Differential lung ventilation assessed by electrical impedance tomography in ultrasound-guided anterior suprascapular nerve block vs. interscalene brachial plexus block: A patient and assessor-blind, randomised controlled trial : European Journal of Anaesthesiology | EJA

Secondary Logo

Journal Logo


Differential lung ventilation assessed by electrical impedance tomography in ultrasound-guided anterior suprascapular nerve block vs. interscalene brachial plexus block

A patient and assessor-blind, randomised controlled trial

Petroff, David; Wiegel, Martin; Pech, Virginia; Salz, Peter; Mrongowius, Julia; Reske, Andreas W.

Author Information
European Journal of Anaesthesiology 37(12):p 1105-1114, December 2020. | DOI: 10.1097/EJA.0000000000001367
  • Free



Ultrasound-guided interscalene brachial plexus block (ISB) is widely used to control pain after shoulder surgery or injury.1,2 Its efficacy notwithstanding, there are drawbacks associated with the technique, which include phrenic nerve block3–5 and motor blocks of the arm or hand.6,7 Blocking the phrenic nerve can become a serious complication if ISB is performed on a patient with compromised lung function. Thus, there are reasons for exploring other options. An alternative to an ISB or to foregoing a regional anaesthetic technique altogether is the ultrasound-guided anterior approach to perform suprascapular nerve block (SSNB),8–12 which may help avoiding problems related to phrenic nerve block and reduce the use of opioids. Although several randomised controlled trials (RCTs) and a recent meta-analysis have examined the analgesic effects of SSNB compared with ISB, few have presented pulmonary outcomes, and the distinction between anterior and posterior approaches to SSNB is not always highlighted.7,12,13

Electrical impedance tomography (EIT) is a noninvasive method for assessing pulmonary ventilation. Regional changes in thoracic impedance are measured in real time using an electrode belt placed around the chest. These impedance changes correlate with changes of intrathoracic gas volume and ventilation, and the technique has been used in the context of phrenic nerve blocks induced by ISB.4 We designed an RCT to test the hypothesis that the ultrasound-guided anterior approach to SSNB is associated with better preservation of lung ventilation than a standard ISB as measured by asymmetries between the operated and nonoperated side with EIT. Secondary endpoints were diaphragmatic motion measured by M-mode5,14 sonography, pain, hand strength, opioid consumption and patient satisfaction.


This study was a patient-blinded and assessor-blinded, 1 : 1 randomised parallel group trial. After approval by the responsible ethics committee (Ethikkommission der Sächsischen Landesärztekammer, EK-BR-23/16-1, chairman Professor Dr Bernd Terhaag), the trial was registered in the WHO listed German Clinical Trials Register (DRKS00011787) on 3 May 2017. Originally, we had planned to see patients 4 days after surgery during a regularly scheduled visit but this proved to be logistically challenging after in-house procedures no longer foresaw a need for this visit and so it was cancelled within the trial as well. No other changes were made to the trial protocol.

Outpatients scheduled for arthroscopic shoulder surgery under general anaesthesia with an American Society of Anesthesiologists (ASA)15 physical status classification score between 1 and 3 were included. Patients had to be over 18 years of age and were excluded if they had known diaphragmatic paresis, were allergic to the local anaesthetic, had coagulation disorders precluding the nerve block, morbid obesity (BMI >35 kg m−2), anatomical anomalies or previous operations on the lung or chest, neuropathies or were pregnant. As the EIT devices work with small applied voltages, patients with an implantable cardioverter defibrillator or pacemaker were also excluded. All patients provided written informed consent.

The study was conducted in an outpatient surgical clinic in Germany between June 2017 and January 2018.

All nerve blocks were performed immediately prior to induction of general anaesthesia with the patient in the supine position. Patients received sufentanil 5 to 10 μg intravenously before the nerve block. Only ultrasound was used for nerve localisation (i.e. peripheral nerve stimulation was not utilised). Flex Focus 500 ultrasound machines equipped with high-frequency linear array probes (model 8870, 6 to 18 MHz, BK Medical, Quickborn, Germany) were used. The colour Doppler was routinely used to help exclude vessels within the area of the nerve block. The SonoPlex Facet tip 50 mm needle (Pajunk, Geisingen, Germany) was used for all blocks.

Interscalene brachial plexus block (ISB)

The ultrasound probe was placed on the neck in a transverse plane and moved in a caudal to cranial direction to identify the anterior and middle scalene muscles as well as the superior trunk of the brachial plexus. The superior trunk of the brachial plexus was identified by tracing the anterior rami of C5 and C6 back from the supraclavicular position cranially, using the differences in anatomy of the transverse processes of the cervical vertebrae C7, C6 and C5. The needle was advanced and visualised in an out-of-plane technique and ropivacaine 10 ml of 1% wt/vol solution injected between the lateral aspect of the superior trunk of the brachial plexus and the middle scalene muscle slightly caudal to the transverse process of C6. In case of anatomical variations, such as the absence of a compact superior trunk,16 the ropivacaine was injected around the anterior rami of C5 and C6 at the position where they could be identified with certainty.

Suprascapular nerve block (SSNB)

The ultrasound-guided anterior subomohyoid suprascapular nerve block used here was described in detail by Siegenthaler and others.7,10,17 The suprascapular nerve originates from the superior trunk of the brachial plexus and runs in a dorsocaudal direction, crossing beneath the omohyoid muscle and finally reaching the suprascapular fossa. It can be visualised by sonography in the supraclavicular region beneath the omohyoid muscle where needle access is usually easy. Identification of the suprascapular nerve beneath the omohyoid muscle can sometimes be facilitated by tracing the nerve from its origin to the puncture site. Advancement of the needle was guided by sonography using an out-of-plane technique. Ten ml of ropivacaine 1% wt/vol were injected. If identification of the suprascapular nerve was insufficient, then the local anaesthetic was injected below the inferior belly of the omohyoid muscle, lateral to the ‘bulk’ of the brachial plexus visualised in the supraclavicular position.

General management and anaesthesia

All operations were performed in the beach chair position. General anaesthesia was induced and maintained using sufentanil (initial bolus 0.2 μg kg−1 body weight, additional boluses of 5 μg if required), propofol (induction dose 1.5 to 2 mg kg−1 body weight, rocuronium bromide (0.6 mg kg−1 body weight) and desflurane (end-expiratory concentration 3.5 to 6.0% in air). For patients receiving total intravenous anaesthesia (n = 2, decided when planning the surgery because of a history of postoperative nausea and vomiting), the infusion rate was 5 to 8 mg kg−1 h−1 body weight). Intra-operative mechanical ventilation was pressure-controlled (Dräger Fabius Tiro) and parameters were as follows: positive end-expiratory pressure (PEEP) 5 cmH2O, tidal volume 6 to 8 ml kg−1 of predicted body weight18 and with FiO2 between 0.4 and 0.5. Every patient received ibuprofen 600 mg up to four times or the synthetic opioid tilidine 100 mg up to twice in 24 h, as needed.

Electrical impedance tomography and ultrasound assessment

EIT is a noninvasive method for monitoring lung ventilation.19 In its functional mode, it visualises the impedance changes in the thorax by inducing an alternating current via electrodes placed on the surface of the thorax. The remaining electrodes measure the resulting voltages, which are used to compute impedance distribution changes.19

We performed EIT measurement using the Swisstom BB 2 device (Landquart, Switzerland). After determination of the chest circumference, a belt with 32 electrodes was applied at the level of the fourth intercostal space and the patient was placed in a supine position and was asked to breathe normally. Data were recorded for 2 min and transferred to a computer for later analysis. The software provides the proportion of impedance change in the left vs. right lung, that is, a surrogate for the percentage of total ventilation in each lung. The percentage on the operated side was used as the primary outcome.

A curved transducer (BK Ultrasound, model 8820e, 2 to 6 MHz) was used for visualisation of the diaphragm. The patient was in a supine position and asked to breathe normally. When examining the right diaphragm, the transducer was positioned below the costal arch between the anterior axillary and midclavicular lines. When examining the left diaphragm, the spleen served as an acoustic window and the transducer was placed in a more posterior position. The direction of diaphragm motion and its amplitude were determined in the M-mode and the maximal motion of four consecutive breaths was taken.14,20 Diaphragm motion was analysed as a secondary outcome.

Further outcomes were grip strength and numbness, pain, use of opioids and pain medication, and patient satisfaction. Grip strength was measured using a dynamometer (TRAILITE, LiteXpress, Coesfeld, Germany), with the patient in a seated position and wearing a sling after surgery. Data on numbness of the lower arm and/or hand as well as pain medication were collected by study personnel. To assess pain, a numerical rating scale (NRS) was chosen ranging from 0 (no pain) to 10 (worst imaginable pain). Regarding pain medication after discharge, ibuprofen 600 mg and tilidine 100 mg were combined based on analogous dose equivalence findings, where tilidine was equated with tramadol.21 Patient satisfaction and whether patients would choose to have the same procedure again were posed as two questions using a 5-point Likert scale on a patient questionnaire. Adverse events were documented using a list of known harms. Dyspnoea was assessed using the American Thoracic Society Scale (grade 0 to 4).22

Diaphragmatic motion, EIT and hand strength were assessed at baseline and 4 and 24 h after the operation. Pain was measured at baseline and at 10 and 30 min as well as 4 and 24 h after the operation. Patients were discharged as per normal practice after anaesthesia had worn off and was not expected to affect ventilation, and returned to the clinic for the final assessment at 24 h.

Sample size calculations were based on case series demonstrating that a very high proportion of patients with ISB suffer from hemidiaphragmatic paresis (14 of 14 patients).3,4 Although larger data sets were unavailable at the time of the trial design,12 anatomical considerations suggest that SSNB patients should have a paresis rate approaching zero. With the conservative assumption that the paresis rate would be 80% in the ISB group and 30% in the SNNB group, a power of 90% is reached with 23 patients per arm using Fisher's exact test (PASS 11). The primary analysis using EIT as a continuous parameter and analysis of covariance (ANCOVA) was expected to have even higher power and the sample size has a power of 90% to detect an effect size of one. This conservative approach to planning the sample size was chosen as published data were unavailable and as sufficient power for the secondary endpoint of paresis was desired. Allowing for two dropouts per arm, we, therefore, planned to randomise a total of 50 patients. One could argue that with the volume of 10 ml of ropivacaine used here, a 60% paresis rate as seen by Lee et al.23 in the ISB group would have been a more appropriate assumption. With the 48 patients in this study, 16% paresis in SSNB would be distinguished from 60% in ISB with 90% power.

Patients were allocated randomly in equal proportions to single-shot ISB or SSNB. Group allocation used block randomisation of block sizes of six and was performed by the study personnel using a web-based randomiser developed and hosted by the University of Graz, thus guaranteeing concealment.24

A team consisting of six experienced anaesthesiologists was responsible for the blocks in all patients. Patients and all study personnel except for the anaesthesiologist in charge were blinded. Patients were unaware of where either puncture site would be, which varies according to nerve block anatomy. A large adhesive dressing extending from the neck to the shoulder covered both potential needle puncture sites, helping to ensure adequate blinding to patients, surgeons and anaesthesiologists.

Statistical analysis

All analyses were performed with the statistical software package R (version 3.4.2, http://www.R-project.org). Mixed models with patient as a random term were used for time series data and post hoc linear multiple comparisons were based on the methods in the multcomp package, where P values were adjusted using Westfall's method.25 Due to the central limit theorem, such methods are appropriate despite the independent variables having bimodal distributions, leading to large standard deviations. In comparisons between groups at various time-points, the baseline comparison was not performed, as the null hypothesis is known to hold because of randomisation but the baseline value was included in the mixed model. As covariates, all analyses of pain data included sex and whether or not a rotator cuff repair26 was performed. Area under the curve for pain data used linear interpolation between time points. Comparison of count data made use of the χ2-test or a Fisher exact test if expected counts were below five and parametric data were compared with a t-test using Welch's approximation. As sensitivity analyses, different correlation structures were considered in the mixed models (unstructured, autoregressive of order 1 and uncorrelated) and diaphragmatic motion was analysed as a nonparametric continuous variable. P values less than 0.05 were considered statistically significant.


A total of 55 patients were randomised instead of 50 to account for a slightly higher dropout rate than anticipated because of technical problems with the EIT measurements. Final analysis included 48 patients (Fig. 1) after seven were excluded because of technical problems with EIT (n=6) or ultrasound (n=1). The technical problems with EIT were because of problems placing the belt and the electrodes slipping. The patients’ baseline characteristics are shown in Table 1. Intra-operative ventilation was conducted with a PEEP of 5 (n = 40) or 6 cmH2O (n = 8), and the mean ± SD tidal volume was 6.3 ± 0.9 ml kg−1 of body weight. Patients in the SSNB groups spent 114 ± 15 min in the anaesthetic recovery room compared with 111 ± 15 min for the ISB group (95% CI for the difference −7 to 11 min, P = 0.61). The maximal and minimal values for oxygen saturation in the recovery room for SSNB vs. ISB were 97.2 ± 1.2 vs. 97.9 ± 1.3% (95% CI for the difference −0.02 to 1.4%, P = 0.058) and 96.1 ± 1.2 vs. 96.4 ± 1.2% (95% CI for the difference −0.3 to 1.0%, P = 0.34), respectively.

Fig. 1:
Study flow chart
Table 1 - Baseline demographic and clinical characteristics
SSNB (n=24) ISB (n=24)
Female 10 (42) 12 (50)
Age (years) 50 ± 12 52 ± 11
Height (cm) 173 ± 11 174 ± 10
Weight (kg) 78 ± 10 79 ± 12
BMI (kg m−2) 25.8 ± 3.0 26.0 ± 2.9
 1 10 (42) 9 (38)
 2 13 (54) 15 (62)
 3 1 (4) 0 (0)
Planned surgery
 Stabilisation 5 (21) 1 (4)
 Decompression/acromioplasty 11 (46) 17 (71)
 Rotator cuff repair 8 (33) 4 (17)
 Diagnostic/debridement 0 (0) 2 (8)
Entries are mean ± standard deviation or number (%). ASA, American Society of Anesthesiologists physical status; ISB, interscalene block; SSNB, suprascapular nerve block.

The target site was visualised successfully in 22 SSNB and all 24 ISB patients. The block was considered unsuccessful (NRS score >6 at any point in time after the operation) in one patient in the ISB group.

Assessment by electrical impedance tomography and M-mode sonography

At baseline, the lung on the operated side received half of the ventilation of the total lung, but with large variance. The median [IQR] values were 50 [42 to 56]% at baseline (SSNB 50 [44 to 55]%, ISB 50 [37 to 57]%). Postoperatively, this value fell to 40 [3 to 50]% (SSNB) vs. 3 [1 to 13]% (ISB) for an adjusted difference of 23 (95 CI, 12 to 34)%, (P < 0.001) and sensitivity analyses show that this result is essentially unaffected by changes to the model's correlation structure. The next day, values had increased considerably, but had not returned to their original values (Fig. 2a). In particular, one patient in the ISB group still had a value of 1% 1 day later. A post hoc power analysis indicates that with the observed differences, the power of the study was greater than 99%.

Fig. 2:
Estimates of percentage of lung ventilation and diaphragmatic movement

Median [IQR] diaphragm motion measured by ultrasound was 3.25 [2.90 to 3.60] cm and 3.40 [3.18 to 3.65] cm at baseline in the SSNB and ISB groups, respectively. Diaphragmatic motion was reduced after surgery to 3.00 [2.45 to 3.32] cm (P = 0.021) in the SSNB arm and to 0.60 [0.30 to 1.25] cm (P < 0.001) in the ISB group for an adjusted difference of 1.90 (95% CI 1.37 to 2.44) cm, (P < 0.001) in the ISB group (Fig. 2b). On the day after surgery, the ISB patient with only 1% lung ventilation according to EIT had a high value for diaphragm motion of 4.6 cm. In a sensitivity analysis, diaphragm motion was significantly different between groups when assessed nonparametrically (P < 0.001).

Pain scores, opioid consumption and pain medication

Mean pain scores were close to 3 before surgery and fell to values below 1 for both groups. The ISB pain score was similar with a difference in NRS points of −0.25 (95% CI −0.83 to 0.33), P = 0.13, Fig. 3a).

Fig. 3:
Pain estimates and hand strength

Opioids in the PACU were used by four patients in the SSNB and one patient in the ISB group, each of whom took oxycodone 10 mg orally (P = 0.35). After discharge, the mean medication equivalent was 2427 mg of ibuprofen over the next 24 h compared with 2406 mg in the ISB group (95% CI −751 to 794) mg for the difference between groups, (P = 0.96). To test robustness, all mathematically possible dose equivalents between tilidine and ibuprofen were considered and the lowest possible P value for a difference between the groups was 0.24.

Hand strength and numbness

Before surgery, hand strength was about 35 kg in each group, but fell considerably in both groups 2 h after surgery. Hand strength in the SSNB group measured at 2 h postsurgery was 11.2 (95% CI, 3.6 to 18.9) kg (P = 0.0024) greater than in the ISB group. After 24 h, hand strength was comparable between the groups, but still lower than at baseline (Fig. 3b). In the ISB group, 11 patients reported numbness in the hand, whereas none did so in the SSNB group (P < 0.001).

Patient perception

In the SSNB group, 7 patients were satisfied and 17 completely satisfied, compared with 8 and 16 in the ISB group (P = 0.76). Similarly, seven patients would probably choose the SSNB procedure again and 17 would certainly choose it compared with 5 and 19 from the ISB group (P = 0.50).


One patient required 0.4 mg of naloxone. During the postoperative period, no patient required oxygen. Horner's syndrome was noted in 5 ISB patients but in none of the SSNB patients (P = 0.0496). No other meaningful differences were noted between the groups and overall the number of adverse events was low (Table 2).

Table 2 - List of harms according to treatment received
SSNB (n=24) ISB (n=24) P
Block failure (NRS >6) 0 1 (4) 1.00
Hoarseness 0 2 (8) 0.49
Horner's syndrome 0 5 (21) 0.0496
Dyspnoea (grade >0) 0 0
Pneumothorax 0 0
LA toxicity 0 0
PONV 0 0
Bleeding 0 0
Haematoma 0 0
Dyspnoea was assessed using the American Thoracic Society Scale (grade 0 to 4). Values are number (%). ISB, interscalene block; LA, local anaesthetic; NRS, numerical rating scale for pain; PONV, postoperative nausea and vomiting; SSNB, suprascapular nerve block.


Using EIT measurements, this study was able to demonstrate that ISB is associated with significantly more imbalanced ventilation than the anterior sub-omohyoid SSNB. This observation was corroborated by a high proportion of hemidiaphragmatic inactivity in the ISB group observed with ultrasound. Pain within the first 6 h after surgery was well controlled by both methods with a very slight, though not significant, benefit for ISB. However, ISB led to numbness in the hand in many of the patients and to reduced hand strength whereas SSNB did not lead to any numbness, and hand strength was considerably greater. Overall, patients were highly satisfied with both procedures.

These findings augment and corroborate published literature and the anatomical explanation for the efficacy of SSNB. Two previous RCTs have used comparable SSNB and ISB techniques to those used here.7,13 In a previous RCT, we ourselves used 10 ml of ropivacaine 1% wt/vol (SSNB) and 20 ml of ropivacaine 0.75% wt/vol (ISB) and found noninferior analgesia in SSNB with significantly less motor block of the hand. Auyong et al.13 used 15 ml of ropivacaine 0.5% wt/vol for both groups and found that SSNB was noninferior to ISB regarding pain and was associated with significantly better preservation of vital capacity and diaphragmatic excursion. A meta-analysis combining 16 RCTs examined pain scores and oral morphine consumption between SSNB (primarily posterior approach) and ISB and concluded that ‘there are no clinically meaningful analgesic differences’, whereas SSNB ‘has fewer side effects’.12

RCTs have also found significant differences in hemidiaphragmatic paresis27 and tended to find somewhat better analgesia in ISB,27–31 although the varieties of anaesthetic doses, concentrations and techniques make generalisations difficult. An RCT exploring 20 vs. 5 ml of ropivacaine 0.5% wt/vol in ISB found that hemidiaphragmatic paresis could be reduced from 100 to 45% with the lower dose without significant differences in pain control, although pain scores tended to be high in both groups and the trial was underpowered to show equivalence in pain.32 Some groups have been able to reduce hemidiaphragmatic paresis to about 20%,33 although even the largest of these had a sample size that could only exclude rates higher than 46%.34 A recent study with 5 cadavers and 10 puncture sites looked at the spread of 5 ml of dye and found that this volume was sufficient to stain the suprascapular nerve and may spread to the phrenic nerve.35 As the scalene muscles are innervated by the ventral rami of C4 to C8, it is also conceivable that impaired scalene activity in both SSNB and ISB can affect ventilation.

We used two methods to evaluate respiratory outcomes in our trial. A recent meta-analysis notes that a number of studies report on respiratory complications but only on one that assessed respiratory function (peak expiratory flow rate).12 EIT can add further information as relative lung ventilation is assessed, whereas ultrasound-based diaphragm motion is a more established tool. Both methods were generally comparable, although there were three instances where EIT demonstrated poor ventilation despite good or at least adequate diaphragmatic motion seen on ultrasound. This suggests that improvements in the technology and software are warranted before EIT is used clinically for evaluating hemidiaphragmatic activity. Preserved diaphragmatic contraction seen in M-mode sonography together with reduced ventilation demonstrated by EIT might, theoretically, be explained by pneumothorax. However, none of the patients with disparate values (three in total) reported any clinical symptoms or signs suggestive of pneumothorax, either in the PACU or during the subsequent postoperative days. In the absence of clinical indications, chest X-ray or other diagnostics (pleural sonography) were obviously not performed. As spontaneous breathing insufficiently excludes residual neuromuscular blockade [train-of-four (TOF) response <90%]36 and as TOF was not quantified routinely in our study, we cannot exclude an effect of residual neuromuscular blockade on our measurements early after the operation. However, neuromuscular monitoring was used routinely and patients breathed spontaneously before extubation. Recording additional respiratory parameters, such as oxygen saturation and respiratory rate would have provided useful information on clinical relevance. However, the aim of this study was to explore reductions in diaphragmatic activity and consequent ventilation impairment, which would gain clinical relevance in patients with (unknown) preexisting respiratory disease.

Our data clearly show that hand strength is much reduced in ISB within the first few hours of surgery. Figure 3b suggests that some loss in strength is attributable to the surgery itself so that grip strength is about 10 kg less in both groups 24 h after surgery compared with baseline. The large difference in numbness between the two groups could arise from the ISB blocking larger parts of the plexus, although observations in cadavers suggest relatively large spread even for small volumes of local anaesthetic in SSNB.35 Block of the hand shortly after surgery can be a hindrance, particularly in the outpatient setting, and may affect patient satisfaction. Although such aspects are less important than pain management and hemidiaphragmatic paresis, they could still tip the scales when choosing an anaesthetic procedure.

This study has some limitations. Our trial would have benefited from longer observation times regarding pain, and from sensory assessment regarding blocks, which could help assess proximal spread.37 We also refrained from using nerve stimulation in addition to ultrasound guidance for confirming correct needle placement, considering the routine that all anaesthetists have with both techniques. Moreover, standardised assessment of inspiration and ventilation would have required sniffing or spirometry and the study would have benefited considerably from contralateral evaluation of diaphragm excursion. It should also be noted that reduced bilateral diaphragm activity may result from general anaesthesia. Volume could perhaps be reduced for ISB and SSNB alike; however, reduction has its limits, especially if procedures are to remain viable even when performed by nonspecialists.

Due to a lack of data, the sample size calculation was based on a binary secondary outcome, although a continuous one was chosen as the primary endpoint where a strong association was expected as hemidiaphragmatic inactivity would induce marked imbalances in ventilation between the operated and nonoperated sides. However, this was expected to be conservative and lead to an underestimate of the true power. However, the proportion of hemidiaphragmatic paresis assumed for the ISB group (80%) relied on data where higher volumes of ropivacaine were used than in our trial. This could have resulted in an underpowered trial but the conservative approach taken suggests that this was not the case and the post hoc power analysis showed that power exceeded 99%. On the other hand, adverse event rates in such a small trial are certainly underpowered but can contribute to the body of knowledge assimilated in meta-analyses.

The current study helps close a gap noted recently in data on the anterior suprascapular nerve block and other diaphragm-sparing techniques.38 In a three-arm trial conducted at the same time as ours, Auyong et al.13 demonstrated noninferior pain control in SSNB vs. ISB and better preservation of vital capacity.13 Our trial shows similar results regarding pain and looks at hemidiaphragmatic activity, quantified using standard ultrasound and noninvasive EIT methods.

The data we present are relevant when faced with choosing the optimal anaesthetic method for managing postoperative pain in outpatient shoulder surgery. The ultrasound-guided anterior approach to SSNB provides similar pain control and preserves ipsilateral lung ventilation and phrenic function better than a standard interscalene block.

Acknowledgements relating to this article

Assistance with the study: none.

Financial support and sponsorship: support was provided solely from institutional and/or departmental sources.

Conflicts of interest: none.

Presentation: none.


1. Fredrickson MJ, Krishnan S, Chen CY. Postoperative analgesia for shoulder surgery: a critical appraisal and review of current techniques. Anaesthesia 2010; 65:608–624.
2. Borgeat A, Ekatodramis G. Anaesthesia for shoulder surgery. Best Pract Res Clin Anaesthesiol 2002; 16:211–225.
3. Urmey WF, Talts KH, Sharrock NE, et al. One hundred percentage incidence of hemidiaphragmatic paresis associated with interscalene brachial plexus anesthesia as diagnosed by ultrasonography. Anesth Analg 1991; 72:498–503.
4. Wiegel M, Hammermüller S, Wrigge H. Electrical impedance tomography visualizes impaired ventilation due to hemidiaphragmatic paresis after interscalene brachial plexus block. Anesthesiology 2016; 125:807.
5. Borgeat A, Perschak H, Bird P, et al. Patient-controlled interscalene analgesia with ropivacaine 0.2% versus patient-controlled intravenous analgesia after major shoulder surgery: effects on diaphragmatic and respiratory function. Anesthesiology 2000; 92:102–108.
6. Fredrickson MJ, Smith KR, Wong AC. Importance of volume and concentration for ropivacaine interscalene block in preventing recovery room pain and minimizing motor block after shoulder surgery. Anesthesiology 2010; 112:1374–1381.
7. Wiegel M, Moriggl B, Schwarzkopf P, et al. Anterior suprascapular nerve block versus interscalene brachial plexus block for shoulder surgery in the outpatient setting: a randomized controlled patient- and assessor-blinded trial. Reg Anesth Pain Med 2017; 42:310–318.
8. Skillern PG. Suprascapular nerve syndrome as revealed by new (anterior) approach in induction of block. AMA Arch Neurol Psychiatry 1954; 71:185–188.
9. Chan CW, Peng PWH. Suprascapular nerve block: a narrative review. Reg Anesth Pain Med 2011; 36:358–373.
10. Siegenthaler A, Moriggl B, Mlekusch S, et al. Ultrasound-guided suprascapular nerve block, description of a novel supraclavicular approach. Reg Anesth Pain Med 2012; 37:325–328.
11. Rothe C, Steen-Hansen C, Lund J, et al. Ultrasound-guided block of the suprascapular nerve - a volunteer study of a new proximal approach. Acta Anaesthesiol Scand 2014; 58:1228–1232.
12. Hussain N, Goldar G, Ragina N, et al. Suprascapular and interscalene nerve block for shoulder surgery: a systematic review and meta-analysis. Anesthesiology 2017; 127:998–1013.
13. Auyong DB, Hanson NA, Joseph RS, et al. Comparison of anterior suprascapular, supraclavicular, and interscalene nerve block approaches for major outpatient arthroscopic shoulder surgery: a randomized, double-blind, noninferiority trial. Anesthesiology 2018; 129:47–57.
14. Lloyd T, Tang Y-M, Benson MD, et al. Diaphragmatic paralysis: the use of M mode ultrasound for diagnosis in adults. Spinal Cord 2006; 44:505–508.
15. ASA House of Delegates/Executive Committee. ASA Physical Status Classification System: American Society of Anesthesiologists 2014. Available at: https://www.asahq.org/standards-and-guidelines/asa-physical-status-classification-system (Accessed 21 September 2020)
16. Harry WG, Bennett JD, Guha SC. Scalene muscles and the brachial plexus: anatomical variations and their clinical significance. Clin Anat 1997; 10:250–252.
17. Stimpson JA, Chen P, Fox B, et al. A good SOS(nB) is also good for your OSS. Reg Anesth Pain Med 2017; 42:795–796.
18. Brower RG, Matthay MA, Morris A, et al. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308.
19. Frerichs I, Amato MBP, van Kaam AH, et al. TREND study group. Chest electrical impedance tomography examination, data analysis, terminology, clinical use and recommendations: consensus statement of the TRanslational EIT developmeNt stuDy group. Thorax 2017; 72:83–93.
20. Gerscovich EO, Cronan M, McGahan JP, et al. Ultrasonographic evaluation of diaphragmatic motion. J Ultrasound Med 2001; 20:597–604.
21. Bandolier. The Oxford League Table of Analgesic Efficacy; 2007. Available at: http://www.bandolier.org.uk/booth/painpag/Acutrev/Analgesics/lftab.html. (Accessed 22 September 2020)
22. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease. American Thoracic Society. Am J Resp Crit Care Med 1995; 152:S77–S121.
23. Lee J-H, Cho S-H, Kim S-H, et al. Ropivacaine for ultrasound-guided interscalene block: 5 mL provides similar analgesia but less phrenic nerve paralysis than 10 mL. Can J Anaesth 2011; 58:1001–1006.
24. University of Graz. Randomizer. Available at: www.randomizer.at. (Accessed 22 September 2020)
25. Westfall PH. Multiple testing of general contrasts using logical constraints and correlations. J Amer Statist Assoc 1997; 92:299.
26. Calvo E, Torres MD, Morcillo D, et al. Rotator cuff repair is more painful than other arthroscopic shoulder procedures. Arch Orthop Trauma Surg 2019; 139:669–674.
27. Aliste J, Bravo D, Finlayson RJ, et al. A randomized comparison between interscalene and combined infraclavicular and suprascapular blocks for arthroscopic shoulder surgery. Can J Anaesth 2018; 65:280–287.
28. Dhir S, Sondekoppam RV, Sharma R, et al. A comparison of combined suprascapular and axillary nerve blocks to interscalene nerve block for analgesia in arthroscopic shoulder surgery: an equivalence study. Reg Anesth Pain Med 2016; 41:564–571.
29. Kumara AB, Gogia AR, Bajaj JK, et al. Clinical evaluation of postoperative analgesia comparing suprascapular nerve block and interscalene brachial plexus block in patients undergoing shoulder arthroscopic surgery. J Clin Orthop Trauma 2016; 7:34–39.
30. Neuts A, Stessel B, Wouters PF, et al. Selective suprascapular and axillary nerve block versus interscalene plexus block for pain control after arthroscopic shoulder surgery: a noninferiority randomized parallel-controlled clinical trial. Reg Anesth Pain Med 2018; 43:738–744.
31. Singelyn FJ, Lhotel L, Fabre B. Pain relief after arthroscopic shoulder surgery: a comparison of intraarticular analgesia, suprascapular nerve block, and interscalene brachial plexus block. Anesth Analg 2004; 99:589–592.
32. Riazi S, Carmichael N, Awad I, et al. Effect of local anaesthetic volume (20 vs 5 ml) on the efficacy and respiratory consequences of ultrasound-guided interscalene brachial plexus block. Br J Anaesth 2008; 101:549–556.
33. Tran DQH, Elgueta MF, Aliste J, Finlayson RJ. Diaphragm-sparing nerve blocks for shoulder surgery. Reg Anesth Pain Med 2017; 42:32–38.
34. Palhais N, Brull R, Kern C, et al. Extrafascial injection for interscalene brachial plexus block reduces respiratory complications compared with a conventional intrafascial injection: a randomized, controlled, double-blind trial. Br J Anaesth 2016; 116:531–537.
35. Sehmbi H, Johnson M, Dhir S. Ultrasound-guided subomohyoid suprascapular nerve block and phrenic nerve involvement: a cadaveric dye study. Reg Anesth Pain Med 2019; 44:561–564.
36. Cammu G, Witte Jde, Veylder Jde, et al. Postoperative residual paralysis in outpatients versus inpatients. Anesth Analg 2006; 102:426–429.
37. Laumonerie P, Ferre F, Cances J, et al. Ultrasound-guided proximal suprascapular nerve block: a cadaveric study. Clin Anat 2018; 31:824–829.
38. Tran DQ, Layera S, Bravo D, et al. Diaphragm-sparing nerve blocks for shoulder surgery, revisited. Reg Anesth Pain Med 2020; 45:73–78.

David Petroff and Martin Wiegel contributed equally.

Copyright © 2020 European Society of Anaesthesiology and Intensive Care. Unauthorized reproduction of this article is prohibited.