Shoulder surgery is associated with postoperative pain that is often significant enough to interfere with initial recovery and rehabilitation. Although opioid analgesics are effective in relieving postoperative pain, they may negatively impact recovery by increasing the likelihood of nausea and vomiting, causing sedation, or affecting sleep. Thus, an opioid-sparing technique for analgesia after shoulder surgery is desirable. The adjunctive use of a suprascapular nerve block (SSNB) improved early postoperative outcome measures after general anesthesia for arthroscopic shoulder surgery (1). These beneficial effects of SSNB are most likely explained by blocking sensory innervation to the posterior shoulder joint and surrounding tissues (2), with consequent reduction in pain and opioid requirements. However, the benefits of SSNB-induced analgesia may be specifically related to the placement of the posterior arthroscopic port and the absence of another form of long-acting brachial plexus conduction anesthesia. The extent to which similar benefits may accrue in the setting of an interscalene brachial plexus block (ISB) used for nonarthroscopic anterior shoulder surgery has not been studied; in our practice, placement of an adjunctive SSNB was subjectively observed to improve analgesia in some patients. This randomized clinical trial was therefore designed to determine the role of SSNB in improving analgesia and other outcome measures after ambulatory nonarthroscopic shoulder surgery performed under a standardized ISB with supplemental general anesthesia.
Materials and Methods
After IRB approval and informed consent, 50 ASA class I–III outpatients scheduled for nonarthroscopic shoulder surgery were entered into this prospective, randomized, double-blinded, placebo-controlled trial. Patients were excluded if they were pregnant, <18 or >65 yr of age, had a history of opioid tolerance or allergy to medications used in the study, were currently taking opioids, or refused regional anesthesia. Acromioplasty, rotator cuff repair, or a combination of the 2 was performed in a similar fashion by 1 of 3 orthopedic surgeons (one of whom performed 80% of surgeries).
In an induction room, all patients received IV ketorolac 30 mg and were sedated to comfort with a maximum of 2 μg/kg of fentanyl, 0.03 mg/kg of midazolam, or both. No prophylactic antiemetics were given. Each patient received a standardized ISB (3) with 30 mL of mepivacaine 1.25% with epinephrine 2.5 μg/mL. A peripheral nerve stimulator was used with a 50-mm insulated stimulating needle (Braun Medical, Bethlehem, PA). Incremental injection of local anesthetic was administered through the fixed needle after either paresthesia to the arm or anterior shoulder or a similarly distributed motor response (4) at a current ≤0.5 mA. The patient then received either a SSNB with 10 mL of bupivacaine 0.25% with epinephrine 5 μg/mL (SSNB group) using the infiltration technique described by Moore (5) or a 5-mL subcutaneous saline sham injection (placebo group). All regional techniques were performed by or under the direct supervision of the primary investigator. Patients were taken to the operating room and monitored in accordance with standard ASA guidelines. Approximately 10–15 min after block placement, patients were asked to externally rotate their shoulder against resistance (supraspinatus and infraspinatus muscle function) to evaluate SSNB adequacy (6). Strength was graded on a 4-point scale as compared with the nonblocked arm (1 = no movement, 2 = significant weakness, 3 = mild weakness, and 4 = no weakness). Patients were unaware that the SSNB was being tested by this maneuver. After the induction of general anesthesia with propofol and placement of a laryngeal mask airway, propofol-nitrous oxide with supplemental sevoflurane was used to maintain blood pressure and heart rate within 20% of preoperative values. Total perioperative midazolam was limited to 5 mg; total fentanyl was limited to 3.5 μg/kg. The surgeon infiltrated the wound with 10 mL of subcutaneous bupivacaine 0.25% at the end of surgery. Patients were instructed not to take supplemental analgesics other than those prescribed by their surgeon (oxycodone 5 mg with acetaminophen 500 mg) and to take these only upon the onset of pain.
In the postanesthesia recovery unit (PACU), nurses who were blinded to patient group recorded the following time data points: admission to PACU Phase 1 and PACU Phase 2, first oral intake, first ambulation, and readiness for discharge (fulfillment of Aldrete criteria ≥9 (7)). Further data points included: total PACU fentanyl use, oral analgesic use, or both (oxycodone 5 mg with acetaminophen 500 mg) and the presence of nausea and vomiting, including the need to treat with ondansetron 1 mg IV. Static visual analog pain scores (VAS; 0 = no pain; 10 = worst imaginable pain) were recorded on admission to Phase 1 and every 30 min until discharge. A static and dynamic VAS were assessed just before discharge. Approximately 24 h after discharge, a blinded observer conducted a standardized telephone interview with patients to assess variables of early postoperative recovery. Patients had been educated at the time of enrollment regarding the nature of these questions. The following time intervals were calculated (time 0 = time admitted to PACU Phase 1): time to PACU Phase 2, time to readiness for discharge, and time to first significant pain (“When did you first notice pain in your shoulder?”).
The primary end-point of this study was time to first significant pain. Secondary outcome measures assessed at 24 h included: postoperative oral analgesic use, static and dynamic verbal pain scores (0 = no pain; 10 = worst imaginable pain), degree of nausea or vomiting, satisfaction with pain treatment, activity level, and quality of first night’s sleep. The latter four quality-of-life data points were assessed by a verbal 10-point scale (0 = worst quality; 10 = best quality). Because this was an observational study and there are no comparative data, no power analysis was performed. The study size of 50 patients corresponded to a similar study performed by Ritchie et al. (1). Age, time from PACU Phase 1 to first pain, and all other variables were compared between groups using the t-test. The time from PACU Phase 1 to first pain is slightly skewed, thus the statistical significance of the difference between treatment groups was also calculated using the Mann-Whitney test, which yielded the same P value. Ordinal scores for nausea and other pain and satisfaction measures were compared between groups using the Mann-Whitney test. Sex composition of the treatment groups was compared using χ2. Probability values <0.05 were considered significant.
Fifty patients (25 per group) were enrolled in and completed the study. Groups were comparable with regard to physical characteristics, distribution of surgical procedures, and preoperative-intraoperative medication use (Table 1). External rotation strength data confirmed SSNB. The SSNB group had a mean score of 1.5 (no movement), whereas the placebo group’s mean score was 2.4 (mild weakness) (P = 0.02). The 25th–75th percentile range of strength was 1–2 for the SSNB group versus 1–4 for the placebo group. Based on PACU admission VAS scores of 0 to 1, brachial plexus block seemed to be satisfactory in all patients.
The PACU Phase 1 and Phase 2 experiences such as time to first oral intake (mean, 70 min) and first ambulation (mean, 100 min), incidence of nausea and vomiting, and the use of ondansetron did not differ significantly between groups. Patients in both groups were discharged from PACU Phase 2 with mean static VAS of 0.4 (range, 0–3) and dynamic VAS of 0.7 (range, 0–5), which were not significantly different between groups.
Table 2 illustrates relevant time data. The SSNB group experienced approximately 3.6 h longer of analgesia (time to first significant pain) than the placebo group (594 ± 369 min versus 375 ± 273 min, respectively;P = 0.02). Of note is the wide variation of these times (Fig. 1). There were no other significant differences in outcome measures at 24-h follow-up (Table 3). There were no unplanned admissions. Two patients from each group made postoperative phone calls regarding pain control to their surgeon within the first 24 h.
The principal finding of this study is that outpatients undergoing nonarthroscopic anterior shoulder surgery may benefit from moderately prolonged analgesia when SSNB is used as an adjunct to an ISB-general anesthesia protocol. However, because the duration of analgesia varies widely among patients, it is difficult to predict if an individual patient will experience prolonged pain relief. Furthermore, adjunctive SSNB has no other impact on 24-hour outcome.
The SSN (C5/C6) arises from the superior trunk of the brachial plexus and has no clinically important cutaneous branches. Along with small branches of the axillary nerve, the SSN innervates up to 70% of the posterior shoulder joint. It also joins with the lateral pectoral nerve to supply sensory innervation of the acromioclavicular joint, subacromial bursa, and coracoclavicular ligament. It is therefore reasonable to assume that SSNB would be a valuable analgesic adjunct for shoulder surgery. The risks of SSNB are the rare (<1%) (5) chance of pneumothorax if the anesthetizing needle traverses the suprascapular notch and punctures the pleura and the equally unlikely possibility of IV local anesthetic injection.
It was previously unknown whether selective SSNB would provide additional benefit to patients undergoing ISB. Ritchie et al. (1) demonstrated that a preemptive SSNB improved analgesia and 24-hour quality-of-life outcomes when used as an adjunct to general anesthesia for shoulder arthroscopy. Patients receiving a SSNB had a 51% reduction in demand for and 31% reduced consumption of postoperative opioids, a lower 24-hour static and dynamic verbal pain scores, a five-fold reduction in nausea and vomiting, and a 24% shorter hospital stay. In contrast, our SSNB group only experienced a mean 220 minutes prolongation of analgesia that varied widely among individuals. Moreover, adjunctive SSNB had no effect on 24-hour outcome measures. There are two important differences in study design that may account for these incongruous results: anesthetic technique and surgical technique. The primary anesthetic in our study was an intermediate duration ISB that would be expected to provide analgesia to the entire shoulder, including the SSN distribution. Therefore, whereas all patients likely benefited from global brachial plexus anesthesia, the relatively limited prolongation of analgesia in the SSNB group was probably not of sufficient duration to significantly spare opioid use or affect 24 h outcome measures. Surgically, our patients underwent anterior procedures that violated tissues outside of the SSN sensory field yet did not produce the posterior port stimulation associated with an arthroscopic approach. Thus, it is not surprising that our patients received relatively little analgesic or immediate postoperative outcome benefit from SSNB.
The ideal anesthetic regimen for shoulder surgery is enigmatic. Pain after shoulder surgery may be severe, even with arthroscopic approaches (1). Pain is a major factor influencing hospital discharge in ambulatory patients, yet eliminating it only reduces mean length of stay by 4.7% after general anesthesia (8). Both Ritchie et al. (1) and our study are consistent with this observation. In the former, reducing pain with adjunctive SSNB positively impacted length of stay after general anesthesia. Conversely, the moderate increase in analgesic duration in our study had little effect over and above the analgesia provided by ISB regional anesthesia. Other methods to reduce postoperative shoulder pain have been evaluated. ISB was retrospectively shown to reduce admission rate, length of stay, and side effects as compared with general anesthesia (9,10). Small-dose ISB (10 mL of bupivacaine 0.125% with 1:400,000 epinephrine) for arthroscopic shoulder surgery reduced time to discharge and improved patient satisfaction, although VAS was only better during the first two hours of recovery (11). Continuous intraarticular local anesthetic infusion was superior to saline control for up to 24 hours after arthroscopic shoulder surgery (12), whereas continuous interscalene local anesthetic improved analgesia and patient satisfaction as compared with patient-controlled analgesia with nicomorphine after open procedures (13). Indeed, most outcome studies support the use of single-dose or continuous local anesthetic, alone or combined with general anesthesia, in the management of postoperative shoulder pain. Ours is the first study to consider selective peripheral nerve block as a supplement to a primary plexus block. The above reports and our data, which confirm the presence of moderately high VAS scores 24 hours after surgery, imply that pain control after shoulder surgery remains a challenging problem.
Our study design may have several limitations. First, we used mepivacaine for the ISB and bupivacaine for the SSNB to help differentiate the analgesic effect of each approach. Because both groups received an ISB and thus presumably had identical blockade of anterior shoulder afferents, we consider it unlikely that the global analgesia observed in Group SSNB is entirely explainable by the prolonged, isolated effects of bupivacaine on SSN innervation. Second, we only followed patient outcomes for 24 hours. Whereas it is possible that SSNB may have provided more long-term benefits, we think this unlikely because others have failed to demonstrate outcome differences after 24 hours (1,11,14). Third, because of the limited sample size, there could be some substantial differences in outcomes between the treatment groups that this study did not have the power to detect. Finally, time to first significant pain is a subjective end-point by which to measure effect, yet we believe it to be the best tool we have to measure block duration in the outpatient.
In conclusion, this is the first randomized clinical trial to compare the benefits of a selective peripheral nerve block used as a supplement with a primary plexus block for shoulder surgery. SSNB extended analgesia by an average of 3.6 hours in patients undergoing nonarthroscopic shoulder surgery under an ISB-general anesthetic regimen. Despite this moderate and statistically significant increase in analgesic duration, there were no other differences in 24-hour clinical outcomes. Because overall outcome is not clearly improved, variability in analgesic duration is large, and the small associated risk of pneumothorax, we cannot recommend SSNB as a routine adjunct to ISB for nonarthroscopic anterior shoulder surgery in outpatients.
The authors express their appreciation to Paul Benca, MD, and Derek Gallichotte, PAC, for their cooperation with this study and to our partners Drs Peter Hodgson, Dan Kopacz, Spencer Liu, Mike Mulroy, and Julie Pollock for their contributions.
1. Ritchie E, Tong D, Chung F, et al. Suprascapular nerve block for postoperative pain relief in arthroscopic shoulder surgery: a new modality? Anesth Analg 1997; 84: 1306–12.
2. Aszmann O, Dellon A, Birely B, McFarland E. Innervation of the human shoulder joint and its implications for surgery. Clin Orthop 1996; 330: 202–7.
3. Winnie AP. Interscalene brachial plexus block. Anesth Analg 1970; 49: 455–66.
4. Silverstein WB, Saiyed MU, Brown AR. Interscalene block with a nerve stimulator: a deltoid motor response is a satisfactory endpoint for successful block. Reg Anesth Pain Med 2000; 25: 356–9.
5. Moore DC. Regional block. 4th ed. Springfield, IL: Charles C. Thomas, 1965.
6. Howell S, Imobersteg A, Seger D, Marone P. Clarification of the role of the supraspinatus muscle in shoulder function. J Bone Joint Surg 1986; 68-A: 398–404.
7. Aldrete JA. The post-anaesthesia recovery score revisited. J Clin Anesth 1995; 7: 89–91.
8. Chung F, Mezei G. Factors contributing to a prolonged stay after ambulatory surgery. Anesth Analg 1999; 89: 1352–9.
9. Brown AR, Weiss R, Greenberg C, et al. Interscalene block for shoulder arthroscopy: comparison with general anesthesia. Arthroscopy 1993; 9: 295–300.
10. D’Alessio J, Rosenblum M, Shea K, Freitas D. A retrospective comparison of interscalene block and general anesthesia for ambulatory surgery shoulder arthroscopy. Reg Anesth 1995; 20: 62–8.
11. Al-Kaisy A, McGuire G, Chan V, et al. Analgesic effect of interscalene block using low-dose bupivacaine for outpatient arthroscopic shoulder surgery. Reg Anesth Pain Med 1998; 23: 469–73.
12. Klein SM, Nielsen KC, Martin A, et al. Interscalene brachial plexus block with continuous intraarticular infusion of ropivacaine. Anesth Analg 2001; 93: 601–5.
13. 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: 102–8.
14. Horn E-P, Schroeder F, Wilhelm S, et al. Wound infiltration and drain lavage with ropivacaine after major shoulder surgery. Anesth Analg 1999; 89: 1461–6.