- Email: [email protected]
Low-Volume Interscalene Brachial Plexus Block for Post-Thoracotomy Shoulder Pain
Low-Volume Interscalene Brachial Plexus Block for Post-Thoracotomy Shoulder Pain Michal Barak, MD,*‡ Dmitri Iaroshevski, MD,* Eleonardo Poppa, MD,* Alon Ben-Nun, MD, PhD,†‡ and Yeshayahu Katz, MD, DSc*‡ Objectives: This study was designed to evaluate the effectiveness of low-volume interscalene brachial plexus block for post-thoracotomy ipsilateral shoulder pain and to compare it with nonsteroidal anti-inflammatory drug treatment. Design: Prospective nonblinded study. Setting: University hospital. Participants: Sixty adult patients. Intervention: Patients who underwent elective thoracic surgery under combined epidural and general anesthesia, and after surgery were free of incisional pain but complaining of shoulder pain, were included in the study. They were selected in a sequential manner and placed into 2 groups of 30 patients each. Group 1 had a low-volume interscalene brachial plexus block, using 10 mL of bupivacaine 0.5%. Group 2 had an intramuscular injection of diclofenac sodium, 75 mg. Measurements and Main Results: Pain was measured during their stay in the postanesthesia care unit (PACU) by using a visual analog score (VAS). Opioids were adminis-
T
HORACIC OPERATIONS are known to be painful procedures.1,2 Patients report moderate-to-severe pain at the incisional site and in the shoulder. Although thoracic epidural is an effective tool for incisional pain, it is ineffective for the control of postoperative shoulder pain.3,4 The incidence of ipsilateral shoulder pain varies. Burgess et al5 observed shoulder pain in 42% of patients, which was severe in 25%. Pennefather et al6 found an incidence of 60%, and Tan et al7 reported an incidence of 68%. In the authors’ earlier study, the incidence of shoulder pain was 37%.3 During the past 2 decades, several solutions have been suggested in an effort to control shoulder pain, reducing it to some extent.5-10 The effectiveness of nonsteroidal anti-inflammatory drugs (NSAIDs) in treating postthoracotomy shoulder pain was shown in several studies. Burgess et al5 reported complete elimination of shoulder pain in 80% of patients using repeating doses of ketorolac, whereas the authors found substantial relief in patients following rectal indomethacin.3 In this prospective study, a low-volume interscalene brachial plexus block was used for pain treatment in patients who suffered post-thoracotomy ipsilateral shoulder pain and compared it with treatment with an intramuscular injection of diclofenac (NSAID).
From the Departments of *Anesthesiology and †Thoracic Surgery, Rambam Medical Center, Haifa, Israel; and ‡The Bruce Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel. Address reprint requests to Yeshayahu Katz, MD, DSc, Department of Anesthesiology, Rambam Medical Center, Haifa 31096, Israel. Email: [email protected] © 2007 Elsevier Inc. All rights reserved. 1053-0770/07/2104-0013$32.00/0 doi:10.1053/j.jvca.2006.08.013 554
tered when pain relief was incomplete. Pain intensity was re-estimated the next morning and patient satisfaction was scored. VAS was found to be significantly lower in the lowvolume interscalene block group than in the diclofenac group at 30 minutes after treatment and when leaving PACU (p < 0.001 for both). Patients in the interscalene block group stayed longer in the PACU (p ⴝ 0.019), and significantly fewer required rescue opioids (p ⴝ 0.03). There was no significant difference between the groups in patient satisfaction with the pain treatment. Conclusions: The authors concluded that low-volume interscalene brachial plexus block is a superior treatment for post-thoracotomy shoulder pain compared with diclofenac injection, although it requires a slightly longer stay in the PACU. © 2007 Elsevier Inc. All rights reserved. KEY WORDS: analgesia, epidural, interscalene block, postoperative pain, thoracic surgery PATIENTS AND METHODS With the approval of the local ethics committee, a prospective study was conducted of 153 patients who were scheduled for elective thoracic surgery for lobe resection with combined epidural-general anesthesia (GA), and all consented to participate in the study prior to anesthesia and surgery. Exclusion criteria included chronic use of analgesics, preoperative shoulder pathology, and contraindications for conducting a brachial plexus block, such as coagulopathy and infection at the site of injection. Only patients who complained of ipsilateral shoulder pain in the postanesthesia care unit (PACU) and whose incisional site was not painful were included in this study. All patients had standard pre- and intraoperative care. Under monitoring with noninvasive blood pressure, electrocardiogram, and pulse oximeter, intravenous (IV) access was established. A thoracic epidural catheter (T5-T8) was placed before the induction of GA. The midline approach was used in all patients, and the epidural space was identified by the loss-of-resistance technique; then, the epidural catheter was inserted. A test dose of 3 mL of lidocaine 2% with epinephrine 1:200,000 was used followed by a bolus dose of 5 mL of bupivacaine 0.125% with fentanyl, 0.025 mg, via the epidural catheter, and then by a continuous infusion of 5 mL/h of bupivacaine 0.125% with fentanyl, 0.002 mg/mL. Before the induction of GA, sensory block was assessed by pinprick. GA was identical for all patients by using IV propofol (2-3 mg/kg) and fentanyl (0.001-0.002 mg/kg), followed by vecuronium (0.1 mg/kg). An inhalation anesthetic (isoflurane) was used for maintenance of anesthesia. All patients were placed in the standard (horizontal) lateral decubitus position. Padded rolls were placed under the dependent axilla, and the arms were flexed and supported by a soft-surface pillow. The operations were performed by a team of 3 surgeons who used a standard technique with separation and retraction of the ribs. No ribs were resected or broken. After surgery, the patients arrived at the PACU where the nurses assessed their pain within minutes of arrival by using a visual analog score (VAS 0-10). A patient who complained of shoulder pain with VAS ⱖ4 while incisional pain was estimated as VAS ⱕ3 was included in the study. The first 30 patients were assigned to have a low-volume interscalene brachial plexus block (group 1). The block was performed by an experienced senior anesthesiologist within 10 to 20 minutes.
Journal of Cardiothoracic and Vascular Anesthesia, Vol 21, No 4 (August), 2007: pp 554-557
POST-THORACOTOMY SHOULDER PAIN
555
Table 1. Patients’ Demographic and Medical Data
Age Sex ASA classification Side of surgery Duration of PACU stay Morphine consumption Number of patients required morphine Single patient’s dose Pain at incision Entering PACU POD 1 Patient satisfaction (scale of 1-5)
Group 1: LowVolume Interscalene Brachial Plexus Block
Group 2: Intramuscular Diclofenac
p Value
63 (44-82) 10/20 18/12 16/14 2.9 ⫾ 1.0
56 (18-81) 13/17 16/14 14/16 2.4 ⫾ 0.6
0.56 0.43 0.60 0.85 0.019
15 (50)
23 (77)
0.03
Mg, mean (⫾SD)
3.8 ⫾ 5.4
6.0 ⫾ 5.6
0.07
VAS, mean (⫾SD) VAS mean (⫾SD)
1.5 ⫾ 0.5 1.8 ⫾ 1.2
1.6 ⫾ 0.5 1.5 ⫾ 0.5
0.51 0.77
Mean (⫾SD)
4.5 ⫾ 0.5
4.0 ⫾ 0.7
0.17
Years, median (range) Male/female (n) 2/3 (n) Left/right Hours, mean (⫾SD) Number (%)
Abbreviation: POD 1, first postoperative day.
Under sterile conditions, the interscalene groove was identified, and an interscalene brachial plexus block was performed by using a nerve stimulator in a standard technique,11 modified by using a low volume of local anesthetic (10 mL of bupivacaine 0.5%). The subsequent 30 patients (group 2) had intramuscular injections of diclofenac sodium, 75 mg (Abitren; ABIC Manufacturer, Biogal Pharmaceutical Works Ltd, Debrecen, Hungary). Thirty minutes after the treatment and every 30 minutes thereafter, the PACU nurse reassessed the pain at the shoulder and incisional sites. Rescue opioids (IV morphine in increments of 2 mg) were administered as long as pain was estimated as VAS ⱖ4. Patients from both groups were examined by the attending anesthesiologist, and all had arterial blood gases and chest x-rays 1 hour after admission to the PACU. Complications such as motor block, Horner’s syndrome, and pneumothorax were ruled out before the patient left the PACU. The anesthesiologist and PACU nurse were not blinded to the patients’ intervention. Criteria for discharge from the PACU included 8 to 10 points according to the Aldrete Score,12 pain estimation in all sites as VAS ⱕ3, and no analgesic requirement or administration in the previous 30 minutes. Data collected included patient’s age, sex, American Society of Anesthesiologists score, type and side of surgery, VAS of pain at incision and shoulder on arriving in the PACU, VAS 30 minutes after treatment and at the time of leaving the PACU, duration of PACU stay, use of opioids in the PACU, VAS regarding shoulder pain, and incisional pain on the first postoperative day, satisfaction from the treatment on a 1 to 5 scale, and complications. Statistical analysis was performed with Statistica (StatSoft, Tulsa, OK). Parametric variables were compared by analysis of variance of repeated measures. Intergroup comparison of nonparametric variables used the Kruksal-Wallis analysis of variance and the median test. For dependent nonparametric paired samples, a Wilcoxon matched pairs test was applied. The p value was adjusted by using Fisher least significant difference. Differences were deemed significant at p ⬍ 0.05. RESULTS
During the 8-month period in which the study was conducted, 153 patients underwent thoracic surgery for lobe resection. Nine were excluded because of the chronic use of analgesics. Of the remaining 144 patients, 60 experienced ipsilateral shoulder pain; thus, the incidence of shoulder pain was about 41%. The demographic and medical data of the patients in both groups were similar (Table 1). Patients who had
the low-volume interscalene brachial plexus block stayed in the PACU longer than patients who were treated with intramuscular diclofenac (2.9 ⫾ 1.0 hours and 2.4 ⫾ 0.6 hours, respectively; p ⫽ 0.019). Significantly more patients who were treated with diclofenac required rescue opioids (23 patients, 77%) than those with the low-volume interscalene block (15 patients, 50%; p ⫽ 0.03). No statistically significant difference was found between the groups in the mean dose of rescue opioids of a single patient (p ⫽ 0.07). Shoulder pain is summarized in Figure 1. VAS was found to be significantly lower in the low-volume interscalene block group than in the diclofenac group at 30 minutes after treatment and when leaving the PACU (p ⬍ 0.001 for both). There was no significant difference in the shoulder pain between the groups on postoperative day 1 (group 1: VAS ⫽ 1.6 ⫾ 0.9 and group 2: VAS ⫽ 2.3 ⫾ 1.3, p ⫽ 0.14). Patient satisfaction with pain management was similar in both groups. No complications were found in patients from either group as evident by physical examinations, arterial blood gases, and chest x-rays. DISCUSSION
Ipsilateral shoulder pain is a common complaint in patients undergoing thoracotomy, occurring in up to 68% of patients.7 The cause of pain is unknown, although some explanations have been proposed. Burgess et al5 hypothesized that the pain results from transection of a major bronchus. Other explanations include the lateral decubitus position, pleural irritation caused by the chest drain, or referred pain from the diaphragm.8 The pain is usually described by patients as moderate-to-severe,4,5 similar to the findings in this study; in these circumstances, additional epidural analgesia is ineffective in alleviating the pain.3 Although the cause of the shoulder pain is unidentified, clinicians have searched for an effective treatment. Traditional analgesia with opioids should be given with caution because opioids are respiratory depressants, and these patients have reduced pulmonary reserves and often suffer from chronic respiratory disease. NSAIDs are another option; their favorable effect was reported previously by Perttunen et al13 who used IV
556
BARAK ET AL
Fig 1.
VAS of shoulder pain in groups 1 and 2 during PACU stay and on the first postoperative day. *p < 0.05.
diclofenac in post-thoracotomy pain. Other studies have described complete elimination of shoulder pain in 80% of patients with repeating doses of ketorolac5 and substantial relief in all patients after rectal indomethacin.3 The problem with NSAIDs is their nephrotoxicity in dehydrated patients,14 such as in patients who undergo thoracotomy.15 Acetaminophen was studied recently, given preemptively and regularly after thoracic surgery, with good effect in alleviating shoulder pain.4 Some novel solutions for treatment of post-thoracotomy shoulder pain have been tried as well. Garner and Coats9 suggested a stellate ganglion block as a possible effective treatment, and phrenic nerve infiltration was tried by Scawn et al.8 Suprascapular nerve block7 and intrapleural bupivacaine6 were tried, all with limited success. Ng and Chow10 described a case series of five patients who had an interscalene brachial plexus block as treatment for post-thoracotomy shoulder pain. This study followed on their experience and tried reduced-volume interscalene brachial plexus blocks in a prospective study. Assuming that the pain was at least partially caused by excessive strain on the shoulder joint capsule and ligaments with the patient in the lateral position, it was thought that regional analgesia to the shoulder might induce pain relief. The reduced volume of local anesthetic was chosen to minimize motor block, as was achieved in group 1, and also to decrease the possible side effects of the local anesthetics. The low-volume interscalene brachial plexus block was an effective treatment for post-thoracotomy shoulder pain, reducing VAS notably within 30 minutes, and continuing its analgesic effect for hours. Intramuscular diclofenac injection was less effective, judging by the patient’s VAS. On the first postoperative day, the difference between the groups became sta-
tistically insignificant. Shoulder pain decreased gradually and disappeared the day after surgery. A similar trend was found by Mac et al4 in a study in which the pain was reduced within 20 hours. However, the possibility of chronic shoulder pain after thoracic surgery exists, as Stammberger et al16 reported. In their study of early- and long-term complaints after video-assisted thoracoscopic surgery (173 patients), 1 patient was unable to work 12 weeks after the operation because of shoulder pain. Postoperative shoulder function may improve with early mobilization and effective physiotherapy,17 which are greatly dependent on good pain control. The exact mechanism by which the interscalene block relieves the pain can be explained in more than one way. The brachial plexus block prevents nociceptive impulses of the shoulder region from reaching the spinal cord. However, involvement of the phrenic nerve in the conduction of stimuli from the thoracic viscera cannot be ruled out. Standard-volume interscalene brachial plexus block carries a 100% phrenic nerve blockade as a side effect.11 This may be somewhat lower using the reduced volume of local anesthetics. Nevertheless, phrenic nerve block could explain the results of this study, supporting the hypothesis of Scawn et al8 regarding the influence of phrenic nerve infiltration on shoulder pain. The pharmacologic effect of local anesthetics may be relevant as well. The systemic absorption of bupivacaine has a hypnotic effect that may alter the patient’s pain perception.18 The group of patients treated with diclofenac showed improvement in shoulder pain, as seen in the decline of VAS (Fig 1). The beneficial effect of NSAIDs in treating post-thoracotomy shoulder pain described by Burgess et al5 with ketorolac and in the authors’ earlier study with indomethacin3 supports
POST-THORACOTOMY SHOULDER PAIN
557
the theory that an inflammatory process is involved in the pain occurrence. Interestingly, both groups were satisfied with pain treatment, which is a known outcome in reports of pain treatment.19,20 Patients often report satisfaction with their pain management, regardless of pain intensity or treatment mode. This can be explained by the patients’ expectations of postoperative pain, their little awareness of the efficacy of analgesics, and the tendency to base their response on the sympathy and friendliness of staff rather than on the outcome of pain relief.19,20 The patients in this study were selected in a sequential manner, which could have introduced bias, because the study was completed in 1 group before starting the other. Admittedly, randomization of treatment would have made for a stronger power of results. The study would also have been better with a double-blind saline control. Although the patients’pain was monitored often by the nurses in the PACU and in the ward, data were collected only at specific time points. Moreover, the interscalene block patients stayed longer in the PACU; thus, a time gap of about 30 minutes was created between the time points of the groups. This may give an inaccurate description of the real pain alteration over time. Another problem with the study was the population size, which is characteristic of other
studies in this field.3-10 Larger definitive trials are required to be more certain as to which technique should be used to effectively treat post-thoracotomy shoulder pain. The application of an invasive treatment, such as interscalene block, carries a higher risk than the intramuscular injection of an NSAID. In this study, there were no complications from the brachial plexus block performance. However, this is a relatively small study (regarding sample size) that does not have enough power to allow meaningful comments about complication rate. Nevertheless, complications such as pneumothorax, Horner’s syndrome, transient phrenic nerve paresis with diaphragm dysfunction, and systemic toxicity of local anesthetics should be kept in mind. In particular, phrenic nerve paresis may affect postoperative pulmonary function and recovery in a patient after lung resection. The use of a reduced concentration of local anesthesia may decrease the risk of complications, in addition to the block being conducted by an experienced, vigilant anesthesiologist. Moreover, patients who underwent interscalene block stayed longer in the PACU because the block performance was a time-consuming procedure. Because interscalene block is a higher-risk time-consuming procedure, it is suggested to reserve this treatment for selected patients whose shoulder pain is severe.
REFERENCES 1. Kaplan JA, Miller ED Jr, Gallagher EG Jr: Postoperative analgesia for thoracotomy patients. Anesth Analg 54:773-777, 1975 2. Lubenow TR, Faber LP, McCarthy RJ, et al: Postthoracotomy pain management using continuous epidural analgesia in 1324 patients. Ann Thorac Surg 58:924-929, 1994 3. Barak M, Ziser A, Katz Y: Thoracic epidural local anesthetics are ineffective in alleviating post-thoracotomy ipsilateral shoulder pain. J Cardiothorac Vasc Anesth 18:458-460, 2004 4. Mac TB, Girard F, Chouinard P, et al: Acetaminophen decreases early post-thoracotomy ipsilateral shoulder pain in patients with thoracic epidural analgesia: A double-blind placebo-controlled study. J Cardiothorac Vasc Anesth 19:475-478, 2005 5. Burgess FW, Anderson DM, Colonna D, et al: Ipsilateral shoulder pain following thoracic surgery. Anesthesiology 78:365-368, 1993 6. Pennefather SH, Akrofi ME, Kendall JB, et al: Double-blind comparison of intrapleural saline and 0.25% bupivacaine for ipsilateral shoulder pain after thoracotomy in patients receiving thoracic epidural analgesia. Br J Anaesth 94:234-238, 2005 7. Tan N, Agnew NM, Scawn ND, et al: Suprascapular nerve block for ipsilateral shoulder pain after thoracotomy with thoracic epidural analgesia: A double-blind comparison of 0.5% bupivacaine and 0.9% saline. Anesth Analg 94:199-202, 2002 8. Scawn ND, Pennefather SH, Soorae A, et al: Ipsilateral shoulder pain after thoracotomy with epidural analgesia: The influence of phrenic nerve infiltration with lidocaine. Anesth Analg 93:260-264, 2001 9. Garner L, Coats RR: Ipsilateral stellate ganglion block effective for treating shoulder pain after thoracotomy. Anesth Analg 78:11951196, 1994
10. Ng KP, Chow YF: Brachial plexus block for ipsilateral shoulder pain after thoracotomy. Anaesth Intensive Care 25:74-76, 1997 11. Wedel DJ and Horlocker TT: Nerve blocks, in Miller RD (ed): Miller’s Anesthesia (ed 6). Philadelphia, PA, Elsevier, 2005, pp 16851717 12. Aldrete JA, Kroulik D: A postanesthetic recovery score. Anesth Analg 49:924-934, 1970 13. Perttunen K, Kalso E, Heinonen J, et al: IV diclofenac in post-thoracotomy pain. Br J Anaesth 68:474-480, 1992 14. Whelton A: Nephrotoxicity of nonsteroidal anti-inflammatory drugs: Physiologic foundations and clinical implications. Am J Med 106:13s-24s, 1999 15. Brodsky JB, Fitzmaurice B: Modern anesthetic techniques for thoracic operations. World J Surg 25:162-166, 2001 16. Stammberger U, Steinacher C, Hillinger S, et al: Early and long-term complaints following video-assisted thoracoscopic surgery: evaluation in 173 patients. Eur J Cardiothorac Surg 18:7-11, 2000 17. Li WW, Lee TW, Yim AP: Shoulder function after thoracic surgery. Thorac Surg Clin 14:331-343, 2004 18. Ben-Shlomo I, Tverskoy M, Fleyshman G, et al: Hypnotic effect of i.v. propofol is enhanced by IM administration of either lignocaine or bupivacaine. Br J Anaesth 78:375-377, 1997 19. Apfelbaum JL, Chen C, Mehta SS, et al: Postoperative pain experience: Results from a national survey suggest postoperative pain continues to be undermanaged. Anesth Analg 97:534-540, 2003 20. McNeill JA, Sherwood GD, Starck PL, et al: Assessing clinical outcomes: Patient satisfaction with pain management. J Pain Symptom Manage 16:29-40, 1998
T
HORACIC OPERATIONS are known to be painful procedures.1,2 Patients report moderate-to-severe pain at the incisional site and in the shoulder. Although thoracic epidural is an effective tool for incisional pain, it is ineffective for the control of postoperative shoulder pain.3,4 The incidence of ipsilateral shoulder pain varies. Burgess et al5 observed shoulder pain in 42% of patients, which was severe in 25%. Pennefather et al6 found an incidence of 60%, and Tan et al7 reported an incidence of 68%. In the authors’ earlier study, the incidence of shoulder pain was 37%.3 During the past 2 decades, several solutions have been suggested in an effort to control shoulder pain, reducing it to some extent.5-10 The effectiveness of nonsteroidal anti-inflammatory drugs (NSAIDs) in treating postthoracotomy shoulder pain was shown in several studies. Burgess et al5 reported complete elimination of shoulder pain in 80% of patients using repeating doses of ketorolac, whereas the authors found substantial relief in patients following rectal indomethacin.3 In this prospective study, a low-volume interscalene brachial plexus block was used for pain treatment in patients who suffered post-thoracotomy ipsilateral shoulder pain and compared it with treatment with an intramuscular injection of diclofenac (NSAID).
From the Departments of *Anesthesiology and †Thoracic Surgery, Rambam Medical Center, Haifa, Israel; and ‡The Bruce Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel. Address reprint requests to Yeshayahu Katz, MD, DSc, Department of Anesthesiology, Rambam Medical Center, Haifa 31096, Israel. Email: [email protected] © 2007 Elsevier Inc. All rights reserved. 1053-0770/07/2104-0013$32.00/0 doi:10.1053/j.jvca.2006.08.013 554
tered when pain relief was incomplete. Pain intensity was re-estimated the next morning and patient satisfaction was scored. VAS was found to be significantly lower in the lowvolume interscalene block group than in the diclofenac group at 30 minutes after treatment and when leaving PACU (p < 0.001 for both). Patients in the interscalene block group stayed longer in the PACU (p ⴝ 0.019), and significantly fewer required rescue opioids (p ⴝ 0.03). There was no significant difference between the groups in patient satisfaction with the pain treatment. Conclusions: The authors concluded that low-volume interscalene brachial plexus block is a superior treatment for post-thoracotomy shoulder pain compared with diclofenac injection, although it requires a slightly longer stay in the PACU. © 2007 Elsevier Inc. All rights reserved. KEY WORDS: analgesia, epidural, interscalene block, postoperative pain, thoracic surgery PATIENTS AND METHODS With the approval of the local ethics committee, a prospective study was conducted of 153 patients who were scheduled for elective thoracic surgery for lobe resection with combined epidural-general anesthesia (GA), and all consented to participate in the study prior to anesthesia and surgery. Exclusion criteria included chronic use of analgesics, preoperative shoulder pathology, and contraindications for conducting a brachial plexus block, such as coagulopathy and infection at the site of injection. Only patients who complained of ipsilateral shoulder pain in the postanesthesia care unit (PACU) and whose incisional site was not painful were included in this study. All patients had standard pre- and intraoperative care. Under monitoring with noninvasive blood pressure, electrocardiogram, and pulse oximeter, intravenous (IV) access was established. A thoracic epidural catheter (T5-T8) was placed before the induction of GA. The midline approach was used in all patients, and the epidural space was identified by the loss-of-resistance technique; then, the epidural catheter was inserted. A test dose of 3 mL of lidocaine 2% with epinephrine 1:200,000 was used followed by a bolus dose of 5 mL of bupivacaine 0.125% with fentanyl, 0.025 mg, via the epidural catheter, and then by a continuous infusion of 5 mL/h of bupivacaine 0.125% with fentanyl, 0.002 mg/mL. Before the induction of GA, sensory block was assessed by pinprick. GA was identical for all patients by using IV propofol (2-3 mg/kg) and fentanyl (0.001-0.002 mg/kg), followed by vecuronium (0.1 mg/kg). An inhalation anesthetic (isoflurane) was used for maintenance of anesthesia. All patients were placed in the standard (horizontal) lateral decubitus position. Padded rolls were placed under the dependent axilla, and the arms were flexed and supported by a soft-surface pillow. The operations were performed by a team of 3 surgeons who used a standard technique with separation and retraction of the ribs. No ribs were resected or broken. After surgery, the patients arrived at the PACU where the nurses assessed their pain within minutes of arrival by using a visual analog score (VAS 0-10). A patient who complained of shoulder pain with VAS ⱖ4 while incisional pain was estimated as VAS ⱕ3 was included in the study. The first 30 patients were assigned to have a low-volume interscalene brachial plexus block (group 1). The block was performed by an experienced senior anesthesiologist within 10 to 20 minutes.
Journal of Cardiothoracic and Vascular Anesthesia, Vol 21, No 4 (August), 2007: pp 554-557
POST-THORACOTOMY SHOULDER PAIN
555
Table 1. Patients’ Demographic and Medical Data
Age Sex ASA classification Side of surgery Duration of PACU stay Morphine consumption Number of patients required morphine Single patient’s dose Pain at incision Entering PACU POD 1 Patient satisfaction (scale of 1-5)
Group 1: LowVolume Interscalene Brachial Plexus Block
Group 2: Intramuscular Diclofenac
p Value
63 (44-82) 10/20 18/12 16/14 2.9 ⫾ 1.0
56 (18-81) 13/17 16/14 14/16 2.4 ⫾ 0.6
0.56 0.43 0.60 0.85 0.019
15 (50)
23 (77)
0.03
Mg, mean (⫾SD)
3.8 ⫾ 5.4
6.0 ⫾ 5.6
0.07
VAS, mean (⫾SD) VAS mean (⫾SD)
1.5 ⫾ 0.5 1.8 ⫾ 1.2
1.6 ⫾ 0.5 1.5 ⫾ 0.5
0.51 0.77
Mean (⫾SD)
4.5 ⫾ 0.5
4.0 ⫾ 0.7
0.17
Years, median (range) Male/female (n) 2/3 (n) Left/right Hours, mean (⫾SD) Number (%)
Abbreviation: POD 1, first postoperative day.
Under sterile conditions, the interscalene groove was identified, and an interscalene brachial plexus block was performed by using a nerve stimulator in a standard technique,11 modified by using a low volume of local anesthetic (10 mL of bupivacaine 0.5%). The subsequent 30 patients (group 2) had intramuscular injections of diclofenac sodium, 75 mg (Abitren; ABIC Manufacturer, Biogal Pharmaceutical Works Ltd, Debrecen, Hungary). Thirty minutes after the treatment and every 30 minutes thereafter, the PACU nurse reassessed the pain at the shoulder and incisional sites. Rescue opioids (IV morphine in increments of 2 mg) were administered as long as pain was estimated as VAS ⱖ4. Patients from both groups were examined by the attending anesthesiologist, and all had arterial blood gases and chest x-rays 1 hour after admission to the PACU. Complications such as motor block, Horner’s syndrome, and pneumothorax were ruled out before the patient left the PACU. The anesthesiologist and PACU nurse were not blinded to the patients’ intervention. Criteria for discharge from the PACU included 8 to 10 points according to the Aldrete Score,12 pain estimation in all sites as VAS ⱕ3, and no analgesic requirement or administration in the previous 30 minutes. Data collected included patient’s age, sex, American Society of Anesthesiologists score, type and side of surgery, VAS of pain at incision and shoulder on arriving in the PACU, VAS 30 minutes after treatment and at the time of leaving the PACU, duration of PACU stay, use of opioids in the PACU, VAS regarding shoulder pain, and incisional pain on the first postoperative day, satisfaction from the treatment on a 1 to 5 scale, and complications. Statistical analysis was performed with Statistica (StatSoft, Tulsa, OK). Parametric variables were compared by analysis of variance of repeated measures. Intergroup comparison of nonparametric variables used the Kruksal-Wallis analysis of variance and the median test. For dependent nonparametric paired samples, a Wilcoxon matched pairs test was applied. The p value was adjusted by using Fisher least significant difference. Differences were deemed significant at p ⬍ 0.05. RESULTS
During the 8-month period in which the study was conducted, 153 patients underwent thoracic surgery for lobe resection. Nine were excluded because of the chronic use of analgesics. Of the remaining 144 patients, 60 experienced ipsilateral shoulder pain; thus, the incidence of shoulder pain was about 41%. The demographic and medical data of the patients in both groups were similar (Table 1). Patients who had
the low-volume interscalene brachial plexus block stayed in the PACU longer than patients who were treated with intramuscular diclofenac (2.9 ⫾ 1.0 hours and 2.4 ⫾ 0.6 hours, respectively; p ⫽ 0.019). Significantly more patients who were treated with diclofenac required rescue opioids (23 patients, 77%) than those with the low-volume interscalene block (15 patients, 50%; p ⫽ 0.03). No statistically significant difference was found between the groups in the mean dose of rescue opioids of a single patient (p ⫽ 0.07). Shoulder pain is summarized in Figure 1. VAS was found to be significantly lower in the low-volume interscalene block group than in the diclofenac group at 30 minutes after treatment and when leaving the PACU (p ⬍ 0.001 for both). There was no significant difference in the shoulder pain between the groups on postoperative day 1 (group 1: VAS ⫽ 1.6 ⫾ 0.9 and group 2: VAS ⫽ 2.3 ⫾ 1.3, p ⫽ 0.14). Patient satisfaction with pain management was similar in both groups. No complications were found in patients from either group as evident by physical examinations, arterial blood gases, and chest x-rays. DISCUSSION
Ipsilateral shoulder pain is a common complaint in patients undergoing thoracotomy, occurring in up to 68% of patients.7 The cause of pain is unknown, although some explanations have been proposed. Burgess et al5 hypothesized that the pain results from transection of a major bronchus. Other explanations include the lateral decubitus position, pleural irritation caused by the chest drain, or referred pain from the diaphragm.8 The pain is usually described by patients as moderate-to-severe,4,5 similar to the findings in this study; in these circumstances, additional epidural analgesia is ineffective in alleviating the pain.3 Although the cause of the shoulder pain is unidentified, clinicians have searched for an effective treatment. Traditional analgesia with opioids should be given with caution because opioids are respiratory depressants, and these patients have reduced pulmonary reserves and often suffer from chronic respiratory disease. NSAIDs are another option; their favorable effect was reported previously by Perttunen et al13 who used IV
556
BARAK ET AL
Fig 1.
VAS of shoulder pain in groups 1 and 2 during PACU stay and on the first postoperative day. *p < 0.05.
diclofenac in post-thoracotomy pain. Other studies have described complete elimination of shoulder pain in 80% of patients with repeating doses of ketorolac5 and substantial relief in all patients after rectal indomethacin.3 The problem with NSAIDs is their nephrotoxicity in dehydrated patients,14 such as in patients who undergo thoracotomy.15 Acetaminophen was studied recently, given preemptively and regularly after thoracic surgery, with good effect in alleviating shoulder pain.4 Some novel solutions for treatment of post-thoracotomy shoulder pain have been tried as well. Garner and Coats9 suggested a stellate ganglion block as a possible effective treatment, and phrenic nerve infiltration was tried by Scawn et al.8 Suprascapular nerve block7 and intrapleural bupivacaine6 were tried, all with limited success. Ng and Chow10 described a case series of five patients who had an interscalene brachial plexus block as treatment for post-thoracotomy shoulder pain. This study followed on their experience and tried reduced-volume interscalene brachial plexus blocks in a prospective study. Assuming that the pain was at least partially caused by excessive strain on the shoulder joint capsule and ligaments with the patient in the lateral position, it was thought that regional analgesia to the shoulder might induce pain relief. The reduced volume of local anesthetic was chosen to minimize motor block, as was achieved in group 1, and also to decrease the possible side effects of the local anesthetics. The low-volume interscalene brachial plexus block was an effective treatment for post-thoracotomy shoulder pain, reducing VAS notably within 30 minutes, and continuing its analgesic effect for hours. Intramuscular diclofenac injection was less effective, judging by the patient’s VAS. On the first postoperative day, the difference between the groups became sta-
tistically insignificant. Shoulder pain decreased gradually and disappeared the day after surgery. A similar trend was found by Mac et al4 in a study in which the pain was reduced within 20 hours. However, the possibility of chronic shoulder pain after thoracic surgery exists, as Stammberger et al16 reported. In their study of early- and long-term complaints after video-assisted thoracoscopic surgery (173 patients), 1 patient was unable to work 12 weeks after the operation because of shoulder pain. Postoperative shoulder function may improve with early mobilization and effective physiotherapy,17 which are greatly dependent on good pain control. The exact mechanism by which the interscalene block relieves the pain can be explained in more than one way. The brachial plexus block prevents nociceptive impulses of the shoulder region from reaching the spinal cord. However, involvement of the phrenic nerve in the conduction of stimuli from the thoracic viscera cannot be ruled out. Standard-volume interscalene brachial plexus block carries a 100% phrenic nerve blockade as a side effect.11 This may be somewhat lower using the reduced volume of local anesthetics. Nevertheless, phrenic nerve block could explain the results of this study, supporting the hypothesis of Scawn et al8 regarding the influence of phrenic nerve infiltration on shoulder pain. The pharmacologic effect of local anesthetics may be relevant as well. The systemic absorption of bupivacaine has a hypnotic effect that may alter the patient’s pain perception.18 The group of patients treated with diclofenac showed improvement in shoulder pain, as seen in the decline of VAS (Fig 1). The beneficial effect of NSAIDs in treating post-thoracotomy shoulder pain described by Burgess et al5 with ketorolac and in the authors’ earlier study with indomethacin3 supports
POST-THORACOTOMY SHOULDER PAIN
557
the theory that an inflammatory process is involved in the pain occurrence. Interestingly, both groups were satisfied with pain treatment, which is a known outcome in reports of pain treatment.19,20 Patients often report satisfaction with their pain management, regardless of pain intensity or treatment mode. This can be explained by the patients’ expectations of postoperative pain, their little awareness of the efficacy of analgesics, and the tendency to base their response on the sympathy and friendliness of staff rather than on the outcome of pain relief.19,20 The patients in this study were selected in a sequential manner, which could have introduced bias, because the study was completed in 1 group before starting the other. Admittedly, randomization of treatment would have made for a stronger power of results. The study would also have been better with a double-blind saline control. Although the patients’pain was monitored often by the nurses in the PACU and in the ward, data were collected only at specific time points. Moreover, the interscalene block patients stayed longer in the PACU; thus, a time gap of about 30 minutes was created between the time points of the groups. This may give an inaccurate description of the real pain alteration over time. Another problem with the study was the population size, which is characteristic of other
studies in this field.3-10 Larger definitive trials are required to be more certain as to which technique should be used to effectively treat post-thoracotomy shoulder pain. The application of an invasive treatment, such as interscalene block, carries a higher risk than the intramuscular injection of an NSAID. In this study, there were no complications from the brachial plexus block performance. However, this is a relatively small study (regarding sample size) that does not have enough power to allow meaningful comments about complication rate. Nevertheless, complications such as pneumothorax, Horner’s syndrome, transient phrenic nerve paresis with diaphragm dysfunction, and systemic toxicity of local anesthetics should be kept in mind. In particular, phrenic nerve paresis may affect postoperative pulmonary function and recovery in a patient after lung resection. The use of a reduced concentration of local anesthesia may decrease the risk of complications, in addition to the block being conducted by an experienced, vigilant anesthesiologist. Moreover, patients who underwent interscalene block stayed longer in the PACU because the block performance was a time-consuming procedure. Because interscalene block is a higher-risk time-consuming procedure, it is suggested to reserve this treatment for selected patients whose shoulder pain is severe.
REFERENCES 1. Kaplan JA, Miller ED Jr, Gallagher EG Jr: Postoperative analgesia for thoracotomy patients. Anesth Analg 54:773-777, 1975 2. Lubenow TR, Faber LP, McCarthy RJ, et al: Postthoracotomy pain management using continuous epidural analgesia in 1324 patients. Ann Thorac Surg 58:924-929, 1994 3. Barak M, Ziser A, Katz Y: Thoracic epidural local anesthetics are ineffective in alleviating post-thoracotomy ipsilateral shoulder pain. J Cardiothorac Vasc Anesth 18:458-460, 2004 4. Mac TB, Girard F, Chouinard P, et al: Acetaminophen decreases early post-thoracotomy ipsilateral shoulder pain in patients with thoracic epidural analgesia: A double-blind placebo-controlled study. J Cardiothorac Vasc Anesth 19:475-478, 2005 5. Burgess FW, Anderson DM, Colonna D, et al: Ipsilateral shoulder pain following thoracic surgery. Anesthesiology 78:365-368, 1993 6. Pennefather SH, Akrofi ME, Kendall JB, et al: Double-blind comparison of intrapleural saline and 0.25% bupivacaine for ipsilateral shoulder pain after thoracotomy in patients receiving thoracic epidural analgesia. Br J Anaesth 94:234-238, 2005 7. Tan N, Agnew NM, Scawn ND, et al: Suprascapular nerve block for ipsilateral shoulder pain after thoracotomy with thoracic epidural analgesia: A double-blind comparison of 0.5% bupivacaine and 0.9% saline. Anesth Analg 94:199-202, 2002 8. Scawn ND, Pennefather SH, Soorae A, et al: Ipsilateral shoulder pain after thoracotomy with epidural analgesia: The influence of phrenic nerve infiltration with lidocaine. Anesth Analg 93:260-264, 2001 9. Garner L, Coats RR: Ipsilateral stellate ganglion block effective for treating shoulder pain after thoracotomy. Anesth Analg 78:11951196, 1994
10. Ng KP, Chow YF: Brachial plexus block for ipsilateral shoulder pain after thoracotomy. Anaesth Intensive Care 25:74-76, 1997 11. Wedel DJ and Horlocker TT: Nerve blocks, in Miller RD (ed): Miller’s Anesthesia (ed 6). Philadelphia, PA, Elsevier, 2005, pp 16851717 12. Aldrete JA, Kroulik D: A postanesthetic recovery score. Anesth Analg 49:924-934, 1970 13. Perttunen K, Kalso E, Heinonen J, et al: IV diclofenac in post-thoracotomy pain. Br J Anaesth 68:474-480, 1992 14. Whelton A: Nephrotoxicity of nonsteroidal anti-inflammatory drugs: Physiologic foundations and clinical implications. Am J Med 106:13s-24s, 1999 15. Brodsky JB, Fitzmaurice B: Modern anesthetic techniques for thoracic operations. World J Surg 25:162-166, 2001 16. Stammberger U, Steinacher C, Hillinger S, et al: Early and long-term complaints following video-assisted thoracoscopic surgery: evaluation in 173 patients. Eur J Cardiothorac Surg 18:7-11, 2000 17. Li WW, Lee TW, Yim AP: Shoulder function after thoracic surgery. Thorac Surg Clin 14:331-343, 2004 18. Ben-Shlomo I, Tverskoy M, Fleyshman G, et al: Hypnotic effect of i.v. propofol is enhanced by IM administration of either lignocaine or bupivacaine. Br J Anaesth 78:375-377, 1997 19. Apfelbaum JL, Chen C, Mehta SS, et al: Postoperative pain experience: Results from a national survey suggest postoperative pain continues to be undermanaged. Anesth Analg 97:534-540, 2003 20. McNeill JA, Sherwood GD, Starck PL, et al: Assessing clinical outcomes: Patient satisfaction with pain management. J Pain Symptom Manage 16:29-40, 1998