Knowledge Key:

Postoperative

Expand all

Systemic analgesia

alpha 2-adrenergic receptor agonists

PROSPECT Recommendations

Despite positive but limited procedure-specific evidence and positive transferable evidence, postoperative systemic dexmedetomidine is not recommended at this time because of potential side-effects (Grade D)

Clinical Practice

The risk/benefit ratio for dexmedetomidine is unclear. Potential side-effects include hypotension, sedation, dizziness and bradycardia

Transferable Evidence from other Procedures

Postoperative intravenous dexmedetomidine reduced morphine use in two studies (various types of surgery; n=295, n=401) Herr et al 2003
Postoperative intravenous dexmedetomidine was associated with increased incidence of hypotension and bradycardia in one study (various types of surgery; n=401) Martin et al 2003
Study details Martin et al 2003 Click here for more information

Thoracotomy-specific Evidence

Postoperative intravenous dexmedetomidine, as an adjunct to thoracic epidural analgesia, reduced pain scores at 3 and 4 h (p=0.03), but there was no significant difference at 0, 1, 2, 8, 12, 16 and 24 h (n=28) Wahlander et al 2005
Postoperative intravenous dexmedetomidine reduced supplementary analgesic requirements compared with placebo Wahlander et al 2005 Click here for more information
Postoperative intravenous dexmedetomidine was associated with a reduction in PaCO2 compared with placebo from 8–24 h (p<0.05), but there was no significant difference from 0–4 h (n=28) Wahlander et al 2005
Postoperative intravenous dexmedetomidine decreased heart rates from 0–24 h (p<0.05), and increased the incidence of hypotension (p=0.04), compared with placebo (n=28) Wahlander et al 2005
Study details Wahlander et al 2005 Click here for more information

Conventional NSAIDs

PROSPECT Recommendations

Postoperative conventional NSAIDs are recommended based on a reduction in pain scores and analgesic use (Grade A)
There is evidence that conventional NSAIDs do not provide additional pain relief in patients receiving epidural analgesia. Therefore it is recommended that conventional NSAIDs are used only if analgesia is inadequate in patients receiving epidural analgesia (Grade A), or paravertebral block (Grade D)
Conventional NSAIDs are not recommended in patients who have an increased risk of bleeding, or who are at risk of gastroduodenal ulcer/erosion (Grade B)
The use of conventional NSAIDs should depend upon assessment of individual patient risks (bleeding complications including epidural haematoma, cardiovascular morbidity, actual or recent gastroduodenal ulcer history, aspirin-sensitive asthma, renal function and hepatic function) (Grade B)

Clinical Practice

Caution is required when conventional NSAIDs are used in patients receiving epidural analgesia concomitantly with medications for thromboprophylaxis, due to the risk of spinal haematoma

Transferable Evidence from Other Procedures

Conventional NSAIDs have proven analgesic efficacy in a variety of surgical procedures Barden et al 2004
A meta-analysis of randomised controlled trials, which was performed to evaluate the risk of morphine side-effects in patients treated with NSAIDs, demonstrated that NSAIDs decreased the incidence of nausea, vomiting and sedation, but not pruritis, urinary retention or respiratory depression Marret et al 2005
One study in major upper abdominal surgery showed that rectal piroxicam plus thoracic epidural bupivacaine plus morphine provided no significant benefit over epidural analgesia alone for reducing postoperative pain scores and supplementary analgesic use (n=44) Mogensen et al 1992b
Randomised trials in healthy volunteers have shown that COX-2-selective inhibitors are associated with a lower incidence of upper gastrointestinal ulceration compared with conventional NSAIDs for short-term use Harris et al 2001
Randomised trials in healthy volunteers showed that conventional NSAIDs (ketorolac, naproxen or diclofenac) reduced the platelet aggregation response compared with placebo; ketorolac and naproxen also prolonged bleeding time compared with placebo Greer et al 1999
Meta-analyses of randomised, controlled trials showed that peri-operative conventional NSAIDs increased the risk of postoperative bleeding requiring treatment and/or the risk of re-operation for haemostasis after tonsillectomy compared with control Marret et al 2003
A randomised trial in healthy volunteers showed that three conventional NSAIDs (diclofenac, ketorolac and ketoprofen) caused a reversible platelet dysfunction Niemi et al 1997
A large randomised controlled trial showed that three conventional NSAIDs (diclofenac, ketorolac and ketoprofen) were associated with a similar incidence of surgical site bleeding after elective surgery Forrest et al 2002
Aspirin and conventional NSAIDs can induce asthma attacks in patients with aspirin-exacerbated respiratory disease Stevenson 2004
Conventional NSAIDs and COX-2-selective inhibitors have been associated with an increased risk of transient hepatotoxicity. Cases of acute hepatic failure have also been reported. Elderly females, with autoimmune disease, and taking other potentially hepatotoxic drugs, may be most susceptible O'Connor et al 2003
Short-term use of conventional NSAIDs or COX-2-selective inhibitors can increase the risk of transient renal impairment, especially in patients with existing renal dysfunction, and both types of agent have been associated with cases of acute renal failure in high-risk patients Cheng et al 2004
Conventional NSAIDs and COX-2-selective inhibitors may elevate blood pressure, particularly in hypertensive patients receiving antihypertensive medication Cheng et al 2004
Conventional NSAIDs and COX-2-selective inhibitors have been associated with impaired bone healing in various animal models and in vitro studies Gajraj 2003

Thoracotomy-specific Evidence

In seven of twelve studies, conventional NSAIDs were superior to placebo or control for reducing VAS pain scores Jain et al 1998 Click here for more information
Postoperative indomethacin was associated with greater PF values on days 1 and 2 compared with cryoanalgesia (p<0.001) (n=60) Keenan et al 1983
Postoperative indomethacin, but not cryoanalgesia, significantly reduced pain scores on movement compared with placebo (p<0.05); there was no significant difference for pain at rest; indomethacin also improved observer grading of pain compared with cryoanalgesia (p<0.05) (n=60) Keenan et al 1983
Postoperative metamizol was associated with better pain relief and reduced supplementary analgesic requirements compared with ketorolac Tulunay et al 1996 Click here for more information
Nimesulide was associated with less reduction in PEFR compared with ibuprofen (significance not stated), when spinal or epidural analgesia were used (four treatment arms; n=30) McCrory et al 2002
Nimesulide and ibuprofen were not significantly different for the incidence of PONV (four treatment arms; n=30) McCrory et al 2002
Nimesulide reduced spinal morphine use compared with ibuprofen, for patients receiving spinal analgesia (significance not stated), but bupivacaine/fentanyl use by patients receiving epidural analgesia was not reduced (four treatment arms; n=30) McCrory et al 2002
Nimesulide reduced VAS pain scores compared with ibuprofen, at rest and on coughing, for patients receiving spinal analgesia (significance not stated), but not for patients receiving epidural analgesia (four treatment arms; n=30) McCrory et al 2002
Conventional NSAIDs did not significantly decrease platelet counts (n=30) Perttunen et al 1992
Conventional NSAIDs plus epidural analgesia were superior to epidural analgesia alone for increasing the time between the first and second requests for analgesia (p=0.000001) (n=26) Jain et al 1998
Conventional NSAIDs were significantly superior to placebo for increasing the time between the first and second requests for analgesia (p=0.000001) (n=26) Jain et al 1998
Ketorolac 10 mg (p<0.05) or 30 mg (p<0.01) significantly reduced the proportion of patients withdrawn for lack of analgesia compared with placebo (n=75) Power et al 1994
In seven of eleven studies, conventional NSAIDs were superior to placebo or control for reducing supplementary analgesic use Carretta et al 1996 Click here for more information
In two of four studies, conventional NSAIDs plus epidural analgesia were similar to epidural analgesia alone for reducing pain scores Bigler et al 1992 Click here for more information
In three of four studies, conventional NSAIDs plus epidural analgesia were similar to epidural analgesia alone for reducing supplementary analgesic requirements Bigler et al 1992 Click here for more information
Conventional NSAIDs did not significantly reduce the incidence of PONV compared with placebo or control (n=45; n=19, n=41, n=30, n=42) McCrory et al 2002 Click here for more information
Results for pulmonary function parameters were mixed in studies of conventional NSAIDs versus placebo or control Bigler et al 1992 Click here for more information
Conventional NSAIDs plus epidural analgesia were associated with similar pulmonary function to epidural analgesia alone Bigler et al 1992 Click here for more information
Study details Bigler et al 1992 Click here for more information

COX-2-selective inhibitors

PROSPECT Recommendations

Postoperative COX-2-selective inhibitors are recommended based on evidence that they provide similar postoperative analgesia to conventional NSAIDs (Grade B)
In patients receiving epidural analgesia or paravertebral block, it is recommended that COX-2-selective inhibitors are used only if analgesia is inadequate (Grade B, see Postoperative Conventional NSAIDs)
COX-2-selective inhibitors may be preferred to conventional NSAIDs in the peri-operative setting in patients who are at increased risk of bleeding or who are at risk of gastroduodenal ulcer/erosion (Grade B)
The use of COX-2-selective inhibitors should depend upon assessment of individual patient risks (cardiovascular morbidity, actual or recent gastroduodenal ulcer history, renal function and hepatic function (Grade B) or aspirin-sensitive asthma (Grade D))

Clinical Practice

Thoracic surgery patients may frequently have coronary artery disease or cardiac contra-indications

Transferable Evidence from Other Procedures

COX-2-selective inhibitors provide similar postoperative analgesia to conventional NSAIDs Rømsing et al 2004
Peri-operative use of rofecoxib (50 mg at 24 hand at 1 to 2 h before total knee arthroplasty, followed by 50 mg daily for 5 days postoperatively, and 25 mg for another 8 days), as part of a multi-analgesic regimen, reduced pain, opioid use, vomiting, sleep disturbance and improved knee range of motion, compared with placebo (n=68) Buvanendran et al 2003
Randomised trials in healthy volunteers have shown that COX-2-selective inhibitors are associated with a lower incidence of upper gastrointestinal ulceration compared with conventional NSAIDs for short-term use Harris et al 2001
Studies in healthy volunteers demonstrated that COX-2-selective inhibitors had no effect on platelet aggregation Greenberg et al 2000
Clinical studies investigating the response to oral challenge with COX-2-selective inhibitors in patients with aspirin-induced asthma have demonstrated that COX-2-selective inhibitors do not have an effect on respiratory function Bavbek et al 2004
A study to assess the safety of the COX-2-selective inhibitors parecoxib and valdecoxib following noncardiac general surgery (including gastrointestinal, orthopaedic, gynaecological, urological, and thoracic surgeries) showed no difference in the incidence of cardiovascular thromboembolic events, renal dysfunction/failure, gastrointestinal ulcer complications, and surgical wound-healing complications, compared with placebo (n=1062) Nussmeier et al 2006
Two clinical trials showed that, in patients who had undergone CABG surgery, COX-2-selective inhibitors (valdecoxib and parecoxib) were associated with a higher rate of serious cardiovascular thromboembolic events (including myocardial infarction) compared with placebo Nussmeier et al 2005
Hypersensitivity reactions and serious skin reactions (e.g. toxic epidermal necrolysis, Stevens-Johnson syndrome, and erythema multiforme) can occur with all COX-2-selective inhibitors. Serious skin reactions have been reported in association with valdecoxib at a higher rate than with other COX-2-selective agents EMEA 2004a
Conventional NSAIDs and COX-2-selective inhibitors have been associated with an increased risk of transient hepatotoxicity. Cases of acute hepatic failure have also been reported. Elderly females, with autoimmune disease, and taking other potentially hepatotoxic drugs, may be most susceptible O'Connor et al 2003
Short-term use of COX-2-selective inhibitors or conventional NSAIDs can increase the risk of transient renal impairment, especially in patients with existing renal dysfunction, and both types of agent have been associated with cases of acute renal failure in high-risk patients Cheng et al 2004
COX-2-selective inhibitors and conventional NSAIDs may elevate blood pressure, particularly in hypertensive patients receiving antihypertensive medication Cheng et al 2004
Although there is some concern that COX-2-selective inhibitors may impair wound healing, evidence from animal and clinical studies is conflicting Blomme et al 2003
Conventional NSAIDs and COX-2-selective inhibitors have been associated with impaired bone healing in various animal models and in vitro studies Gajraj 2003

Thoracotomy-specific Evidence

None cited

Ketamine

PROSPECT Recommendations

Low-dose ketamine cannot be recommended at this time due to a lack of procedure-specific evidence (Grade D), although analgesic data from other procedures are promising

Clinical Practice

Ketamine may be of benefit for prevention of chronic pain conditions following thoracotomy, where 40–60% patients may develop chronic pain and/or allodynia. Additionally, it may be of benefit in patients who have opioid tolerance. Ketamine is most often used after induction of GA, but before incision (pre-operatively)

Transferable Evidence

Studies of intravenous or neuraxial ketamine in a variety of surgeries, including abdominal, gynaecological, orthopaedic, gastric, hepatic, and genito-urinary surgery, showed a reduction in postoperative pain and opioid use when used as an adjunct to morphine Bell et al 2005

Thoracotomy-specific Evidence

Postoperative ketamine and pethidine were similarly effective for reducing pain scores (n=30) Dich-Nielsen et al 1992
Postoperative ketamine and pethidine were similar for supplementary analgesic use during 0–5 h (n=30) Dich-Nielsen et al 1992
Postoperative ketamine and pethidine were similar for the proportion of patients requiring an extra analgesic injection within 2 h, but during 2–3 h ketamine was associated with a lower proportion of patients requiring supplementary analgesia than pethidine (p<0.05) (n=30) Dich-Nielsen et al 1992
Postoperative ketamine was associated with a lower PaCO₂ than pethidine after 30 min and throughout the study (p<0.05), but there was no difference in PaO₂ (n=30) Dich-Nielsen et al 1992
Postoperative ketamine and pethidine were not significantly different for the incidence of PONV (n=30) Dich-Nielsen et al 1992
Study details Dich-Nielsen et al 1992 Click here for more information

Gabapentin/pregabalin

PROSPECT Recommendations

Gabapentin/pregabalin cannot be recommended at this time due to a lack of procedure-specific evidence (Grade D), although analgesic data from other procedures are promising

Clinical Practice

Gabapentin may be of benefit for prevention of chronic pain conditions following thoracotomy, where 40–60% of patients may develop chronic pain and/or allodynia

Transferable Evidence

Studies of gabapentin and pregabalin in mastectomy, abdominal surgery, laparoscopic cholecystectomy, spinal surgery, knee surgery, ear-nose-throat surgery, and nephrectomy showed reductions in postoperative pain and supplementary analgesic requirements for at least 24 h Dahl et al 2004
One study showed no significant benefit of gabapentin 800 mg for reducing pain or opioid use compared with placebo in spinal surgery (n=60) Radhakrishnan et al 2005

Thoracotomy-specific Evidence

None cited

Opioid, strong

PROSPECT Recommendations

Intravenous PCA strong opioid is recommended if regional analgesic techniques fail or are not possible (Grade A)
Even though intravenous PCA strong opioids showed no consistent analgesic benefit over intramuscular PRN opioids (grade B), they are recommended based on greater patient satisfaction compared with regular (fixed-interval) or PRN dosing (grade B); however, fixed-interval intravenous administration titrated to pain intensity is also recognised as an effective mode of administration (grade D)
Intramuscular strong opioids are not recommended due to the pain associated with these injections (grade D)
There is insufficient evidence to recommend any opioid in preference to another

Clinical Practice

Regular administration, titrated for pain intensity, is generally accepted as an effective method of administering strong opioids
Most clinical trials showing benefits of intramuscular strong opioids use nurse-administered regimens. In regular clinical practice, full adherence to nurse-administered regimens is not usually achievable, and the full analgesic benefits of intramuscular strong opioids are also not achieved
Intramuscular administration of strong opioids is considered to be more painful than intravenous administration; however, the dose and rapidity of intravenous administration should be assessed to minimise the risk of respiratory depression

Transferable Evidence

Bolus plus infusion IV PCA morphine conferred a significant benefit over bolus IV PCA morphine alone for reducing postoperative pain scores in abdominal hysterectomy, with no significant difference in overall analgesic consumption El-Falaki et al 2000 Click here for more information
Studies in abdominal surgery comparing IV PCA and IM regular/on-request administration of strong opioids showed inconsistent results for postoperative pain scores Choiniere et al 1998 Click here for more information
There is mixed evidence in abdominal surgery for a benefit of PCA compared with regular/on-request IM administration of strong opioids for reducing overall opioid consumption, although one study suggests that they produced different patterns of dosing Thomas et al 1995 Click here for more information
The incidence of PONV was not significantly different between PCA and IM morphine, in three studies in abdominal surgery (n=126, n=22, n=41) Choiniere et al 1998
A systematic review comparing intravenous PCA opioids with intravenous, intramuscular or subcutaneous opioids by injection showed that PCA opioids were associated with greater pain relief, reduced supplementary analgesic requirements (analysis of eleven studies, n=691), and more patients preferred PCA opioids (analysis of four trials, n=352) compared with traditional opioid analgesia Walder et al 2001
Strong opioids are associated with adverse effects, including nausea, vomiting, sedation, confusion, paralytic ileus and urinary retention Wheeler et al 2002

Thoracotomy-specific Evidence

Postoperative nalbuphine reduced supplementary fentanyl requirements compared with placebo Baxter et al 1989 Click here for more information
An intravenous morphine infusion and intramuscular morphine on demand were similar for the total dose of morphine used (n=30) Orr et al 1981
An intravenous morphine infusion reduced pain scores at rest compared with intramuscular morphine on demand, between 3 h and day 2 (p=0.001), but not at 1.5 h, or on day 8, or on movement or on physiotheraphy (n=30) Orr et al 1981
Intravenous PCA and intramuscular meperidine were similar for duration of hospital stay, but intravenous PCA meperidine increased the proportion of patients leaving hospital within the first week compared with intramuscular meperidine (p<0.05) (n=40) Boulanger et al 1993
Intravenous PCA meperidine reduced the proportion of patients requiring an increase in meperidine dose compared with intramuscular meperidine (p<0.05) (n=40) Boulanger et al 1993
Intravenous PCA meperidine reduced meperidine use in the recovery room compared with intramuscular meperidine (p<0.001), but not during 0–24 h or 24–48 h (n=40) Boulanger et al 1993
PCA-administered strong opioids were associated with better pain relief than intramuscularly administered strong opioids Lehmann et al 1991 Click here for more information
Postoperative PCA fentanyl was associated with a significantly higher PaO₂ at 13 h, and a significantly lower PaCO₂ at 7, 13 and 15 h, than PCA buprenorphine (no p-values reported) (two treatment arms, n=40) Lehmann et al 1991
Postoperative PCA buprenorphine was similar to PCA fentanyl for pain scores during 0–24 h, and for the number of analgesic requests (two treatment arms, n=40) Lehmann et al 1991
Postoperative nalbuphine 200 µg/kg bolus plus 50 µg/kg/h infusion reduced the incidence of therapeutic failures (defined as at least one measurement of PaCO₂ >50 mmHg) compared with placebo or lower concentrations of nalbuphine (p<0.01) (three treatment arms; n=63) Baxter et al 1989
Postoperative nalbuphine did not significantly reduce pain scores compared with placebo (three treatment arms; n=63) Baxter et al 1989
Postoperative nalbuphine did not significantly reduce the incidence of PONV compared with placebo (three treatment arms; n=63) Baxter et al 1989
VAS pain scores and the incidence of PONV were similar for all concentrations of postoperative nalbuphine (three arms; n=63) Baxter et al 1989
PCA buprenorphine was similar to PCA fentanyl for the incidence of PONV (two treatment arms, n=40) Lehmann et al 1991
Intravenous PCA opioid showed no significant benefit compared with intramuscular opioid for reducing the incidence of PONV Boulanger et al 1993 Click here for more information
Intravenous PCA strong opioids were not associated with improved pulmonary function compared with intramuscular strong opioids: FEV1, FVC and arterial PaCO₂ (n=40) arterial blood gas levels (two PCA arms, n=60) Boulanger et al 1993
Study details Baxter et al 1989 Click here for more information

Opioid, weak

PROSPECT Recommendations

Postoperative weak opioids are recommended to control moderate- (VAS>30<50) or low- (VAS £30) intensity pain in the late postoperative period, only if COX-2-selective inhibitors/conventional NSAIDs plus paracetamol are insufficient or are contra-indicated (grade D)

Clinical Practice

It cannot be concluded from procedure-specific evidence that intravenous tramadol is better or worse than intravenous morphine, as the study is under-powered Erolcay et al 2003
Opioid adverse effects, especially nausea and dizziness, limit the usefulness of tramadol, but many of the effects are less than for strong opioids

Transferable Evidence

None cited

Thoracotomy-specific Evidence

Intra-operative plus postoperative intravenous tramadol significantly reduced pain scores compared with placebo at rest at 2, 3, 4 and 8 h (p<0.05) but not at 1, 12, 16, 20 and 24 h, and reduced pain scores on coughing at 2, 3 and 4 h (p<0.05) (n=59) Bloch et al 2002
Intra-operative plus postoperative intravenous tramadol significantly reduced intravenous PCA morphine use compared with placebo up to 24 h Bloch et al 2002 Click here for more information
Intra-operative plus postoperative intravenous tramadol was associated with a greater vital capacity at 20 h (p<0.05) compared with placebo (n=59) Bloch et al 2002
In two studies, intravenous tramadol was similar to epidural morphine for reducing pain scores: at rest and on coughing during the first 24 h (n=59) Bloch et al 2002
In two studies, intravenous tramadol was similar to epidural morphine for morphine use Bloch et al 2002 Click here for more information
In two studies, intravenous tramadol was associated with a higher PaO₂ compared with epidural morphine Bloch et al 2002 Click here for more information
Postoperative tramadol and morphine were similar for pain scores at rest and during deep inspiration at 2–24 h, and for measures of pulmonary function (PaCO₂ and PaO₂) (n=40) Erolcay et al 2003
Intra-operative plus postoperative intravenous tramadol did not significantly reduce the incidence of PONV compared with placebo (n=59) Bloch et al 2002
Intravenous tramadol was similar to epidural morphine for the incidence of PONV (n=59) Bloch et al 2002
Postoperative intravenous PCA tramadol and morphine were similar for the incidence of PONV (n=40) Erolcay et al 2003
Study details Bloch et al 2002 Click here for more information

Paracetamol

PROSPECT Recommendations

Paracetamol alone is not recommended for high-intensity pain (VAS³50 mm) (grade B)
Paracetamol is recommended if analgesia is inadequate with regional techniques (Grade B)
Paracetamol is recommended as part of a multianalgesic regimen including intravenous PCA strong opioid (grade D)
The additional analgesic effect of paracetamol, when combined with conventional NSAIDs/COX-2-selective inhibitors, is limited in major surgery

Clinical Practice

None cited

Transferable Evidence

Paracetamol is an effective analgesic for the treatment of postoperative pain of moderate intensity Rømsing et al 2002
New formulations of intravenous paracetamol may aid accurate administration of paracetamol by producing more predictable plasma concentrations in the immediate postoperative period Holmer Pettersson et al 2004
There is evidence that concurrent use of paracetamol and conventional NSAIDs improves pain relief compared with paracetamol alone, but there is no evidence for a superior analgesic effect of the combination compared with conventional NSAIDs alone Altman 2004

Thoracotomy-specific Evidence

None cited

Epidural Analgesia

Thoracic epidural

PROSPECT Recommendations

Thoracic epidural infusion of LA plus strong opioid, continued for 2–3 days after surgery, is recommended based on a reduction of pain compared with systemic analgesia (Grade A)
There are not enough data to recommend one specific combination of LA and opioid over another
There are not enough data to recommend a specific concentration or volume of LA and strong opioid
There are not enough data to recommend lipophilic opioids in preference to hydrophilic opioids or vice versa, in combination with LA

Clinical Practice

Postoperative thoracic epidural infusion of strong opioid alone may be used when a catheter is already in place, and hypotension is a problem (e.g. due to blood loss during the procedure)
If opiate side-effects are a problem then epidural LA alone may be used
PCEA is becoming a more commonly used technique for administering postoperative analgesics via the epidural route
Thoracic epidural bolus of strong opioids is a technique that is no longer used
Use of heparin or conventional NSAIDs may increase the risk of spinal haematoma due to epidural analgesia

Transferable Evidence

A meta-analysis of randomised controlled trials found that both continuous epidural infusion and PCEA analgesia provided superior postoperative analgesia compared with intravenous PCA analgesia Wu et al 2005
A meta-analysis found that continuous epidural infusion provided superior analgesia and reduced the incidence of pruritis compared with PCEA analgesia, but was associated with a higher incidence of PONV and motor block Wu et al 2005
A meta-analysis of randomised controlled trials found that epidural analgesia significantly decreased the incidence of pulmonary morbidity compared with systemic opioids Ballantyne et al 1998 Click here for more information
Epidural administration of strong opioids is associated with side effects including pruritis, PONV, urinary retention, and respiratory depression Chaney 1995

Thoracotomy-specific Evidence

Two studies showed a similar incidence of PONV for high volume/low concentration and low volume/high concentration bupivacaine plus sufentanil (n=30; n=60) Laveaux et al 1993 Click here for more information
Bupivacaine 0.06% and 0.125%, and 0.1% and 0.2% for peri-operative thoracic epidural infusion were similar for reducing pain scores up to 24 h Burgess et al 1994 Click here for more information
In one study, thoracic epidural bupivacaine 0.05% and 0.1% was superior to 0.01% for reducing pain scores on physiotherapy, but not at rest or on coughing, while another study showed thoracic epidural bupivacaine 0.1% and 0.01% were similar Etches et al 1996 Click here for more information
In two of four studies, thoracic epidural lipophilic strong opioid was superior to systemic lipophilic strong opioid for reducing the incidence of PONV Guinard et al 1992 Click here for more information
One study reported that postoperative thoracic epidural lipophilic strong opioid significantly reduced the duration of hospital stay compared with systemic strong opioid (p=0.02) (n=32) Guinard et al 1992
In two of four studies, thoracic epidural lipophilic strong opioid was associated with superior pulmonary function to systemic lipophilic strong opioid Salomaki et al 1991 Click here for more information
Thoracic epidural diamorphine provided analgesia of longer duration than intramuscular diamorphine (p<0.05) (n=18) Jacobson et al 1983
One of two studies showed that that thoracic epidural hydrophilic strong opioid was associated with superior pulmonary function to systemic hydrophilic strong opioid Hasenbos et al 1986 Click here for more information
Tetracaine, for thoracic epidural injection via PCEA, was similar to ropivacaine for pain scores on days 1 and 2, for the total use of LA solution, for the duration of the intervals between PCEA bolus injections, and for measures of pulmonary function (FEV₁, FVC, PEFR) (n=40) Guo et al 2003
Ropivacaine, for thoracic epidural injection, was superior to bupivacaine for PaO₂ (p<0.05) (n=30), but similar for PaCO₂, FVC, PEFR and FEV₁ (n=52; n=30) Shorrab et al 2003
Ropivacaine and bupivacaine, for thoracic epidural injection, were similar for the incidence of PONV (n=52; n=30) Shorrab et al 2003
Ropivacaine and bupivacaine, for thoracic epidural injection, were similar for the proportion of patients requesting supplementary analgesics (n=30) Shorrab et al 2003
Ropivacaine and bupivacaine, for thoracic epidural injection, were similar for pain scores: at rest at 4, 8, 12, and 24 h (n=30) (Shorrab 2003); at rest and on measurement of PEF during 2–48 h (n=52) Macias et al 2002
One study reported that high volume/low concentration and low volume/high concentration bupivacaine plus sufentanil were associated with similar PaCO₂ (n=30) Laveaux et al 1993 Click here for more information
Two of three studies showed that thoracic epidural strong opioid plus LA was similar to LA alone for the incidence of PONV Bloch et al 2002 Click here for more information
High volume/low concentration bupivacaine plus sufentanil was superior to low volume/high for reducing supplementary analgesic requirements Snijdelaar, Hasenbos et al 1994 Click here for more information
For peri-operative thoracic epidural infusion of bupivacaine plus sufentanil, high volume/low concentration was similar to low volume/high concentration for reducing pain scores at rest and on exercise during days 0–3 (n=30; n=60) Laveaux et al 1993 Click here for more information
In three of four studies, thoracic epidural bupivacaine and thoracic epidural morphine were similar for pain scores El-Baz et al 1984 Click here for more information
Thoracic epidural morphine was superior to thoracic epidural bupivacaine for extending the duration of pain relief per injection during 0–72 h (p<0.01), 24–48 h (p<0.02) and 48–72 h (p<0.01), but not during 0–24 h (three treatment arms; n=60) El-Baz et al 1984
Thoracic epidural LA and thoracic epidural strong opioid were similar for supplementary analgesic use (n=22; n=30; n=18) Logas et al 1987
Thoracic epidural LA and thoracic epidural strong opioid were similar for the incidence of PONV (n=22; n=30) Logas et al 1987
Thoracic epidural bupivacaine was associated with similar pulmonary function measures to thoracic epidural morphine El-Baz et al 1984 Click here for more information
A continuous infusion of thoracic epidural morphine was superior to intermittent injections for reducing the mean dose of morphine/24-h period (p<0.01) (n=60) El-Baz et al 1984
Thoracic epidural bupivacaine plus fentanyl was similar to spinal morphine, or spinal morphine plus systemic ibuprofen, for pain scores (at rest or cough) (n=35) McCrory et al 2002
Peri-operative thoracic epidural bupivacaine plus fentanyl with or without systemic nimesulide or ibuprofen was associated with similar pain scores at rest or on coughing to spinal morphine (n=35) McCrory et al 2002
Thoracic epidural bupivacaine plus fentanyl was similar to spinal morphine for the use of rescue morphine (n=35) McCrory et al 2002
Thoracic epidural bupivacaine plus fentanyl and spinal morphine were associated with a similar incidence of PONV (n=35) McCrory et al 2002
Thoracic epidural bupivacaine plus fentanyl was associated with similar PEFR to spinal morphine (n=35) McCrory et al 2002
Thoracic epidural strong opioid plus LA was superior for reducing morphine use compared with epidural LA alone Bloch et al 2002 Click here for more information
Six of seven studies showed that thoracic epidural LA plus strong opioid was superior to systemic analgesia for reducing pain scores Azad et al 2000 Click here for more information
Thoracic epidural LA plus strong opioid reduced supplementary analgesic requirements compared with systemic analgesia Brichon et al 1994 Click here for more information
Thoracic epidural LA plus strong opioid was superior to systemic analgesia for reducing the incidence of nausea Azad et al 2000 Click here for more information
One of two studies showed that thoracic epidural LA plus strong opioid was superior to systemic analgesia for reducing the duration of hospital stay (p<0.05) (n=563) Boisseau et al 2001 Click here for more information
Thoracic epidural bupivacaine plus fentanyl, with or without systemic nimesulide or ibuprofen, was associated with similar PEFR compared with spinal morphine (n=35) McCrory et al 2002
For peri-operative thoracic epidural analgesia, fentanyl 2, 5 and 10 µg/ml were similar for the incidence of PONV (three treatment arms; n=89) Tan et al 2004
Two of three studies showed that thoracic epidural bupivacaine infusion was superior to systemic analgesia for reducing pain scores Bachmann-Mennenga et al 1993 Click here for more information
One of two studies showed that thoracic epidural LA infusion was superior to systemic analgesia for reducing the dose of strong opioid Bachmann-Mennenga et al 1993 Click here for more information
Thoracic epidural bupivacaine for anaesthesia and analgesia was superior to intravenous propofol plus fentanyl anaesthesia with postoperative piritramide, for maintaining intra-operative PaO₂ and cardiac output (p<0.05) (n=50) Von Dossow et al 2001
Thoracic epidural bupivacaine for anaesthesia and analgesia was superior to intravenous propofol plus fentanyl anaesthesia with postoperative piritramide for reducing time to extubation and duration of intensive care unit stay (p<0.04) (n=50) Von Dossow et al 2001
For peri-operative thoracic epidural analgesia, fentanyl 5 or 10 µg/ml significantly reduced the proportion of patients with VAS pain score >30 mm (p<0.01) compared with 2 µg/ml (three treatment arms; n=89) Tan et al 2004
For peri-operative thoracic epidural analgesia, there was no significant difference between morphine 1 mg and 0.5 mg for PaCO₂ (n=60) Geurts et al 1995
For peri-operative thoracic epidural analgesia, morphine 1 mg and 0.5 mg were similar for the incidence of PONV (n=60) Geurts et al 1995
Nine studies compared the effect of thoracic epidural LA plus strong opioid versus thoracic epidural strong opioid alone on postoperative pain scores Click here for more information
Thoracic epidural LA plus lipophilic strong opioid was superior to epidural lipophilic strong opioid alone for reducing pain George et al 1991 Click here for more information
Three of six studies showed that thoracic epidural LA plus lipophilic strong opioid reduced supplementary analgesic use compared with epidural lipophilic strong opioid alone Burgess et al 1994 Click here for more information
Peri-operative thoracic epidural LA plus lipophilic strong opioid was associated with better maintenance of pulmonary function compared with epidural lipophilic strong opioid alone Burgess et al 1994 Click here for more information
For peri-operative thoracic epidural analgesia, morphine 1 mg significantly reduced the proportion of patients who required additional bolus doses compared with morphine 0.5 mg on day 1 (p<0.001), day 2 (p<0.05), and day 3 (p<0.001) (n=60) Geurts et al 1995
For peri-operative thoracic epidural analgesia, morphine 1 mg significantly reduced pain scores compared with morphine 0.5 mg, and significantly reduced the proportion of patients experiencing pain on deep breathing, on coughing and on physiotherapy Geurts et al 1995 Click here for more information
Bupivacaine 0.06% and 0.125%, for peri-operative thoracic epidural infusion, were associated with similar blood gas levels (n=20) Burgess et al 1994
Three studies showed no significant difference between different concentrations of bupivacaine (0.1% and 0.01%; 0.01%, 0.05% and 0.1%; or 0.1% and 0.2%, respectively) for the incidence of PONV (n=18; n=44; n=62) Etches et al 1996
One study showed a significant benefit of thoracic epidural bupivacaine 0.05% and 0.1%, compared with 0.01% for reducing opioid use on day 1 (p<0.03) while another study showed no significant difference Liu et al 1995b Click here for more information
Bupivacaine 0.06% and 0.125% for peri-operative thoracic epidural infusion were similar for use of fentanyl and proportion of patients requiring fentayl or ketorolac; bupivacaine 0.1% and 0.2% were similar for the amount of epidural solution required Burgess et al 1994 Click here for more information
In two studies plus one arm of a third study, thoracic epidural strong opioid plus LA was superior to epidural LA alone for reducing pain scores Bloch et al 2002 Click here for more information
Thoracic epidural bupivacaine plus fentanyl was associated with a lower PEFR compared with spinal morphine plus systemic nimesulide (p<0.001) (n=35) McCrory et al 2002
Thoracic epidural morphine was similar to intravenous tramadol for the incidence of PONV (n=59) Bloch et al 2002
A continuous infusion of thoracic epidural morphine was similar to intermittent injections for reduction in pain scores up to 72 h (n=60) El-Baz et al 1984
A continuous infusion of thoracic epidural morphine was associated with similar outcomes of pulmonary function tests compared with intermittent injections: PaO₂, PaCO₂, FEV₁/FVC and inspiratory flow rate (n=60)
In two of four studies, thoracic epidural lipophilic strong opioid was similar to systemic strong opioid for supplementary analgesic use Benzon et al 1993 Click here for more information
Two of three studies showed that epidural LA plus hydrophilic strong opioid was similar to epidural hydrophilic strong opioid alone for reducing VAS pain scores Burgess et al 1994 Click here for more information
Two of three studies showed that epidural LA plus hydrophilic strong opioid was similar to epidural hydrophilic strong opioid alone for supplementary analgesic requirements Etches et al 1996 Click here for more information
Thoracic epidural bupivacaine plus hydrophilic strong opioid showed no significant benefit for reducing the incidence of PONV compared with epidural hydrophilic strong opioid alone (n=66; n=20; n =43) Etches et al 1996
Thoracic epidural bupivacaine plus lipophilic strong opioid was of no significant benefit for reducing the incidence of PONV compared with epidural lipophilic strong opioid alone (n=21; n=37; n=24; n=40; n=65) George et al 1991
Thoracic epidural LA plus hydrophilic strong opioid was associated with similar pulmonary function (FVC, FEV₁, FEF, and PEFR), on days 1 and 2, to epidural hydrophilic strong opioid alone (n=43) Singh et al 1997
Postoperative thoracic epidural bupivacaine was similar to intravenous buprenorphine for PaO₂ and PaCO₂ (n=20) Bachmann-Mennenga et al 1993
Studies of thoracic epidural strong opioid plus LA versus epidural LA alone showed inconsistent results for pulmonary function parameters Bloch et al 2002 Click here for more information
Thoracic epidural bupivacaine was similar to intramuscular morphine on request for the incidence of PONV (two systemic analgesia arms; n=30) Logas et al 1987
In three out of four studies, thoracic epidural lipophilic strong opioid and systemic lipophilic strong opioid were similar for pain scores Guinard et al 1992 Click here for more information
In three of four studies, thoracic epidural hydrophilic strong opioid and systemic hydrophilic strong opioid were similar for pain scores Hasenbos et al 1986 Click here for more information
Thoracic epidural hydrophilic strong opioid was similar to systemic hydrophilic strong opioid for supplementary analgesic use (n=20; n=22) Larsen et al 1986
Thoracic epidural hydrophilic strong opioid was similar to systemic hydrophilic strong opioid for the incidence of PONV (n=22) Logas et al 1987
Thoracic epidural morphine was of no significant benefit compared with intravenous tramadol for reducing pain scores at rest or on coughing during 0–24 h, or supplementary morphine use (n=59; n=39) Bloch et al 2002
Thoracic epidural morphine was associated with a greater PaCO₂ at 2 and 6 h compared with intravenous tramadol (no p-value) (n=59) Bloch et al 2002
Four of six studies showed that thoracic epidural LA plus strong opioid was associated with similar pulmonary function to systemic analgesia Boisseau et al 2001 Click here for more information
Thoracic epidural bupivacaine plus fentanyl was inferior to spinal morphine plus systemic nimesulide for reducing pain scores on days 1 and 2 (p<0.001) (n=35) McCrory et al 2002
Study details Azad et al 2000 Click here for more information
Thoracic epidural LA + strong opioid versus systemic analgesia
Thoracic epidural LA versus systemic analgesia
Thoracic epidural LA + strong opioid versus thoracic epidural strong opioid alone
Thoracic epidural LA + strong opioid versus thoracic epidural LA alone
Thoracic epidural strong opioid versus systemic opioid
Thoracic epidural LA versus thoracic epidural strong opioid
Thoracic epidural analgesia versus other regional analgesic techniques
Thoracic epidural analgesia, miscellaneous studies

Lumbar epidural

PROSPECT Recommendations

Lumbar epidural strong opioid is not recommended as the first choice of epidural technique, based on evidence that the thoracic epidural route is more effective for pain relief following thoracotomy (Grade A). However, there is procedure-specific evidence that lumbar epidural hydrophilic strong opioid reduces pain compared with systemic analgesia

Clinical Practice

Infusion techniques are more convenient for use in clinical practice and adequate analgesia is more likely to be maintained than with bolus administration of lumbar epidural strong opioid
Unlike morphine, there is a ceiling effect to the respiratory depressant effect of nalbuphine; however, less morphine is required for comparable analgesia
Use of heparin or conventional NSAIDs may increase the risk of spinal haematoma due to epidural analgesia

Transferable Evidence

A meta-analysis of randomised controlled trials found that both continuous epidural infusion and PCEA analgesia provided superior postoperative analgesia compared with intravenous PCA analgesia Wu et al 2005
A meta-analysis found that continuous epidural infusion provided superior analgesia and reduced the incidence of pruritis compared with PCEA analgesia, but was associated with a higher incidence of PONV and motor block Wu et al 2005
A meta-analysis of randomised controlled trials found that epidural analgesia significantly decreased the incidence of pulmonary morbidity compared with systemic opioids Ballantyne et al 1998 Click here for more information
Epidural administration of strong opioids is associated with side effects including pruritis, PONV, urinary retention, and respiratory depression Chaney 1995

Thoracotomy-specific Evidence

Lumbar and thoracic epidural bupivacaine plus fentanyl were similar for the incidence of PONV (n=82) Sahin et al 1994
One study showed that morphine and nalbuphine, for intra-operative lumbar epidural injection, were associated with a similar incidence of nausea (five treatment arms; n=52) Baxter et al 1991
One study showed that lumbar epidural morphine was superior to all concentrations of lumbar epidural nalbuphine for reducing PCA fentanyl use (p<0.01) (five treatment arms; n=52) Baxter et al 1991
Lumbar epidural morphine was superior to all concentrations of lumbar epidural nalbuphine for reducing pain Baxter et al 1991 Click here for more information
Lumbar epidural fentanyl plus morphine significantly reduced the proportion of patients with pains (other than lower back pain, i.e. incisional pain, radiculopathy) (p=0.034) compared with epidural morphine alone (n=60) Radpay et al 2003
Lumbar epidural nalbuphine was superior to lumbar epidural morphine for reducing the proportion of patients requiring naloxone for PaCO₂>50 mmHg Baxter et al 1991 Click here for more information
One study showed that epidural infusion was associated with a higher PaCO₂ compared with epidural boluses of diamorphine (p=0.005) (n=22) Patrick et al 1991
High volume fentanyl for postoperative lumbar epidural analgesia was superior to low volume fentanyl for overall VRS pain scores at certain time points, levels of activity and incision points Thomson et al 1995 Click here for more information
High volume and low volume fentanyl, for postoperative lumbar epidural analgesia, were associated with similar use of ketorolac and fentanyl (n=66) Thomson et al 1995
High volume and low volume fentanyl, for postoperative lumbar epidural analgesia, were associated with a similar incidence of PONV (n=66) Thomson et al 1995
Thoracic and lumbar epidural hydrophilic strong opioid were associated with similar pulmonary function parameters: FVC, FEV₁ and PEFR (n=20) Grant et al 1993
In three of four studies, lumbar and thoracic epidural lipophilic strong opioid were associated with similar pulmonary function Guinard et al 1992 Click here for more information
Thoracic and lumbar epidural hydrophilic strong opioid were similar for the incidence of PONV (n=20) Grant et al 1993
Six studies showed that lumbar and thoracic epidural lipophilic strong opioid were similar for the incidence of PONV (n=29; n=52; n=32; n=23; n=30; n=22) Bouchard et al 1995 Click here for more information
Lumbar administration was superior to thoracic administration of epidural hydrophilic strong opioid for reducing epidural morphine use, but not epidural fentanyl use, during 0–24 h (p<0.05) (n=20) Grant et al 1993
In four of six studies, lumbar and thoracic epidural lipophilic strong opioid were similar for strong opioid use Bouchard et al 1995 Click here for more information
In five of six studies, lumbar epidural lipophilic strong opioid was similar to thoracic lipophilic epidural strong opioid for pain scores at rest Bouchard et al 1995 Click here for more information
For postoperative lumbar epidural injection, sufentanil 30, 50 and 75 µg were similar for pain scores 15 and 30 min after each dose, for the mean number of doses of sufentanil during 24 h, and for the mean duration of analgesia (three treatment arms; n=22) Whiting et al 1988
Two of three studies showed that lumbar epidural hydrophilic strong opioid was associated with superior pulmonary function to systemic hydrophilic strong opioid Shulman et al 1984 Click here for more information
Lumbar epidural hydrophilic strong opioid was superior to systemic hydrophilic strong opioid for reducing pain scores Miguel et al 1993 Click here for more information
In two of four studies, lumbar epidural lipophilic strong opioid was superior to systemic lipophilic strong opioid for reducing supplementary analgesic use Grant et al 1992 Click here for more information
In two of four studies, lumbar epidural lipophilic strong opioid was superior to systemic lipophilic strong opioid for reducing pain scores Baxter et al 1994 Click here for more information
Ten studies compared lumbar with thoracic epidural analgesia; scatter plots to illustrate the spread of average pain scores between studies show a trend towards increased pain scores with lumbar epidural analgesia Click here for more information
Epidural infusion of diamorphine did not significantly reduce pain scores during 0–18 h, or the total use of diamorphine, compared with epidural boluses of diamorphine (n=22) Patrick et al 1991
One study showed that epidural infusion and epidural boluses of diamorphine were similar for the incidence of PONV (n=22) Patrick et al 1991
Postoperative lumbar epidural fentanyl plus morphine was similar to lumbar epidural morphine alone for the incidence of PONV (n=60) Radpay et al 2003
Nalbuphine 10 mg and 20 mg lumbar epidural were not significantly different for pain scores at 1 and 2 h; 2/4 patients who received nalbuphine 10 mg, and 5/5 patients who received nalbuphine 20 mg were withdrawn from the study within 3 h due to inadequate analgesia (n=9) Etches et al 1991
For postoperative lumbar epidural injection, sufentanil 75 µg was associated with a greater PaCO₂ than sufentanil 30 µg (p<0.05) (three treatment arms; n=22) Whiting et al 1988
One of two studies showed that lumbar was inferior to thoracic epidural hydrophilic strong opioid for reducing pain scores Yang et al 1993 Click here for more information
Lumbar was inferior to thoracic epidural bupivacaine plus fentanyl, for reducing the epidural infusion rate during 0–8 h (p=0.028) but not at 12 or 16 h (n=46) Hurford et al 1993
In one of two studies, lumbar was inferior to thoracic administration of epidural bupivacaine plus fentanyl for reducing pain scores and for reducing the epidural infusion rate Sahin et al 1994 Click here for more information
Two studies showed that lumbar epidural hydrophilic strong opioid was similar to systemic hydrophilic strong opioid for the duration of hospital stay (n=30; n=30) Slinger et al 1995
One study showed no significant benefit of lumbar epidural hydrophilic strong opioid for reducing the incidence of PONV compared with systemic hydrophilic strong opioid (n=30) Shulman et al 1984
In three of four studies, lumbar epidural lipophilic strong opioid was associated with similar pulmonary function to systemic lipophilic strong opioid Grant et al 1992 Click here for more information
Lumbar epidural lipophilic strong opioid did not reduce the incidence of PONV compared with systemic lipophilic strong opioid in three of four studies Baxter et al 1994 Click here for more information
Study details Baxter et al 1994 Click here for more information
Lumbar epidural analgesia, miscellaneous studies
Lumbar versus thoracic epidural analgesia
Lumbar strong opioid versus systemic strong opioid

Epidural analgesia: adjuvants/other agents

PROSPECT Recommendations

Thoracic epidural corticosteroid is not recommended because there are limited data (Grade D)
Epidural epinephrine is recommended if a low dose of epidural LA and/or opioid is used (Grade B)

Clinical Practice

None cited

Transferable Evidence

Three studies in major thoracic or abdominal surgery showed that addition of epinephrine 1.5–2 µg/ml to thoracic epidural local anaesthetic plus strong opioid reduced pain intensity Niemi et al 1998 Click here for more information
Epinephrine 1.5–2 µg/ml was associated with reduced frequency of pruritis (p<0.002; n=36) Niemi et al 2003

Thoracotomy-specific Evidence

Addition of epinephrine to thoracic epidural strong opioid was superior to epidural strong opioid alone for reducing use of epidural fentanyl, and for extending the duration of analgesia Click here for more information
Postoperative epidural methylprednisolone did not significantly reduce pain scores at rest and on mobilisation, or total morphine requirement over 48 h, compared with placebo (n=24) Blanloeil et al 2001
Postoperative epidural methylprednisolone did not significantly reduce the incidence of PONV compared with placebo (n=24) Blanloeil et al 2001
Addition of epinephrine to postoperative thoracic epidural strong opioid did not significantly reduce pain scores compared with epidural strong opioid alone Baron et al 1996 Click here for more information
Addition of epinephrine to postoperative thoracic epidural strong opioid did not significantly reduce the incidence of PONV compared with epidural strong opioid alone (n=34; n=23) Baron et al 1996
Addition of epinephrine to postoperative thoracic epidural strong opioid was associated with similar pulmonary function to epidural strong opioid alone: FEV₁, FVC (n=34) PaCO₂ (n=23) Baron et al 1996
Study details Blanloeil et al 2001 Click here for more information

Spinal Analgesia

Spinal strong opioid

PROSPECT Recommendations

Spinal strong opioid should be administered only as a single pre-operative bolus dose for postoperative analgesia (grade A) (see Pre-operative Spinal Analgesia)
Repeated peri-operative doses by the spinal route are not recommended because they are not considered to be safe or practical (grade D)

Clinical Practice

Although one study shows that postoperative spinal fentanyl was superior to placebo or no treatment for VAS pain scores, the catheters used in the technique are no longer available
Spinal anaesthetics are generally administered as a single pre-operative bolus, and repeated peri-operative doses are not considered to be safe or practical

Transferable Evidence

Spinal administration of strong opioids is associated with side effects including pruritis, PONV, urinary retention, and respiratory depression Chaney 1995

Thoracotomy-specific Evidence

Postoperative spinal fentanyl was superior to placebo or no treatment for reducing VAS pain scores at 1 h (p<0.001) and during 2–10 h (p=0.03), and VRS pain scores at rest (p<0.001), on coughing (p<0.001) and on movement (p=0.025) (two control arms; n=30) Sudarshan et al 1995
Postoperative spinal fentanyl was superior to placebo or no treatment for reducing intravenous morphine use during 0–4 h (p=0.002), but not at 10 h (two control arms; n=30) Sudarshan et al 1995
Spinal morphine was similar to thoracic epidural bupivacaine plus fentanyl with or without systemic nimesulide or ibuprofen for reducing pain scores at rest or on coughing (n=35) McCrory et al 2002
Spinal morphine was similar to thoracic epidural bupivacaine plus fentanyl for reducing rescue morphine use (n=35) McCrory et al 2002
One study showed that spinal morphine was similar to thoracic epidural bupivacaine plus fentanyl for the incidence of PONV (n=35) McCrory et al 2002
Postoperative spinal fentanyl was associated with a greater PEFR than placebo or no treatment at 1 h (p<0.001) and 2–10 h (p=0.009) but there was no significant difference between groups for PaCO₂ at 30 min or up to 180 min (two control arms; n=30) Sudarshan et al 1995
Spinal morphine was associated with a similar PEFR to thoracic epidural bupivacaine plus fentanyl (n=35) McCrory et al 2002
Spinal morphine plus systemic nimesulide (but not ibuprofen) was superior to epidural bupivacaine plus fentanyl for reducing pain scores McCrory et al 2002 Click here for more information
Spinal morphine plus systemic nimesulide (but not ibuprofen) was associated with a higher PEFR than epidural bupivacaine plus fentanyl (p<0.001) (n=35) McCrory et al 2002
Study details Sudarshan et al 1995 Click here for more information

Analgesic Peripheral Nerve Blocks

Paravertebral block

PROSPECT Recommendations

Paravertebral block with LA, as a bolus followed by a continuous infusion for 2–3 days, is recommended, based on evidence that the technique provides comparable postoperative analgesia to thoracic epidural with LA (Grade A), and may be associated with fewer adverse effects (Grade A)
Paravertebral block cannot be recommended in preference to thoracic epidural with LA plus opioid, and vice versa, because of limited data
There is not enough evidence to recommend one LA in preference to another, or any particular concentration or volume of any LA

Clinical Practice

A paravertebral block can be used in combination with other analgesic techniques, as part of a multimodal analgesic regimen
Paravertebral LA may be administered as a bolus at the end of surgery
Paravertebral block is used less frequently than epidural analgesia in clinical practice

Transferable Evidence

Paravertebral block improved pain relief, reduced opioid use and was associated with improved pulmonary function compared with placebo in pleurectomy Mozell et al 1991
Four studies in breast surgery found that paravertebral block was associated with analgesic benefits compared with control Kairaluoma et al 2004 Click here for more information
Bilateral paravertebral block combined with general anaesthesia reduced pain scores, supplementary analgesic use, and the incidence of PONV, compared with general anaesthesia alone in laparoscopic cholecystectomy (n=60) Naja et al 2004

Thoracotomy-specific Evidence

Paravertebral bupivacaine was superior to interpleural bupivacaine plus wound infiltration for reducing pain scores on day 2 (p<0.05) but not on day 1 (n=40) Wedad et al 2004
Paravertebral bupivacaine plus fentanyl was associated with superior pulmonary function to thoracic epidural bupivacaine plus fentanyl for FEV₁ and FVC during 0–72 h (n=50) Bimston et al 1999
Paravertebral LA was associated with reduced frequency of postoperative complications compared with thoracic epidural LA Dhole et al 2001 Click here for more information
Fewer patients receiving paravertebral LA compared with thoracic epidural LA + opioid suffered pruritis (0 and 2 patients, respectively) or urinary retention (0 and 3 patients, respectively) in one study De Cosmo et al 2002
A systematic review comparing paravertebral and epidural blockade for thoracotomy found no significant difference for postoperative pain scores or morphine use, but showed a significant benefit of paravertebral block for reducing postoperative pulmonary complications and side-effects Davies et al 2006 Click here for more information
Paravertebral bupivacaine was similar to intercostal bupivacaine for pain scores at rest and on coughing, or supplementary morphine use during 0–48 h (n=30) Perttunen et al 1995
Paravertebral bupivacaine was similar to intercostal bupivacaine for the incidence of PONV (n=30) Perttunen et al 1995
Paravertebral bupivacaine was similar to intercostal bupivacaine for PaO₂, and FEV₁ (n=30) Perttunen et al 1995
Paravertebral and interpleural bupivacaine were similar for pain scores and morphine use during 0–48 h (n=45; n=11) Richardson et al 1995
Paravertebral bupivacaine was associated with superior pulmonary function to interpleural bupivacaine during 0–48 h: FVC, FEV₁ (p=0.001; n=45); FVC, FEV₁ (p<0.05; n=11) Richardson et al 1995
One of two studies showed that paravertebral bupivacaine was associated with superior pulmonary function to thoracic epidural bupivacaine plus fentanyl Kaiser et al 1998 Click here for more information
Paravertebral bupivacaine was superior to interpleural bupivacaine plus wound infiltration for reducing pethidine use (p<0.05) (n=40) Wedad et al 2004
Paravertebral bupivacaine was associated with superior pulmonary function (FVC, FEV₁, PEFR) to interpleural bupivacaine plus wound infiltration on days 1 and 2 (p<0.05) (n=40) Wedad et al 2004
One of two studies showed that paravertebral bupivacaine was superior to lumbar epidural morphine for reducing pain scores Richardson et al 1993 Click here for more information
Paravertebral bupivacaine was superior to lumbar epidural morphine for reducing the incidence of vomiting (p<0.05) but not the incidence of nausea (n=20) Richardson et al 1993
Paravertebral bupivacaine was associated with similar pulmonary function to lumbar epidural morphine (PEFR, FEV₁ and FVC) (n=20) Richardson et al 1993
For paravertebral block, lidocaine and bupivacaine were associated with similar pulmonary function: PEFR, FVC and FEV₁ (n=43) Barron et al 1999
For paravertebral block, lidocaine and bupivacaine were similar for pain scores and morphine use: during 0–72 h (n=43) Barron et al 1999
Continuous infusion and intermittent boluses of postoperative paravertebral bupivacaine were associated with similar PaO₂ (n=30) Catala et al 1996
A postoperative infusion of paravertebral bupivacaine was superior to intermittent boluses for reducing pain scores at rest and on movement at 4, 10, 20 and 48 h (p=0.003), but not at 0 and 1 h (n=30) Catala et al 1996
Studies comparing paravertebral LA versus thoracic epidural LA plus strong opioid showed mixed results for postoperative pain scores Kaiser et al 1998 Click here for more information
For paravertebral block, ropivacaine 0.5% was associated with greater PEFR compared with ropivacaine 0.25% and 0.375% (p=0.004) (n=60) Gamal et al 2003
For paravertebral block, ropivacaine 0.5% was superior to 0.25% and 0.375% for reducing the incidence of PONV (p<0.05) (n=60) Gamal et al 2003
Six studies plus one arm of a study, out of eight studies, showed that paravertebral bupivacaine was superior to control for reducing pain scores (systemic analgesia available to all patients) Barron et al 1999 Click here for more information
Seven studies showed that paravertebral bupivacaine was superior to control for reducing supplementary analgesic use (systemic analgesia available to all patients) Barron et al 1999 Click here for more information
Five studies, plus two arms from a further two studies, showed that paravertebral bupivacaine was associated with superior pulmonary function to control (systemic analgesia available to all patients) Barron et al 1999 Click here for more information
For paravertebral block, ropivacaine 0.5% was superior to 0.25% and 0.375% for reducing the use of supplementary morphine during 0–24 h (p=0.008) and 24–48 h (p=0.005) (n=60) Gamal et al 2003
For paravertebral block, ropivacaine 0.5% was superior to 0.25% (p=0.001) and 0.375% (p=0.02) for reducing pain scores at rest and on coughing during 0–48 h (n=60) Gamal et al 2003
For paravertebral block, lidocaine and bupivacaine were associated with a similar duration of hospital stay (n=43) Barron et al 1999
Paravertebral LA and thoracic epidural LA were similar for reducing postoperative pain Dhole et al 2001 Click here for more information
Paravertebral and thoracic epidural administration of LA were not significantly different for the incidence of persistent chest pains at 6 months (n=95) Richardson et al 1999
Paravertebral bupivacaine was similar to thoracic epidural bupivacaine for supplementary analgesic requirements Dhole et al 2001 Click here for more information
Paravertebral LA was similar to thoracic epidural LA plus strong opioid for reducing supplementary analgesic use De Cosmo et al 2002 Click here for more information
Paravertebral bupivacaine plus fentanyl was similar to thoracic epidural bupivacaine plus fentanyl for the number of requests for analgesia (n=50) Bimston et al 1999
One of two studies showed that paravertebral bupivacaine reduced the incidence of PONV compared with thoracic epidural bupivacaine (no p-value) (n=95) Richardson et al 1999 Click here for more information
Paravertebral bupivacaine (± fentanyl) was similar to thoracic epidural bupivacaine plus fentanyl for the incidence of PONV (n=50, n=50) Bimston et al 1999
Paravertebral bupivacaine was similar to thoracic epidural bupivacaine for the duration of hospital stay (n=95) Richardson et al 1999
Paravertebral bupivacaine (± fentanyl) was similar to thoracic epidural bupivacaine plus fentanyl for the duration of intensive care unit or hospital stay (n=50; n=50; n=30) Bimston et al 1999
Two of four studies showed that paravertebral bupivacaine was associated with superior pulmonary function to thoracic epidural bupivacaine Richardson et al 1995 Click here for more information
Continuous infusion did not significantly reduce morphine use compared with intermittent boluses of postoperative paravertebral bupivacaine (n=30) Catala et al 1996
In one of two studies, paravertebral bupivacaine was inferior to peri-operative lumbar epidural morphine for reducing cumulative morphine use at 14–70 h (p<0.05) (n=72) Richardson et al 1993 Click here for more information
Paravertebral LA plus strong opioid increased pain scores compared with thoracic epidural LA plus strong opioid at 8 h (p=0.007), 16 h (p=0.023), 24 h (p=0.02), and 32 h (p=0.033), but not at 0 or 40–96 h (n=50) Bimston et al 1999
Paravertebral bupivacaine did not significantly reduce the duration of hospital stay compared with control (systemic analgesia available to all patients) (two treatment arms; n=63) Barron et al 1999
Paravertebral bupivacaine did not significantly reduce the incidence of PONV compared with control (systemic analgesia available to all patients) (n=20) Carabine et al 1995
Study detail Richardson et al 1994 Click here for more information
Paravertebral block, miscellaneous studies
Paravertebral block versus other regional analgesic techniques
Paravertebral block versus control
Most studies of paravertebral block included postoperative administration, with or without an additional pre- or intra-operative bolus; all results (with the exception of Richardson et al 1994

Intercostal nerve block

PROSPECT Recommendations

Intercostal nerve block with LA is recommended, if epidural analgesia and paravertebral block are not possible, as an infusion via a catheter, continued for 2–3 days postoperatively, which should be preceded by an intra-operative bolus dose (Grade D)
There is not enough evidence to recommend one LA in preference to another, or any particular concentration or volume of any LA

Clinical Practice

Infusion techniques for intercostal nerve block are more convenient for use in clinical practice and adequate analgesia is more likely to be maintained than with intermittent bolus administration

Transferable Evidence

Two studies in open cholecystectomy and one study in upper abdominal surgery showed that intercostal nerve block did not significantly reduce pain scores compared with control (n=37, n=40, n=66) Maidatsi et al 1998
In open cholecystectomy, intercostal nerve block reduced supplementary opioid use compared with control in one study (n=37) Maidatsi et al 1998
A meta-analysis found that intercostal nerve block tended to reduce the incidence of pulmonary complications compared with systemic opioids, but these differences did not achieve statistical significance. There were no significant differences in surrogate measures of pulmonary function (FEV₁, FVC, and PEFR) Ballantyne et al 1998
The incidence of pneumothorax following intercostal nerve block in thoracic and upper abdominal surgery has been reported in the range of 0.073% to 19% Shanti et al 2001

Thoracotomy-specific Evidence

Two of three studies showed that repeat dose intercostal bupivacaine was superior to control for reducing pain scores Bachmann-Mennenga et al 1993 Click here for more information
Continuous infusion intercostal LA was superior to control for reducing pain scores during 0–24 h or during 24–48 h (p=0.03), whenever bupivacaine was administered (n=20) Dryden et al 1993
Two of three studies showed that repeat dose intercostal bupivacaine reduced supplementary analgesic requirements compared with control Bachmann-Mennenga et al 1993 Click here for more information
Continous infusion intercostal LA reduced morphine use compared with control, during 0–24 h or 24–48 h (p=0.04), whenever bupivacaine was administered (n=20) Dryden et al 1993
Repeat dose intercostal bupivacaine was associated with superior maintenance of pulmonary function compared with control Bachmann-Mennenga et al 1993 Click here for more information
Repeat dose intercostal LA was associated with similar postoperative pain scores to thoracic epidural analgesia Asantila et al 1986 Click here for more information
Three of three studies showed that repeat dose intercostal LA was similar to thoracic epidural analgesia for supplementary analgesic use (n=30; n=20; n=29) Asantila et al 1986 Click here for more information
Continuous infusion intercostal LA was similar to thoracic epidural analgesia for postoperative fentanyl use (n=47) Debreceni et al 2003
Repeat dose intercostal LA was associated with similar pulmonary function to thoracic epidural analgesia Asantila et al 1986 Click here for more information
Continuous infusion intercostal LA was associated with similar pulmonary function (PEFR, FVC, FEF, and FEV₁/FVC) to thoracic epidural analgesia (n=47) Debreceni et al 2003
Continuous infusion intercostal bupivacaine was similar to thoracic epidural analgesia for the incidence of PONV (n=47) Debreceni et al 2003
Three studies showed that repeated intercostal bupivacaine injections were associated with similar pulmonary function to single intra-operative injections Asantila et al 1986 Click here for more information
Two of three studies showed that repeated intercostal bupivacaine injections were similar to single intra-operative injections for reducing supplementary analgesic use (n=20; n=22) Scheinin et al 1987b Click here for more information
Repeat intercostal bupivacaine injections, pre- or intra- plus postoperatively, were similar to single intra-operative injections for reducing pain scores (n=20; n=21; n=22) Asantila et al 1986 Click here for more information
Continuous infusion intercostal LA was inferior to thoracic epidural analgesia for reducing pain scores at 4, 8, and 12 h (p<0.05) but not at 16 and 20 h (n=47) Debreceni et al 2003
Limited data comparing intercostal LA with thoracic epidural analgesia could be meta-analysed; however, a scatter plot shows that both techniques were associated with similar postoperative pain scores at 1 h, although pain scores tended to be marginally higher at day 1/24 h following intercostal LA Click here for more information
Continuous infusion intercostal bupivacaine was similar to control for the incidence of nausea (n=20) Dryden et al 1993
Study detail Bachmann-Mennenga et al 1993 Click here for more information
Intercostal nerve block versus other regional techniques
Intercostal nerve block, repeat versus single injection
Intercostal nerve block versus control
Studies of intercostal LA included single dose, repeated dose, and continuous infusion regimens. Single dose, repeat dose, and continuous infusion regimens have been analysed separately; all results for repeat dose and continuous infusion studies are listed below

Interpleural nerve block

PROSPECT Recommendations

Interpleural LA is not recommended due to lack of efficacy (Grade A) as well as potential toxicity associated with high absorption of LA (Grade D)

Clinical Practice

Interpleural nerve blocks are associated with a high absorption of LA, because of the large surface area exposed to the drug

Transferable Evidence from other Procedures

None cited

Thoracotomy-specific Evidence

Postoperative interpleural bupivacaine was associated with similar spirometric test results (FVC, FEV₁) during days 0–5 and at 12 weeks compared with intra-operative cryoanalgesia plus intravenous morphine (n=24) Miguel et al 1993
Postoperative interpleural bupivacaine was similar to intra-operative cryoanalgesia plus intravenous morphine for supplementary analgesic use (n=24) Miguel et al 1993
Postoperative interpleural bupivacaine was superior to intra-operative cryoanalgesia for reducing the proportion of patients who required papaveretum (no p value), but not the mean dose of papaveretum or the number of requests for oral analgesia (n=31) Shafei et al 1990
Postoperative interpleural bupivacaine was similar to intra-operative cryoanalgesia for average pain scores (n=31) Shafei et al 1990
Postoperative interpleural bupivacaine was associated with similar blood gas levels to postoperative intercostal bupivacaine (n=20) Bachmann-Mennenga et al 1993
Postoperative interpleural bupivacaine was superior to intra-operative intercostal bupivacaine for reducing papaveretum use (p=0.0012), the proportion of patients requiring papaveretum (no p value) and the number of requests for oral analgesia (p=0.0047) (n=32) Shafei et al 1990
Postoperative interpleural bupivacaine was similar to intra-operative intercostal bupivacaine for average pain scores (n=32) Shafei et al 1990
Interpleural bupivacaine was similar to paravertebral bupivacaine for pain scores and opioid use during 0–48 h (n=45; n=11) Richardson et al 1995
Postoperative interpleural bupivacaine and lumbar epidural morphine were associated with similar pulmonary function (FVC and FEV₁) (n=20) Miguel et al 1993
Interpleural block plus wound infiltration was associated with superior pulmonary function compared with thoracic epidural bupivacaine on day 1 (p<0.05) but on day 2 thoracic epidural bupivacaine was superior to interpleural block plus wound infiltration (p<0.05) (n=40) Wedad et al 2004
Interpleural bupivacaine was associated with similar levels of blood gases compared with thoracic epidural bupivacaine (n=30) Bachmann-Mennenga et al 1993
Use of both paravertebral and lateral catheters for administration of postoperative interpleural bupivacaine was superior to use of a single paravertebral catheter for reducing supplementary opioid use Ferrante et al 1991 Click here for more information
Bupivacaine 0.25% and 0.5% for interpleural injection were similar for pain scores after each injection, and for supplementary analgesic use (n=18) Elman et al 1993
Postoperative interpleural bupivacaine was associated with superior pulmonary function compared with control Bachmann-Mennenga et al 1993 Click here for more information
Postoperative interpleural bupivacaine was superior to control for extending the duration of analgesia: after the 2nd injection (p<0.005), but not the 1st injection (n=15) Symreng et al 1989
In five of ten studies, postoperative interpleural LA was superior to control for reducing supplementary analgesic use Broome et al 1993 Click here for more information
Five of ten studies showed that postoperative interpleural LA was superior to control (systemic analgesia available to all patients) for reducing pain Broome et al 1993 Click here for more information
Interpleural bupivacaine was associated with inferior pulmonary function compared with paravertebral bupivacaine: FVC and FEV₁ during days 0–2 (p<0.003) (n=45) FVC at 12 and 36 h (p<0.05), FEV₁ at 12, 24, 36 and 48 h (p<0.05) (n=11) Richardson et al 1995
Interpleural bupivacaine was inferior to intercostal bupivacaine for reducing pain scores (no p-value) and supplementary buprenorphine use (p<0.01) (n=20) Bachmann-Mennenga et al 1993
Interpleural bupivacaine plus wound infiltration was associated with inferior pulmonary function compared with paravertebral bupivacaine (FVC, FEV₁, PEFR) on days 1 and 2 (p<0.05) (n=40) Wedad et al 2004
Addition of epinephrine to the LA solution for interpleural nerve block was of no significant benefit over LA alone, for extending the duration of analgesia (n=24) Kambam et al 1989
Postoperative interpleural bupivacaine was inferior to intra-operative cryoanalgesia plus intravenous morphine for reducing pain scores and persistent postthoracotomy pain Miguel et al 1993 Click here for more information
For administration of postoperative interpleural bupivacaine, use of both paravertebral and lateral catheters was of no significant benefit over use of a single paravertebral catheter for reducing pain scores at 0–4 h (n=14) Ferrante et al 1991
Addition of epinephrine to the LA solution for interpleural nerve block was of no significant benefit over LA alone, for reducing pain scores (n=14; n=24) Ferrante et al 1991
Addition of epinephrine to the LA solution for interpleural nerve block was of no significant benefit over LA alone, for reducing opioid use or the frequency of interpleural injections during 24 h (n=14) Ferrante et al 1991
Interpleural bupivacaine plus wound infiltration was inferior to paravertebral bupivacaine for reducing pain scores on day 2 (p<0.05) but not on day 1, and for reducing pethidine use (p<0.05) (n=40) Wedad et al 2004
Two studies showed that interpleural LA was inferior to lumbar epidural strong opioid for reducing pain scores Miguel et al 1993 Click here for more information
Interpleural block plus wound infiltration was inferior to thoracic epidural bupivacaine for reducing pain scores on day 2 (p<0.05) but not on day 1, and for reducing pethidine use (p<0.05) (n=40) Wedad et al 2004
One of two studies showed that interpleural bupivacaine was inferior to thoracic epidural bupivacaine for reducing pain scores (no p value) and buprenorphine use (p<0.01) (n=30) Brockmeier et al 1994 Click here for more information
Interpleural bupivacaine increased persistent postthoracotomy pain at 12 weeks (p<0.007) compared with intravenous morphine (n=21) Miguel et al 1993
One study showed that postoperative interpleural morphine and intravenous morphine were associated with similar pulmonary function (FEV₁) and blood gas levels (n=17) Welte et al 1992
Postoperative interpleural morphine and intravenous morphine were similar for pain scores and supplementary morphine use at 0–6 h, 22–24 h and at 1 week (n=17) Welte et al 1992
Postoperative interpleural bupivacaine did not significantly reduce the duration of hospital stay compared with control (n=16; n=19; n=40) Raffin et al 1994
Study detail Bachmann-Mennenga et al 1993 Click here for more information
Interpleural nerve block versus other regional analgesic techniques
Interpleural nerve block, miscellaneous studies
Interpleural nerve block versus control

Alternate Analgesic Techniques

Acupuncture

PROSPECT Recommendations

Auricular acupuncture is not recommended due to limited evidence (Grade D)

Clinical Practice

None cited

Transferable Evidence from other Procedures

None cited

Thoracotomy-specific Evidence

Auricular acupuncture significantly reduced pain scores on coughing on days 1–5 (p<0.05), and supplementary analgesic requirements (no statistical analysis), compared with no acupuncture (n=36) Wang et al 1988
Auricular acupuncture was associated with superior pulmonary function to no acupuncture: the negative inspiratory force on days 1, 3, 4, and 5 (p<0.05), FVC and FEV₁ on days 3 and 6 (p<0.05) (n=36) Wang et al 1988
Study details Wang et al 1988 Click here for more information

TENS

PROSPECT Recommendations

TENS is not recommended because of limited evidence (Grade D)

Clinical Practice

TENS could be considered as an adjuvant to pharmacological pain management, for those with sufficient expertise in the technique

Transferable Evidence from other Procedures

A metaanalysis of randomised, placebo-controlled trials, found that TENS, when administered with a strong, subnoxious intensity at an adequate frequency in the wound area, significantly reduced analgesic use after various types of surgery (including abdominal, gynaecological, and thoracic procedures) Bjordal et al 2003
TENS reduced pain scores after minor rib fracture on days 1 and 3 (p<0.05), but not day 0, compared with NSAID, or NSAID plus inactive TENS, or control (placebo tablet, no TENS) (n=100) Oncel et al 2002
TENS reduced pain scores during activity (p<0.05), but not at rest, compared with placebo TENS or no TENS, after abdominal surgery (n=30) Rakel et al 2003

Thoracotomy-specific Evidence

One of two studies showed that TENS significantly reduced pain scores compared with sham TENS on day 1 (p=0.014) but not on day 2 (n=24) Warfield et al 1985
TENS significantly increased tolerance of deep coughing and percussion on day 1 (p=0.018) and day 2 (p=0.0064) compared with sham TENS (n=24) Warfield et al 1985
One of two studies showed that TENS significantly reduced requirement for anti-emetics compared with no TENS (n=40) Stubbing et al 1988
One of two studies showed that TENS significantly reduced the duration of intensive care unit stay compared with sham TENS (p=0.013) (n=24) Warfield et al 1985
TENS did not significantly reduce supplementary analgesic requirements compared with no TENS: during 0–24 h (n=40) Stubbing et al 1988
TENS did not significantly extend the time to oral analgesia compared with no TENS (n=40) Stubbing et al 1988
One study showed that TENS was not associated with an increased PEFR compared with no TENS (n=40) Stubbing et al 1988
Study details Stubbing et al 1988 Click here for more information

Analgesia for Shoulder Pain

Nerve block

PROSPECT Recommendations

Suprascapular nerve block is not recommended because of limited evidence (grade D)

Clinical Practice

Suprascapular nerve block is not associated with impaired respiratory function, in contrast to the phrenic nerve block

Transferable Evidence from other Procedures

None cited

Thoracotomy-specific Evidence

Postoperative suprascapular nerve block with bupivacaine was superior to placebo for reducing shoulder pain scores at 4 h (p<0.05) but not at any other time during 0–6 h (n=30) Tan et al 2002
Postoperative suprascapular nerve block with bupivacaine was not significantly different from placebo for the amount of epidural solution used (n=30) Tan et al 2002
Study details Tan et al 2002 Click here for more information

Analgesia for Chest Tube Removal

Ice pack

PROSPECT Recommendations

An ice pack is not recommended because of lack of analgesic benefit (Grade A)

Clinical Practice

None cited

Transferable Evidence from other Procedures

None cited

Thoracotomy-specific Evidence

For chest tube removal, an ice pack did not significantly reduce pain scores compared with a tepid water pack (n=50) Sauls 2002
Study details Sauls 2002 Click here for more information

Interpleural nerve block

PROSPECT Recommendations

Interpleural local anaesthestic is not recommended because of lack of analgesic benefit (Grade A)

Clinical Practice

None cited

Transferable Evidence from other Procedures

None cited

Thoracotomy-specific Evidence

Interpleural bupivacaine was not superior to placebo for reducing pain scores for intensity, sensation, affect or distress on chest tube removal Puntillo 1996 Click here for more information
Study details Puntillo 1996 Click here for more information

Topical agent

PROSPECT Recommendations

Topical LA is not recommended, based on limited evidence (Grade D)

Clinical Practice

Topical LA should be applied 3 h before removal to achieve full analgesic benefit

Transferable Evidence from other Procedures

None cited

Thoracotomy-specific Evidence

Topical lidocaine/prilocaine cream was superior to intravenous morphine for reducing pain scores during chest tube removal (p<0.01), but not after removal (n=52) Valenzuela et al 1999
Study details Valenzuela et al 1999 Click here for more information