Secondary Logo

Journal Logo

Subpleural block is less effective than thoracic epidural analgesia for post-thoracotomy pain: a randomised controlled study

Kanazi, Ghassan E.; Ayoub, Chakib M.; Aouad, Marie; Abdallah, Faraj; Sfeir, Pierre M.; Adham, Almoataz-Billah F.; El-Khatib, Mohamad F.

European Journal of Anaesthesiology: April 2012 - Volume 29 - Issue 4 - p 186–191
doi: 10.1097/EJA.0b013e32834fcef7

Context Thoracic epidural and paravertebral blocks provide adequate analgesia for postoperative thoracotomy pain. Both procedures are usually performed percutaneously with considerable failure rates. A subpleural catheter placed in the space posterior to the parietal pleura and alongside the paravertebral area may provide superior postoperative pain relief.

Objective To compare subpleural analgesia with thoracic epidural analgesia in patients undergoing thoracotomy.

Design Randomised, double-blind study.

Setting A tertiary care University Medical Centre between 26 June 2008 and 21 March 2011.

Patients Forty-two patients scheduled for elective posterolateral thoracotomy. Patients with American Society of Anesthesiologists physical status ≥4, with a previous history of thoracotomy, on chronic pain medications or with a contraindication to receiving local anaesthetics or thoracic epidural block were excluded from the study.

Interventions Patients were randomised to receive either subpleural analgesia or thoracic epidural analgesia for 24-h post-thoracotomy pain control.

Main outcome measures A visual analogue scale was used to assess pain at rest and on coughing during the first 24 h postoperatively and the incidence of hypotension was recorded.

Results Patients who received subpleural analgesia had higher visual analogue scores at rest and on coughing than those who received thoracic epidural analgesia. Seven patients who started with subpleural analgesia were treated with thoracic epidural analgesia at a mean (SD) of 3.9 (4.8) h. The remaining 14 patients had a median (IQR [range]) visual analogue score of 5 cm (4–5 [3–6]) at rest and were maintained on subpleural analgesia until the end of the study. The visual analogue score at rest was <7 cm in all 21 patients who received thoracic epidural analgesia and none was switched to subpleural analgesia during the study. None of the patients in the subpleural analgesia group experienced hypotension compared with five of the 21 patients in the thoracic epidural analgesia group (P = 0.047).

Conclusion Thoracic epidural analgesia is superior to subpleural analgesia in relieving post-thoracotomy pain.

From the Department of Anesthesiology, Medical Center, American University of Beirut, Beirut, Lebanon (GEK, CMA, MA, MFE), Department of Anesthesiology, Toronto Western Hospital, University Health Network, Toronto, Canada (FA), Department of Surgery, Medical Center, American University of Beirut, Beirut, Lebanon (PMS) and Department of Critical Care Medicine, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Canada (ABFA)

Correspondence to Dr Mohamad F. El-Khatib, Department of Anesthesiology, Medical Center, American University of Beirut, Hamra Street, P.O. Box 11–0236, Beirut 1107–2020, Lebanon E-mail:

Published online 14 February 2012

Back to Top | Article Outline


During posterolateral thoracotomy, the pain resulting from damage to ribs, muscles and peripheral nerves is severe and intense.1 The pain can prevent deep breathing and effective coughing, which are essential in the postoperative period to restore lung function and reduce pulmonary complications.2 Pain control is a key element in attenuating the stress response and decreasing morbidity after major thoracic surgery.1,3

Thoracic epidural analgesia (TEA) has long been considered as the mainstay of the management of postoperative thoracotomy pain4 despite morbidities such as dural puncture, hypotension and urinary retention.5,6 Paravertebral analgesia emerged as a comparable analgesic technique to TEA for patients undergoing thoracotomy with the advantage of reducing the incidence of hypotension seen with TEA, but with the possibility of neurological complications and inadvertent pleural puncture.7 The percutaneous technique for insertion of thoracic epidural catheters and paravertebral blocks is associated with an incidence of failure.7 Insertion of a paravertebral catheter under direct vision has been described and its use provides comparable analgesia to percutaneous paravertebral and TEA.8 To our knowledge, the analgesic effect of subpleural analgesia via a catheter placed under direct vision in the space posterior to the parietal pleura and alongside the paravertebral area has not been evaluated in the context of posterolateral post-thoracotomy pain.

The aim of the current study was to compare subpleural analgesia with TEA in patients undergoing posterolateral thoracotomy.

Back to Top | Article Outline


Ethical approval for this study (Ethical Committee number ANES.GK.03) was provided by the Institutional Review Board of the American University of Beirut (Chairperson Professor I. Salti) on 1 April 2008. Written informed consent was obtained from each patient before inclusion in the study.

Forty-two patients scheduled for elective posterolateral thoracotomy for lung cancer between 26 June 2008 and 21 March 2011 were included in the study. Patients with American Society of Anesthesiologists status ≥4, with a previous history of thoracotomy, on chronic pain medication or with a contraindication to receiving local anaesthetics or thoracic epidural block were excluded from the study. All surgical procedures were performed by the same surgeon through a posterolateral thoracotomy incision through the fifth or sixth rib.

On arrival in the operating room and with patient in the sitting position, a thoracic epidural catheter was inserted by an experienced anaesthesiologist in the thoracic epidural space at the T5-T7 level using a 16-gauge Tuohy needle (B. Braun, Melsungen AG, Melsungen, Germany) with a loss of resistance technique under sterile conditions. The catheter was advanced 4–5 cm inside the epidural space and a test dose of 3 ml lidocaine 2% with epinephrine 1 : 200 000 was given to exclude misplacement of the catheter.

All patients received midazolam 2 mg and glycopyrronium 0.2 mg as premedication. Anaesthesia was induced with lidocaine 1 mg kg−1, propofol 2 mg kg−1 and fentanyl 3 μg kg−1. Rocuronium 0.6 mg kg−1 was given to facilitate tracheal intubation with a left-sided double lumen tracheal tube (Mallinckrodt Medical, Athlone, Ireland) and the position of the tube was confirmed with a fibrescope. Anaesthesia was maintained with sevoflurane in a mixture of oxygen and air, supplemented by incremental doses of fentanyl and rocuronium. All patients were monitored with two-lead ECG (leads II and V5) for heart rate and ST segment changes, with non-invasive pulse oximetry (SpO2), oesophageal temperature and an arterial catheter for invasive blood pressure monitoring and blood gas analysis. Airway pressures, ventilation parameters, inspired oxygen concentration (FiO2), expired end-tidal carbon dioxide concentration (EtCO2) and end-tidal sevoflurane concentration were also monitored.

Before the surgical wound was closed, the parietal pleura was raised bluntly from the posterior chest wall down to the vertebral body three intercostal spaces above the thoracotomy incision. An 18-gauge epidural catheter (B. Braun) was advanced into the space at the level of the neck of the ribs and laid on the endothoracic fascia under direct vision. The catheter was secured with 4–0 Prolene sutures to maintain its position during lung expansion and extruded through the chest wall.

After insertion of the epidural and subpleural catheters in all patients, they were randomised to one of two groups using a computer-generated series of random numbers. Patients randomised to start with TEA received 10 ml of 0.125% bupivacaine with 5 μg ml−1 of adrenaline and patients randomised to start with subpleural analgesia received a loading dose of 20 ml of 0.25% bupivacaine with 5 μg ml−1 of adrenaline via the subpleural catheter. The doses of bupivacaine were based on the clinically used concentrations for epidural and subpleural analgesia rather than equipotency.8,9 After receiving the bolus dose, a continuous infusion of 0.125% bupivacaine 8 ml h−1 was started through the appropriate catheter and maintained for 24 h postoperatively. In all patients, the trachea was extubated at the end of surgery.

Postoperatively, patients were assessed by a blinded observer every 15 min for the first 2 h and then every 6 h for the next 24 h. The function of this observer was limited to data collection. The following variables were monitored throughout the study: heart rate, blood pressure, respiratory rate and oxygen saturation. Hypotension was defined as a decrease in blood pressure of more than 25% from baseline. Pain intensity was measured at rest (VASR) and on coughing (VASC) using a visual analogue scale for pain (VAS) in which 0 cm is no pain and 10 cm is the worst pain possible. Sedation was evaluated using a sedation score in which 0 = alert; 1 = occasionally drowsy, easy to arouse; 2 = frequently drowsy, easy to arouse; and 3 = somnolent, difficult to arouse. Nausea/vomiting was evaluated as 0 = absent, 1 = mild nausea and 2 = severe nausea and/or vomiting.

A VASR score of 4 cm or less was considered to be an acceptable level of pain. If the VASR score was more than 4 cm but less than 7 cm, patients in the TEA group received a bolus dose of 10 ml of 0.125% bupivacaine; patients in the subpleural analgesia group received a bolus dose of 20 ml of 0.25% bupivacaine. The infusion rate in both groups was increased in 2-ml increments following each intervention to a maximum of 12 ml h−1 to maintain the VASR at less than 7 cm. Patients were maintained on either TEA or subpleural analgesia as long as the VASR was less than 7 cm. If the VASR score was ≥7 cm, the patient was transferred to the other analgesic technique and if either technique failed to achieve adequate pain relief with (VASR at least 7 cm), patient-controlled morphine analgesia was started. All changes were made by the pain service team, whose members were not blinded to the type of analgesia which each patient received.

All patients received intravenous paracetamol 1 g every 6 h as a standing dose, and intravenous ondansetron 4 mg for nausea or vomiting if required. The study was terminated 24 h after the time of randomisation.

Back to Top | Article Outline

Power analysis and statistics

A power analysis indicated that 21 patients were needed in each group. The sample size determination was based on a power of 80%, a type I error of 5%, a clinically significant change in the VASR value of 2 cm and a VASR standard deviation of 2 cm. The secondary outcomes were haemodynamic variables, sedation score and nausea/vomiting score. Mean and SD were calculated for continuous data. Median (IQR [range]) for VAS values were computed and the analysis of variance (ANOVA) (for continuous data), the Fisher's exact test (for categorical data) and the nonparametric Mann–Whitney U-test were used as appropriate for statistical analysis. Data analysis was based on an intention-to-treat basis. The level of significance was considered at P value less than 0.05.

Back to Top | Article Outline


Forty-two patients were included in the study (Fig. 1). Twenty-one patients started with TEA and the remainder started with subpleural analgesia. Patients’ characteristics were similar in both groups [subpleural analgesia vs. TEA: age (years), 60 (12) vs. 58 (18); sex (male/female), 15/6 vs. 14/7; weight (kg), 85 (18) vs. 80 (19); height (cm), 172 (7) vs. 169 (9); total intraoperative fentanyl (μg), 838 (42) vs. 822 (33); time between induction and emergence from anaesthesia (min), 260 (70) vs. 261 (95); intraoperative fluid input (ml), 125 (19) vs. 132 (21); intraoperative blood loss (ml), 1483 (111) vs. 1503 (137)]. No patient withdrew and there was no malfunctioning or blockages of the catheter in either group.



Patients with subpleural analgesia had higher VASR (Fig. 2) and VASC (Fig. 3) at all intervals in comparison to those who started with TEA. Seven of those patients (33%) were converted to TEA at 3.9 (4.8) h because of an inability to maintain the VASR score <7 cm (VASR 7) (7–8 [7–8]). After the switch, the VASR decreased significantly to 2 (2–3 [2–3]) (P = 0.016). Similarly, the VASC decreased significantly from 8 (8–9 [7–10]) to 6 (5–7 [5–7]) (P = 0.014). The remaining 14 patients (67%) in the subpleural analgesia group had a VASR of 5 (4–5 [3–6]) and were maintained on subpleural analgesia until the end of the study.





The VASR was <7 cm in all 21 patients who started with TEA and none were changed to subpleural analgesia during the study.

The incidence of hypotension within the first 6 h was higher in the TEA group (five of 21) than in the subpleural analgesia group (none) (P = 0.047) despite comparable intraoperative fluid intake and blood loss in the two groups.

Sedation and nausea and vomiting scores were similar in both groups and the arterial oxygen saturation was consistently above 97% in all patients while receiving oxygen 3–5 l min−1 via a simple oxygen face mask.

Back to Top | Article Outline


Our current findings indicate that TEA provides better pain relief at rest and during coughing than subpleural analgesia in patients undergoing posterolateral thoracotomy despite using a dose of local anaesthetic for subpleural analgesia four times greater than for TEA. The pain scores at rest and during coughing were significantly lower with TEA compared with subpleural analgesia during the first 24 h postoperatively. Subpleural analgesia failed to provide adequate postoperative analgesia in 33% of patients.

Our findings with TEA are similar to those reported in the literature.7,8,10,11 Bachmann-Mennenga et al.,10 Von Dossow et al.11 and Davies et al.7 showed that VAS at rest in the presence of thoracic epidural ranged between 1 and 6 cm which is comparable with the range of VASR in the current study. Furthermore, in a recent meta-analysis, Davies et al.7 showed that the incidence of hypotension was 19% during TEA which is comparable with the 24% incidence of hypotension reported in the current study.

Subpleural analgesia was successful in providing adequate pain control in 14 of 21 patients. This might be explained by the spread of local anaesthetic into the extrapleural space and its diffusion into the paravertebral space.12 Injection of radio-opaque material through the subpleural catheter revealed a localised and longitudinal spread of contrast into the extrapleural area (Fig. 4) with no evidence of interpleural spread.13 The failure to provide adequate post-thoracotomy pain relief in the remaining seven of the patients receiving subpleural analgesia could be attributed to dislodgement of the subpleural catheter or inadequate and limited diffusion of local anaesthetic into the paravertebral space. Murphy showed that India ink injected via the intercostal space can spread posteriorly to the subpleural space.13 Conacher and Korki14 showed that contrast material injected into the paravertebral space remains localised and spreads up and down the vertebral space. The subpleural space is separated from the paravertebral space by the endothoracic or extrapleural fascia,15,16 and the presence of the extrapleural fascia may have prevented adequate diffusion of local anaesthetic to the nerve endings,17 as evidenced by the higher VASR in the 14 patients in whom VASR was maintained at <7 cm compared with the TEA patients. This is further acknowledged by Sabanathan et al.,18 who described the approach for the surgical insertion of a paravertebral catheter. They stressed the importance of interrupting the extrapleural fascia and the insertion of the catheter under direct vision into the paravertebral space.



The incidence of hypotension was lower in the SPA group than in the TEA group. This finding is in agreement with the results of Davies et al.,7 who reported that the incidence of hypotension was higher with thoracic epidural than with paravertebral analgesia. There were no differences between techniques in the incidence of nausea and vomiting or sedation levels.

In the current study, the pain scores during coughing continued to be higher than those at rest in all patients and at all times whether on TEA or subpleural analgesia. However, the coughing VAS scores were always lower with TEA than with subpleural analgesia. Boisseau et al.19 showed that decreased VAS scores during coughing did not necessarily decreased atelectasis after thoracotomy. Consequently, VASR rather than VASC was used as the criterion for changing treatment between the two analgesic techniques.

Our study did not evaluate postoperative pulmonary complications through the use of pulmonary function tests because of the inability of our patients to perform pulmonary function testing within the first 24 h postoperatively. However, our patients were monitored continuously with pulse oximetry and none showed any clinically significant decrease in oxygen saturation.

To our knowledge, there are no published data which have identified equipotent doses for bupivacaine when given for TEA and subpleural analgesia. Previous studies8,9 reported doses in a ratio of TEA-to-subpleural analgesia of 1 : 2 and, in our study, the doses and concentrations of bupivacaine for the thoracic epidural and subpleural analgesia techniques were based on the concentrations used clinically for epidural and subpleural analgesia rather than equipotency. A limitation of the current study is that we did not try higher analgesic concentrations and volumes with subpleural analgesia.

Back to Top | Article Outline


Our study demonstrates that TEA is better than subpleural analgesia in providing pain relief at rest and during coughing for control of post-thoracotomy pain. This may be attributed to the limited diffusion of the local analgesic to the paravertebral space with the subpleural analgesia technique. After open thoracotomy, subpleural analgesia was successful in providing adequate pain control in two-thirds of the patients. Whether increasing the volume or the concentration in the subpleural analgesia group would further improve analgesia could be the subject for future studies.

Back to Top | Article Outline


Assistance with study: none.

This work was supported by the Department of Anesthesiology, American University of Beirut, Beirut, Lebanon.

Conflict of interest: none.

Back to Top | Article Outline


1. Peeters-Asdourian C, Gupta S. Choices in pain management following thoracotomy. Chest 1999; 115:122–124S.
2. Warner D. Preventing postoperative pulmonary complications. Anesthesiology 2000; 92:1467–1472.
3. Richardson J, Sabanathan S, Shah R. Postthoracotomy spirometric lung function: the effect of analgesia. J Cardiovasc Surg 1999; 40:445–456.
4. Lubenow T, Faber L, McCarthy R, et al. Postthoracotomy pain management using continuous epidural analgesia in 1 324 patients. Ann Thor Surg 1994; 58:924–930.
5. Savage C, McQuitty C, Wang D, Zwischenberger J. Postthoracotomy pain management. Chest Surg Clin N Am 2002; 12:251–263.
6. Powell E, Cook D, Pearce A, et al. A prospective, multicenter, observational cohort study of analgesia and outcome after pneumonectomy. Br J Anaesth 2011; 106:364–370.
7. Davies RG, Myles PS, Graham JM. A comparison of the analgesic efficacy and side-effects of paravertebral vs epidural blockade for thoracotomy: a systematic review and meta-analysis of randomized trials. Br J Anaesth 2006; 96:418–426.
8. Richardson J, Sabanathan S, Jones J, et al. A prospective, randomized comparison of preoperative and continuous balanced epidural or paravertebral bupivacaine on postthoracotomy pain, pulmonary function and stress responses. Br J Anaesth 1999; 83:387–392.
9. Joshi G, Bonnet F, Shah R, et al. A systematic review of randomized trials evaluating regional techniques for postthoracotomy analgesia. Anesth Analg 2008; 107:1026–1040.
10. Bachmann-Mennenga B, Biscoping J, Kuhn DF, et al. Intercostal nerve block, interpleural analgesia, thoracic epidural block or systemic opioid application for pain relief after thoracotomy? Eur J Cardiothorac Surg 1993; 7:12–18.
11. Von Dossow V, Welte M, Zaune U, et al. Thoracic epidural anesthesia combined with general anesthesia: the preferred anesthetic technique for thoracic surgery. Anesth Analg 2001; 92:848–854.
12. Richardson J, Lönnqvist PA, Naja Z. Bilateral thoracic paravertebral block: potential and practice. Br J Anaesth 2011; 106:164–171.
13. Murphy D. Continuous intercostal nerve blockade: anatomical study to elucidate its mode of action. Br J Anaesth 1984; 56:627–630.
14. Conacher ID, Korki M. Postoperative paravertebral blocks for thoracic surgery. A radiological appraisal. Br J Anaesth 1987; 59:155–161.
15. Richardson J, Lönnqvist PA. Thoracic paravertebral block. Br J Anaesth 1998; 81:230–238.
16. Karmakar M. Thoracic paravertebral block. Anesthesiology 2001; 95:771–780.
17. McKenzie AG, Mathe S. Interpleural local anaesthesia. Anatomical basis for mechanism of action. Br J Anaesth 1996; 76:297–299.
18. Sabanathan S, Bickford Smith P, Narsing P, et al. Continuous intercostal nerve block for pain relief after thoracotomy. Ann Thorac Surg 1988; 46:425–426.
19. Boisseau N, Rabary O, Padovani B, et al. Improvement of ‘dynamic analgesia’ does not decrease atelectasis after thoracotomy. Br J Anaesth 2001; 87:546–549.

analgesia; epidural; subpleural; thoracotomy

© 2012 European Society of Anaesthesiology