International Journal of Surgery 22 (2015) 79e85

Contents lists available at ScienceDirect

International Journal of Surgery

journal homepage: www.journal-surgery.net

Review

Effect of wound infiltration with ropivacaine or bupivacaine analgesia

in breast cancer surgery: A meta-analysis of randomized controlled
trials
Ka-Wai Tam a, b, c, d, e, Shin-Yan Chen f, g, Tsai-Wei Huang h, Chao-Chun Lin c,
Chih-Ming Su b, Ching-Li Li b, Yuan-Soon Ho i, j, Wan-Yu Wang b, **, 1,
Chih-Hsiung Wu b, d, *, 1
a
Graduate Institute of Clinical Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
b
Division of General Surgery, Department of Surgery, Taipei Medical University, Shuang Ho Hospital, New Taipei City, Taiwan
c
Center for Evidence-based Health Care, Taipei Medical University, Shuang Ho Hospital, New Taipei City, Taiwan
d
Department of Surgery, School of Medicine, College of Medicine, Taipei Medical University, Taiwan
e
Center for Evidence-based Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
f
Department of Anesthesiology, Taipei Medical University, Shuang Ho Hospital, New Taipei City, Taiwan
g
Department of Anesthesiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
h
Department of Nursing, College of Medicine and Nursing, HungKuang University, Taiwan
i
School of Medical Laboratory Science and Biotechnology, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan
j
Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taipei, Taiwan

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

1. Introduction 2. Materials and methods

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
was accepted by the online PROSPERO international prospective
register of systematic reviews of the National Institute for Health
Research (CRD42015017030).
2.3. Data extraction
Baseline and outcome data were independently extracted by 2
reviewers (WYW and KWT). In particular, the reviewers extracted
data about the study design, study population characteristics, inclu-
sion and exclusion criteria, surgical and anesthetic techniques used,
drug administration methods, operative parameters and complica-
tions, and postoperative parameters. The individually recorded de-
cisions of the 2 reviewers were compared, and any disagreements
were resolved by a third reviewer (SYC). The authors of the studies
were contacted for additional information when necessary.
2.4. Methodological quality appraisal
Two of us (WYW and KWT) independently assessed the meth-
odological quality of each study. The quality of the studies was
assessed using the “risk of bias” method recommended by the
Cochrane Collaboration [8]. Several domains were assessed, such as
the adequacy of randomization, allocation concealment, blinding of
the patients and outcome assessors, length of follow-up, informa-
tion provided on study withdrawals and whether intention-to-treat
(ITT) analysis was performed, and freedom from other biases.
2.5. Outcomes and statistical analysis
The primary outcome was the occurrence and severity of post-
operative pain, which was measured at 1, 2, 12, and 24 h after
operation. The secondary outcomes included complications and
analgesic consumption.
All data were entered and analyzed using the Review Manager,
Version 5 (Cochrane Collaboration, Oxford, England). A meta-
analysis was performed according to the PRISMA guidelines [9].
When necessary, standard deviations were estimated from the
provided confidence interval (CI) limits or standard errors. Effect
sizes of continuous outcomes are reported as the weighted mean
difference (WMD). The precision of the effect sizes is reported as
95% CIs. A pooled estimate of the WMD was computed using the
DerSimonian and Laird random effects model [10].
Cochrane Q tests and I2 statistics were used to evaluate the
statistical heterogeneity and inconsistency of treatment effects
among the studies, respectively. Statistical significance was set at
P < 0.1 for the Cochrane Q tests. Statistical heterogeneity was
assessed by performing the I2 test, with I2 quantifying the propor-
tion of the total outcome variability that was attributable to vari-
ability among the studies. Sensitivity analyses were performed to
assess any impact of the study quality on the effect estimates.
Subgroup analyses were also performed by pooling estimates for
similar patient subsets among trials, where available.
3. Results
3.1. Characteristics of the trials
Fig. 1 presents a flowchart describing the process by which we
screened and selected trials. The initial search strategy used yielded
759 citations, 689 of which were ineligible on the basis of the
criteria used in screening the titles and abstracts; thus, we retrieved
the full text of 70 studies. Most of these were excluded from our
final review for the following reasons: 9 were retrospective studies
or prospective articles; 40 enrolled patients with various condi-
tions, such as plastic surgery of the breast; 1 evaluated the effects of
81
Studies identified through Additional studies identified through
PubMed, Embase, Cochrane, Scopus and by searching references
databases (n = 755) (n = 4)
Search for potentially relevant trials (n = 759)
Studies excluded after reading
titles and abstracts
Not relevant (n = 682)
Review (n = 7)
Studies retrieved for further
evaluation (n = 70)
Studies excluded
Animal study (n = 5)
Different comparison (n = 2)
Different disease (n = 40)
Different intervention (n = 1)
Included studies (n = 13) Not randomized (n = 9)
Fig. 1. Flowchart describing the selection of studies.
local anesthesia using various interventions, such as the infusion
pump [11]; 5 were animal studies; and 2 compared the outcomes of
local anesthesia with those of thoracic paravertebral block [12,13].
Thus, 13 studies were finally eligible [14e26], and their character-
istics are provided in Table 1.
These 13 trials had been published between 2000 and 2015 and
had sample sizes ranging from 30 to 238 patients. Ten trials recruited
patients who had undergone mastectomy with ALND
[14e16,19e23,25,26]. Most of the trials reported that the breast sur-
geries were performed in a single unit. Six trials compared the effects
of ropivacaine with those observed in a control group
[14,15,17,18,22,25]. Among these studies, Johansson el al. (2003) also
investigated the outcomes of ropivacaine when combined with fen-
tanyl [17], and Rica et al. evaluated the effects of pre- and postemptive
local infiltration with ropivacaine [22]. Seven trials assessed the
anesthetic effects of bupivacaine [16,19e21,23,24,26], and among
these trials, Mohamed et al. also compared the analgesic effect of
clonidine with that of bupivacaine [20], Pettersson et al. evaluated the
outcome of topical application of lignocaine/prilocaine cream [21], and
Vallejo et al. determined the difference in outcomes between pre- and
postoperative wound infiltration with bupivacaine [24]. The method
of administration of ropivacaine or bupivacaine varied considerably
among trials: experimental drugs were injected into the breast wound
in most trials; however, in some studies, they were injected into the
axilla [16,18], into the skin and subcutaneous tissue [21], into the drain
[19,23], or into the 2nd and 3rd intercostal spaces and humeral
insertion of the pectoralis [14]. The dosages of ropivacaine or bupi-
vacaine and a placebo were adjusted according to various protocols.
Baseline characteristics and the type of breast surgery were balanced
between the treatment groups in the 13 included RCTs.
The methodological quality of the included trials is summarized in
Table 2. Eight studies reported acceptable methods of randomization
and clearly described the method of allocation concealment
[14,15,19,20,23e26]. All studies, apart from that conducted by
Campbell [16], reported the blinding of both the patients and outcome
assessors. Five studies used ITT analysis [14,16,19,21,22], and 1 [16]
also performed a per-protocol analysis. The number of patients lost
to follow-up was acceptable (<20%) in all studies; however, the loss to
follow-up rate in 1 study was unclear [24]. Other biases that existed in
the studies included no control between group comparisons [22]; no
reporting of patient characteristics [21,23]; nonuniform perioperative
analgesia [17]; and some patients receiving extra fentanyl doses [18].
82 K.-W. Tam et al. / International Journal of Surgery 22 (2015) 79e85

Table 1
Characteristics of the selected randomized controlled trials.

Study Institutions/ Surgical technique Patient Age, years Intervention

Surgeons number

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

ITT, Intention-to-treat; PP, Per-protocol.

 K.-W. Tam et al. / International Journal of Surgery 22 (2015) 79e85 83

Experimental Control Mean Difference Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
1.1.1 Postoperative pain scores at 1 hour
Albi-Feldzer 2013 1.2 0.25 117 1.6 0.25 119 55.8% -0.40 [-0.46, -0.34]
Baudry 2008 1 2.75 40 0.75 1.3 38 14.8% 0.25 [-0.70, 1.20]
Johansson 2000 2.6 2.9 30 2.3 2.3 29 8.6% 0.30 [-1.03, 1.63]
Johansson 2003 2.1 2.7 15 1.5 1.2 15 7.0% 0.60 [-0.90, 2.10]
Vallejo 2006 1 2.1 30 2 1.8 30 13.8% -1.00 [-1.99, -0.01]
Subtotal (95% CI) 232 231 100.0% -0.26 [-0.68, 0.17]
Heterogeneity: Tau² = 0.08; Chi² = 5.98, df = 4 (P = 0.20); I² = 33%
Test for overall effect: Z = 1.19 (P = 0.24)

1.1.2 Postoperative pain scores at 2 hours

Albi-Feldzer 2013 0.55 0.25 117 0.75 0.2 119 69.3% -0.20 [-0.26, -0.14]
Baudry 2008 0.4 1.1 40 0.5 0.9 38 15.3% -0.10 [-0.55, 0.35]
Johansson 2000 2.1 2.6 30 1.5 1.6 29 3.0% 0.60 [-0.50, 1.70]
Johansson 2003 0.8 1.1 15 0.9 1.1 15 5.7% -0.10 [-0.89, 0.69]
Vigneau 2011 2.2 1 24 3 1.5 24 6.7% -0.80 [-1.52, -0.08]
Subtotal (95% CI) 226 225 100.0% -0.19 [-0.39, 0.00]
Heterogeneity: Tau² = 0.01; Chi² = 4.95, df = 4 (P = 0.29); I² = 19%
Test for overall effect: Z = 1.95 (P = 0.05)

1.1.3 Postoperative pain scores at 12 hours

Albi-Feldzer 2013 0.65 0.2 117 0.7 0.2 119 50.4% -0.05 [-0.10, 0.00]
Baudry 2008 0.8 1.4 40 0.5 0.9 38 7.6% 0.30 [-0.22, 0.82]
Johansson 2000 0.2 0.9 30 0.2 0.8 29 10.2% 0.00 [-0.43, 0.43]
Johansson 2003 0.3 0.5 15 0.1 0.2 15 20.0% 0.20 [-0.07, 0.47]
Vigneau 2011 1.2 0.7 24 1.5 0.7 24 11.8% -0.30 [-0.70, 0.10]
Subtotal (95% CI) 226 225 100.0% 0.00 [-0.15, 0.16]
Heterogeneity: Tau² = 0.01; Chi² = 6.45, df = 4 (P = 0.17); I² = 38%
Test for overall effect: Z = 0.03 (P = 0.98)

1.1.4 Postoperative pain scores at 24 hours

Albi-Feldzer 2013 0.4 0.15 117 0.45 0.2 119 50.4% -0.05 [-0.10, -0.00]
Baudry 2008 0.2 0.5 40 0.1 0.3 38 32.5% 0.10 [-0.08, 0.28]
Campbell 2014 2.3 4.48 45 1.7 3.96 34 0.8% 0.60 [-1.27, 2.47]
Johansson 2000 0.8 1.2 30 0.4 0.8 29 8.8% 0.40 [-0.12, 0.92]
Vallejo 2006 1.7 2.3 30 1.1 1.3 30 3.0% 0.60 [-0.35, 1.55]
Vigneau 2011 2.4 1.2 24 2.9 1.5 24 4.4% -0.50 [-1.27, 0.27]
Subtotal (95% CI) 286 274 100.0% 0.04 [-0.13, 0.21]
Heterogeneity: Tau² = 0.01; Chi² = 8.81, df = 5 (P = 0.12); I² = 43%
Test for overall effect: Z = 0.50 (P = 0.61)

-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
outcome measures at 1, 2, 12, and 24 h postoperatively. We used
data obtained from 8 of these RCTs in our analysis but excluded 5
because the studies provided only medians and ranges [23,26],
mean visual analog scores without standard deviations [19,20], or
the 25th and 75th centiles for the pain scores [21]. We used only the
results of ropivacaine or bupivacaine groups versus control groups
for the meta-analysis. Combined drug groups, such as clonidine with
bupivacaine and fentanyl with ropivacaine, were excluded before
data pooling. The pooled mean differences in the pain score
were 0.26 (95% CI: 0.68 e 0.17) at 1 h, 0.00 (95% CI: 0.15 e 0.16) at
12 h, and 0.04 (95% CI: 0.13 e 0.21) at 24 h postoperatively; no
significant difference was observed in the postoperative pain scores
between the groups (Fig. 2). However, the severity of pain was
significantly reduced in the experimental group, and the pooled mean
difference was 0.19 (95% CI: 0.39 e 0.00) at 2 h postoperatively. The
I2 value ranged from 19 to 43% at various time periods postoperatively,
indicating low to moderate heterogeneity among studies (Fig. 2).
The studies that we did not include in the meta-analysis showed
various results. Pettersson et al. compared postoperative pain after
84 K.-W. Tam et al. / International Journal of Surgery 22 (2015) 79e85

Preemptive analgesia Postoperative analgesia Mean Difference Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
Rica 2007 34.25 21.91 15 26.03 20.23 15 34.3% 8.22 [-6.87, 23.31]
Vallejo 2006 16 20 30 17 23 30 65.7% -1.00 [-11.91, 9.91]

Total (95% CI) 45 45 100.0% 2.16 [-6.68, 11.00]

Heterogeneity: Tau² = 0.00; Chi² = 0.94, df = 1 (P = 0.33); I² = 0%
-100 -50 0 50 100
Test for overall effect: Z = 0.48 (P = 0.63) Favours preemptive Favours postoperative

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-
vacaine and those receiving topical application of lignocaine/pri-
locaine cream [21]. Local anesthesia slightly but not significantly
reduced the overall pain scores and morphine consumption compared
with those of the control group. Talbot et al. also revealed no statistical
differences in pain scores between the 2 groups [23]. However, Zie-
linski et al. showed a significantly lower pain intensity assessed 4 and
12 h after surgery (P ¼ 0.004 and 0.02, respectively) for the treatment
group [26]. Mohamed et al. and Jonavithula et al. also indicated a sig-
nificant postoperative analgesic efficacy in treatment groups [19,20].
3.2.2. Preemptive versus postoperative local infiltration
Two studies compared the analgesic efficacy between preemp-
tive and postoperative local wound infiltration at 24 h after surgery
[22,24]. The severity of pain did not differ between the groups
(WMD 2.16; 95% CI: 6.68 e 11.00) (Fig. 3)
3.3. Analgesic consumption
Analgesic consumption was reported in 10 trials [15e21,23,24,26].
We used data obtained from 6 of these RCTs in our analysis but
excluded 5 studies because 2 did not provide the standard deviation
for opioid use at 24 and 48 h postoperatively [16,26], 1 reported only
the number of patients requiring opioids [24], and the other 2 pro-
vided only the 25th and 75th centiles for morphine consumption
[21]. No significant difference in the requirements for postoperative
analgesia between the experimental and control groups was reported
(WMD 0.81; 95% CI: 1.82 e 0.19). However, the I2 value was 94%,
indicating severe heterogeneity among the studies. Therefore, a
sensitivity analysis was performed using the data after excluding 1 trial
with unexplained high mean tramadol consumption (216.38 ± 39.05
and 369.26 ± 33.81 mg in the bupivacaine and control groups,
respectively) [20]; the result showed no significant difference in the
analgesic consumption (WMD 0.12; 95% CI: 0.38 e 0.13) (Fig. 4).
The studies that we did not include in the meta-analysis re-
ported various outcomes. The average amount of morphine con-
sumption was significantly lower in the bupivacaine group than in
the control group in the studies conducted by Campbell (3.42 mg
versus 7.33 mg; P ¼ 0.02) [16] and Zielinski (P ¼ 0.02) [26]. How-
ever, no significant difference in analgesic consumption was
observed in 2 other trials [21,24].
3.4. Complications
No significant differences in postoperative complications asso-
ciated with wound healing [16] and postoperative nausea and
vomiting [17,18,24] between the experimental and control groups
were observed in the included studies. However, an increased
incidence of superficial bruising (bupivacaine 20/39 versus control
16/40) was reported in 1 trial [16].
4. Discussion
Persistent pain with sensory disturbances is a major clinical
problem following breast cancer surgery. Surgical wound infiltration
with a local anesthetic solution is speculated to reduce this pain.
Overall, our meta-analysis indicated that wound infiltration with
bupivacaine or ropivacaine following breast cancer surgery reduces
pain significantly at only 2 h postoperatively. We found no difference
in postoperative pain at 1, 12, and 24 h and in the requirements for
postoperative analgesia between the group receiving infiltration of a
local anesthetic solution and the control group.
The optimal location for instilling local anesthetics to reduce
pain during breast cancer surgery remains controversial. In most of
our included studies, experimental drugs infiltrated the breast or
axillary space. However, bupivacaine infiltrated the skin and sub-
cutaneous tissue in 1 RCT [21], and the drug was instilled into the
drain in 2 other trials [19,23]. The results indicated that the oper-
ative wound instillation of bupivacaine did not confer any addi-
tional analgesic benefits compared with local wound infiltration
with bupivacaine in the drain.
Preemptive anesthesia in the operative wound before further
surgical dissection may have prevented central sensitization caused
by incisional and inflammatory injury [27,28]. Although the onset
of action of ropivacaine is quick and its duration of action is longer
than that of bupivacaine [29], 2 of our included trials that evaluated

Experimental Control Std. Mean Difference Std. Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
Baudry 2008 6.9 8.7 40 6.2 6.8 38 31.7% 0.09 [-0.36, 0.53]
Johansson 2000 4.2 4.3 30 4.2 2.6 29 24.1% 0.00 [-0.51, 0.51]
Johansson 2003 2.4 1.8 15 2.9 2 15 12.3% -0.26 [-0.97, 0.46]
Jonnavithula 2015 36 4.3 20 67.5 65 20 15.5% -0.67 [-1.31, -0.03]
Talbot 2004 12 11 19 13 9.5 21 16.4% -0.10 [-0.72, 0.53]

Total (95% CI) 124 123 100.0% -0.12 [-0.38, 0.13]

Heterogeneity: Tau² = 0.00; Chi² = 4.05, df = 4 (P = 0.40); I² = 1%
-4 -2 0 2 4
Test for overall effect: Z = 0.95 (P = 0.34) Favours experimental Favours control

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
the outcome of preemptive and postoperative pain control revealed
that the pain score did not differ significantly between the groups
when ropivacaine [22] or bupivacaine [24] was used.
The optimal dose of bupivacaine or ropivacaine for instillation in
appropriate places during mastectomy is inconclusive. In our
included trials, a higher concentration or larger volume of ropiva-
caine (4.75 mg/mL R 40 mL) [15] did not yield stronger analgesic
effects than did lower doses (7.5 mg/mL solution 20 mL) [25].
Although a potential disadvantage of local anesthetic instillation is
the risk of increased seroma formation, no excess seroma formation
was reported in our included trials. No evidence of bupivacaine or
ropivacaine toxicity, such as postoperative nausea and vomiting,
cardiac arrhythmias, or delayed wound healing, was observed in
any of the studies.
The studies included in our analysis exhibited considerable
heterogeneity, which was attributable to various clinical factors.
First, the surgical technique of breast cancer surgery was not
identical in all the studies; some recruited patients received ALND,
but some did not. Second, some trials did not standardize the
perioperative analgesia. Finally, the timing and location of drug
instillation and the dosage of experimental drugs for instillation
were slightly different among studies. Such diversity among the
studies resulted in the heterogeneity.
Our study has several limitations. First, some trials were small,
with only 30 patients being recruited per treatment group. Second,
several studies did not report details on the characteristics of the
patients and types of surgery. Finally, several of our primary and
secondary outcomes were variably reported, thus potentially
limiting the inferences based on our analysis.
In conclusion, the results of our meta-analysis revealed that the
infusion of ropivacaine or bupivacaine following breast cancer
surgery decreased immediate postoperative pain but did not
reduce pain at 12, and 24 h postoperatively. Although some trials
showed that wound infiltration with ropivacaine or bupivacaine
reduced immediate postoperative pain, the analgesic consumption
did not differ between the groups. According to the results of this
meta-analysis, the potential clinical value of local anesthetic infil-
tration for postoperative pain relief can be questioned.
Ethical approval
No needed.
Sources of funding
The authors Ka-Wai Tam, Shin-Yan Chen, Tsai-Wei Huang, Chao-
Chun Lin, Chih-Ming Su, Ching-Li Li, Yuan-Soon Ho, Wan-Yu Wang,
and Chih-Hsiung Wu have no financial ties to disclose.
Author contribution
Wan-Yu Wang and Ka-Wai Tam devised and designed the study;
Wan-Yu Wang and Ka-Wai Tam extracted the data; Wan-Yu Wang,
Chih-Hsiung Wu, Shin-Yan Chen, and Ka-Wai Tam analyzed and
interpreted data; Wan-Yu Wang and Ka-Wai Tam wrote the first
draft; Chih-Hsiung Wu, Yuan-Soon Ho, Tsai-Wei Huang, Chao-Chun
Lin, Chih-Ming Su, Ching-Li Li, and Ka-Wai Tam reviewed critically
the analyses and academic writing; all authors contributed to
subsequent versions and approved the final article.
Conflict of interest
All authors have no conflicts of interest or financial ties to
disclose.
85
Guarantor
Ka-Wai Tam is the guarantor of the study.
References
[1] A. Jemal, F. Bray, M.M. Center, et al., Global cancer statistics, CA, Cancer. J. Clin.
61 (2011) 69e90.
[2] S. Darby, P. McGale, C. Correa, et al., the Early Breast Cancer Trialists' Collaborative
Group (EBCTCG), Effect of radiotherapy after breast-conserving surgery on 10-year
recurrence and 15-year breast cancer death: meta-analysis of individual patient
data for10801 women in 17 randomised trials, Lancet 378 (2011) 1707e1716.
[3] M. Koivuranta, E. La €a
€ra €, L. Snåre, et al., A survey of postoperative nausea and
vomiting, Anaesthesia 52 (1997) 443e449.
[4] N.B. Scott, Wound infiltration for surgery, Anaesthesia 65S (2010) 67e75.
[5] Y.S. Tong, C.C. Wu, C.H. Bai, et al., Effect of extraperitoneal bupivacaine anal-
gesia in laparoscopic inguinal hernia repair: a meta-analysis of randomized
controlled trials, Hernia 18 (2014) 177e183.
[6] D.B. Scott, A. Lee, D. Fagan, et al., Acute toxicity of ropivacaine compared with
that of bupivacaine, Anesth. Analg. 69 (1989) 563e569.
[7] G. Baudry, A. Steghens, D. Laplaza, et al., Ropivacaine infiltration during breast
cancer surgery: postoperative acute and pain effect, Ann. Fr. Anesth. Reanim.
27 (2008) 979e986.
[8] Higgins JPT SG, Green S, Cochrane Handbook for Systematic Reviews of In-
terventions Version 5.1.0 [Updated March 2011]. The Cochrane Collaboration.
Available from: www.cochrane-handbook.org.
[9] A. Liberati, D.G. Altman, J. Tetzlaff, et al., The PRISMA statement for reporting
systematic reviews and meta-analyses of studies that evaluate health care in-
terventions: explanation and elaboration, J. Clin. Epidemiol. 62 (2009) e1e34.
[10] R. DerSimonian, N. Laird, Meta-analysis in clinical trials, Control Clin. Trials 7
(1986) 177e188.
[11] L. Ferreira Laso, A. Lopez-Picado, L. Lamata, et al., Postoperative analgesia by
infusion of local anesthetic into the surgical wound after modified radical mas-
tectomy: a randomized clinical trial, Plast. Reconstr. Surg. 134 (2014) 862ee870e.
[12] M. Chiu, G.L. Bryson, A. Lui, et al., Reducing persistent postoperative pain and
disability 1 year after breast cancer surgery: a randomized, controlled trial
comparing thoracic paravertebral block to local anesthetic infiltration, Ann.
Surg. Oncol. 21 (2014) 795e801.
[13] T. Sidiropoulou, O. Buonomo, E. Fabbi, et al., A prospective comparison of continuous
wound infiltration with ropivacaine versus single-injection paravertebral block
after modified radical mastectomy, Anesth. Analg. 106 (2008) 997e1001.
[14] A.1 Albi-Feldzer, E.E. Mouret-Fourme, S. Hamouda, et al., A double-blind
randomized trial of wound and intercostal space infiltration with ropivacaine
during breast cancer surgery: effects on chronic postoperative pain, Anes-
thesiology 118 (2013) 318e326.
[15] G. Baudry, A. Steghens, D. Laplaza, et al., Ropivacaine infiltration during breast
cancer surgery: postoperative acute and chronic pain effect, Ann. Fr. Anesth.
Reanim. 27 (2008) 979e986.
[16] I. Campbell, S. Cavanagh, J. Creighton, et al., To infiltrate or not? Acute effects
of local anaesthetic in breast surgery, ANZ. J. Surg. 22 (2014), http://dx.doi.org/
10.1111/ans.12541 [Epub ahead of print].
[17] A. Johansson, J. Kornf€ alt, L. Nordin, et al., Wound infiltration with ropivacaine
and fentanyl: effects on postoperative pain and PONV after breast surgery,
J. Clin. Anesth. 15 (2003) 113e118.
[18] A. Johansson, J. Axelson, C. Ingvar, et al., Preoperative ropivacaine infiltration
in breast surgery, Acta. Anaesthesiol. Scand. 44 (2000) 1093e1098.
[19] N. Jonnavithula, H. Khandelia, P. Durga, et al., Role of wound instillation with
bupivacaine through surgical drains for postoperative analgesia in modified
radical mastectomy, Indian J. Anaesth. 59 (2015) 15e20.
[20] S.A. Mohamed, H.S. Abdel-Ghaffar, Effect of the addition of clonidine to locally
administered bupivacaine on acute and chronic postmastectomy pain, J. Clin.
Anesth. 25 (2013) 20e27.
[21] N. Pettersson, L. Perbeck, R.G. Hahn, Efficacy of subcutaneous and topical local
anaesthesia for pain relief after resection of malignant breast tumours, Eur. J.
Surg. 167 (2001) 825e830.
[22] M.A. Rica, A. Norlia, M. Rohaizak, et al., Preemptive ropivacaine local anaesthetic
infiltration versus postoperative ropivacaine wound infiltration in mastectomy:
postoperative pain and drain outputs, Asian J. Surg. 30 (2007) 34e39.
[23] H. Talbot, S.P. Hutchinson, D.L. Edbrooke, et al., Evaluation of a local anaes-
thesia regimen following mastectomy, Anaesthesia 59 (2004) 664e667.
[24] M.C. Vallejo, A.L. Phelps, N. Sah, et al., Preemptive analgesia with bupivacaine
for segmental mastectomy, Reg. Anesth. Pain Med. 31 (2006) 227e232.
[25] A. Vigneau, A. Salengro, J. Berger, et al., A double blind randomized trial of
wound infiltration with ropivacaine after breast cancer surgery with axillary
nodes dissection, BMC Anesthesiol. 11 (2011) 23.
[26] J. Zielinski, R. Jaworski, I. Smietanska, et al., A randomized, double-blind, placebo-
controlled trial of preemptive analgesia with bupivacaine in patientsundergoing
mastectomy for carcinoma of the breast, Med. Sci. Monit. 17 (2011) CR589e597.
[27] J. Ballantyne, Preemptive analgesia: an unsolved problem, Curr. Opin. Anes-
thesiol. 14 (2001) 499e504.
[28] Kissin, Preemptive analgesia, Anesthesiology 93 (2000) 1139e1142.
[29] D.L. Moffit, D.A. De Berker, C.T.K. Kennedy, et al., Assessment of ropivacaine as a local
anaesthetic for skin infiltration in skin surgery, Dermatol. Surg. 27 (2001) 437e440.