INTRODUCTION
Annually, more than 500,000 shoulder and proximal humerus surgeries are performed in the United States (USA) [1]. Most of these procedures involve the repair of labral and rotator cuff injuries [2]. Although minimally invasive techniques have advanced substantially in recent decades, these operations remain associated with considerable postoperative pain [3]. In this context, brachial plexus blocks are widely used because they provide effective postoperative analgesia for upper limb procedures [4].
The interscalene brachial plexus block (ISB) is the most commonly used regional anesthesia technique for shoulder surgery and has been shown to significantly reduce postoperative pain scores and opioid consumption [4]. This technique involves the injection of local anesthetic at the level of the cricoid cartilage, between the anterior and middle scalene muscles, targeting the C5 and C6 nerve roots [5]. Despite its analgesic efficacy, ISB is associated with a high incidence of unilateral diaphragmatic paresis due to the close proximity of the injection site to the phrenic nerve, which courses superficially along the anterior scalene muscle [6].
In contrast, the superior trunk block (STB) has emerged as a promising alternative, providing effective analgesia with a lower incidence of diaphragmatic paralysis and fewer postoperative neurological risks [7]. This technique, first described by Burkett-St. Laurent et al. [6], targets the C5 and C6 nerve roots distal to the interscalene groove, near their convergence into the superior trunk and before the origin of the suprascapular nerve, which may explain the decreased involvement of the phrenic nerve.
Based on this evidence, the primary aim of this systematic review was to compare the incidence of phrenic nerve involvement between STB and interscalene block (ISB) performed for upper limb surgeries. Secondary outcomes included postoperative pain intensity, opioid consumption, patient satisfaction, motor block duration, and the incidence of Horner’s syndrome (oculosympathetic paresis).
MATERIALS AND METHODS
This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [8]. The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420250654685. This study was exempt from Institutional Review Board approval.
Inclusion and exclusion criteria
We included only randomized controlled trials (RCTs) involving adults (≥ 18 years) classified as American Society of Anesthesiologists (ASA) I-III that directly compared STB with interscalene nerve block performed for upper limb surgeries and reported the incidence of hemidiaphragmatic paralysis (HDP) assessed by ultrasound. In this review, a reduction of more than 75% in diaphragmatic excursion, or an equivalent validated ultrasound threshold, was defined as complete HDP; a reduction between 25% and 75% was defined as partial HDP, and less than 25% as no paralysis [9,10]. Both single-shot and continuous catheter techniques were considered eligible. Secondary outcomes included postoperative opioid consumption, pain intensity (measured using scales such as the visual analog scale [VAS] or numeric rating scale [NRS]), motor block duration, patient satisfaction, and the occurrence of oculosympathetic paresis.
Studies were excluded if they involved patients with ASA physical status greater than III, additional or alternative surgical procedures, other methods of HDP assessment (e.g., chest X-ray), or pediatric populations. Trials were also excluded if the intervention consisted solely of STB or compared it with alternative nerve blocks, sham, or placebo, involved multiple-dose comparisons, or failed to describe the method used to assess diaphragmatic function postoperatively. Observational studies, case reports, reviews, or publications with incomplete outcome data that could not be retrieved were excluded.
Study selection
To identify relevant studies, the following databases were searched through June 2025: PubMed, Embase, Scopus, Cochrane, and Web of Science. No restrictions on language or year of publication were applied. The detailed search strategy included the terms “superior trunk,” “superior trunk block,” and “block,” combined with the Boolean operators “OR” and “AND,” as described in the supplementary Data 1. Study selection was conducted independently by two reviewers (DFC and EBSN) in two phases. First, titles and abstracts were screened; then, full-text articles were assessed to determine eligibility. Any disagreement between reviewers was resolved through discussion with a third reviewer (JPAC).
Data extraction
Data extraction was conducted independently by two reviewers (DFC and EBSN). Any discrepancies were resolved by a third reviewer (JPAC), who verified the entire extraction process. Data were collected using standardized forms in Microsoft Excel (version 2503, Microsoft) and included information on study author, publication year, country, total sample size, participant characteristics (age and type of surgery), interventions, outcomes, and criteria used to define complete HDP. When data were missing or incomplete, the original study authors were contacted for clarification.
Assessment of study quality
The Cochrane Risk of Bias 2 (RoB 2, Cochrane) tool was used to assess the methodological quality of the included studies [11]. This tool evaluates five domains of potential bias: randomization process, allocation concealment, blinding of participants and assessors, completeness of outcome data, and selective reporting of results. Quality assessment was conducted independently by two reviewers (DFC and EBSN), with a third reviewer (JPAC) consulted to resolve any discrepancies.
Data synthesis
The meta-analysis was conducted using the meta package in R (version 4.4.2, R Foundation) with a random-effects model [12]. Binary outcomes were analyzed using risk ratios (RRs), and continuous outcomes were analyzed using mean differences (MDs), both with 95% confidence intervals (CIs). Statistical significance was defined as a P value < 0.05. Heterogeneity was assessed using the χ2 (Q) test and the I2 statistic, with substantial heterogeneity considered when I2 > 50%. Publication bias was evaluated using funnel plots and Egger’s test for asymmetry, with a P value < 0.05 indicating significant bias.
Because pain assessment scales are highly correlated, pain scores reported on the VAS were converted to equivalent NRS values when necessary. For studies reporting data as medians and interquartile ranges, conversions were performed using the methods described by Wan et al. [13] and Luo et al. [14]. Postoperative opioid consumption was standardized to milligrams of oral morphine equivalents (MME) according to the conversion table provided by the National Institutes of Health [15].
A meta-regression was performed to evaluate the effect of anesthetic volume used in STB as a covariate on study outcomes. A mixed-effects model was applied, with τ2 estimated using the DerSimonian and Laird method. Residual heterogeneity, the proportion of variance explained, and tests for residual heterogeneity and moderators were calculated to assess model fit and the moderating influence of anesthetic volume. Detailed information available in Supplementary Data 2. Sensitivity analyses were conducted using a leave-one-out approach and by restricting the analysis to studies with low RoB. To address heterogeneity related to local anesthetic volume, subgroup analyses were performed based on the total volume used for STB, categorized as ≤ 10 ml and > 10 ml.
To determine whether the cumulative evidence had sufficient statistical power and to prevent premature conclusions, a Trial sequential analysis (TSA, version 9.5.10, Copenhagen Trial Unit) was performed [16]. The predefined intervention effects were established as follows: a 50% reduction in the incidence of complete or partial HDP, a five-fold increase in the absence of this event, a 0.5-point reduction in 24-h pain scores, a 5 mg difference in opioid consumption, and a 1-h reduction in motor block duration. These parameters were based on findings from previous studies with a low RoB [17,18]. Two-tailed tests were applied with a type I error rate of 5% and a power of 90% (type II error rate of 10%).
To compare the two techniques, conventional and sequential monitoring limits were constructed. Heterogeneity adjustment was based on variance using the DerSimonian and Laird random-effects model. To assess the strength of evidence, a cumulative Z-curve was plotted. Additionally, the required information size was calculated, representing the number of participants needed in the meta-analysis to confirm or reject a specified intervention effect.
Confidence in evidence
The certainty of evidence was evaluated using the GRADE approach, which considered the consistency, precision, applicability, and methodological rigor of the included studies [19].
RESULTS
Study selection
The database search identified 640 records, of which 376 were excluded during the initial screening due to duplication (Fig. 1). After screening titles and abstracts, 251 records were excluded for not meeting the eligibility criteria. Thirteen articles were retrieved for full-text review. Among the five studies excluded at this stage, two evaluated STB at different anesthetic doses, two compared it with other regional techniques, and one used an alternative method to assess HDP [20].
Study characteristics
A total of eight RCTs were included in this systematic review [17,18,21-26], published between 2019 and 2025. Regarding geographic distribution, two studies were conducted in South Korea, three in China, and the remaining three in India, Egypt, and the USA. Together, these studies included 597 patients—297 in the STB group and 300 in the ISB group. Most participants underwent unilateral shoulder arthroscopy for rotator cuff repair, with mean ages ranging from 41.1 to 63.0 years. The general characteristics of the included studies are summarized in Table 1.
RoB and publication bias
Six studies were classified as having a low RoB [17,18,21-24], while two studies raised some concerns related to randomization procedures, outcome measurement, or lack of protocol registration [25,26]. Assessment using the funnel plot and Egger’s test revealed no evidence of publication bias (Fig. 2; Supplementary Fig. 1).
Incidence of HDP
Eight studies (597 patients) reported the incidence of complete HDP [17,18,21-26]. The pooled analysis demonstrated an approximately 90% reduction in the incidence of HDP (RR, 0.10; 95% CI, 0.07-0.17; I2 = 0%; P < 0.001, Fig. 3A). A significant increase in the absence of HDP was also observed among patients who received STB (RR, 6.97; 95% CI, 3.13-15.53; I2 = 70.3%, P < 0.001; Fig. 3B).
Quality of analgesia
Five studies (408 patients) reported total postoperative opioid consumption (P = 0.2621; Fig. 4A) [17,18,22,24,26]. Four studies (338 patients; STB, 167; ISB, 171) reported pain scores at rest at 24 h [17,18,22,26]. Overall, no significant difference was detected (P = 0.6610; Fig. 4B). The duration of motor block was comparable between groups (466 patients; P = 0.6243; Fig. 4C).
Leave-one-out analysis
For the 24-h pain outcome, heterogeneity became negligible after exclusion of the study by Kim et al. [18]. A notable finding was identified for opioid consumption: a significant reduction favoring STB (MD, -5.49 mg MME; 95% CI, -7.21 to -3.77; I2 = 0%, P < 0.001; Supplementary Fig. 2C) was observed when the study by Zhang et al. [22], which employed a subparaneural technique for STB using a low anesthetic volume, was excluded from the analysis. No significant differences were found for the remaining outcomes.
Meta-regression
A meta-regression analysis was performed to examine the association between anesthetic volume and the risk of HDP. The analysis showed no significant relationship between the volume administered in STB and the incidence of the primary outcome (P = 0.548; Supplementary Fig. 3).
Trial sequential analysis
The TSA for the outcomes of complete HDP, absence of HDP, and Horner’s syndrome indicated that the required information size was reached and the monitoring boundaries were crossed, confirming a high level of certainty regarding the reduction in the incidence of these events (Fig. 3). For other outcomes, the required information size and monitoring boundaries were not attained (Supplementary Fig. 4).
Sensitivity analysis
The sensitivity analysis restricted to studies with a low RoB confirmed the robustness of the reduction in HDP incidence favoring STB, with no change in effect size for the other outcomes (Supplementary Table 1). Subgroup analysis based on anesthetic volume revealed no significant differences between the >10 ml and ≤10 ml subgroups (Supplementary Table 2).
Quality of evidence
The GRADE assessment rated the certainty of evidence as high for the incidence of complete HDP and Horner’s syndrome, moderate for the absence of HDP and patient satisfaction with analgesia, and low for oral morphine consumption, 24-h postoperative pain at rest, and duration of motor block (Table 2).
DISCUSSION
In this study, STB was compared with ISB to evaluate an adverse effect of clinical importance. Although effective, ISB delivers local anesthetic in close proximity to the phrenic nerve and frequently causes HDP with respiratory impairment, which can be critical in patients with pulmonary disease [27,28]. Respiratory complications are relatively common after ISB, particularly among patients with preexisting pulmonary conditions [29], and neurological symptoms such as numbness, weakness, or paresthesia may also occur in the days following the procedure [30].
Sripriya et al. [31] demonstrated that, in STB, the distance between the phrenic nerve and the brachial plexus is approximately ten times greater than in ISB, thereby reducing the risk of diaphragmatic paresis. Because the C5 and C6 nerve roots converge at this level, STB is theoretically expected to provide effective shoulder analgesia. In a subsequent study refining the STB technique, Shin et al. [32] evaluated a modified ISB approach at the C5 level and reported comparable analgesia to the conventional technique, with a lower incidence of postoperative numbness and weakness in the upper limb. Despite these advantages, oxygen saturation decreased in both groups, indicating that even the modified ISB at the C5 level does not entirely prevent phrenic nerve paresis [6,32].
The TSA confirmed adequate statistical power for the primary outcome and for Horner’s syndrome, strongly supporting the observed benefit. Conversely, no significant differences were identified between techniques for opioid consumption, 24-h pain scores, satisfaction with analgesia, or motor block duration, and the TSA did not confirm statistical reliability for these secondary outcomes.
The certainty of evidence varied across outcomes due to methodological and statistical limitations. Evidence for HDP and Horner’s syndrome was rated as high, supported by consistent and precise estimates. In contrast, the certainty of evidence for the absence of HDP and patient satisfaction was rated as moderate because heterogeneity and wide CIs introduced inconsistency and imprecision. Evidence for postoperative pain at rest and motor block duration was downgraded to low certainty owing to imprecision. A similar downgrade applied to oral morphine milligram equivalent consumption on the first postoperative day, driven by substantial heterogeneity.
Our results are consistent with the findings of a prior review of four RCTs by Amaral et al. [33], which identified STB as an effective alternative to ISB. By incorporating more recent trials, a larger pooled sample, and applying both TSA and GRADE methodologies, our meta-analysis provides stronger and more reliable evidence. Unlike another recent review that included only three studies and failed to achieve TSA significance [34], our findings offer more robust support for reduced HDP and for analgesia-related outcomes.
Phrenic nerve paralysis following ISB—particularly in patients with preexisting respiratory impairment—can have a significant clinical impact and result in unforeseen complications such as prolonged hospital stay and unplanned intensive care unit admission [35,36]. Patients receiving continuous blocks exhibit an even higher risk of respiratory failure. Therefore, careful selection of regional anesthesia techniques is essential, especially in patients with risk factors such as high ASA physical status scores, pulmonary disease, obesity, or other conditions associated with reduced tolerance to HDP [36-38]. It is important to emphasize, however, that although STB substantially reduces the risk of HDP, it does not completely eliminate it.
Conversely, trials with larger sample sizes, lower anesthetic volumes (≤ 10 ml), and greater methodological standardization demonstrated more consistent findings [17,18,21-24]. Similar to ISB, smaller volumes in STB may further decrease the incidence of respiratory complications. Wang et al. [39] reported lower rates of HDP with 10 ml vs. 15 ml of ropivacaine, without compromising analgesic efficacy or block quality in STB. Two included studies employed smaller volumes using the subparaneural technique for STB compared with ISB, showing an improved safety profile while maintaining comparable efficacy [22,24].
An important limitation of this review was the lack of comparisons with other emerging strategies beyond STB, such as the effect of volume reduction in ISB. Previous studies have shown that low-volume ISB techniques (e.g., 5 ml) can substantially reduce the incidence of HDP while maintaining effective analgesia [40,41]. Another alternative is the extrafascial ISB technique using either standard (20 ml) or reduced volumes (10 ml), which decreases the incidence of HDP but is associated with shorter block duration and increased postoperative opioid consumption [42]. Despite these modifications, HDP rates often remain high in ISB, and the risk of block failure increases [40-43].
Another limitation is the limited availability of direct comparative studies between STB and other regional anesthesia techniques. Preliminary evidence suggests that the costoclavicular block results in a lower incidence of HDP than STB but provides inferior analgesia and a higher need for rescue blocks [44]. Continuous suprascapular block has also been proposed as an alternative, offering analgesia comparable to ISB with a lower risk of HDP [45,46]. However, in a clinical trial comparing continuous STB with continuous suprascapular block, STB provided superior postoperative analgesia despite a higher incidence of HDP [47].
Future studies should determine the optimal minimum volume of local anesthetic for STB that preserves phrenic nerve function without compromising analgesic efficacy. Direct comparative trials with other diaphragm-sparing techniques are also essential to establish the best risk-benefit profile. Furthermore, additional research is warranted to clarify the effectiveness and safety of the subparaneural approach.
This systematic review and meta-analysis demonstrate that, based on evidence ranging from high to low certainty across outcomes, STB is a safer and equally effective alternative to ISB for shoulder surgery. It provides comparable analgesia while causing fewer neurological adverse effects, including phrenic nerve dysfunction and oculosympathetic paresis. Overall, these findings support STB as a regional anesthesia technique that offers effective pain control with a superior safety profile in this clinical context.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
