Share this article on:

Does Patient-Controlled Continuous Interscalene Block Improve Early Functional Rehabilitation After Open Shoulder Surgery?

Section Editor(s): Horlocker, Terese T.Hofmann-Kiefer, Klaus MD*; Eiser, Tim MD*; Chappell, Daniel MD*; Leuschner, Stephan MD; Conzen, Peter MD*; Schwender, Dirk MD

doi: 10.1213/ane.0b013e31816151ab
Regional Anesthesia: Research Report

BACKGROUND: Early mobilization after shoulder surgery plays a vital role in successful functional rehabilitation. However, postoperative pain often reduces, or even prevents, effective physiotherapy. We investigated the effect of analgesia via patient-controlled interscalene technique on early functional rehabilitation after open shoulder surgery.

METHODS: Eighty-seven patients were randomly assigned to one of two groups: patient-controlled continuous interscalene block (PCISB) and patient-controlled IV (opioid) analgesia (PCA). Interscalene block was performed preoperatively; otherwise analgesic protocols were started in the postanesthesia care unit and were continued for 72 h. Physiotherapy was performed for 60 min a day on day 2 and 3 after surgery according to a standardized protocol. Maximum mobility was defined as the range of motion that could be achieved with pain as the limiting factor. Efficiency of functional rehabilitation was evaluated 1 day before and 3 days after surgery with the help of a multimodal scoring system (Constant-Score) that evaluates pain, daily life activity, strength and range of motion. Maximum intensity of pain was also monitored via Visual Analog Scales for the first 72 h after surgery and during in-hospital physiotherapy.

RESULTS: Constant-Score rates were significantly improved by the interscalene block. However, no significant differences in mobility and strength sub-scores were observed between the groups. Compared with PCA, PCISB proved to be beneficial concerning pain at rest at 6 h (P < 0.001), 24 h (P = 0.044), and 72 h (P = 0.013) and for pain during physiotherapy at 48 h after surgery (P = 0.016).

CONCLUSION: Compared with opioid-based PCA, PCISB improved analgesia, but not function, during early rehabilitation of the shoulder joint.

IMPLICATIONS: Compared with opioid-based patient-controlled analgesia, patient-controlled continuous interscalene block improved analgesia, but not function, during early rehabilitation of the shoulder joint in patients undergoing open shoulder procedures.

From the *Clinic of Anesthesiology/Critical Care Medicine and Pain Therapy (M.A.B., P.C.), Ludwig-Maximilians-University, City of Munich, Germany; †Department of Anesthesia, Municipal Hospital Friedrichshafen, City of Friedrichshafen, Germany; and ‡Department of Anesthesia, Municipal Hospital Luebeck, City of Luebeck, Germany.

Accepted for publication November 20, 2007.

Supported by Astra Zeneca Int.

The results of the study presented in the current work were worked out in the Clinic of Anesthesia, Ludwig-Maximilians-University, City of Munich, Germany.

All authors declare that there is no conflict of interest.

Reprints will not be available from the author.

Address correspondence to Dr. Klaus Hofmann-Kiefer, Klinik für Anästhesiologie der Ludwig-Maximilians-Universität München, Klinikum Innenstadt, Maistr. 11, D-80337 München, Germany. Address e-mail to

Early mobilization after trauma and orthopedic joint surgery is one of the basic principles of successful rehabilitation. Unfortunately, postoperative pain limits early mobilization, especially within the first 72 h after surgery.1 Severe postoperative pain is often observed after shoulder surgery, but can be successfully controlled by regional analgesic techniques, especially interscalene brachial plexus block.2–4 Recently, Capdevila et al.5 demonstrated that patient-controlled regional anesthesia optimized functional recovery and pain relief in patients undergoing acromioplasty (ACP) in an ambulatory orthopedic setting. Their primary objective was to determine whether patient-controlled perineural analgesia provided better patient functional capacity in general and facilitates daily activity after open shoulder surgery. In contrast to Capdevila et al.'s work, the goal of the current study was to evaluate whether early physiotherapy combined with patient-controlled regional analgesia specifically improves mobility and strength of the upper extremity compared with physiotherapy combined with patient-controlled opioid pain therapy.

Back to Top | Article Outline


After approval of the IRB, 87 consecutive adult patients (ASA physical status I–III) with rotator cuff defects requiring surgery (Ellman Grade II–III or Bateman Grade II–IV),6–8 who were scheduled to undergo open rotator cuff repair or ACP were included in this randomized, prospective clinical trial. All patients gave written informed consent. Exclusion criteria were a history of allergy to local anesthetics or opioids, blood clotting impairment, atrioventricular block higher than first degree,9,10 severe bronchopulmonary disorders, neuraxial diseases or plexus neuropathies and inability to cooperate or understand the study protocol. Surgical procedures were performed by two similarly experienced surgeons.

A team member who was not involved in the clinical investigations performed the randomization. Using a computer-generated table, each patient was assigned to either the patient-controlled continuous interscalene block (PCISB) or patient-controlled (opioid) analgesia (PCA) group. The clinical investigators belonged to the acute pain service of our clinic and were not involved in the anesthesia procedures, i.e., performing general or regional anesthesia. Thus, group designation was first revealed to one of the investigators with arrival of the patient in the postanesthesia care unit. However, this was not disclosed to the team member measuring the pre- and postoperative Constant-Scores (CS) (see below) until the study was completed.

On the day before surgery, mobility and muscular strength of the shoulder were measured with the help of a scoring system introduced by Constant and Murley in 1987.11 The CS is a method of functional assessment that provides reproducible subjective and objective data. It includes a questionnaire asking for pain intensity, sleep disturbances and the quality of the patient's daily life activity. In addition, it includes a range of motion measurement. The patient moves his upper arm actively into four clearly defined directions with the range of motion being measured in comparison to the neutral position. According to the guidelines of the European Society for Shoulder and Elbow Surgery, these movements should be pain-free, i.e., the number of degrees at which the pain starts determines the range of motion.12 The objective section of the score is completed by measuring the strength of the abducted upper arm (Table 1).13–16

Table 1

Table 1

Because most of our (postoperative) patients were not completely pain-free, even at rest, one item of the CS had to be adopted for our purposes. In the current study, the number of degrees at which pain became subjectively intolerable for the patient determined the range of active motion.

Patients were orally premedicated with midazolam 0.1 mg/kg 1 h before anesthesia. After arrival in the operating room, two peripheral IV catheters were inserted and afterwards the patients in the PCISB group were placed in the supine position. Continuous interscalene brachial plexus block was performed by experienced anesthesiologists with the assistance of a peripheral nerve stimulator (needle: “Contiplex D®”, stimulator: “Stimuplex® HNS11”; Braun, Melsungen, Germany) according to the technique originally described by Winnie,17 which was modified by Meier et al.18,19 The needle was connected to the nerve stimulator initially set at 1.0 mA and 2 Hz. It was redirected until biceps or other more distal upper extremity muscles were elicited with a minimum current between 0.3 and 0.4 mA. Afterwards, the catheter (20G polyamide catheter, Braun, Melsungen, Germany) was inserted through the needle and advanced 4 cm past the needle tip. After exclusion of endovascular placement, the catheter was injected with 40 mL of 0.75% ropivacaine as an intermittent injection over 10 min. To test the spread of sensory block, we used the pinprick method and, in addition, evaluated the motor block by means of the Simon-Scale, which is a modified Bromage-Scale for the upper extremity.20 Exclusion before surgery was planned in case the developing block should be incomplete (sensory block of <3 dermatomes or a missing sensory block of the upper arm region dermatomes C5 and C6). We abstained from placing placebo catheters in the PCA group because of ethical reasons21 and we did not perform single-injection blocks.

General anesthesia, standardized for the two study groups, was induced with thiopental 5 mg/kg, fentanyl 0.002 mg/kg and rocuronium 0.6 mg/kg or atracurium 0.5 mg/kg. To maintain general anesthesia, we administered sevoflurane (1%–2% end-tidal concentration) in air. If necessary, additional doses of fentanyl and muscle relaxants were given without restrictions. During the surgical procedure no further local anesthetic was administered in the PCISB group.

Back to Top | Article Outline

Postoperative Analgesic Management

The postoperative analgesia protocol was started in the postanesthesia care unit 30 min after patient arrival (study t = 0) and continued on the surgical ward. In the PCISB group, a patient-controlled infusion system (IVAC PCAM, Alaris Medical Systems, Hampshire, Great Britain) was started with 0.2% ropivacaine (10 mg/h as a basal infusion rate) via the interscalene catheter. Patient-controlled boluses of 8 mg were available every 20 min up to a maximum of 450 mg of ropivacaine in 24 h. Patient-controlled regional analgesia was maintained for 72 h after surgery in the PCISB group. In patients in the PCA group, a PCA pump (IVAC PCAM) was connected, providing patient-controlled boluses of 2 mg piritramide within a lockout period of 10 min, no background basal infusion rate and a 4 h maximum dose of 30 mg. Corresponding to the PCISB group, patient-controlled analgesia was maintained for 72 h.

It was our intention to achieve comparative levels (<30 mm on a 100 mm Visual Analog Scale [VAS]) in both of the study groups before starting pain therapy with help of infusion pumps in the postanesthesia care unit. In the PCISB group this usually was provided by the interscalene block given before surgery, but IV rescue doses of piritramide were possible. In the PCA group, this was achieved by an initial loading dose of piritramide. On the ward, IV rescue doses of 3.75–7.5 mg of piritramide were possible in both groups without restriction if pain exceeded a score of >40 mm on the VAS, except during physiotherapy. Consumption of local anesthetics and opioids as well as PCA variables (number of boluses demanded, number of boluses delivered) were recorded for 72 h after surgery.

Mobilization was performed on the ward on day two and three after surgery. Two physiotherapists mobilized the shoulder according to a standardized protocol for 1 h per day. Patients were trained in extension and adduction, as well as flexion and abduction, by the physiotherapeutic staff. According to the physiotherapeutic protocol, passive motion (in contrast to active movements during assessments of CS) was stopped in order to avoid damage to the operated shoulder joint if pain became intolerable or if pain exceeded a level of 75 points on a 100 points VAS. Independently of pain intensity, flexion and abduction were both limited to an angle of 60 to 70 degrees.

Pain was evaluated by an (unblinded) investigator using a VAS at 6 h, 24 h, 48 h, and 72 h after the end of surgery. In addition, pain was evaluated during early mobilization via VAS. About 80 h after surgery, PCISB and PCA pumps were discontinued and the catheters removed. Without further physiotherapy, CS were assessed again 96 h after the end of surgery, 16 h after the catheters had been removed, by a team member blinded to the previous analgesic technique.

Back to Top | Article Outline

Statistical Analysis

Sample size calculations were performed according to the following considerations: we considered a 25% reduction in CS (primary end-point) between groups as clinically relevant (mean decrease from 20 to 15 on a scale of 0–100). Assuming a standard deviation (sd) of ±7.022,23 for CS, a two-sided type I error protection of 0.05 and a power of more than 0.80 to avoid a type II error, 34 patients were required in each group. Secondary end-points were perioperative consumption of opioids and local anesthetics, postoperative pain intensity (VAS), and postoperative nausea and vomiting.

Normality was assessed using the Kolmogorov-Smirnov test. Normally distributed variables were analyzed using parametric methods, summarizing data using mean ± sd and comparing groups using Student's t-test. Otherwise, variables were analyzed using nonparametric methods, summarizing using median (5th–95th percentiles) and comparing groups using the Mann–Whitney rank-sum test. Fisher's exact test was used to compare categorical data where appropriate. Statistical analysis was performed using Sigma Stat software Version 3.1 (RockWare, Golden), and Microsoft Excel Version 2003 (Microsoft Deutschland GmbH, Unterschleiβheim, Germany).

Back to Top | Article Outline


Enrollment commenced in October 2004 and was concluded in July 2006. Seventy patients completed the study (PCISB n = 36; PCA n = 34). With β = 0.86, (α = 0.05, two-sided; sd = ±7.4) the power of the investigation proved to be above the desired level of 80%. Patient characteristics, type and duration of surgery were similar in all groups (Table 2).

Table 2

Table 2

Back to Top | Article Outline

Primary End-Point

CS decreased from preoperative baseline measurements in both groups (Fig. 1). However, this decrease was significantly less distinct in the PCISB group (P = 0.021). The results of the CS’ subsections are displayed in Table 3. Significant differences between the groups were found only in the subsection “pain” (P = 0.047). The maximum postoperative range of movements was considerably limited in both of the groups. Many patients in both groups were completely unable to move their upper extremities actively in one of the intended directions (abduction: PCA 52.9%, PCISB 64.9%; retroflexion: PCA 61.7%, PCISB 56.7%; external rotation: PCA 97.0%, PCISB 97.3%; internal rotation: PCA 50.0%, PCISB 48.6%). Only 5.9% of the patients in the PCA group and 13.5% in the PCISB group were able to perform the strength test. However, there were no differences between groups.

Figure 1

Figure 1

Table 3

Table 3

Back to Top | Article Outline

Secondary End-Points

Mean intraoperative fentanyl consumption was significantly lower in the PCISB group compared with the PCA group (P < 0.01; Student's t-test) (Table 2). In the PCISB group, mean VAS scores were significantly lower at rest at 6 h (P < 0.001), 24 h (P = 0.044), and 72 h (P = 0.013), as well as during physiotherapy on day 2 (P = 0.016) compared with the PCA group (Fig. 2). As a consequence, rescue doses of piritramide were required significantly more often in the PCA group (P < 0.001). Nausea and/or vomiting occurred in 26.5% (PCA group) and 5.5% (PCISB group) (P < 0.01) and were treated successfully in all cases with odansetron. There were no catheter infections.

Figure 2

Figure 2

Back to Top | Article Outline

Protocol Violations

Incomplete blocks according to our exclusion criteria did not occur. Therefore, none of the patients were excluded before surgery. However, one patient was dropped from the study because the block did not include the dermatome C5 when re-evaluated in the postanesthesia care unit. Nine patients were excluded due to catheter dislocation or early catheter removal: One of these patients wanted the catheter to be removed because of a Horner's syndrome that had developed as a consequence of PCISB. Four patients were excluded because of postoperative dyspnea, caused by a paresis of the diaphragm in the PCISB group. A violation of the study protocol occurred three times in both groups. In the PCA group, one patient had to be excluded because of intolerable pruritus. Thus, 70 patients completed the study and remained in the final analysis. The withdrawn subjects’ characteristics did not differ significantly from those of the patients who completed the study.

Back to Top | Article Outline


The current study focused on the influence of PCISB on early rehabilitation after open shoulder surgery. We found that PCISB results in superior postoperative analgesia in comparison to PCA. However, we did not observe significant differences between the groups in terms of shoulder joint mobility and muscular strength after perioperative physiotherapy.

The CS is an accepted tool for assessing functional disorders of the shoulder joint. Nevertheless, tissue trauma and swelling can considerably restrict some patients’ capacity to perform active movements, including abduction of the upper arm. In these patients, the practicability of measuring active range of motion can be limited. Because of this, other investigators may have developed alternative methods of assessment that are easier to accomplish but which only measure subjective end-points, such as sleep quality or patient satisfaction.5,24 However, these subjective criteria, which are principally pain-dependent, are not convenient to determine the influence of postoperative trauma or the efficacy of physiotherapy on shoulder mobility. We believe that only objective data, as demonstrated by Ilfeld et al.25 in a recent investigation on total shoulder arthroplasty, or range of motion measurements, as provided by the CS, are suitable in this context, even though they are difficult to obtain. Alternative tests, such as the Oxford Shoulder Score or the Sf36-test12 do not actively test mobility.

In contrast to some other investigations, the maximum range of motion achieved in both groups was surprisingly low in our study. Ilfeld et al.,24,25 observed significant improvements in external rotation in patients receiving interscalene analgesia for shoulder arthroplasty compared with placebo. However, patients requiring arthroplasty usually have osseous lesions, unlike our study population, which suffered from muscular defects. In addition, Ilfeld et al. did not use a test measuring active range of motion, like the CS. These differences in study design hinder an objective comparison of the data. Nevertheless, in the current study the decrease in overall postoperative CS was significantly less extensive in the PCISB group in comparison to PCA. Thus, PCISB seemed at first to provide an advantage in early rehabilitation. However, a subgroup analysis clearly demonstrated that the differences between groups only derived from significantly improved pain scores in the PCISB group. Superior analgesia did not lead to improved results of early rehabilitation, as defined by strength and range of motion (Table 3). One can assume that postoperative swelling, stiffness and inflammation are crucial factors in this context. Otherwise the improved analgesic quality provided by PCISB should have had a more relevant influence. This is emphasized by the fact that the postoperative scores in the pain-subsection were even higher than the preoperative values, demonstrating the effectiveness of both analgesic techniques evaluated in the current study.

In 1999, Capdevila et al.26 successfully used regional anesthetic techniques up to 12 h a day for early rehabilitation after knee surgery. One can speculate that with several hours of training a day, “best possible” analgesia, provided by PCISB, could also have had more influence in the current investigation. Therefore, it may be possible that with longer rehabilitation intervals, the advantages of regional analgesia could have become more evident. However, we tried to evaluate the benefits of PCISB under regular clinical conditions and therefore the clinic's physiotherapeutic protocol remained unchanged for the study population.

In conclusion, compared with PCA, PCISB improved pain management, as well as postoperative nausea and vomiting, but not early rehabilitation of the shoulder joint in the current investigation. Further studies are necessary to determine whether a longer interval of regional anesthesia would have a favorable influence on the outcome of early rehabilitation of patients undergoing rotator cuff repair or open ACP.

Back to Top | Article Outline


1. Julien RE, Williams BA. Regional anesthesia procedures for outpatient shoulder surgery. Int Anesthesiol Clin 2005;43: 167–75
2. Borgeat A, Schappi B, Biasca N, Gerber C. Patient-controlled analgesia after major shoulder surgery: patient-controlled interscalene analgesia versus patient-controlled analgesia. Anesthesiology 1997;87:1343–7
3. Borgeat A, Tewes E, Biasca N, Gerber C. Patient-controlled interscalene analgesia with ropivacaine after major shoulder surgery: PCIA vs PCA. Br J Anaesth 1998;81:603–5
4. Kean J, Wigderowitz CA, Coventry DM. Continuous interscalene infusion and single injection using levobupivacaine for analgesia after surgery of the shoulder: a double-blind, randomised controlled trial. J Bone Joint Surg Br 2006;88:1173–7
5. Capdevila X, Dadure C, Bringuier S, Bernard N, Biboulet P, Gaertner E, Macaire P. Effect of patient-controlled perineural analgesia on rehabilitation and pain after ambulatory orthopedic surgery: a multicenter randomized trial. Anesthesiology 2006;105:566–73
6. Bateman JE. The diagnosis and treatment of ruptures of the rotator cuff. Surg Clin North Am 1963;43:1523–30
7. Ellman H. Diagnosis and treatment of incomplete rotator cuff tears. Clin Orthop Relat Res 1990;64–74
8. Habermeyer P, Lehmann L, Lichtenberg S. Rotator cuff tears: diagnosis and therapy. Orthopade 2000;29:196–208
9. Vyas A, O'Connell FM, Vacanti CA. Third-degree heart block complicating supraclavicular brachial plexus block. Anesthesiology 1996;85:675–7
10. Matta BF, Magee P. Wenckebach type heart block following spinal anaesthesia for caesarean section. Can J Anaesth 1992; 39:1067–8
11. Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. Clin Orthop Relat Res 1987;160–4
12. Eur Society for Surgery of the Shoulder and the Elbow. Internet Communication, 2007.
13. Conboy VB, Morris RW, Kiss J, Carr AJ. An evaluation of the Constant-Murley shoulder assessment. J Bone Joint Surg Br 1996;78:229–32
14. Constant CR. Assessment of shoulder function. Orthopade 1991;20:289–94
15. Constant CR. An evaluation of the Constant-Murley shoulder assessment. J Bone Joint Surg Br 1997;79:695–6
16. Fokter SK, Cicak N, Skorja J. Functional and electromyographic results after open rotator cuff repair. Clin Orthop Relat Res 2003;121–30
17. Winnie AP. Interscalene brachial plexus block. Anesth Analg 1970;49:455–66
18. Meier G, Bauereis C, Maurer H, Meier T. Interscalene plexus block. Anatomic requirements–anesthesiologic and operative aspects. Anaesthesist 2001;50:333–41
19. Meier G, Bauereis C, Heinrich C. Interscalene brachial plexus catheter for anesthesia and postoperative pain therapy. Experience with a modified technique. Anaesthesist 1997;46:715–19
20. Simon MA, Gielen MJ, Alberink N, Vree TB, van EJ. Intravenous regional anesthesia with 0.5% articaine, 0.5% lidocaine, or 0.5% prilocaine. A double-blind randomized clinical study. Reg Anesth 1997;22:29–34
21. Brull R, McCartney CJ, Chan VW, El-Beheiry H. Neurological complications after regional anesthesia: contemporary estimates of risk. Anesth Analg 2007;104:1009–11
22. Hsu SL, Ko JY, Chen SH, Wu RW, Chou WY, Wang CJ. Surgical results in rotator cuff tears with shoulder stiffness. J Formos Med Assoc 2007;106:452–61
23. Gartsman GM, Khan M, Hammerman SM. Arthroscopic repair of full-thickness tears of the rotator cuff. J Bone Joint Surg Am 1998;80:832–40
24. Ilfeld BM, Morey TE, Enneking FK. Continuous infraclavicular brachial plexus block for postoperative pain control at home: a randomized, double-blinded, placebo-controlled study. Anesthesiology 2002;96:1297–304
25. Ilfeld BM, Vandenborne K, Duncan PW, Sessler DI, Enneking FK, Shuster JJ, Theriaque DW, Chmielewski TL, Spadoni EH, Wright TW. Ambulatory continuous interscalene nerve blocks decrease the time to discharge readiness after total shoulder arthroplasty: a randomized, triple-masked, placebo-controlled study. Anesthesiology 2006;105:999–1007
26. Capdevila X, Barthelet Y, Biboulet P, Ryckwaert Y, Rubenovitch J, d'Athis F. Effects of perioperative analgesic technique on the surgical outcome and duration of rehabilitation after major knee surgery. Anesthesiology 1999;91:8–15
© 2008 International Anesthesia Research Society