm these results. Second, to ensure technical consistency and minimize performance bias, all nerve blocks were performed by a single regional anesthesiologist (YYZ). While this approach enhances internal validity, it may limit the generalizability and reproducibility of the results. Therefore, future studies are warranted to evaluate the learning curve and reproducibility of the described “stepwise” localization and “sandwich” injection techniques when performed by other clinicians. Third, the lack of systematic pulmonary function assessment using bedside spirometry represents another limitation. Although no patients showed clinical signs of respiratory distress, spirometry could have provided more sensitive and objective measures of asymptomatic diaphragmatic dysfunction and changes in lung volumes. Incorporating spiometric evaluation in future studies would allow a more comprehensive assessment of respiratory outcomes. Finally, the use of postoperative intra-articular analgesic infiltration may have attenuated intergroup differences in pain scores, potentially favoring the LD group. Moreover, intraoperative intravenous dexmedetomidine might have influenced postoperative analgesia and extended block duration, thereby reducing variability in pain and block-related outcomes between the groups.

In summary, our modified superior trunk block technique, utilizing a low-dose regimen of ropivacaine (10 mL of 0.25%) combined with refined injection methods, significantly reduces the incidence of both partial and complete hemidiaphragmatic paralysis while still delivering effective surgical anesthesia for shoulder arthroscopy, despite yielding less effective postoperative analgesia.

Abbreviations

ASA

American Society of Anaesthesiologists

BMI

body mass index

CA

carotid artery

CI

confidence interval

CONSORT

Consolidated Standards of Reporting Trials

HDP

hemi diaphragmatic paralysis

LA

local anaesthetic

IJV

internal jugular vein

IQR

interquartile range

ISB

Interscalene brachial plexus block

MAC

monitored anaesthesia care

MT

middle trunk

NRS

numerical rating scale

PACU

postanaesthetic care unit

PONV

postoperative nausea and vomiting

RCR

rotator cuff repair

RR

relative risk

SA

scalenus anterior

SCM

sternocleidomastoid muscle

ScA

subclavian artery

ScV

subclavian vein

SD

standard deviation

SM

scalenus medius

SSN

suprascapular nerve

ST

superior trunk

SpO₂

pulse oxygen saturation

STB

superior trunk block

TP

transverse process

VRS

verbal rating scale

Authors’ contributions

Name: Jinxiang Zhang, MD. Contribution: This author helped conduct the study, write and revise the manuscript; and read and approve the final version of the manuscript.Name: Yangyang Zhou, MD.Contribution: This author helped design the study, develop the protocol, enroll patients, read and approve the final version of the manuscript, and contribute equally to the first author.Name: Peng Ding, MD. Contribution: This author helped conduct the study, develop the protocol, revise the manuscript, read and approve the final version of the manuscript, and contribute equally to the first author.Name: Dongyu Zheng, MD.Contribution: This author helped analyze data, revise the manuscript, read and approve the final version of the manuscript, and contributed equally to the first author.Name: Hongwei Zhu, MD.Contribution: This author helped enroll patients, and read and approve the final version of the manuscript.Name: Ming Gong, MD.Contribution: This author helped analyze data, and read and approve the final version of the manuscript.Name: Liye Yang, MD.Contribution: This author helped design the study, develop the protocol, oversee study activities, review final data, and critically review the manuscript.Name: Yonghua Li, MD.Contribution: This author served as the study director and helped design the study, develop the protocol, oversee study activities, review final data, and critically review the manuscript.

Funding

This work was supported by the Medical Innovation Research Project of Shanghai Science and Technology Commission (Grant Numbers: 21Y11906400, 22Y11904000), the second round of the Shanghai Shenkang Hospital Development Centre’s “Three Year Action Plan to Promote Clinical Skills and Clinical Innovation in Municipal Hospitals” research-oriented physician innovation and transformation ability training project (Grant Number: SHDC2023CRD024), the Clinical research projects initiated by researchers in demonstration research wards (Grant Number: 2023YJBF-PY11), and the Youth Talent Program of Zhengzhou Joint Logistic Support Center (No.202310).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was a prospective, randomized controlled trial that received approval from the Ethics Committee of Shanghai Changzheng Hospital of Naval Medical University (Approval No. CZEC2020-10). All methods were performed in accordance with relevant guidelines and regulations and with CONSORT recommendations. Before participation, all the patients and/or their legal guardians provided written informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.