Laparoscopic cholecystectomy has to a large extent replaced traditional cholecystectomy as the surgical method of choice for biliary surgery. Laparoscopic cholecystectomy has been claimed to be associated with reduced morbidity compared to the traditional open surgical approach , and has also been found to result in less postoperative respiratory dysfunction , a better cosmetic result  and, perhaps most importantly, a shorter hospital stay . Nevertheless, even if laparoscopic cholecystectomy is less painful than open cholecystectomy, pain remains an issue in the postoperative period [5-7].
A number of different regional anaesthetic techniques have been attempted in order to improve postoperative analgesia following laparoscopic cholecystectomy. However, neither intraperitoneal instillation of bupivacaine alone, nor when combined with local infiltration of the abdominal incisions, has been able to achieve appropriate postoperative pain relief [8-12].
The use of paravertebral blockade has previously been shown to improve postoperative analgesia and to reduce the hormonal stress response following open cholecystectomy [13,14]. This method has also been shown to preserve lung function and to provide superior pain relief after thoracotomy compared to other regional anaesthetic techniques . A modified paravertebral blockade technique, including the use of a nerve stimulator, has been used at our institution since 1998 for different types of surgery and has been found capable of producing excellent and surprisingly long-lasting postoperative pain relief [16-19].
The objective of the present study was, thus, to compare a bilateral nerve-stimulator guided paravertebral blockade (T5-6) combined with general anaesthesia with general anaesthesia alone for laparoscopic cholecystectomy, using a prospective, randomized study design. The primary outcome parameter was postoperative pain as assessed by visual analogue scale (VAS). Consumption of supplemental analgesics was a secondary outcome measure.
Following Ethics Committee approval and informed written consent, 60 adult patients, ASA I-III, scheduled for laparoscopic cholecystectomy were prospectively assigned to receive either a combined technique using a general anaesthesia and a bilateral paravertebral block (active) or a general anaesthesia alone (control). Patients were assigned to either group using sealed envelopes containing randomly generated instructions to perform or not to perform the paravertebral block prior to anaesthesia. Surgeons were kept blinded to the patient's group and did not enter the operating room until the patient had received a general anaesthetic alone or the combined technique. Authors were not involved in data collection; nurses who were blinded to the patient's group collected the data. The study was conducted at our institution from December 2001 to April 2002.
The primary study end-point was postoperative pain intensity assessed by VAS pain scores, and registered at predetermined time points, both at rest and on attempted movement. Secondary outcome measures were the consumption of supplemental analgesics, the number of patients who were free from postoperative nausea and vomiting, and duration of hospital stay. Patient characteristics and clinical factors were also noted.
General anaesthesia technique
General anaesthesia was induced with intravenous (i.v.) fentanyl (1.5 μg kg−1) and thiopental (3-5 mg kg−1) followed by endotracheal intubation (ETI), facilitated by atracurium (0.5 mg kg−1). Anaesthesia was subsequently maintained with isoflurane 1-3%, fentanyl (3-4 μg kg−1), nitrous oxide 70% and oxygen 30%. The isoflurane concentration was adjusted with the intention of keeping heart rate (HR) and arterial pressure within ±25% of pre-induction values. At the end of the operation, residual neuromuscular blockade was antagonized with neostigmine (0.05 mg kg−1) and atropine (0.01 mg kg−1).
Paravertebral nerve block technique
Prior to induction of anaesthesia a bilateral paravertebral block was performed at the T5-6 level with the patient in the lateral left decubitus position. Two sites of injection were marked 2.5-3 cm lateral to the midline on both sides (Fig. 1).
Following aseptic preparation of the skin, the injection sites was infiltrated with 0.5-1.0 mL of lidocaine 1%, using a 29-G needle. A nerve stimulator (Stimuplex®; B. Braun AG, Melsungen, Germany) was used to identify a muscular response appropriate for the T5-6 level. A 10 cm 21-G insulated needle (Stimuplex®) was introduced perpendicularly to the skin at the site of the upper injection point using the following nerve stimulator settings: 5 mA, 9 V and 1 Hz. Initially contractions of the paraspinal muscles were seen as a result of direct muscle stimulation. As the tip of the needle reached the costo-transverse ligament the contractions of the paraspinal muscles ceased. When the needle subsequently entered the paravertebral space, contractions of the ipsilateral rectus abdominal muscle was seen. The depth between the skin and the paravertebral space was usually between 5 and 7 cm, depending on the weight and height of the patient.
After the paravertebral space had been entered, the stimulating needle was gently manipulated into a position to allow a muscular response while the stimulating current was reduced to 0.4-0.8 mA. The manipulation of the needle tip within the paravertebral space is not an in-out movement; it is rather an angular manipulation and circumferential rotation around the axis of the needle in order to orient the tip of the needle into close proximity of the corresponding nerve root/intercostal nerve. At this point 0.30 mL kg−1 of a mixture of local anaesthetic-clonidine-fentanyl was injected. Each 20 mL of the injected solution contained lidocaine 2% 6 mL, lidocaine 2% 6 mL with epinephrine 1/200 000, bupivacaine 0.5% 5 mL, fentanyl 1 mL (50 μg mL−1) and clonidine 2 mL (150 μg mL−1). As previously described the use of this mixture has been found to produce surprisingly long-lasting post-operative pain relief [17-19].
To block the lower side a similar approach was used with one exception. When the needle was inserted, an oblique downward angle of 45° in relation to the spine was used (Fig. 1). The depth from the skin to the paravertebral space was usually 0.5-1.5 cm less than the corresponding distance for the upper injection site.
The time taken to position the patient, to perform aseptic preparation of the injection site and to perform both paravertebral injections normally ranges between 10 and 15 min.
Postoperative pain assessment
Pain was assessed during the first three postoperative days at predetermined time intervals (6, 12, 24, 36, 48 and 72 h). Pain scores were always performed before giving the patient the supplemental analgesics. A VAS (0: no pain and 10: worst possible pain) was used during the hospital stay. Patients discharged from the hospital before 72 h were contacted daily by telephone to assess their postoperative pain. VAS pain scores were obtained both at rest and during attempted activity. Pain during activity was assessed by asking the patient (1) to move from the supine to the sitting position, (2) to walk approximately 5 m inside the room and (3) to take a deep breath and to cough. The type of pain experienced by the patient whether abdominal, shoulder or both was also recorded.
During the first 12 postoperative hours, intramuscular meperidine (1 mg kg−1; Rhône-Poulenc Rorer, Specia, France) was administered if the VAS pain score was ≥4. After the first 12 postoperative hours, if VAS scores were ≥4, oral dextropropoxyphene (two tablets every 6 h and maximum eight tablets per day) (Hoechst, Mirian, Roussel, France) was administered as the primary analgesic. If pain relief was not achieved (VAS > 4) within 30 min, intramuscular meperidine was subsequently given.
Postoperative nausea and vomiting
The number of patients free of nausea or vomiting was recorded at the same predetermined time intervals as used for the pain assessments (see above). No distinction was made between vomiting and retching. The degree of nausea was measured by the use of a verbal rating scale (0: no nausea and 10: worst possible nausea). Patients experiencing either nausea or vomiting were given metoclopramide (Synthelabo, Mersaco, France) 10 mg i.v.
Discharge from hospital
The discharge decision was entirely made by the surgeon in charge of the patient according to established clinical routine. Thus, none of the investigators were involved with, or could influence this decision-making.
Data are reported as mean, standard deviation (SD), median (range) or percentage. Data were statistically analysed by contingency tables, U-test or t-test as appropriate. Two-way analysis of variance for repeated measurements was used to compare pain scores within the follow-up period between the two groups. P < 0.05 were considered significant.
In order to detect a 25% reduction of the primary end-point of the study (VAS score on activity 6 h post-operatively) with an α and β value of 0.05 and 0.10, respectively, a sample size of 30 patients in each study group was necessary.
Sixty patients were included in the study, 30 patients received the combined technique and 30 patients were controls. No patient was withdrawn from the study.
Patients' characteristics and preoperative data are presented in Table 1. No significant differences were found between the groups with regard to age, weight, height, concomitant pathology or body mass index (BMI).
All patients were haemodynamically stable during surgery (data not shown) and the duration of the surgical intervention was similar in both groups (Table 2). The duration of recovery room stay did not differ between the groups (Table 2). Both groups were also found to be similar with regard to time to first oral intake, time to first micturition and duration of hospital stay (Table 2).
The two-way analysis of variance for repeated measurements of pain scores within the follow-up period indicated that there was a significant difference between the two groups (P < 0.05). At each observed interval time, VAS pain scores, both at rest and during activity, were found to be lower in the active compared to the control group (P < 0.05) (Table 3).
The number of patients who recorded VAS ≤3 at rest, during movement, on coughing and on walking at the different time intervals are presented in Table 4.
With regard to the type of postoperative pain experienced by the patients, subjects in the control group were found to have greater problems both with abdominal pain and shoulder pain compared to patients in the active group (Table 5).
The need for supplemental postoperative analgesia was also found to be higher in the control group compared to the active group during the first 36 postoperative hours (Fig. 2).
No significant difference with regard to vomiting could be observed during the postoperative observation period between the two groups (Table 2). The number of nausea-free patients were found to be significantly higher in the active group compared to the control at 6 h (77% vs. 30%) and at 12 h (97% vs. 63%) (P < 0.05) (Table 2).
The main finding of the present prospective, randomized study was a significant improvement of postoperative pain relief following laparoscopic cholecystectomy in patients treated with general anaesthesia and bilateral paravertebral nerve blockade compared to patients who received general anaesthesia alone. Postoperative pain was found to be reduced during the entire 72 h postoperative observation period, both with regard to pain at rest and pain on activity, and the incidence of both abdominal pain and shoulder pain was reduced in patients treated with paravertebral blocks. Furthermore, a higher proportion of patients receiving paravertebral block were free from nausea during the first 12 postoperative hours.
Despite the various advantages associated with laparoscopic cholecystectomy, pain remains an issue in the postoperative period . VAS scores for patients receiving only an general anaesthetic were very similar to those recorded by Tsertelli and colleagues, and O'Boyle and colleagues [6,7]. Treatment attempts with intra-abdominal instillation of local anaesthetics alone [18,19] or combined with local infiltration at the laparoscopic access sites [9,20] have not resulted in satisfactory postoperative analgesia [8,11,12]. The reason for the lack of efficacy for these treatments is probably multi-factorial but two possible explanations can be speculated. First, the duration of postoperative analgesia following infiltration of local anaesthetics has been reported to be limited [21,22] and systematic reviews have questioned the efficacy of this treatment modality for postoperative analgesia . A limited duration of action can also be speculated for intra-abdominal instillation of local anaesthetics since the surface area of the peritoneum is large and rapid systemic absorption of the local anaesthetic can, thus, potentially limit the duration and efficacy of this treatment . Second, cholecystectomy is associated with a substantial component of visceral pain and the incidence of shoulder pain probably reflects a component of diaphragmatic referred pain . Thus, in order to provide adequate pain relief, both high quality afferent somatic and visceral pain blockade are most likely necessary in order to successfully treat cholecystectomy pain.
Contrary to intra-abdominal instillation of local anaesthetics and local infiltration, paravertebral blockade has been reported to provide high quality afferent blockade with abolishment of somatosensory evoked potentials  and has also been found capable of attenuating the postoperative stress response associated with traditional cholecystectomy . A combined somatic and visceral analgesic effect has also been described for intrapleural anaesthesia , a nerve blocking technique that to a large extent exerts its effect by diffusion of local anaesthetics into the paravertebral space.
Adequate somatic and visceral blockade will require that the bilateral injections at T5-6 are able to spread to multiple adjacent thoracic segments. Although not specifically studied in the present study, such spread between segments has been verified in both patient and cadaver studies [27,28] and single level injections of local anaesthetics have been also found to result in multi-segment anaesthesia [29,30]. We have previously reported that a nerve-stimulator guided technique appears to improve the accuracy and success rate of the paravertebral technique , a finding that is further underscored by the fact that no block failures were noted in the present study. Thus, it can be speculated that the significantly improved pain relief found in patients treated with bilateral vertebral block might reflect a dense afferent blockade of both somatic and visceral pain pathways.
Furthermore, the duration of improved postoperative analgesia was extended well beyond the expected duration of the local anaesthetic component of the mixture used to accomplish the block in the present study. This is in keeping with our previously published experiences with this particular anaesthetic mixture, where surprisingly long duration of analgesia has been found after different types of surgery [16-19]. Addition of clonidine to local anaesthetic solutions has been found capable of prolonging the duration of peripheral nerve blocks . However, the duration of postoperative analgesia observed in the present study would be more consistent with a pre-emptive analgesic effect. Support for a pre-emptive analgesic action of a local anaesthetic-clonidine mixture has recently been reported after tonsillectomy in children. The use of local infiltration of the tonsillar fossae prior to surgery with a ropivacaine-clonidine mixture was found to result in improved analgesia both at rest and on swallowing during a 5 day post-operative observation period compared to ropivacaine alone or placebo . Injection of morphine accurately around the nerves has been shown to improve pain relief for a long time [33,34].
Different studies showed that opioid consumption might cause nausea . Hence the significant reduction of opioid consumption in the active group may explain the greater number of patients free of nausea. Another factor that may play a role in reducing the nausea incidence is the use of clonidine .
Based on our previously reported risk rate of pneumothorax following paravertebral blockade, Tighe indicated a lower risk rate for pneumothorax and accidental pleural puncture equal to 0.125% and 0.2% per single injection, respectively . However, in this current study none of these complications were observed but one minor complication is haematoma at site of injection. We also wish to mention that no motor nerve function was compromised during the present study.
Despite the fact that patients who received a general anaesthetic with a paravertebral block experienced significantly better postoperative pain relief and a reduced problem with nausea and vomiting, this could not be converted into a shorter duration of hospital stay. Hospital discharge was not shortened due to other factors, such as surgery difficulties and pre-existing pathology that influenced the surgeons' decision to discharge their patients. However, despite the lack of effect on hospital stay we are convinced, based on the results of the present study, that the combined technique deserves more widespread use in this patient category due to its superior effect on postoperative pain.
1. Luman W, Adams WH, Nixon SN, et al.
Incidence of persistent symptoms after laparoscopic cholecystectomy: a prospective study. Gut
2. Frazee RC, Roberts JW, Okeson GC, et al.
Open versus laparoscopic cholecystectomy. A comparison of postoperative pulmonary function. Ann Surg
651-653; discussion 653-654.
3. Wastell C. Laparoscopic cholecystectomy. BMJ
4. Grace PA, Quereshi A, Coleman J, et al.
Reduced post-operative hospitalization after laparoscopic cholecystectomy. Br J Surg
5. Joris J, Cigarini I, Legrand M, et al.
Metabolic and respiratory changes after cholecystectomy performed via laparotomy or laparoscopy. Br J Anaesth
6. O'Boyle CJ, deBeaux AC, Watson DI, et al.
Helium vs carbon dioxide gas insufflation with or without saline lavage during laparoscopy. Surg Endosc
7. Tsereteli Z, Terry ML, Bowers SP, et al.
Prospective randomized clinical trial comparing nitrous oxide and carbon dioxide pneumoperitoneum for laparoscopic surgery. J Am Coll Surg
8. Joris J, Thiry E, Paris P, Weerts J, Lamy M. Pain after laparoscopic cholecystectomy: characteristics and effect of intraperitoneal bupivacaine. Anesth Analg
9. Bisgaard T, Klarskov B, Kristiansen VB, et al.
Multi-regional local anesthetic infiltration during laparoscopic cholecystectomy in patients receiving prophylactic multimodal analgesia: a randomized, double-blinded, placebo-controlled study. Anesth Analg
10. Dath D, Park AE. Randomized, controlled trial of bupivacaine injection to decrease pain after laparoscopic cholecystectomy. Can J Surg
11. Rademaker BM, Kalkman CJ, Odoom JA, de Wit L, Ringers J. Intraperitoneal local anaesthetics after laparoscopic cholecystectomy: effects on postoperative pain, metabolic responses and lung function. Br J Anaesth
12. Scheinin B, Kellokumpu I, Lindgren L, Haglund C, Rosenberg PH. Effect of intraperitoneal bupivacaine on pain after laparoscopic cholecystectomy. Acta Anaesthesiol Scand
13. Bigler D, Dirkes W, Hansen R, Rosenberg J, Kehlet H. Effects of thoracic paravertebral block with bupivacaine versus combined thoracic epidural block with bupivacaine and morphine on pain and pulmonary function after cholecystectomy. Acta Anaesthesiol Scand
14. Giesecke K, Hamberger B, Jarnberg PO, Klingstedt C. Paravertebral block during cholecystectomy: effects on circulatory and hormonal responses. Br J Anaesth
15. Richardson J, Lönnqvist PA. Thoracic paravertebral block. Br J Anaesth
16. Naja Z, Lönnqvist PA. Somatic paravertebral nerve blockade. Incidence of failed block and complications. Anaesthesia
17. Naja Z, el Hassan MJ, Khatib H, Ziade MF, Lönnqvist PA. Combined sciatic-paravertebral nerve block vs. general anaesthesia for fractured hip of the elderly. Middle East J Anesthesiol
18. Naja MZ, el Hassan MJ, Oweidat M, Zbibo R, Ziade MF, Lönnqvist PA. Paravertebral blockade vs. general anesthesia or spinal anesthesia for inguinal hernia repair. Middle East J Anesthesiol
19. Naja Z, Ziade MF, Lönnqvist PA. Bilateral paravertebral somatic nerve block for ventral hernia repair. Eur J Anaesthesiol
20. Alexander DJ, Ngoi SS, Lee L, et al.
Randomized trial of periportal peritoneal bupivacaine for pain relief after laparoscopic cholecystectomy. Br J Surg
21. Szem JW, Hydo L, Barie PS. A double-blinded evaluation of intraperitoneal bupivacaine vs saline for the reduction of postoperative pain and nausea after laparoscopic cholecystectomy. Surg Endosc
22. Mraovic B, Jurisic T, Kogler-Majeric V, Sustic A. Intraperitoneal bupivacaine for analgesia after laparoscopic cholecystectomy. Acta Anaesthesiol Scand
23. Wills VL, Hunt DR. Pain after laparoscopic cholecystectomy. Br J Surg
24. Tsimoyiannis EC, Siakas P, Tassis A, Lekkas ET, Tzourou H, Kambili M. Intraperitoneal normal saline infusion for postoperative pain after laparoscopic cholecstectomy. World J Surg
25. Richardson J, Jones J, Atkinson R. The effect of thoracic paravertebral blockade on intercostal somatosensory evoked potentials. Anesth Analg
26. Pettersson N, Perbeck L, Brismar B, Hahn RG. Sensory and sympathetic block during interpleural analgesia. Reg Anesth
27. Conacher ID. Resin injection of thoracic paravertebral spaces. Br J Anaesth
28. Conacher ID, Kokri M. Postoperative paravertebral block for thoracic surgery. A radiological appraisal. Br J Anaesth
29. Saito T, Den S, Cheema SP, et al.
A single-injection, multi-segmental paravertebral block-extension of somatosensory and sympathetic block in volunteers. Acta Anaesthesiol Scand
30. Pusch F, Freitag H, Weinstabl C, Obwegeser R, Huber E, Wildling E. Single-injection paravertebral block compared to general anaesthesia in breast surgery. Acta Anaesthesiol Scand
31. Singelyn FJ, Gouverneur JM, Robert A. A minimum dose of clonidine added to mepivacaine prolongs the duration of anaesthesia and analgesia after axillary brachial plexus block. Anesth Analg
32. Giannoni C, White S, Enneking FK, Morey T. Ropivacaine with or without clonidine improve pediatric tonsillectomy pain. Arch Otolaryngol Head Neck Surg
33. Sanchez R, Nielsen H, Heslet L, Iversen AD. Neuronal blockade with morphine. A hypothesis. Anaesthesia
34. Mays KS, Lipman JJ, Schnapp M. Local analgesia without anesthesia using peripheral perineural morphine injections. Anesth Analg
35. Vinson-Bonnet B, Coltat JC, Fingerhut A. Local Infiltration with ropivacaine improves immediate postoperative pain control after hemorrhoidal surgery. Dis Colon Rectum
36. Fazi L, Jantzen EC, Rose JB, Kurth CD, Watcha MF. A comparison of oral clonidine and oral midazolam as pre-anesthetic medications in the pediatric tonsillectomy patient. Anesth Analg
37. Tighe SQ. Paravertebral block. Anaesthesia