Professional Documents
Culture Documents
Review
h i g h l i g h t s
Efficacy of bupivacaine or ropivacaine for pain relief in breast cancer surgery was evaluated.
Analgesic consumption was not significantly different between experimental and control groups.
Bupivacaine or ropivacaine infiltration caused less pain at 2 h postoperatively.
No difference in postoperative pain reduction at 1, 12, and 24 h between two groups.
a r t i c l e i n f o a b s t r a c t
Article history:
Background: Although not completely painless, breast-conserving surgery is considerably less painful
Received 23 June 2015
than modified radical mastectomy. Local anesthetics are speculated to reduce postoperative pain when
Received in revised form
22 July 2015
placed at the surgical site. Thus, we conducted a systematic review of randomized controlled trials to
Accepted 29 July 2015 evaluate the efficacy of bupivacaine or ropivacaine analgesia for pain relief in breast cancer surgery.
Available online 12 August 2015 Methods: PubMed, Embase, the Cochrane Library, Scopus, and the ClinicalTrials.gov registry were
searched for studies published up to July 2015. Individual effect sizes were standardized, and a meta-
Keywords: analysis was performed to calculate a pooled effect size by using random effects models. Pain was
Breast cancer assessed using a visual analog scale at 1, 2, 12, and 24 h postoperatively. The secondary outcomes
Bupivacaine included complications and analgesic consumption.
Local anesthesia Results: We reviewed 13 trials with 1150 patients. We found no difference in postoperative pain
Meta-analysis reduction at 1, 12, and 24 h after breast cancer surgery between the experimental and control groups. The
Postoperative pain severity of pain was significantly reduced in the experimental group (weighted mean difference 0.19;
Ropivacaine
95% confidence interval: 0.39e0.00) at 2 h postoperatively. Moreover, postoperative analgesic con-
Wound infiltration
sumption did not differ significantly between the groups. No major drug-related complication was
observed in any study.
* Corresponding author. Division of General Surgery, Department of Surgery, Taipei Medical University, Shuang Ho Hospital, 291, Zhongzheng Road, Zhonghe District, New
Taipei City, 23561, Taiwan.
** Corresponding author. Division of General Surgery, Department of Surgery, Taipei Medical University, Shuang Ho Hospital, 291, Zhongzheng Road, Zhonghe District, New
Taipei City, 23561, Taiwan.
E-mail addresses: kelvintam@h.tmu.edu.tw (W.-Y. Wang), 13259@s.tmu.edu.tw (C.-H. Wu).
1
Wan-Yu Wang and Chih-Hsiung Wu contributed equally to this work.
http://dx.doi.org/10.1016/j.ijsu.2015.07.715
1743-9191/© 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
80 K.-W. Tam et al. / International Journal of Surgery 22 (2015) 79e85
Conclusion: Administration of the local anesthetics bupivacaine or ropivacaine during breast cancer
surgery decreased pain significantly at only 2 h but did not reduce pain at 12, and 24 h
postoperatively.
© 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2. Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.1. Inclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.2. Search strategy and study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.3. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
2.4. Methodological quality appraisal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
2.5. Outcomes and statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.1. Characteristics of the trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.2. Pain score . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
3.2.1. Local wound infiltration versus control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
3.2.2. Preemptive versus postoperative local infiltration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.3. Analgesic consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.4. Complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Ethical approval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Sources of funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Author contribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Guarantor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Breast cancer is the most common female malignancy in 2.1. Inclusion criteria
Western countries [1]. Surgical resection with or without axillary
lymph node dissection (ALND), followed by adjuvant radiation To be included in the analysis, studies had to be RCTs evaluating
therapy and chemotherapy, has greatly improved the oncological the outcome of wound infiltration with local analgesics in breast
outcome of breast cancer [2]. In the recent decade, surgery evolu- cancer surgery. In addition, studies had to clearly report the in-
tion from modified radical mastectomy to breast-conserving sur- clusion and exclusion criteria for patients, the anesthetic technique,
gery has emphasized cosmetic advantages and the concept of small the surgical technique used to treat breast cancer, and the defini-
wounds and pain reduction. Reducing postoperative pain following tion and evaluation of postoperative pain. We excluded trials that
breast cancer surgery to achieve a more rapid recovery and shorten met at least one of the following criteria: (1) the patients had not
the hospital stay has become a crucial objective. received partial or modified radical mastectomy for breast cancer,
Traditionally, postoperative pain could be reduced by taking such as in studies that only included breast tumor sampling; (2) the
narcotics or nonsteroidal anti-inflammatory drugs; however, these patients had undergone non-cancer-related breast surgery; (3) the
drugs can cause adverse effects such as nausea, vomiting, and clinical outcomes had not been clearly stated; or (4) duplicate
dyspepsia [3]. Alternative analgesic regimens in the peri- and reporting of patient cohorts had occurred.
postoperative periods may reduce unwanted side effects. Surgical
wound infiltration with a local anesthetic solution is currently 2.2. Search strategy and study selection
performed in many surgical procedures, including abdominal hys-
terectomy, cesarean section, and inguinal hernia repair [4,5]. Relevant studies published before the end of July 2015 were
Bupivacaine is an ideal choice for local analgesia in breast cancer identified by conducting a computer search of the PubMed,
surgery through instillation in the dissection space on the basis of its Embase, Scopus, and Cochrane databases. The following Medical
characteristic long-acting efficiency in similar applications. More- Subject Headings search headings were used: breast cancer or
over, ropivacaine is a long-acting local anesthetic that is structurally carcinoma or neoplasm, mastectomy or lumpectomy or surgery or
related to bupivacaine. Recent data show that ropivacaine produces excision, bupivacaine or ropivacaine or carbostesin or marcaine or
fewer central nervous system and cardiovascular side effects sensorcaine or vivacaine, and local anesthesia. The “related articles”
compared with bupivacaine [6]. Several published randomized function in PubMed was used to broaden the search, and all ab-
controlled trials (RCTs) have investigated the efficacy of intra- stracts, studies, and citations retrieved were reviewed. In addition,
operative instillation of bupivacaine or ropivacaine along the surgical we attempted to identify other studies by searching the reference
wound of mastectomy; however, the results were inconclusive [7]. sections of relevant papers and contacting known experts in the
Therefore, we investigated patient outcomes after wound infiltration field. Finally, unpublished studies were sought using the
with local analgesics in breast cancer surgery by conducting a sys- ClinicalTrials.gov registry (http://clinicaltrials.gov/). No language
tematic review and meta-analysis of the evidence available to date. restrictions were applied. The systematic review described herein
K.-W. Tam et al. / International Journal of Surgery 22 (2015) 79e85 81
was accepted by the online PROSPERO international prospective Studies identified through Additional studies identified through
register of systematic reviews of the National Institute for Health
PubMed, Embase, Cochrane, Scopus and by searching references
Research (CRD42015017030).
databases (n = 755) (n = 4)
Table 1
Characteristics of the selected randomized controlled trials.
Albi-Feldzer [2013] 4 hospitals Conservative surgery with ALND, R: 119 R: 57 (47e65) R: 0.375% R 3 mg/kg mixed with saline
MRM with or without ALND C: 119 C: 56 (48e68) C: Saline solution
Also injection in 2nd & 3rd intercostal spaces and humeral
insertion of pectoralis
Baudry [2008] 1 hospital MRM or partial mastectomy R: 40 R: 52.4 ± 11.2 R: 4.75 mg/mL R 40 mL
with ALND C: 38 C: 57.7 ± 12.6 C: NS 40 mL
Campbell [2014] 1 hospital, Mastectomy, wide local excision or B: 45 B: 60.6 B: 0.25% B 20 mL with or without adrenaline in breast þ 0.25% B
4 surgeons lumpectomy with or without ALND C: 45 C: 60.4 20 mL with adrenaline into axilla
C: Nil
Johansson [2003] 1 hospital Partial mastectomy with or R1: 15 R1: 52 ± 7 R1: 0.375% R 0.3 mL/kg
without ALND R2: 15 R2: 58 ± 9 R2: 0.375% R 0.3 mL/kg þ fentayl 0.5 g/kg
C: 15 C: 54 ± 11 C: Nil
Johansson [2000] 1 hospital Partial mastectomy with or R: 30 R: 54 ± 11 R: R 3.75 mg/mL in breast þ 3.75 mg/mL in axilla before surgery
without ALND C: 29 C: 55 ± 9 C: NS 0.3 mL/kg in breast þ 0.3 mL/kg in axilla before surgery
Jonnavithula [2015] 1 hospital MRM B: 20 C: 45.7 ± 10.5 B: 0.25% B 20 mL into chest wall and axillary drain respectively
C1: 20 C: 44.5 ± 10.5 then clamped 20 min
C2: 20 C2: 47 ± 10.9 C1: NS 20 mL
C2: Nil
Mohamed [2013] Not reported MRM with ALND B1: 35 B1: 39.5 ± 5.9 B1: 0.5% B 5 mL þ clonidine 250 mg diluted with NS to 15 mL
B2: 35 B2: 40.3 ± 5.5 B2: 0.5% B 5 mL þ clonidine 150 mg diluted with NS to 15 mL
B3: 35 B3: 40.1 ± 4.4 B3: 0.5% B 5 mL diluted with NS to 15 mL
C: 35 C: 38.9 ± 6.0 C: NS
Petersson [2001] 1 hospital Mastectomy and ALND B: 29 Not reported B: 0.5% B 20 mL in skin and subcutaneous tissue
L/P: 31 L/P: Topical application of lidocaine/prilocaine mixture 10 g
C: 30 (Emla cream, Astra, Sweden)
C: Nil
Rica [2007] 1 hospital Mastectomy and ALND R1: 15 R1: 58.5 ± 10.5 R1: Preemptive local infiltration 0.2% R 20 mL diluted with NS to
R2: 15 R2: 53.1 ± 11.5 80 mL
R2: Postoperative local infiltration 0.2% R 20 mL diluted with NS
to 80 mL
Talbot [2004] 1 hospital MRM B: 19 Not reported B: 0.5% B 20 mL into the axillary drain at the end of surgery and
C: 21 q4h 6 doses for 24 h postoperatively
C: NS
Vallejo [2006] 1 hospital Segmental mastectomy B1: 30 B1: 47 ± 10 B1: Pre þ post wound infiltration 0.5% B10 mL
without ALND B2: 30 B2: 52 ± 11 B2: Preincisional 0.5% B10 mL þ postoperative NS infiltration
B3: 30 B3: 48 ± 12 B3: Preincisional NS þ postoperative 0.5% B10 mL infiltration
C: 30 C: 51 ± 12 C: Pre þ post wound infiltration NS10 mL
Vigneau [2011] 1 hospital Mastectomy or lumpectomy R: 24 R: 58 ± 13 R: R 7.5 mg/mL solution 20 mL
with ALND C: 24 C: 50 ± 11 C: NS 20 mL
Zielinski 1 hospital, MRM B: 57 B: 58.8 (31e82)y B: Preemptive local infiltration 100 mg B diluted with NS 40 mL
[2011] 5 surgeons C: 55 C: 60.1 (25e83) C: NS
ALND: axillary lymph node dissection; B: bupivacaine; C: control; L/P: lidocaine/prilocaine; MRM: modified radical mastectomy; NS: normal saline; R: ropivacaine.
Data are presented as mean ± standard deviation, except where y indicates median (range).
3.2. Pain score visual analog scale (0 ¼ no pain, 10 ¼ worst possible pain)
[14e16,19e21,23,26], whereas in the other 5, it was assessed using
3.2.1. Local wound infiltration versus control a 0e100-mm scale [17,18,22,24,25]. To compare the outcome
In 8 studies, postoperative pain was assessed using a 10-point measures, we converted all scales to a 10-point scale. Because of
Table 2
Methodological quality assessment of the selected trials.
Study [Year] Country Allocation generation Allocation Double Data Loss to Other bias
concealment blinding analysis follow-up
Albi-Feldzer [2013] France Computer generated Adequate Yes ITT 7.2% Low risk
Baudry [2008] France Random numbers table Adequate Yes PP 3.7% Undefined risk
Campbell [2014] New Zealand Sealed envelope Unclear Patients ITT/PP 12.2% Low risk
blinded
Johansson [2003] Sweden Unclear Unclear Yes PP 8.8% Perioperative analgesia
was not uniform
Johansson [2000] Sweden Unclear Adequate Yes PP 1.6% Preoperative local injection
Jonnavithula [2015] India Computer generated Adequate Yes ITT 0% Low risk
Mohamed [2013] Egypt Computer generated Adequate Yes PP 6.6% Low risk
Pettersson [2001] Sweden Sealed envelope Unclear Yes ITT 0% No patient's characteristic
reporting
Rica [2007] Malaysia Unclear Unclear Yes ITT 1.6% No control group
Talbot [2004] UK Sealed envelope Adequate Yes PP 5% No patient's characteristic
reporting
Vallejo [2006] USA Computer generated Adequate Yes Unclear Unclear Undefined risk
Vigneau [2011] France Random numbers table Adequate Yes PP 8.0% Low risk
Zielinski [2011] Poland Random numbers table Adequate Yes PP 5.6% Low risk
-4 -2 0 2 4
Favours experimental Favours control
Test for subgroup differences: Chi² = 4.62, df = 3 (P = 0.20), I² = 35.1%
Fig. 2. Forest plot of comparison: experimental drug infiltration versus control; outcome: incidence of postoperative pain at 1, 2, 12, and 24 h postoperatively.
different timings in the pain score assessment, we compared these were 0.26 (95% CI: 0.68 e 0.17) at 1 h, 0.00 (95% CI: 0.15 e 0.16) at
outcome measures at 1, 2, 12, and 24 h postoperatively. We used 12 h, and 0.04 (95% CI: 0.13 e 0.21) at 24 h postoperatively; no
data obtained from 8 of these RCTs in our analysis but excluded 5 significant difference was observed in the postoperative pain scores
because the studies provided only medians and ranges [23,26], between the groups (Fig. 2). However, the severity of pain was
mean visual analog scores without standard deviations [19,20], or significantly reduced in the experimental group, and the pooled mean
the 25th and 75th centiles for the pain scores [21]. We used only the difference was 0.19 (95% CI: 0.39 e 0.00) at 2 h postoperatively. The
results of ropivacaine or bupivacaine groups versus control groups I2 value ranged from 19 to 43% at various time periods postoperatively,
for the meta-analysis. Combined drug groups, such as clonidine with indicating low to moderate heterogeneity among studies (Fig. 2).
bupivacaine and fentanyl with ropivacaine, were excluded before The studies that we did not include in the meta-analysis showed
data pooling. The pooled mean differences in the pain score various results. Pettersson et al. compared postoperative pain after
84 K.-W. Tam et al. / International Journal of Surgery 22 (2015) 79e85
Fig. 3. Forest plot of comparison: preemptive versus postoperative local infiltration; outcome: incidence of postoperative pain at 24 h postoperatively.
breast surgery between patients receiving infiltration with bupi- versus 7.33 mg; P ¼ 0.02) [16] and Zielinski (P ¼ 0.02) [26]. How-
vacaine and those receiving topical application of lignocaine/pri- ever, no significant difference in analgesic consumption was
locaine cream [21]. Local anesthesia slightly but not significantly observed in 2 other trials [21,24].
reduced the overall pain scores and morphine consumption compared
with those of the control group. Talbot et al. also revealed no statistical 3.4. Complications
differences in pain scores between the 2 groups [23]. However, Zie-
linski et al. showed a significantly lower pain intensity assessed 4 and No significant differences in postoperative complications asso-
12 h after surgery (P ¼ 0.004 and 0.02, respectively) for the treatment ciated with wound healing [16] and postoperative nausea and
group [26]. Mohamed et al. and Jonavithula et al. also indicated a sig- vomiting [17,18,24] between the experimental and control groups
nificant postoperative analgesic efficacy in treatment groups [19,20]. were observed in the included studies. However, an increased
incidence of superficial bruising (bupivacaine 20/39 versus control
3.2.2. Preemptive versus postoperative local infiltration 16/40) was reported in 1 trial [16].
Two studies compared the analgesic efficacy between preemp-
tive and postoperative local wound infiltration at 24 h after surgery 4. Discussion
[22,24]. The severity of pain did not differ between the groups
(WMD 2.16; 95% CI: 6.68 e 11.00) (Fig. 3) Persistent pain with sensory disturbances is a major clinical
problem following breast cancer surgery. Surgical wound infiltration
3.3. Analgesic consumption with a local anesthetic solution is speculated to reduce this pain.
Overall, our meta-analysis indicated that wound infiltration with
Analgesic consumption was reported in 10 trials [15e21,23,24,26]. bupivacaine or ropivacaine following breast cancer surgery reduces
We used data obtained from 6 of these RCTs in our analysis but pain significantly at only 2 h postoperatively. We found no difference
excluded 5 studies because 2 did not provide the standard deviation in postoperative pain at 1, 12, and 24 h and in the requirements for
for opioid use at 24 and 48 h postoperatively [16,26], 1 reported only postoperative analgesia between the group receiving infiltration of a
the number of patients requiring opioids [24], and the other 2 pro- local anesthetic solution and the control group.
vided only the 25th and 75th centiles for morphine consumption The optimal location for instilling local anesthetics to reduce
[21]. No significant difference in the requirements for postoperative pain during breast cancer surgery remains controversial. In most of
analgesia between the experimental and control groups was reported our included studies, experimental drugs infiltrated the breast or
(WMD 0.81; 95% CI: 1.82 e 0.19). However, the I2 value was 94%, axillary space. However, bupivacaine infiltrated the skin and sub-
indicating severe heterogeneity among the studies. Therefore, a cutaneous tissue in 1 RCT [21], and the drug was instilled into the
sensitivity analysis was performed using the data after excluding 1 trial drain in 2 other trials [19,23]. The results indicated that the oper-
with unexplained high mean tramadol consumption (216.38 ± 39.05 ative wound instillation of bupivacaine did not confer any addi-
and 369.26 ± 33.81 mg in the bupivacaine and control groups, tional analgesic benefits compared with local wound infiltration
respectively) [20]; the result showed no significant difference in the with bupivacaine in the drain.
analgesic consumption (WMD 0.12; 95% CI: 0.38 e 0.13) (Fig. 4). Preemptive anesthesia in the operative wound before further
The studies that we did not include in the meta-analysis re- surgical dissection may have prevented central sensitization caused
ported various outcomes. The average amount of morphine con- by incisional and inflammatory injury [27,28]. Although the onset
sumption was significantly lower in the bupivacaine group than in of action of ropivacaine is quick and its duration of action is longer
the control group in the studies conducted by Campbell (3.42 mg than that of bupivacaine [29], 2 of our included trials that evaluated
Fig. 4. Forest plot of comparison: experimental drug infiltration versus control; outcome: postoperative analgesic consumption.
K.-W. Tam et al. / International Journal of Surgery 22 (2015) 79e85 85