Macrosson et al.

Perioperative Medicine (2024) 13:80 Perioperative Medicine

https://doi.org/10.1186/s13741-024-00437-0

REVIEW Open Access

A systematic review and meta‑analysis

of thoracic epidural analgesia versus other
analgesic techniques in patients
post‑oesophagectomy
Duncan Macrosson1,2*, Adam Beebeejaun1,2 and Peter M. Odor1

Abstract
Background Oesophageal cancer surgery represents a high perioperative risk of complications to patients, such
as postoperative pulmonary complications (PPCs). Postoperative analgesia may influence these risks, but the most
favourable analgesic technique is debated. This review aims to provide an updated evaluation of whether thoracic
epidural analgesia (TEA) has benefits compared to other analgesic techniques in patients undergoing oesophagec-
tomy surgery. Our hypothesis is that TEA reduces pain scores and PPCs compared to intravenous opioid analgesia
in patients post-oesophagectomy.
Methods Electronic databases PubMed, Excerpta Medica Database (EMBASE) and Cochrane Central Register
of Controlled Trials (CENTRAL) were searched for randomised trials of analgesic interventions in patients undergoing
oesophagectomy surgery. Only trials including thoracic epidural analgesia compared with other analgesic techniques
were included. The primary outcome was a composite of respiratory infection, atelectasis and respiratory failure
(PPCs), with pain scores at rest and on movement as secondary outcomes. Data was pooled using random effect
models and reported as relative risks (RR) or mean differences (MD) with 95% confidence intervals (CIs).
Results Data from a total of 741 patients in 10 randomised controlled trials (RCTs) from 1993 to 2023 were included.
Nine trials were open surgery, and one trial was laparoscopic. Relative to intravenous opioids, TEA significantly
reduced a composite of PPCs (risk ratio (RR) 3.88; 95% confidence interval (CI) 1.98–7.61; n = 222; 3 RCTs) and pain
scores (0–100-mm visual analogue scale or VAS) at rest at 24 h (MD 9.02; 95% CI 5.88–12.17; n = 685; 10 RCTs) and 48 h
(MD 8.64; 95% CI 5.91–11.37; n = 685; 10 RCTs) and pain scores on movement at 24 h (MD 14.96; 95% CI 5.46–24.46;
n = 275; 4 RCTs) and 48 h (MD 16.60; 95% CI 8.72–24.47; n = 275; 4 RCTs).
Conclusions Recent trials of analgesic technique in oesophagectomy surgery are restricted by small sample size
and variation of outcome measurement. Despite these limitations, current evidence indicates that thoracic epidural
analgesia reduces the risk of PPCs and severe pain, compared to intravenous opioids in patients following oesopha-
geal cancer surgery. Future research should include minimally invasive surgery, non-epidural regional techniques
and record morbidity, using core outcome measures with standardised endpoints.
Trial registration Prospectively registered on PROSPERO (CRD42023484720).

*Correspondence:
Duncan Macrosson
[email protected]
Full list of author information is available at the end of the article

© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom-
mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 2 of 15

Keywords Thoracic epidural, Oesophagectomy, Analgesia, Postoperative pulmonary complications, Meta-analysis,

Systematic review

Background epidural analgesia versus intravenous opioid analgesia,

Oesophagectomies are considered major and complex Visser et al. (2017) observed no significant difference in
surgery with significant postoperative pain and high pain scores at 24 and 48 h postoperatively, and Hughes
postoperative complication rates which can decrease et al. (2018) observed no significant difference in rest
long-term survival (Booka et al. 2018). Epidural analgesia pain postoperatively (Visser et al. 2017; Hughes et al.
is often considered the gold standard form of postopera- 2018). Both reviews concluding that no benefit could be
tive analgesia for this surgery (Low et al. 2018). Epidur- shown by epidurals regarding postoperative pulmonary
als have been shown to reduce postoperative pain scores complications (PPCs). Since these reviews, further rel-
and some postoperative complications such as respira- evant randomised trials have been published (Xu et al.
tory failure in major abdominal surgery but with uncer- 2023; Zhu et al. 2020; Li et al. 2019; Wang et al. 2019).
tain replication in oesophageal cancer surgery (Pirie et al. During the completion of this review, a network meta-
2020; Rigg et al. 2002). Pulmonary complications are of analysis evaluating analgesic strategies post-oseophagec-
interest as they are one of the most common postopera- tomy by Ramjit et al. (2024) was published showing an
tive complications, especially in high-risk open abdomi- increase in postoperative forced vital capacity; a reduc-
nal surgery; they predict long- and short-term health tion in pain scores, opioid consumption, intensive care
outcomes, admission to critical care and hospital length unit stay and time to extubation in thoracic epidural anal-
of stay (Booka et al. 2018; Odor et al. 2020). gesia (TEA) versus systemic opioids (Ramjit et al. 2024).
Although epidural analgesia may potentially improve The review did not find enough data to analyse morbidity
outcomes, contraindications include patient refusal, anti- including postoperative pulmonary complications.
coagulant use or a patient’s pre-existing anatomical or Our primary aim was to evaluate whether TEA reduced
neurological issues. Epidurals are associated with compli- respiratory morbidity versus other analgesic techniques
cations such as urinary retention, hypotension and par- following oesophagectomy surgery, with a secondary
tial or complete failure and rarer complications such as objective to compare analgesic outcomes.
neurological damage (8.2–17.4 cases of permanent nerve
damage per 100,000 patients receiving epidural analgesia) Methods
(Cook et al. 2009). Postoperative management of epidural This review protocol was prospectively registered on
analgesia also represents a higher resource requirement PROSPERO (CRD42023484720) and followed guid-
(Holtz et al. 2022). Recent evidence has suggested mini- ance from the Preferred Reporting Items for Systematic
mally invasive oesophagectomy surgery is gaining in reviews and Meta-Analysis (PRISMA) statement 2020
popularity compared to open oesophagectomy surgery (Page et al. 2020).
(Mann et al. 2020). Existing evidence from colorectal The review question was as follows: “In adult patients
surgical data shows epidural analgesia achieves superior undergoing elective oesophagectomy, does thoracic
pain relief compared to opioid analgesia for open surgery, epidural analgesia influence postoperative pulmonary
but not for less invasive (laparoscopic) surgery (Borzel- complications in comparison to other analgesic tech-
lino et al. 2024; Turi et al. 2024). Therefore, as minimally niques?” We used the framework of PICOS (Population,
invasive oesophagectomy surgery increases in frequency, Intervention, Comparison, Outcomes and Study design).
epidural analgesia may in turn become less beneficial. Participants included adult patients undergoing elective
Finally, other regional techniques such as paravertebral oesophagectomy. Thoracic epidural analgesia was the
and erector spinae catheters have been gaining favour comparator group. The intervention groups included any
in recent years, having many of the benefits of epidural other form of analgesia such as intravenous opioids or
analgesia but with a more favourable side effect profile, other regional techniques. Study design was restricted to
although randomised clinical trials are lacking (Feenstra randomised clinical trials only.
et al. 2023). Many of these factors may result in a reduc-
tion in the use of epidural analgesia for postoperative Primary outcome
pain management (Pirie et al. 2020). The primary outcome was a composite of postopera-
Two previous meta-analyses compared analgesic tech- tive pulmonary complications (PPCs), including res-
niques in oesophagectomies in 2017 and 2018 but found piratory infection, respiratory failure and atelectasis
a paucity of prospective trials to compare. Regarding within 30 days of surgery. Standard diagnostic criteria
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 3 of 15

were based upon the European Perioperative Clinical Study selection

Outcomes (EPCO) consensus statement (Jammer et al. After de-duplication, the primary author screened titles
2015). However, as the review search dates included and abstracts against the inclusion criteria to identify
a period before the most recent consensus defini- potentially relevant papers. One researcher was used at
tions of PPCs, we categorised explicit descriptions of this stage who erred on the side of over-inclusion. The
PPCs in each trial according to closeness of match to second stage involved full-text review of all potentially
the EPCO definitions. Where a composite PPC was eligible studies by two authors and recording the reason
not reported, we contacted corresponding authors via for the exclusion of a paper (Fig. 1).
email to request additional information, including pri-
mary data. Data extraction
One author extracted data from the selected publica-
Secondary outcomes tions using a pre-piloted data abstract tool. All data was
Secondary outcomes included resting and dynamic 24- checked by a second reviewer. Information included
and 48-h pain scores measured with a 100-mm visual is as per the study characteristic tables below. Data
analogue scale (VAS), technical failure, postoperative not reported in the studies was recorded as “NR” (not
nausea and vomiting (PONV) and length of hospital reported), and non-applicable data was recorded as
stay. Pain not only is important for humane reasons “N/A”. WebPlotDigitizer was used to estimate these
but also it slows progress towards enhanced recovery numerical scores and standard deviations from graphi-
targets and can lead to further postoperative complica- cal data (Rohatgi 2022). Risk-of-bias assessment was
tions (Low et al. 2018). Technical failure was assessed completed after data extraction. Two authors individu-
because epidurals have a high rate of failure (27–32%) ally evaluated the methodological quality of all articles
(Hermanides et al. 2012). This is at odds with the most using the Cochrane risk-of-bias assessment tool version
common intervention group of intravenous opioid 2 (Higgins et al. 2023).
analgesia, which has virtually no failure rate. Techni-
cal failure can be defined as insufficient epidural anal- Data synthesis
gesia which requires removal or switch of analgesic Meta-analysis was performed on any primary or sec-
regimen and also includes accidental catheter dislodge- ondary outcome included by more than one study. For
ment (Hermanides et al. 2012). PONV is included as it the dichotomous outcomes of PPCs and PONV, inci-
is a well-known side effect of opioid analgesia; it can dences of outcomes per group were extracted from
be defined as the 24-h incidence of postoperative nau- each study to allow a pooled meta-analysis of risk
sea or vomiting, as this is commonly reported in trials ratio estimates with 95% confidence intervals. For the
and the most clinically relevant time interval (Dolin continuous outcomes of 24- and 48-h pain scores for
and Cashman 2005). Finally, length of hospital stay is a resting and dynamic pain, mean scores and standard
well-established and important perioperative outcome, deviations for both groups in each study were extracted
measured in time (hours) from admission to discharge. to allow a pooled meta-analysis of mean difference esti-
mates with 95% confidence intervals.
Search strategy Where results were presented in included trials as
We searched PubMed, Embase, and CENTRAL data- mixed data of median (IQR) and mean (SD), we con-
bases, using a combination of relevant keywords and verted to mean and standard deviation throughout, to
medical subject heading terms for oesophagectomy sur- enable pooled comparison. Standard deviations for pain
gery and epidural analgesia. Search limits were applied scores were imputed for one trial that did not report, by
to restrict results to RCTs published from 1 January 2013 combining the mean standard deviations of other tri-
to 31 December 2023. We included all randomised con- als (Fares et al. 2014). One trial measured postoperative
trolled trials of adult (age ≥ 18 years) patients undergo- pain scores twice a day; these morning and afternoon
ing elective oesophagectomy surgery in which one group scores were combined to give a single mean and stand-
received postoperative TEA. Intraoperative TEA was ard deviation for each day (Flisberg et al. 2001). Two
not included as all patients undergo general anaesthe- trials had four groups of participants we combined into
sia; thus, pain control and its sequelae are more relevant two groups according to the method of postoperative
for the postoperative period. The full search strategy analgesia (Zhu et al. 2020; Li et al. 2019). Standard devi-
is detailed in Appendix. No language restrictions were ations were computed using an online calculator (Stat-
placed on eligible studies. sToDo) decomposing the mean and standard deviations
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 4 of 15

Fig. 1 PRISMA flow diagram

of two groups into one single group (CombineMeanSD. open surgery. Three trials recorded PPCs, and only two
2023; Altman 2000). trials recorded length of hospital stay and PONV. How-
All meta-analyses were performed using RevMan ever, all trials recorded some form of pain score at 24 and
(Cochrane and Collaboration 2024). Inverse variance 48 h (Table 2).
was used as the statistical method for both dichotomous Patient characteristics included a mean age of
and continuous outcomes. Statistical heterogeneity was 59.9 years, mean BMI of 22.7 and a proportion of 75.3%
assessed by using both the I2 and χ2 tests. A random males (Table 3). All trials compared TEA to intravenous
effects model was adopted due to the clinical and meth- opioids, with only one trial including a third group of
odological diversity between trials. Formal meta-analyses paravertebral and transversus abdominis plane blocks.
were not possible for other outcomes including technical Regional and intravenous drug regimens varied (Table 4).
failure and length of hospital stay due to insufficient data; PPCs, rest and dynamic pain scores are displayed in
therefore, this data is presented in tabulated form and/or Tables 5, 6 and 7 respectively and PONV and length of
narratively appraised. hospital stay in Tables 8 and 9.
Regarding overall risk of bias, seven trials were judged
Results as having some concerns, with three being judged as high
Description of included studies risk (Fig. 2). Some trials did not mention their randomi-
A total of 330 publications were found over three data- sation or allocation concealment method. Nearly all tri-
bases. After filtering for eligibility criteria, 10 randomised als did not blind their assessors, accounting for zero trials
trials with 741 patients over 5 countries were included able to be judged as a low risk of bias overall.
(Table 1).
The details of inclusion and exclusion criteria varied Postoperative pulmonary complications (PPCs)
between trials, with many excluding comorbid patients Regarding the primary outcome of PPCs within 30 days
based on ASA grade or individual diseases. Only one trial of surgery, one trial (Xu et al. 2023) used the EPCO defi-
used a laparoscopic technique for surgery; all others were nition (Jammer et al. 2015). Another trial (Maghsoudloo
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 5 of 15

Table 1 Study information

Author and Year Country Design Randomisation Allocation Source of funding Number Surgical approach
concealment

Flisberg et al. 2001 Sweden RCT​ NR NR University and Society 33 Open

of Medicine grants
Yokoyama et al. 2005 Japan RCT​ NR NR NR 30 Open
Fares et al. 2014 Egypt RCT​ Computer generated Opaque envelopes None 30 Open
Wang et al. 2017 China RCT​ Random number table NR NR 80 Open
Wang et al. 2019 China RCT​ Computer generated Opaque envelopes Innovation and Sci- 40 Open
ence Foundation
grants
Liu and Wang 2015 China RCT​ NR NR NR 60 Open
Li 2019 China RCT​ Computer generated Opaque envelopes Science Foundation 100 Open
and educational
grants
Zhu et al. 2020 China RCT​ Computer generated NR NR 120 Open
Maghsoudloo et al. Iran RCT​ Based on odd/even NR NR 80 Open
2023 surgical date
Xu et al. 2023 China RCT​ Computer generated Opaque envelopes Cancer, Science Foun- 168 Laparoscopic
dation and research
grants
RCT​randomised controlled trial, NR not reported

et al. 2023) did not state a definition, and the third trial Rest pain was meta-analysed at 24 and 48 h (Fig. 4a
(Fares et al. 2014) reported a composite of PPCs: indi- and b). All trials but one favoured TEA. Both meta-
vidual incidences of pneumonia, pleural effusions and analyses suggest a significant reduction in pain scores
ARDS (acute respiratory distress syndrome). These were in the TEA group regarding rest pain at 24 (MD
all included in the meta-analysis of PPCs. None of these 9.02; 95% CI 5.88–12.17) and 48 h (MD 8.64; 95% CI
trials specified a time limit for recording these compli- 5.91–11.37).
cations. The meta-analysis (Fig. 3) suggests TEA may Four of 10 trials measured dynamic pain scores at 24
reduce the risk of a composite of PPCs (RR 3.88; 95% and 48 h; the meta-analyses were displayed in Fig. 4c and
CI 1.98–7.61), although this is of lower certainty due to d. The summary measures again suggested a significant
some risk of bias and differences in definition of compos- reduction in pain scores in the TEA group at 24 (MD
ites of PPCs. The I2 and χ2 tests suggest little heterogene- 14.96; 95% CI 5.46–24.46) and 48 h (MD 16.60; 95% CI
ity, although there is some uncertainty of these values as 8.72–24.47), with a larger effect but wider confidence
the number of studies and sample sizes is small. interval in comparison to the rest pain meta-analyses.
The mean difference (MD) is measured in millimetre
Postoperative pain scores of the 0–100-mm visual analogue score (VAS). All pain
Some studies reported only pain scores (Zhu et al. 2020; score meta-analyses suggest considerable heterogeneity
Wang et al. 2019; Maghsoudloo et al. 2023; Wang et al. when assessing their I2 and χ2 tests.
2017; Yokoyama et al. 2005; Liu and Wang 2015) and did
not specify rest and dynamic pain scores. In this case, Postoperative nausea and vomiting
these pain scores were included in the meta-analyses for Regarding PONV, only three trials reported data, no
rest pain scores. Nearly all pain scores were stated as the definitions or time limits were given by any trial and
VAS (visual analogue score), apart from one study which two trials recorded nausea and vomiting as a single
used the box scale and was still included in the pain event (Xu et al. 2023; Li et al. 2019). One trial recorded
meta-analyses (Yokoyama et al. 2005). Some dynamic them as two separate events (Wang et al. 2017), which
pain scores did not record their dynamic activity or were combined into a single event by addition, in order
had different activities across trials. All but three trials to include it in the meta-analysis (Fig. 5). This meta-
did not report numerical data for their pain scores and analysis did not suggest any difference in reduction of
only had graphical data (Li et al. 2019; Wang et al. 2019; PONV rates by either method of analgesia and should
Maghsoudloo et al. 2023). be considered of low certainty.
Table 2 Inclusion and exclusion criteria and primary and secondary trial outcomes matching systematic review outcomes
Year and author Inclusion criteria Exclusion criteria Primary outcomes Secondary outcomes

Flisberg et al. 2001 ASA 1–3, thoracoabdominal oesophagectomy NR Nil Pain scores at 24 and 48 h
Yokoyama et al. 2005 ASA 1–2, radical oesophagectomy, oesophageal Preoperative steroid/NSAID Nil Pain scores at 24 and 48 h
cancer
Macrosson et al. Perioperative Medicine

Fares et al. 2014 ASA 1–2, age 20–60 years, Ivor Lewis oesophagec- NYHA 3–4, chronic obstructive pulmonary disease, Pulmonary complications Pain scores at 24 and 48 h
tomy CKD, Child–Pugh B liver failure or worse, bleeding
diathesis. Preoperative steroid
Liu and Wang 2015 ASA 1–2, age 40–65 years, 45–70 kg, radical Preoperative radiotherapy or chemotherapy. Endo- Nil Pain scores at 24 and 48 h
oesophagectomy and thoracotomy crine/metabolic diseases, antiplatelet/anticoagulant
(2024) 13:80

drugs. Perioperative blood transfusion

Wang et al. 2017 ASA 1–2, age 18–65 years, BMI 18–30 Heart/liver/kidney/electrolyte/neuromuscular/ Nil Pain scores at 24 and 48 h, PONV, LOS
immune/endocrine disorder. Postoperative
mechanical ventilation or secondary surgery dur-
ing study
Wang et al. 2019 Elective radical oesophagectomy ASA 4, preoperative opioids/NSAIDs/immunosup- Nil Pain scores at 24 and 48 h
pressive drugs. Cardiovascular/immune/endocrine
disease
Li 2019 ASA 1–2, age 40–65 years, 45–80 kg, thoracoab- Preoperative chemo/radiotherapy, cardiovascular/ Nil Pain scores at 24 and 48 h, PONV
dominal oesophagectomy, gastro-oesophageal autoimmune/endocrine/metabolic/coagulation/
carcinoma kidney/liver disorder. Preoperative opioid/NSAID/
blood transfusion. Postoperative mechanical
ventilation
Zhu et al. 2020 ASA 1–2, thoracotomy Preoperative chemo/radiotherapy/opioid/immu- Nil Pain scores at 24 and 48 h
nosuppressant/steroid. Autoimmune/endocrine/
metabolic diseases. Abnormal liver and kidney func-
tion tests. Perioperative blood transfusion
Maghsoudloo et al. 2023 20–80 years, oesophagectomy ASA 3 or higher, coagulation/severe haemody- Pulmonary complications Pain scores at 24 and 48 h
namic/severe movement/chest/neurological
disorders, substance abuse
Xu et al. 2023 > 17 years, thoraco-laparoscopic (McKeown) Open oesophagectomy, chronic pain with opioid Pulmonary complications Pain scores at 24 and 48 h, LOS
oesophagectomy use, not suitable for one lung ventilation, prior lung
surgery
ASA American Society of Anesthesiologists physical status grade, y years, BMI body mass index, NSAID nonsteroidal anti-inflammatory drug, NYHA New York Heart Association heart failure grade, CKD chronic kidney
disease, PONV postoperative nausea and vomiting, LOS length of stay in hospital
Page 6 of 15
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 7 of 15

Table 3 Patient characteristics

Year and author Analgesic modality N Age (years) Male/female ASA ½/3 Weight (kg) Height (cm) BMI

Flisberg et al. 2001 TEA 18 61 15/3 5/10/3 77 172 26.0

IV 15 68 9/6 1/12/2 65 170 22.5
Yokoyama et al. 2005 Bilevel TEA 15 60 + /– 8 13/2 NR 61 + /– 9 162 + /– 10 23.2
IV 15 62 + /– 9 12/3 NR 60 + /– 7 161 + /– 8 23.2
Fares et al. 2014 TEA 15 53 + /– 10 12/3 9/6/0 NR NR 23.9 + /– 1.5
IV 15 59 + /– 6 11/4 8/7/0 NR NR 22.1 + /– 3.3
Liu and Wang 2015 TEA 30 56 + /– 8 25/9 13/17/0 53 + /– 14 NR NR
IV 30 54 + /– 11 23/7 11/19/0 55 + /– 13 NR NR
Wang et al. 2017 TEA 40 56 + /– 7 24/16 12/28/0 63 + /– 7 165 + /– 5 23.1
IV 40 59 + /– 4 26/14 14/26/0 64 + /– 7 166 + /– 7 23.2
Wang et al. 2019 TEA 20 56 + /– 14 NR NR 62 + /– 10 168 + /– 8 22.0
IV 20 56 + /– 14 NR NR 60 + /– 9 165 + /– 8 22.0
Li 2019 TEA 50 57 + /– 5 42/8 NR NR NR 22 + /– 4
IV 50 58 + /– 5 42/8 NR NR NR 23 + /– 4
Zhu et al. 2020 TEA 60 62 + /– 7 49/11 23/37 61 + /– 7 NR NR
IV 60 61 + /– 7 45/15 19/41 61 + /– 6 NR NR
Maghsoudloo et al. 2023 TEA 40 63 + – 8 17/23 NR NR NR NR
IV 40 63 + /– 9 24/16 NR NR NR NR
Xu et al. 2023 TEA 56 62 48/8 0/54/2 NR NR 22.4
IV 56 61 48/8 0/53/3 NR NR 22.0
PVB + TAP 56 63 43/13 1/46/9 NR NR 22.2

Other outcomes outcome sets (such as those from the StEP — COMPAC
There was not enough data to allow meta-analysis of group) with standardised endpoints, creating more com-
length of hospital stay. Only two trials recorded total hos- parable data for future meta-analyses (Myles et al. 2016;
pital stay, both without standard deviations and measur- Boney et al. 2022). Also, this result was on the basis of a
ing time to the nearest day (Xu et al. 2023; Flisberg et al. small number of small trials, with one trial being judged
2001). Other trials’ measurements varied including criti- as a high risk of bias.
cal care unit stay, postoperative care unit stay and pre- Regarding non-epidural regional techniques, we found
and postoperative hospital stay with different units of only one trial directly comparing these to thoracic epi-
measuring time. dural analgesia. Xu et al. (2023) had 3 groups with 56
Only one trial recorded technical failure of epidurals patients in each group and showed a similar 5 and 7
(Flisberg et al. 2001). This trial recorded 4 failures of epi- PPCs in its TEA group versus its PVB/TAP group, con-
dural analgesia in 18 participants (22%), and the defini- trasting to the larger 17 PPCs in its IV group (Xu et al.
tion of failure and causes was not recorded. Assuming 2023). Therefore, this combined single-shot paravertebral
the definition of failure was similar to other trials, this and transversus abdominus plane block technique (PVB/
would be in keeping with failure rates reported in the TAP) shows promise for the future but would benefit
current literature (27–32%) (Hermanides et al. 2012). from a larger body of evidence. A disadvantage is that
this technique would require expertise and time for two
Discussion separate procedures; an advantage would be no require-
This meta-analysis provides evidence for a significant ment for running a postoperative neuraxial local anaes-
reduction in a composite of PPCs in patients receiving thetic infusion with its associated risks.
TEA compared with IV opioids. However, overall cer- Regarding the secondary outcome of pain scores, this
tainty of evidence is low. Two of the three trials assess- review shows a significant reduction in pain scores for
ing PPCs did not adhere to a standardised definition, and patients receiving TEA compared to intravenous analge-
one trial measured composites of PPCs such as pneumo- sia. This significance is displayed at 24 and 48 h postoper-
nia, atelectasis or pleural effusion. A limitation of com- atively, at rest and during dynamic movement. This is in
posite measurements is the lack of clarity as to which contrast to the two previous systematic reviews in 2017
component is different. The solution to this is to use core and 2018 which did not show a significant difference
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 8 of 15

Table 4 Analgesic regimens

Year and author Analgesic modality N Analgesic location Drug regimen Additional analgesia

Flisberg et al. 2001 TEA 18 T6–12 4 ml/h 0.25% bupiv- 4–6 ml 0.25% bupivacaine clinician
acaine + 0.125 mg/ml morphine bolus + / − clinician SC morphine
IV 15 0.5–4 mg/h IV morphine 0.5–2 mg IV PCA morphine/15-min
lockout + /– clinician SC morphine
Yokoyama et al. 2005 Bilevel TEA 15 T34 + T10–11 4 ml/h 0.2% ropivacaine + 4 mcg/ 5-ml 0.2% ropivacaine clinician bolus
ml fentanyl
IV 15 1 mg/h morphine IV ­ olusa
2.5-mg IV PCA b
Fares et al. 2014 TEA 15 T5–7 0.1 ml/kg/h 0.125% bupivacaine + 5 1-mg IV PCA morphine/5-min
mcg/ml fentanyl lockout
IV 15 1-mg morphine IV PCA/5-min NR
lockout
Liu and Wang 2015 TEA 30 T7–8 4-ml load + 4 ml/h of 0.1% ropiv- 4-ml PCEA bolus/40-min lockout
acaine
IV 30 5-ml load + 1 ml/h IV of 80-ml 2-ml IV PCA bolus/15-min lockout
normal saline with 800-mg trama-
dol + 100-mg flurbiprofen
Wang et al. 2017 TEA 40 T7–8 3–4 ml/h 0.125% bupivacaine + 20 ­ olusa
3–4 ml PCEA b
mcg/ml morphine
IV 40 0.6–1 mg/h morphine ­ olusa
2–3 mg IV PCA b
Wang et al. 2017 TEA 20 NR 3 ml/h 0.125% ropivacaine + 0.4 3 ml/15-min lockout PCEA bolus
mcg/ml sufentanil
IV 20 0.03 mcg/kg/h sufentanil + 0.5 mg/ 3-ml PCA bolus/15-min lockout
ml flurbiprofen at 3 ml/h (sufentanil + flurbiprofen)
Li 2019 TEA 50 T7–8 0.125% ropivacaine + 2 mcg/ml 2-ml PCEA bolus/15-min lockout
fentanyl at 5 ml/h
IV 50 6 mcg/kg fentanyl + 12 mg/kg 2-ml IV PCA bolus/15-min lockout
tramadol in 100 ml at 2 ml/h
Zhu et al. 2020 TEA 60 T4–6 2 ml/h of 100-ml normal saline 0.5-ml PCEA bolus/15-min lockout
with 200–300 mcg fentanyl, 150-mg
ropivacaine and 5-mg droperidol
IV 60 2 ml/h of 100-ml normal saline 2-ml IV PCA bolus/15-min lockout
with 15 mcg/kg fentanyl
Maghsoudloo et al. 2023 TEA 40 T6–8 4 ml/h 0.125% bupivacaine + 1 ml/h 3-mg IV PCA morphine ­bolusa
PCEA up to max 2 ml/h
IV 40 10 mcg/kg/h morphine + ketorolac 3-mg IV PCA morphine ­bolusa
120 mg/day
Xu et al. 2023 TEA 56 T6–9 2 ml/h 0.15% ropiv- 4-ml PCEA bolus/60-min lockout
acaine + 0.12 mg/kg morphine
in 100 ml
IV 56 1 mg/h IV PCA oxycodone 2-mg IV PCA bolus/5-min lockout
PVB + TAP 56 T4–7 PVB 15-ml 0.33% ropivacaine 1 mg/h IV oxycodone + 2-mg IV PCA
at each level. TAP 20-ml 0.25% bolus 5-min lockout
ropivacaine. Both intraoperative
single-shot blocks
N number, IV intravenous, TEA thoracic epidural analgesia, PCA patient-controlled analgesia, PCEA patient-controlled epidural analgesia, SC subcutaneous, PVB
paravertebral block, TAP transversus abdominis plane block. aNo lockout time reported

in pain scores, probably due to a paucity of data at that difference suggested in the literature (Myles et al. 2017).
time (Visser et al. 2017; Hughes et al. 2018). But this is in The single non-epidural regional technique in our review
agreement with a 2024 network meta-analysis of 14 trials, (PVB/TAP group) showed higher pain scores in its trial
which also suggests a statistically significant reduction in than the TEA group but lower pain scores than the IV
pain scores with epidural versus systemic opioids (Ramjit group, with varying levels of significance, and poorer
et al. 2024). However, our results may only be clinically pain control as time progressed (Xu et al. 2023).
significant for the dynamic pain scores, which have a MD We were unable to compare patient-controlled epidural
above that of the 10-mm minimum clinically important analgesia (PCEA) versus continuous epidural analgesia
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 9 of 15

Table 5 Postoperative pulmonary complications (PPCs) in this review, as only two trials ran continuous epidural
Year and author Analgesic modality N Pulmonary
regimens (Maghsoudloo et al. 2023), (Fares et al. 2014).
complications There was also a large amount of methodological diver-
sity within trials that allowed epidural boluses, some
Fares et al. 2014 TEA 15 3
being clinician bolus only (not PCEA) (Flisberg et al.
IV 15 12
2001; Yokoyama et al. 2005). Bolus volumes varied, there
Maghsoudloo et al. 2023 TEA 40 0
were differences in concentration and type of local anaes-
IV 40 5
thetic and some trials had unspecified lockout times
Xu et al. 2023 TEA 56 5
(Flisberg et al. 2001; Wang et al. 2017). Epidurals were
IV 56 17
sited at different thoracic vertebral levels with one trial
PVB + TAP 56 7 siting two thoracic epidurals (Yokoyama et al. 2005).

Table 6 Pain scores at rest (+ / − standard deviations)

Year and author Analgesic modality N Pain score Pain score Pain score method
24-h rest 48-h rest

Flisberg et al. 2001 TEA 18 18.75 + / − 5.23 13.5 + / − 3.80 VAS

IV 15 15.75 + / − 4.57 12.5 + / − 4.42
Yokoyama et al. 2005 Bilevel TEA 15 7+/−7 7+/−7 Box scale
IV 15 12.5 + / − 8.5 11 + / − 7
Fares et al. 2014 TEA 15 9 + / − 6.82 7 + / − 5.34 VAS
IV 15 25 + / − 7.57 25 + / − 6.94
Liu and Wang 2015 TEA 30 6.5 + / − 1.5 5.7 + / − 1.2 VAS
IV 30 17.7 + / − 4.1 15.8 + / − 3.4
Wang et al. 2017 TEA 40 29 + / − 2 25 + / − 2 VAS
IV 40 43 + / − 4 32 + / − 4
Wang et al. 2019 TEA 20 15.7 + / − 5.9 21.5 + / − 4.8 VAS
IV 20 24.7 + / − 7.3 31.5 + / − 5.4
Li 2019 TEA 50 22.5 + / − 7.44 21 + / − 5.83 VAS
IV 50 33 + / − 8.67 30.5 + / − 8.67
Zhu et al. 2020 TEA 60 17.5 + / − 9 13 + / − 6.5 VAS
IV 60 26 + / − 7 22.5 + / − 6
Maghsoudloo et al. 2023 TEA 40 38 + / − 11 28 + / − 7 VAS
IV 40 41 + / − 11 31 + / − 9
Xu et al. 2023 TEA 56 4.3 + / − 7 2.6 + / − 5.8 VAS
IV 56 18.5 + / − 9.5 19 + / − 11
PVB + TAP 56 6+/−9 6.2 + / − 9

Table 7 Pain scores on movement (+ / − standard deviations)

Year and author Analgesic modality N Pain score Pain score Dynamic pain score
24-h dynamic 48-h dynamic

Flisberg et al. 2001 TEA 18 33.5 + / − 8.74 37.75 + / − 7.37 Movement to sitting position
IV 15 35 + / − 9.24 43 + / − 7.85
Fares et al. 2014 TEA 15 28 + / − 9.96 27 + / − 8.78 NR
IV 15 48 + / − 11.32 48 + / − 10.38 NR
Li et al. 2019 TEA 50 28 + / − 7.13 23.5 + / − 5.67 On active coughing
IV 50 40.5 + / − 9.72 38.5 + / − 9.29
Xu et al. 2023 TEA 56 10 + / − 14 11.5 + / − 13 NR
IV 56 36 + / − 15 37 + / − 14 NR
PVB + TAP 56 16.5 + / − 14 19 + / − 14 NR
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 10 of 15

Table 8 Postoperative nausea and vomiting Other limitations include the small numbers of partici-
Year and author Analgesic modality N Postoperative
pants in some trials with no large clinical effectiveness
nausea and trials. Small RCTs can overestimate treatment effects in
vomiting the real world (Dechartres et al. 2024). There was meth-
Wang et al. 2017 TEA 40 13
odological diversity in the inclusion and exclusion cri-
IV 40 14
teria, populations were from different countries and
Li 2019 TEA 50 16
there were differences in surgical approaches. Nearly all
IV 50 5
included studies used an open approach (large abdominal
Xu et al. 2023 TEA 56 0
incision), with one study using a laparoscopic approach.
Although there is a lack of evidence regarding analgesic
IV 56 2
strategies in laparoscopic surgery, omitting this study
PVB + TAP 56 0
does not significantly change the results of the meta-
analyses. These differences may have contributed to the
high statistical heterogeneity in the pain meta-analyses.
Table 9 Length of hospital stay
Trials had little data on morbidity which was also
Year and author Analgesic modality N Length of poorly defined, and many did not assess pain after 48 h.
hospital stay However, within the 48-h postsurgical timeframe, pain
(days)
scores were well reported, with nearly all trials using a
Flisberg et al. 2001 TEA 18 17 (range 9–59) visual analogue scale (VAS) at standardised time inter-
IV 15 16 (range 8–44) vals. Pain measurement with the VAS is a validated, sub-
Xu et al. 2023 TEA 56 15 jective measure in acute pain which is well understood
IV 56 15 by patients (Delgado et al. 2024; Haefeli and Elfering
PVB + TAP 56 14 2006). Its measurement at rest and movement at 24 h is
also considered a key patient-reported outcome measure
(Myles et al. 2018).
Regarding the limitations of this systematic review
process, we did not search for trials with non-epidural
regional techniques, unless they included epidural anal-
gesia as a comparator group. This was to avoid the bias
associated with indirect comparisons within a network
meta-analysis (Feenstra et al. 2023). We searched three
large databases, but others were omitted. Regarding the
data extraction process, the conversion of graphical data
to numerical data using online software was required,
which does not have perfect accuracy. One trial did not
report its standard deviation for pain scores, and this
was imputed in order for inclusion in the meta-analyses
(Fares et al. 2014). Lastly, a lack of data for less invasive
surgical techniques and non-epidural regional techniques
is a limitation of this review. However, these surgical
techniques are not yet developed at many centres, are
not available for all levels of disease progression and have
not yet shown clear short- and long-term benefits (Jebril
et al. 2024).
This review has the benefit of including recent evidence
and being restricted to oesophagectomy patients only,
who have very specific analgesic requirements, compared
to older systematic reviews which included data from
non-randomised trials (Visser et al. 2017) and data from
(non-oesophagectomy) gastric surgery patients (Hughes
et al. 2018). It is the first systematic review to interro-
gate respiratory outcomes, albeit a composite, and it
Fig. 2 Risk-of-bias summary is the first to show evidence to support a reduction in a
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 11 of 15

Fig. 3 Meta-analysis of postoperative pulmonary complications (PPCs)

composite of PPCs with TEA versus intravenous analge- Cochrane Highly Sensitive Search Strategy for identify-
sia. It also supports the recent network meta-analysis by ing randomized trials in MEDLINE: sensitivity-maximiz-
Ramjit et al., by showing a reduction in pain scores with ing version (2008 revision); PubMed format
TEA versus intravenous analgesia (Ramjit et al. 2024). #1 randomized controlled trial [pt]
Contextualising this, our systematic review can sup- #2 controlled clinical trial [pt]
port clinicians in utilising thoracic epidural analgesia for #3 randomized [tiab]
reducing pain and PPCs. It can aid discussions in preop- #4 placebo [tiab]
erative counselling, shared decision-making and during #5 drug therapy [sh]
perioperative risk stratification and planning of postop- #6 randomly [tiab]
erative care. #7 trial [tiab]
#8 groups [tiab]
Conclusions #9 #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8
This meta-analysis provides evidence that TEA should #10 animals [mh] NOT humans [mh]
currently remain the gold standard analgesic technique #11 #9 NOT #10
for reducing pain after elective oesophagectomy. It is also Cochrane Highly Sensitive Search Strategy for identify-
the first review to provide evidence that TEA reduces a ing randomized controlled trials in Embase: (2020 revi-
composite of PPCs following oesophagectomy surgery, sion); Embase.com format
although this conclusion is of low certainty. Future trials #1 ‘randomized controlled trial’/de
are needed to compare TEA administration techniques, #2 ‘controlled clinical trial’/de
including PCEA. Non-epidural regional analgesic tech- #3 random*:ti,ab,tt
niques should also be considered for future research. Tri- #4 ‘randomization’/de
als must include more recent laparoscopic and minimally #5 ‘intermethod comparison’/de
invasive surgical approaches, since the benefit and risk #6 placebo:ti,ab,tt
profile of TEA may not be generalised to these patient #7 (compare:ti,tt OR compared:ti,tt OR comparison:ti,tt)
groups. Appropriate powering to detect clinical effective- #8 ((evaluated:ab OR evaluate:ab OR evaluating:ab
ness is required, as is the use of core outcome sets with OR assessed:ab OR assess:ab) AND (compare:ab OR
standardised endpoints (Myles et al. 2016). compared:ab OR comparing:ab OR comparison:ab))
#9 (open NEXT/1 label):ti,ab,tt
#10 ((double OR single OR doubly OR singly) NEXT/1
Appendix (blind OR blinded OR blindly)):ti,ab,tt
Search Criteria #11 ‘double blind procedure’/de
1# (o)esophagectom* OR (o)esophagogastric resection #12 (parallel NEXT/1 group*):ti,ab,tt
OR (o)esophageal surger* OR Ivor Lewis. #13 (crossover:ti,ab,tt OR ‘cross over’:ti,ab,tt).
AND #14 ((assign* OR match OR matched OR allocation)
2# epidural* NEAR/6 (alternate OR group OR groups OR intervention
AND OR interventions OR patient OR patients OR subject OR
3# Cochrane Highly Sensitive Search Strategy for RCTs subjects OR participant OR participants)):ti,ab,tt
(Lefebvre et al. 2023) #15 (assigned:ti,ab,tt OR allocated:ti,ab,tt)
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 12 of 15

Fig. 4 a Meta-analysis of pain scores at rest at 24 h. b Meta-analysis of pain scores at rest at 48 h. c Meta-analysis of dynamic pain scores at 24 h. d
Meta-analysis of dynamic pain scores at 48 h
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 13 of 15

Fig. 5 Meta-analysis of postoperative nausea and vomiting (PONV)

#16 (controlled NEAR/8 (study OR design OR OR dog:ti,tt OR dogs:ti,tt OR cattle:ti,tt OR bovine:ti,tt

trial)):ti,ab,tt OR monkey:ti,tt OR monkeys:ti,tt OR trout:ti,tt OR
#17 (volunteer:ti,ab,tt OR volunteers:ti,ab,tt) marmoset*:ti,tt) AND ‘animal experiment’/de)
#18 ‘human experiment’/de #33 (‘animal experiment’/de NOT (‘human experi-
#19 trial:ti,tt ment’/de OR ‘human’/de))
#20 #1OR#2OR#3OR#4OR#5OR#6OR#7OR#8OR#9 #34 #21OR#22OR#23OR#24OR#25OR#26OR#27OR#
OR#10OR#11OR#12OR#13 OR #14 OR #15 OR #16 OR 28OR#29OR#30OR#31OR #32 OR #33
#17 OR #18 OR #19 #35 #20 NOT #34
#21 (((random* NEXT/1 sampl* NEAR/8 (‘cross sec-
tion*’ OR questionnaire* OR survey OR surveys OR
Abbreviations
database or databases)):ti,ab,tt) NOT (‘comparative PROSPERO International Prospective Register of Systematic Reviews
study’/de OR ‘controlled study’/de OR ‘randomised PRISMA Preferred Reporting Items for Systematic reviews and Meta-Analyses
controlled’:ti,ab,tt OR ‘randomized controlled’:ti,ab,tt OR PICOS  Population, Intervention, Comparison, Outcomes and Study
design framework for research projects
‘randomly assigned’:ti,ab,tt)) PPC Postoperative pulmonary complications
#22 (‘cross‐sectional study’/de NOT (‘randomized con- EPCO European Perioperative Clinical Outcomes
trolled trial’/de OR ‘controlled clinical study’/de OR ‘con- VAS Visual analogue scale
PONV Postoperative nausea and vomiting
trolled study’/de OR ‘randomised controlled’:ti,ab,tt OR MEDLINE Medical Literature Analysis and Retrieval System Online
‘randomized controlled’:ti,ab,tt OR ‘control group’:ti,ab,tt EMBASE Excerpta Medica Database
OR ‘control groups’:ti,ab,tt)) CENTRAL Cochrane Central Register of Controlled Trials
HSSS Cochrane Highly Sensitive Search Strategy filter
#23 (‘case control*’:ti,ab,tt AND random*:ti,ab,tt RevMan Review Manager
NOT (‘randomised controlled’:ti,ab,tt OR ‘randomized CEBM Centre for Evidence-Based Medicine
controlled’:ti,ab,tt)) TEA Thoracic epidural analgesia
IV Intravenous
#24 (‘systematic review’:ti,tt NOT (trial:ti,tt OR PCA Patient-controlled analgesia
study:ti,tt)) PCEA Patient-controlled epidural analgesia
#25 (nonrandom*:ti,ab,tt NOT random*:ti,ab,tt) SC Subcutaneous
TAP Transversus abdominus plane
#26 ‘random field*’:ti,ab,tt PVB Paravertebral block
#27 (‘random cluster’ NEAR/4 sampl*):ti,ab,tt ASA American Society of Anesthesiologists
#28 (review:ab AND review:it) NOT trial:ti,tt COPD Chronic obstructive pulmonary disease
BMI Body mass index
#29 (‘we searched’:ab AND (review:ti,tt OR review:it)) NSAID Nonsteroidal anti-inflammatory drug
#30 ‘update review’:ab NYHA New York Heart Association heart failure grade
#31 (databases NEAR/5 searched):ab CKD Chronic kidney disease
LOS Length of stay in hospital
#32 ((rat:ti,tt OR rats:ti,tt OR mouse:ti,tt OR mice:ti,tt RCT​ Randomised controlled trial
OR swine:ti,tt OR porcine:ti,tt OR murine:ti,tt OR NR Not reported
sheep:ti,tt OR lambs:ti,tt OR pigs:ti,tt OR piglets:ti,tt n/a Not applicable
CONSORT Consolidated Standards of Reporting Trials
OR rabbit:ti,tt OR rabbits:ti,tt OR cat:ti,tt OR cats:ti,tt
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 14 of 15

MD Mean differences of Anaesthetists. Br J Anaesth. 2009;102(2):179–90. [cited 2023 Sep 13].

SD Standard deviation Available from: http://​www.​bjana​esthe​sia.​org.​uk/​artic​le/​S0007​09121​
RR Risk ratio 73452​82/​fullt​ext.
CI Confidence interval Dechartres A, Trinquart L, Boutron I, Ravaud P. Influence of trial sample
ARDS Acute respiratory distress syndrome size on treatment effect estimates: meta-epidemiological study. BMJ.
StEP Standardised Endpoints in Perioperative Medicine 2013;346(7908). [cited 2024 Feb 1]. Available from: https://​www.​bmj.​
COMPAC Core Outcome Measures for Perioperative and Anaesthetic Care com/​conte​nt/​346/​bmj.​f2304
CINAHL Cumulative Index to Nursing and Allied Health Literature database Delgado DA, Lambert BS, Boutris N, McCulloch PC, Robbins AB, Moreno
LILACs Latin American and Caribbean Health Sciences Literature database MR, et al. Validation of digital visual analog scale pain scoring with a
traditional paper-based visual analog scale in adults. J Am Acad Orthop
Acknowledgements Surg Glob Res Rev. 2018;2(3):p e088. [cited 2024 Feb 14]. Available from:
Not applicable. https://​www.​ncbi.​nlm.​nih.​gov/​pmc/​artic​les/​PMC61​32313/.
Dolin SJ, Cashman JN. Tolerability of acute postoperative pain management:
Authors’ contributions nausea, vomiting, sedation, pruritis, and urinary retention. Evidence from
DM was responsible for study concept, design, literature review, methodology, published data. Br J Anaesth. 2005;95(5):584–91. [cited 2023 Nov 18].
analysis and interpretation of data and writing of the manuscript. AB provided Available from: https://​www.​bja.​oxfor​djour​nals.​org.
assistance with study design, data extraction and quality assessment. PO Fares KM, Muhamed SA, Hamza HM, Sayed DM, Hetta DF. Effect of thoracic
guided study design, methodology and revision of the manuscript. All authors epidural analgesia on proinflammatory cytokines in patients subjected
reviewed the manuscript. to protective lung ventilation during Ivor Lewis esophagectomy. Pain
Physician. 2014;17(4):305–15.
Funding Feenstra ML, van Berge Henegouwen MI, Hollmann MW, Hermanides J,
No funding was provided for this research. Eshuis WJ. Analgesia in esophagectomy: a narrative review. J Thorac
Dis. 2023;15(9):5099–111. [cited 2024 Jan 28]. Available from: https://​jtd.​
Availability of data and materials amegr​oups.​org/​artic​le/​view/​78340/​html.
No datasets were generated or analysed during the current study. Feenstra ML, van Berge Henegouwen MI, Hollmann MW, Hermanides J,
Eshuis WJ. Analgesia in esophagectomy: a narrative review. J Thorac Dis.
2023;15(9):5099–111. [cited 2023 Dec 9]. Available from: https://​jtd.​amegr​
Declarations oups.​org/​artic​le/​view/​78340/​html.
Flisberg P, Törnebrandt K, Walther B, Lundberg J. Pain relief after esophagec-
Ethics approval and consent to participate tomy: thoracic epidural analgesia is better than parenteral opioids. J
Ethics approval and consent were not required as all individual trials had Cardiothorac Vasc Anesth. 2001;15(3):282–7. [cited 2023 Dec 9]. Available
already gained approval prior to publication. from: https://​pubmed.​ncbi.​nlm.​nih.​gov/​11426​356/.
Haefeli M, Elfering A. Pain assessment. European Spine Journal. 2006;15(Suppl
Consent for publication 1):S17. [cited 2024 Feb 14]. Available from: /pmc/articles/PMC3454549/.
Not applicable. Hermanides J, Hollmann MW, Stevens MF, Lirk P. Failed epidural: causes and
management. BJA British Journal of Anaesthesia. 2012;109(2):144–54.
Competing interests https://​doi.​org/​10.​1093/​bja/​aes214. [cited 2023 Nov 18].
The authors declare no competing interests. Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The
Cochrane Collaboration’s tool for assessing risk of bias in randomised
Author details trials. BMJ. 2011;343(7829). [cited 2023 Oct 18]. Available from: https://​
1
Department of Anaesthesia and Perioperative Medicine, University College www.​bmj.​com/​conte​nt/​343/​bmj.​d5928
London Hospitals NHS Foundation Trust, London, England. 2 University College Holtz M, Liao N, Lin JH, Asche CV. Economic outcomes and incidence of post-
London, London, England. surgical hypotension with liposomal bupivacaine vs epidural analgesia
in abdominal surgeries. J Health Econ Outcomes Res. 2022;9(2):86. [cited
Received: 13 May 2024 Accepted: 14 July 2024 2023 Oct 18]. Available from: /pmc/articles/PMC9473799/.
Hughes M, Yim I, Deans DAC, Couper GW, Lamb PJ, Skipworth RJE. System-
atic review and meta-analysis of epidural analgesia versus different
analgesic regimes following oesophagogastric resection. World J Surg.
2018;42(1):204–10. [cited 2023 Sep 13]. Available from: https://​pubmed.​
References ncbi.​nlm.​nih.​gov/​28741​191/.
Altman DG. Statistics with confidence : confidence intervals and statistical Jammer I, Wickboldt N, Sander M, Smith A, Schultz MJ, Pelosi P, et al. Standards
guidelines. 2000;240. for definitions and use of outcome measures for clinical effectiveness
Boney O, Moonesinghe SR, Myles PS, Grocott MPW, Bartoszko J, Beattie research in perioperative medicine: European Perioperative Clinical Out-
WS, et al. Core Outcome Measures for Perioperative and Anaesthetic come (EPCO) definitions: a statement from the ESA-ESICM joint taskforce
Care (COMPAC): a modified Delphi process to develop a core out- on perioperative outcome measures. Eur J Anaesthesiol. 2015;32(2):88–
come set for trials in perioperative care and anaesthesia. Br J Anaesth. 105. [cited 2023 Nov 18]. Available from: https://​pubmed.​ncbi.​nlm.​nih.​
2022;128(1):174–85. [cited 2024 Feb 1]. Available from: http://​www.​bjana​ gov/​25058​504/.
esthe​sia.​org.​uk/​artic​le/​S0007​09122​10062​2X/​fullt​ext. Jebril W, Klevebro F, Rouvelas I, Nilsson M. Open, hybrid or total minimally
Booka E, Takeuchi H, Suda K, Fukuda K, Nakamura R, Wada N, et al. Meta- invasive esophagectomy; a comprehensive review based on a systematic
analysis of the impact of postoperative complications on survival after literature search. Annals of Esophagus. 2021;4(0). [cited 2024 Jun 23].
oesophagectomy for cancer. BJS Open. 2018;2(5):276. [cited 2023 Oct 18]. Available from: https://​aoe.​amegr​oups.​org/​artic​le/​view/​5557/​html
Available from: /pmc/articles/PMC6156161/ Lefebvre C, Glanville J, Briscoe S, Featherstone R, Littlewood A, Marshall C, et al.
Borzellino G, Francis NK, Chapuis O, Krastinova E, Dyevre V, Genna M. Role of Technical supplement to Chapter 4: searching for and selecting studies.
epidural analgesia within an ERAS program after laparoscopic colorectal 2022; [cited 2023 Oct 18]. Available from: https://​metho​ds.​cochr​ane.​org/​
surgery: a review and meta-analysis of randomised controlled studies. mecir
Surg Res Pract. 2016;2016. [cited 2024 Jun 22]. Available from: https://​ Li Y, Dong H, Tan S, Qian Y, Jin W. Effects of thoracic epidural anesthesia/anal-
pubmed.​ncbi.​nlm.​nih.​gov/​27642​630/ gesia on the stress response, pain relief, hospital stay, and treatment
CombineMeanSD. [cited 2023 Dec 10]. Available from: https://​www.​stats​todo.​ costs of patients with esophageal carcinoma undergoing thoracic sur-
com/​Combi​neMea​nsSDs.​php gery: a single-center, randomized controlled trial. Medicine. 2019;98(7).
Cook TM, Counsell D, Wildsmith JAW. Major complications of central neuraxial [cited 2023 Dec 9]. Available from: https://​pubmed.​ncbi.​nlm.​nih.​gov/​
block: report on the Third National Audit Project of the Royal College 30762​735/
Macrosson et al. Perioperative Medicine (2024) 13:80 Page 15 of 15

Liu Y, Wang P. Effect of different analgesia methods on blood coagulation Biomedical Res-India. 2017;28(5):2081–6. . [cited 2023 Dec 9]. Available
in patients who underwent esophageal carcinoma surgery. Chin J Clin from: https://​www.​biome​dres.​info.
Oncol. 2015;42(7):378–81. Wang J, Yin Y, Zhu Y, Xu P, Sun Z, Miao C, et al. Thoracic epidural anaesthesia
Low DE, Allum W, De Manzoni G, Ferri L, Immanuel A, Kuppusamy MK, et al. and analgesia ameliorates surgery-induced stress response and postop-

After Surgery (ERAS®) Society recommendations. World Journal of Sur-

Guidelines for perioperative care in esophagectomy: Enhanced Recovery erative pain in patients undergoing radical oesophagectomy. J Int Med
Res. 2019;47(12):6160–70.
gery. 2018;43(2):299–330. [cited 2023 Sep 13]. Available from: https://​link.​ Xu M, Feng Y, Song X, Fu S, Lu XF, Lai J, et al. Combined ultrasound-guided
sprin​ger.​com/​artic​le/​10.​1007/​s00268-​018-​4786-4. thoracic paravertebral nerve block with subcostal transversus abdominis
Maghsoudloo M, Abbassi S, Samadi S, Beigmohammadi MT, Hajipour A, Atash- plane block for analgesia after total minimally invasive McKeown
gahi M, et al. Comparison between two methods of patient-controlled esophagectomy: a randomized, controlled, and prospective study. Pain
analgesia through intravenous and thoracic epidural to control pain and Ther. 2023;12(2):475–89. [cited 2023 Dec 9]. Available from:https://​pub-
complications after surgery in esophageal cancer patients: a randomized med.​ncbi.​nlm.​nih.​gov/​36648​745/.
controlled trial. Acta Med Iran. 2023;61(6):334–40. [cited 2023 Dec 9]. Yokoyama M, Itano Y, Katayama H, Morimatsu H, Takeda Y, Takahashi T,
Available from: https://​publi​sh.​kne-​publi​shing.​com/​index.​php/​ACTA/​artic​ et al. The effects of continuous epidural anesthesia and analgesia on
le/​view/​13833. stress response and immune function in patients undergoing radical
Mann C, Berlth F, Hadzijusufovic E, Lang H, Grimminger PP. Minimally invasive esophagectomy. Anesth Analg. 2005;101(5):1521–7. [cited 2023 Dec 9].
esophagectomy: clinical evidence and surgical techniques. Langenbecks Available from: https://​pubmed.​ncbi.​nlm.​nih.​gov/​16244​024/.
Arch Surg. 2020;405(8):1061. [cited 2024 Jun 24]. Available from: /pmc/ Zhu R, Xiang J, Tan M. Effects of different anesthesia and analgesia on cellular
articles/PMC7686170/ immunity and cognitive function of patients after surgery for esophageal
Myles PS, Group on behalf of the CS, Grocott MPW, Group on behalf of the CS, cancer. Minerva Chir. 2020;75(6):449–56. [cited 2023 Dec 9]. Available
Boney O, Group on behalf of the CS, et al. Standardizing end points in from: https://​pubmed.​ncbi.​nlm.​nih.​gov/​32773​737/.
perioperative trials: towards a core and extended outcome set. BJA Brit
J Anaesth. 2016;116(5):586–9. https://​doi.​org/​10.​1093/​bja/​aew066 [cited
2024 Feb 1] Publisher’s Note
Myles PS, Myles DB, Galagher W, Boyd D, Chew C, MacDonald N, et al. Measur- Springer Nature remains neutral with regard to jurisdictional claims in pub-
ing acute postoperative pain using the visual analog scale: the minimal lished maps and institutional affiliations.
clinically important difference and patient acceptable symptom state. Br
J Anaesth. 2017;118(3):424–9. [cited 2024 Jun 23]. Available from: http://​
www.​bjana​esthe​sia.​org.​uk/​artic​le/​S0007​09121​73021​2X/​fullt​ext.
Myles PS, Boney O, Botti M, Cyna AM, Gan TJ, Jensen MP, et al. Systematic
review and consensus definitions for the Standardised Endpoints in
Perioperative Medicine (StEP) initiative: patient comfort. Br J Anaesth.
2018;120(4):705–11. [cited 2024 Feb 14]. Available from: http://​www.​
bjana​esthe​sia.​org/​artic​le/​S0007​09121​83000​11/​fullt​ext.
Odor PM, Bampoe S, Gilhooly D, Creagh-Brown B, Ramani Moonesinghe S.
Perioperative interventions for prevention of postoperative pulmonary
complications: systematic review and meta-analysis. BMJ. 2020;368. [cited
2023 Nov 18]. https://​www.​bmj.​com/​conte​nt/​368/​bmj.​m540
Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al.
The PRISMA 2020 statement: an updated guideline for reporting system-
atic reviews. BMJ. 2021;372. [cited 2023 Oct 18]. Available from: https://​
www.​bmj.​com/​conte​nt/​372/​bmj.​n71
Pirie K, Myles PS, Riedel B. A survey of neuraxial analgesic preferences in open
and laparoscopic major abdominal surgery amongst anaesthetists in
Australia and New Zealand. Anaesth Intensive Care. 2020;48(4):314–7.
[cited 2023 Sep 13]. Available from: https://​journ​als.​sagep​ub.​com/​doi/​10.​
1177/​03100​57X20​937315.
Ramjit S, Davey MG, Loo C, Moran B, Ryan EJ, Arumugasamy M, et al. Evaluat-
ing analgesia strategies in patients who have undergone oesophagec-
tomy—a systematic review and network meta-analysis of randomised
clinical trials. Diseases of the Esophagus. 2024; [cited 2024 Apr 20].
https://​doi.​org/​10.​1093/​dote/​doad0​74
Rigg JRA, Jamrozik K, Myles PS, Silbert BS, Peyton PJ, Parsons RW, et al. Epidural
anaesthesia and analgesia and outcome of major surgery: a randomised
trial. Lancet. 2002;359(9314):1276–82. [cited 2023 Sep 13]. Available from:
https://​pubmed.​ncbi.​nlm.​nih.​gov/​11965​272/.
Rohatgi A. https://​autom​eris.​io/​WebPl​otDig​itizer. 2022 [cited 2023 Nov 22].
WebPlotDigitizer Version 4.6, Pacifica, CA, USA. Available from: https://​
apps.​autom​eris.​io/​wpd/
The Cochrane Collaboration. Review Manager (RevMan) version 7.2.0. 2024
[cited 2024 Apr 25]. Available from: revman.cochrane.org
Turi S, Gemma M, Braga M, Monzani R, Radrizzani D, Beretta L. Epidural anal-
gesia vs systemic opioids in patients undergoing laparoscopic colorectal
surgery. Int J Colorectal Dis. 2019;34(5). [cited 2024 Jun 23]. Available
from: https://​pubmed.​ncbi.​nlm.​nih.​gov/​30927​065/
Visser E, Marsman M, van Rossum PSN, Cheong E, Al-Naimi K, van Klei WA, et al.
Postoperative pain management after esophagectomy: a systematic
review and meta-analysis. Dis Esophagus. 2017;30(10):1–11. . [cited 2023
Sep 13]. Available from: https://​pubmed.​ncbi.​nlm.​nih.​gov/​28859​388/.
Wang LJ, Cang J, Xue ZG. Cost and effectiveness comparison of thoracic epi-
dural anaesthesia combined with general anaesthesia in esophagectomy.