SCIENTIFIC REVIEW

Esophagectomy for End-Stage Achalasia: Systematic Reviewand Meta-analysis

Alberto Aiolfi1 • Emanuele Asti1 • Gianluca Bonitta1 • Luigi Bonavina1

Published online: 11 October 2017

� Societe Internationale de Chirurgie 2017

Abstract

Background Indications for surgery and clinical outcomes of esophagectomy in the management of end-stage

achalasia are not clearly defined. The aim of this systematic review and meta-analysis was to provide evidence-based

information to help in the decision-making and in the choice of surgical technique.

Methods An extensive literature search was conducted to identify all reports on esophagectomy for end-stage

achalasia patients over the past three decades. MEDLINE, Embase and Cochrane databases were thoroughly con-

sulted matching the terms ‘‘achalasia,’’ ‘‘end-stage achalasia,’’ ‘‘esophagectomy’’ and ‘‘esophageal resection’’ with

‘‘AND’’ and ‘‘OR.’’ Short- and long-term outcome data were extracted. Pooled prevalence of pneumonia, anasto-

motic leakage and mortality were calculated using Freeman–Tukey double arcsine transformation and DerSimonian–

Laird estimator in random effect meta-analysis. Heterogeneity among studies was evaluated using I2-index and

Cochrane Q test. Meta-regression was used to address the effect of potential confounders.

Results Eight papers published between 1989 and 2014 matched the inclusion criteria. In total, 1307 patients were

included. Esophagectomy was performed through a transthoracic (78.7%) or a transhiatal (21.3%) approach. The

stomach was used as an esophageal substitute in 95% of patients. Pooled prevalence of pneumonia, anastomotic

leakage and mortality were 10% (95% CI 4–18%), 7% (95% CI 4–10%) and 2% (95% CI 1–3%), respectively.

Conclusions Esophagectomy for end-stage achalasia is safe and effective. Based on the results of this study,

esophagectomy should be performed without hesitation in patients who are fit for major surgery and present with

disabling symptoms, poor quality of life and dolichomegaesophagus recalcitrant to multiple endoscopic dilatations

and/or surgical myotomies.

Introduction

Achalasia is a rare disease, with an estimated annual

incidence of 1 case per 100,000/year. It results from a

dysfunction of the esophageal myenteric plexus that causes

impaired lower esophageal sphincter (LES) relaxation and

loss of peristalsis [1]. Dysphagia, regurgitation, weight

loss, respiratory symptoms and aspiration pneumonia are

common. The incidence of squamous cell carcinoma in

achalasia is 312.4 cases per 100,000 patients/year [2].

Treatment of achalasia aims to reduce the pressure

gradient across the LES and improve gravitational eso-

phageal emptying in an effort to palliate symptoms and halt

the natural history of the disease [3]. Pneumatic dilatation

and surgical myotomy are the two most commonly used

first-line treatment modalities [4, 5]. While some patients

achieve excellent long-term symptom control with a single

& Luigi Bonavina

luigi.bonavina@unimi.it

1 Division of General Surgery, Department of Biomedical

Sciences for Health, IRCCS Policlinico San Donato,

University of Milan, Piazza Edmondo Malan, 1,

20097 San Donato Milanese, MI, Italy

123

World J Surg (2018) 42:1469–1476

DOI 10.1007/s00268-017-4298-7

intervention, other patients will require multiple treatments

[6]. In the absence of therapy, or when therapy is inade-

quate, progressive dilatation and increasing tortuosity of

the esophagus lead to the end-stage disease in about 5% of

patients [7].

Esophagectomy may be required in patients with end-

stage achalasia to restore alimentary transit, reverse nutri-

tional deficiencies and decrease the risk of aspiration

pneumonia and carcinoma [8]. The aim of this systematic

review and meta-analysis was to examine the current evi-

dence on the therapeutic role and outcomes of

esophagectomy in the management of end-stage achalasia.

Methods

We conducted this study according to the Preferred

Reporting Items for Systematic Reviews and Meta-analy-

ses (PRISMA) statement [9]. An extensive literature search

between 1987 and 2017 was conducted by two independent

authors (AA and SS) to identify the English-written pub-

lished series on esophagectomy in end-stage achalasia

patients. MEDLINE, Embase and Cochrane databases were

consulted matching the terms ‘‘achalasia,’’ ‘‘end-stage

achalasia,’’ ‘‘esophagectomy’’ and ‘‘esophageal resection’’

with ‘‘AND’’ and ‘‘OR.’’ The search was completed by

consulting the listed references of each article.

Articles with sample size [15 were included in the

systematic review; abstracts, case reports, case series and

non-English-written articles were excluded (Fig. 1). Two

authors (AA and EA) independently extracted data from

eligible studies. Data extracted included study character-

istics (first author name, year and journal of publication),

number of patients included in the series, time frame,

clinical and demographic characteristics of patients’ pop-

ulation, type of surgical procedure and postoperative out-

comes. Disagreements between authors were resolved by

consensus; if no agreement could be reached, a third senior

author (LB) made the decision.

Three investigators independently assessed the

methodological quality of the papers using the Newcastle–

Ottawa Scale (NOS) [10]. Each study is judged on a ‘‘star

system’’ based on the selection of the study groups and the

ascertainment of outcome of interest. Each study could

earn a maximum of 9 stars. Studies with low-quality score

(NOS\6) were excluded.

Statistical analysis

Proportions were transformed via the Freeman–Tukey

double arcsine method [11] and with the corresponding

back-transformation equation [12]. Then an inverse-vari-

ance-weighted random effect meta-analysis was performed

by conventional methods using DerSimonian–Laird esti-

mator for estimate between-study variance (s2) [13].

Clopper–Pearson 95% confidence interval for individual

was computed [14]. Heterogeneity among the studies was

evaluated by I2-index, Cochran Q test and using visual

inspection of radial plot [15]. Small study and publication

bias effects were assessed by trim and fill funnel plot visual

inspection [16]. Egger tests were applied [17]. Prediction

interval for treatment effect of a new study is calculated

according to Borestein [15]. As sample size is not the same

in all studies, we gradually removed small sample size to

perform a sensitivity analysis to assess stability of results.

Two-sided p value were considered statistically significant

when\0.05. All analyses were carried out using R version

3.2.2 software [18].

Results

Systematic review

Eight studies published between 1989 and 2014 met the

inclusion criteria. The total number of patients was 1307;

the sample size of the individual studies ranged from 15 to

963. All reports were observational, cohort studies; each

study earned a NOS score of 6 or 7 (median 6.4), sug-

gesting a good quality level. Demographic, clinical and

operative variables of the patient sample are shown in

Table 1. The largest report included 963 patients registered

in a national administrative database. Five papers included

more than 30 patients. The age of the included patients

ranged from 41 to 54.6. Half of the patients were females

(50.5%), and the duration of symptoms before

esophagectomy ranged from 0.3 to 21 years.

The majority of patients had multiple previous endo-

scopic and/or surgical attempts to relieve long-lasting

achalasia symptoms. The indication for esophageal resec-

tion was mostly driven by recurrent and disabling symp-

toms and radiographic findings of sigmoid esophagus. All

patients were operated through an open surgical approach.

Overall, 1029 patients (78.7%) were managed through a

transthoracic approach while 278 (21.3%) underwent a

transhiatal esophagectomy. The stomach was most com-

monly (95.3%) used to replace the esophagus. The post-

operative morbidity ranged from 19 to 50%. Pneumonia

and anastomotic leaks were the two most commonly

reported complications. The mortality ranged from 0 to

5.4%. The hospital length of stay ranged from 10 to

16 days. Overall, 27.5% of the patients required endo-

scopic dilation of the anastomosis within 1 year. At a

follow-up ranging from 12 to 75 months, 7 of the 8 studies

reported that the proportion of patients on an unrestricted

diet ranged from 75 to 100%. One study reported a median

1470 World J Surg (2018) 42:1469–1476

123

weight gain of 6.3 kg at a mean follow-up of 72 months

[21]. No validated questionnaires for quality of life

assessment were used in any of the studies.

Meta-analysis

In addition to a systematic review, we performed a fre-

quentist meta-analysis. Considering random effect model,

the estimated pooled prevalence of pneumonia resulting

from 7 studies, which include a total of 1185 patients, is

10% (95% CI 4–18%). The prediction lower and upper

limits are 0 and 42%, respectively. The heterogeneity index

is high (I2 = 82%, 95% CI 64–91%; p\ 0.01). Funnel plot

shows that publication and small study bias should be

considered. The Egger test does not show evidence of

publication bias (p = 0.20). The meta-regression, using the

age as a covariate, does not significantly reduce the

heterogeneity of the study (I2 = 75%, p\ 0.01) (Fig. 2).

The estimated pooled prevalence of anastomotic leak

resulting from 7 studies, which include a total of 344

patients, is 7% (95% CI 4–10%). The prediction lower and

upper limits are 4 and 12%, respectively. The heterogeneity

index is zero (I2 = 0%, 95% CI 0–66%; p = 0.52). Funnel

plot shows that publication and small study bias effect

could be rejected according to Egger test (p = 0.63)

(Fig. 3).

The estimated pooled prevalence of mortality resulting

from 8 studies, which include a total of 1307 patients, is

Records iden�fied through MEDLINE searching

(n = 32)Screen

ing

Includ

edEligibility

Iden

�fica�o

n

Addi�onal records iden�fied through Embase, Cochrane

database and ar�cle’s references (n = 25 )

Records a�er duplicates removed (n = 43)

Records screened(n = 43)

Excluded (n = 27)

(treated by endoscopic dilata�on or Heller myotomy)

Full-text ar�cles assessed for eligibility

(n = 16) Excluded (n = 8)

• Case report/case series with <15pa�ents (n = 6)

• Abstract (n = 1)• Not English-wri�en (n = 1)

Studies included in qualita�ve synthesis

(n = 8)

Studies included in quan�ta�ve synthesis

(meta-analysis)(n = 8)

Fig. 1 PRISMA diagram

World J Surg (2018) 42:1469–1476 1471

123

Table

1D

emo

gra

ph

ican

dcl

inic

ald

ata

of

13

07

pat

ien

tsu

nd

erg

oin

ges

op

hag

ecto

my

for

end

-sta

ge

ach

alas

ia

Ref

eren

ces

No

.

pat

ien

ts

Mea

n

age

M/F

Mea

nd

ura

tio

no

f

sym

pto

ms

(yea

rs)

Su

rgic

al

app

roac

h

Co

nd

uit

Pn

eum

on

ia

(%)

An

asto

mo

tic

leak

age

(%)

Ov

eral

l

mo

rbid

ity

(%)

Mo

rtal

ity

(%)

Mea

nfo

llo

w-u

p

(mo

nth

s)

Orr

ing

eran

d

Sti

rlin

g[1

9]

26

49

10

/16

21

TH

(24

)S

tom

ach

3.8

41

93

.83

0

TT

(2)

Pin

ott

iet

al.

[20]

12

24

56

3/5

9N

RT

H(1

22

)S

tom

ach

NR

8.2

19

4.1

60

Pet

ers

[21]

19

44

13

/61

0T

T(1

9)

Co

lon

5.3

02

10

72

Mil

ler

etal

.[2

2]

37

56

20

/17

0.3

TH

(9)

Sto

mac

h(3

1)

colo

n(6

)

5.4

5.4

32

.45

.47

5

TT

(28

)

Ban

bu

ryet

al.

[23]

32

48

15

/17

13

TH

(21

)S

tom

ach

22

13

NR

34

2

TT

(11

)

Dev

aney

etal

.

[24]

93

51

43

/50

NR

TH

(87

)S

tom

ach

(91

)

colo

n(2

)

21

03

02

38

TT

(6)

Tan

ket

al.

[25]

15

41

10

/56

.5T

H(1

5)

Sto

mac

h2

6.6

13

.35

05

12

Mo

len

aet

al.

[26]

96

35

4.6

47

2/

49

1

NR

TT

(96

3)

Sto

mac

h(9

29

)

colo

n(3

4)

17

NR

NR

2.7

NR

TH

tran

shia

tal,TT

tran

sth

ora

cic,

NR

no

tre

po

rted

1472 World J Surg (2018) 42:1469–1476

123

2% (95% CI 1–3%). The prediction lower and upper limits

are 1 and 3%, respectively. The heterogeneity index is zero

(I2 = 0%, 95% CI 0–43%; p = 0.78). Funnel plot shows

that publication and small study bias effect could not be

rejected. The Egger test do not show evidence of publi-

cation bias (p = 0.09) (Fig. 4).

The Galbraith plot confirmed the level of heterogeneity

for each outcome (Fig. 5). The sensitivity analysis showed

the robustness of the results. In particular, regarding

pneumonia complications, we removed the study by

Devaney et al. [24], without alteration, where I2 value

decreased (I2 = 58.9%), indicating that the result was

stable.

Discussion

The systematic review presented in this study shows that

73% of esophagectomies for end-stage achalasia were

performed in patients with a mean age around 55 years.

However, there is lack of information regarding the natural

history and progression of the disease, and one-third of the

studies did not report the age of onset and the time that

elapsed to reach the end-stage disease phenotype. In most

circumstances, dysphagia, regurgitation and weight loss

were present, combined with massive dilatation and tor-

tuosity of the esophagus, and the disease was unresolved

despite multiple endoscopic and/or surgical treatments.

Whether possible reluctance by the patient or the surgeon

Fig. 3 Forest (a) and funnel (b) plot of anastomotic leakage

Fig. 2 Forest (a) and funnel (b) plot of pneumonia

World J Surg (2018) 42:1469–1476 1473

123

to undertake such a radical surgical intervention may have

delayed the date of operation remains unspecified.

Although we have excluded from our search papers

focused on concomitant achalasia and esophagectomy

cancer, we found two patients with incidental diagnosis of

squamous cell carcinoma in the resected specimen and 3

more patients with preoperative diagnosis of squamous cell

carcinoma. In addition, an emergency esophagectomy was

performed in eight patients due to bleeding or to esopha-

geal perforation secondary to endoscopic dilatation

[19, 21–23]. For the above reasons, the pooled incidence of

postoperative complications, namely pneumonia, anasto-

motic leaks and deaths, may have been overestimated by

including in the analysis patients with a potentially higher-

risk profile.

The heterogeneity of pooled pneumonia prevalence was

high in the meta-analysis. The meta-regression, adjusted

for the age, does not explain this heterogeneity. Other

possible sources of heterogeneity may be the generic

definition of pneumonia, the unknown patients’ smoking

habit and the unspecified effect of surgical approach

(transhiatal vs. transthoracic). This does not allow a sub-

group analysis. Despite the heterogeneity, the prevalence

of pneumonia was 9% with a upper limit of the prediction

interval reaching 42%. This is in accordance with the lit-

erature data indicating that pneumonia is a major compli-

cation of esophagectomy. Apparently, the incidence of

postoperative respiratory complications in achalasia

patients was similar compared to that of patients under-

going esophagectomy for cancer. Although this may be

surprising, since one would have expected a chronic pul-

monary damage from repetitive aspiration in achalasia,

outcomes of esophagectomy were not influenced by the

disease.

On the other hand, the heterogeneity of pooled anasto-

motic leakage and mortality prevalence were low. The

pooled prevalence of anastomotic leak and mortality was

7% (95% CI 4–10%) and 2% (95% CI 1–3%), respectively,

Fig. 4 Forest (a) and funnel (b) plot of mortality

Fig. 5 Galbraith plot of pneumonia (a), anastomotic leakage (b) and mortality (c)

1474 World J Surg (2018) 42:1469–1476

123

which is comparable with the mean incidence of anasto-

motic leak and mortality after esophagectomy for carci-

noma. Although there was no heterogeneity, it was not

possible to analyze more in depth the outcomes stratified

by type of surgical approach and type of esophageal sub-

stitute because of lack of comparative data.

Unfortunately, functional outcomes such as dietary

changes were not reported in a standardized manner. For

this reason, we were unable to calculate a pooled preva-

lence with meta-analysis; it was only possible to describe

these results in a qualitative manner observing that the

majority of patients were able to eat a normal diet. It is also

unfortunate that no formal assessment of quality of life was

provided by any of the included studies.

From the present study, no recommendations can be

made regarding the ideal method of surgical approach and

reconstruction for end-stage achalasia because of the lack

of direct comparative data. All patients in this analysis

have been operated through an open surgical approach.

However, surgical mobilization of the esophagus can be

successfully accomplished also by a thoracoscopic or a

laparoscopic transhiatal route. There are only two small

published case series reporting the results of thoracoscopic

and transhiatal esophagectomy, respectively, in patients

with end-stage achalasia. Unfortunately, these reports did

not meet the criteria of inclusion in our meta-analysis

because of the limited number of patients included. The

thoracoscopic approach resulted in zero operative mortal-

ity, one anastomotic leak and no pneumonia [27]. The

laparoscopic transhiatal approach resulted in 9% mortality,

anastomotic leak 18% and no pneumonia [28]. There is

further recent evidence showing that the risk of respiratory

infection after esophagectomy can be reduced by threefold

if a minimally invasive approach is used [29, 30].

Regarding the choice of the esophageal substitute, some

authors argue that gastric interposition is the procedure of

choice because a single anastomosis is needed and because

of the more constant pattern of blood supply of the stomach

compared to the colon [31, 32]. Use of small-sized gastric

tubes and long-term therapy with proton pump inhibitors

may reduce the risk of reflux esophagitis and Barrett’s

esophagus. Other authors argue that the colon is the best

replacement conduit [21]. Colon interposition is more

complex compared to gastroplasty and requires three

anastomoses. For long interpositions, ischemia at the distal

extremity of the graft is perceived as a risk factor for fistula

and stricture formation. In addition, redundancy of the

conduit may cause mechanical dysfunction and reinter-

vention [33]. Again, no conclusive evidence can be drawn

from our review because no comparative studies are

reported.

Conclusions

The main finding of this meta-analysis is that esophagec-

tomy for end-stage achalasia can be safely performed with

acceptable morbidity, low mortality and satisfactory long-

term alimentary function. It is likely that such results can

be obtained only in specialized centers. Due to the lack of

conclusive evidence, the choice of the operative approach

and the type of reconstruction should be based on a tailored

patient evaluation and surgeon’s preference. Esophagec-

tomy should probably be performed without hesitation in

patients who are fit for major surgery, complain of long-

lasting disabling symptoms and present with sigmoid

esophagus recalcitrant to multiple endoscopic and laparo-

scopic interventions. The role of minimally invasive

esophagectomy in the management of end-stage achalasia

should be further investigated. Future studies should

address patients’ quality of life assessment with validated,

generic and disease-specific questionnaires, in order to

provide patients, surgeons and payers with more compre-

hensive outcome data for a better decision-making.

References

1. Mayberry JF (2001) Epidemiology and demographics of achala-

sia. Gastrointest Endosc Clin N Am 11:235–248

2. Tustumi F, Bernardo WM, da Rocha JRM et al (2017) Esopha-

geal achalasia: a risk factor for carcinoma. A systematic review

and meta-analysis. Dis Esophagus 30:1–8

3. Boeckxstaens GE, Zaninotto G, Richter JE (2014) Achalasia.

Lancet 383(9911):83–93

4. Campos GM, Vittinghoff E, Rabl C et al (2009) Endoscopic and

surgical treatments for achalasia: a systematic review and meta-

analysis. Ann Surg 249(1):45–57

5. Asti E, Sironi A, Lovece A et al (2017) Health-related quality of

life after laparoscopic Heller myotomy and Dor fundoplication

for achalasia. Surgery 161:977–983

6. Moonen A, Annese V, Belmans A et al (2016) Long-term results

of the European achalasia trial: a multicentre randomised con-

trolled trial comparing pneumatic dilation versus laparoscopic

Heller myotomy. Gut 65(5):732–739

7. Vela MF, Richter JE, Wachsberger D et al (2004) Complexities

of managing achalasia at a tertiary referral center: use of pneu-

matic dilatation, Heller myotomy, and botulinum toxin injection.

Am J Gastroenterol 99(6):1029–1036

8. Watson TJ (2015) Esophagectomy for end-stage achalasia. World

J Surg 39(7):1634–1641. doi:10.1007/s11605-007-0226-8

9. Moher D, Liberati A, Tetzlaff J et al (2009) Preferred reporting

items for systematic reviews and meta-analyses: the PRISMA

statement. PLoS Med 6(7):e1000097

10. Stang A (2010) Critical evaluation of the Newcastle-Ottawa scale

for the assessment of the quality of nonrandomized studies in

meta-analyses. Eur J Epidemiol 25(9):603–605

11. Freeman MF, Tukey JW (1950) Transformations related to the

angular and the square root. Ann Math Stat 21(4):607–611

12. Miller JJ (1978) The inverse of the Freeman–Tukey double arc-

sine transformation. Am Stat 32:138

World J Surg (2018) 42:1469–1476 1475

123

13. DerSimonian R, Laird N (1986) Meta-analysis in clinical trials.

Control Clin Trials 7(3):177–188

14. Clopper CJ, Pearson ES (1934) The use of confidence or fiducial

limits illustrated in the case of the binomial. Biometrika

26(4):404–413

15. Borenstein M, Hedges LV, Higgins JPT et al (2009) Introduction

to meta-analysis. Wiley, Chichester

16. Anzures-Cabrera J, Higgins JP (2010) Graphical displays for

meta-analysis: an overview with suggestions for practice. Res

Synth Methods 1(1):66–80

17. Egger M, Davey Smith G, Schneider M et al (1997) Bias in meta-

analysis detected by a simple, graphical test. BMJ

315(7109):629–634

18. R Development Core Team (2015) A language and environment

for statistical computing. R Foundation for Statistical Computing,

Vienna. ISBN 3-900051-07-0

19. Orringer MB, Stirling MC (1989) Esophageal resection for

achalasia: indications and results. Ann Thorac Surg

47(3):340–345

20. Pinotti HW, Cecconello I, da Rocha JM et al (1991) Resection for

achalasia of the esophagus. Hepatogastroenterology

38(6):470–473

21. Peters JH, Kauer WK, Crookes PF et al (1995) Esophageal

resection with colon interposition for end-stage achalasia. Arch

Surg 130(6):632–636

22. Miller DL, Allen MS, Trastek VF et al (1995) Esophageal

resection for recurrent achalasia. Ann Thorac Surg 60(4):922–925

23. Banbury MK, Rice TW, Goldblum JR et al (1999) Esophagec-

tomy with gastric reconstruction for achalasia. J Thorac Cardio-

vasc Surg 117(6):1077–1084

24. Devaney EJ, Iannettoni MD, Orringer MB et al (2001)

Esophagectomy for achalasia: patient selection and clinical

experience. Ann Thorac Surg 72(3):854–858

25. Tank AK, Kumar A, Babu TL et al (2009) Resectional surgery in

achalasia cardia. Int J Surg 7(2):155–158

26. Molena D, Mungo B, Stem M et al (2014) Outcomes of

esophagectomy for esophageal achalasia in the United States.

J Gastrointest Surg 18(2):310–317

27. Schuchert MJ, Luketich JD, Landreneau RJ et al (2009) Mini-

mally invasive surgical treatment of sigmoidal esophagus in

achalasia. J Gastrointest Surg 13(6):1029–1035

28. Palanivelu C, Rangarajan M, Jategaonkar PA et al (2008)

Laparoscopic transhiatal esophagectomy for ‘sigmoid’ megae-

sophagus following failed cardiomyotomy: experience of 11

patients. Dig Dis Sci 53(6):1513–1518

29. Yibulayin W, Abulizi S, Lv H et al (2016) Minimally invasive

oesophagectomy versus open esophagectomy for

resectable esophageal cancer: a meta-analysis. World J Surg

Oncol 14(1):304

30. Biere SS, van Berge Henegouwen MI, Bonavina L et al (2017)

Predictive factors for post-operative respiratory infections after

esophageal cancer: outcome of randomized trial. J Thorac Dis

9(Suppl8):S861–S867

31. Duranceau A, Liberman M, Martin J et al (2012) End-stage

achalasia. Dis Esophagus 25(4):319–330

32. Felix VN (2016) Esophagectomy for end-stage achalasia. Ann N

Y Acad Sci 1381(1):92–97

33. Bonavina L, Chella B, Segalin A (1998) Surgical treatment of the

redundant interposed colon after retrosternal esophagoplasty. Ann

Thorac Surg 65(5):1446–1448

1476 World J Surg (2018) 42:1469–1476

123