The American Journal of Surgery 194 (2007) 678 – 682

Clinical surgery–International

Reoperation for dysphagia after cardiomyotomy for achalasia

Brechtje A. Grotenhuis, M.D.a, Bas P.L. Wijnhoven, M.D., Ph.D.a, Jennifer C. Myers, B.Sc.b,
Glyn G. Jamieson, M.S., F.R.A.C.S.b, Peter G. Devitt, M.S., F.R.A.C.S.b,
David I. Watson, M.D., F.R.A.C.S.a,*
a
Flinders University Department of Surgery, Flinders Medical Centre, Bedford Park, South Australia 5042, Australia
b
University of Adelaide Discipline of Surgery, Royal Adelaide Hospital, Adelaide, South Australia 5000, Australia

Manuscript received November 24, 2006; revised manuscript January 14, 2007

Abstract
Background: Although laparoscopic cardiomyotomy is considered to be the treatment of choice for achalasia,
there is no consensus about how persistent or recurrent dysphagia after myotomy should be treated. In the
current study we evaluated our experience with reoperation following previous cardiomyotomy.
Methods: Between 1992 and 2006, 19 patients underwent re-myotomy: 7 for persistent dysphagia, 12 for
recurrent dysphagia. Different operative approaches were used, depending on surgeon’s preference and the
technique used for the first operation.
Results: The mean interval between the cardiomyotomies was 81 months. In 12 patients, the alternative
body cavity to that used for the first operation was used for access in the revision operation. This was
associated with a shorter operation time. Mean follow-up after the revision operation was 3.6 years. Mean
satisfaction score was 7 (out of 10), and 89% of patients had an improvement in symptoms.
Conclusion: Reoperation for persistent or recurrent achalasia achieves a satisfactory outcome in most
patients. Using the alternative body cavity to that used in the original procedure facilitates minimal access
techniques, and gives easier access to the operative field. © 2007 Excerpta Medica Inc. All rights reserved.

Keywords: Achalasia; Cardiomyotomy; Reoperation; Operation technique

Achalasia is an uncommon esophageal motility disorder char-
acterized by the inability of the lower esophageal sphincter
(LES) to relax, and loss of esophageal body peristalsis. The
predominant symptoms of dysphagia are secondary to progres-
sive relative obstruction at the gastroesophageal junction.
Treatment is directed at lowering the resistance of the LES. In
recent years, minimal access approaches to this problem have
become the standard of care. Cardiomyotomy via either the
laparoscopic or thoracoscopic approach has been shown to be
safe and effective, with the laparoscopic approach being the
more popular surgical procedure [1–3].
Although laparoscopic myotomy is effective in reducing
symptoms in more than 90% of patients, persistent or re-
current symptoms occur in approximately 10% to 20% of
cases [3–5]. An incomplete myotomy, dense adhesions, and
fibrosis around the gastroesophageal junction, and the ad-
dition of a tight fundoplication to the cardiomyotomy have
all been reported as causes of failure [1,4]. Either pneumatic
* Corresponding author. Tel.: ⫹61 8 8204 6086; fax: ⫹61 8 8204
6130.
E-mail address: david.watson@flinders.edu.au
balloon dilation or surgical reoperation can be undertaken
for these patients. However, only a few published reports
have focused on surgical treatment following a failed car-
diomyotomy [6 –9]. In this study, we evaluated our experi-
ence with reoperation, including a re-myotomy, for persis-
tent or recurrent dysphagia after previous cardiomyotomy.
Patients and Methods
From July 1992 to September 2006, 204 patients under-
went a laparoscopic cardiomyotomy for achalasia by sur-
geons from 2 university teaching hospitals in Adelaide,
South Australia (Royal Adelaide Hospital and Flinders
Medical Centre). The details for each patient (symptoms,
preoperative tests, operation details, and follow-up) have
been collected prospectively and recorded in a database.
From this database, all patients who underwent a reopera-
tion for persistent or recurrent symptoms were identified
and included in this study. The patients who underwent their
primary surgery elsewhere, but their redo cardiomyotomy in
our departments, were also included in the database.
Patients who underwent their original operation in either of
our departments underwent preoperative investigation with

0002-9610/07/$ – see front matter © 2007 Excerpta Medica Inc. All rights reserved.
doi:10.1016/j.amjsurg.2007.01.035
 B.A. Grotenhuis et al. / The American Journal of Surgery 194 (2007) 678 – 682
esophageal manometry, barium swallow x-ray, and upper gas-
trointestinal endoscopy. The preoperative assessment for pa-
tients who underwent their first operation elsewhere depended
on the preferences of the original surgeon. Work-up for revi-
sion surgery similarly included esophageal manometry, barium
swallow x-ray, and upper gastrointestinal endoscopy.
The original surgical procedure was undertaken according
to the surgeon’s preferred technique. This included open and
minimal access techniques via an abdominal or thoracic ap-
proach. For the procedures undertaken in our departments from
1992 onwards, a laparoscopic approach for the first operation
was routine. When an open operation or a thoracoscopic ap-
proach had been undertaken previously, the operation had
either been performed before 1992, or it had been performed
by another surgeon, with the patient subsequently referred to us
for revision surgery. The reoperative strategy was determined
by the preferences of the operating surgeon. When a laparo-
scopic abdominal approach was undertaken, the technique
used was similar to that described previously [10]. This en-
tailed access via 5 laparoscopic ports, reversal of any previous
fundoplication, mobilization of the anterior and lateral aspects
of the distal esophagus (but not the posterior esophagus), and
removal of the fat pad overlying the anterior cardia. A re-
myotomy was performed and extended distally from approx-
imately 5 mm below to 5 to 6 cm above the gastroesophageal
junction. The adequacy of the myotomy was then checked
visually using a gastroscope. Complete opening of the LES,
easy passage of the scope and trans-illumination of the mucosa
under laparoscopic/thoracoscopic vision with no evidence of
intact circular muscle fibers was considered to indicate an
adequate myotomy. Subsequently, an anterior 180° partial fun-
doplication was performed by suturing the anterior fundus to
the left and right sides of the esophagus and the right hiatal
pillar, and the fundus to the anterior rim of the hiatus. If an
open abdominal approach was required, a similar set of steps
were undertaken, with access obtained via an upper midline
abdominal incision.
When a thoracoscopic approach was undertaken, the patient
was positioned fully prone, and access was obtained via 3 or 4
ports that were inserted between the ribs of the left posterior
chest. The lower left lateral and posterior esophagus was then
exposed, and the left hiatal pillar was identified. The distal
extent of dissection extended through the esophageal hiatus to
the posterior aspect of the cardia. If a previous anterior partial
fundoplication had been performed, this was left intact, as the
dissection was confined to the postero-lateral esophagus, so
that dissection was in a tissue plane that had not been dissected
previously. The myotomy was then performed. This extended
distally from approximately 5 mm below to at least 5 to 6 cm
above the gastroesophageal junction, and proximally even as
far as the aortic arch if required. If an open thoracic approach
was required, a similar set of steps were undertaken but with
access obtained via a left postero-lateral thoracotomy incision.
After reoperation, a structured questionnaire was sent out
to all patients 3 months, 6 months, and 1 year after the
operation, and yearly thereafter. If patients did not return the
questionnaire, a second questionnaire was mailed out. If this
was not returned, an attempt was made to interview the
patient by telephone, using the same questionnaire. Patients
were asked to indicate the presence or absence of the fol-
lowing symptoms: dysphagia, painful swallowing, chest
679
pain, heartburn, regurgitation, and cough, and to indicate the
frequency of any symptoms (scale ⫽ never, once a month,
few times a week, and daily). Overall satisfaction with the
outcome of the procedure was also determined using a
visual analog scale (0 ⫽ totally dissatisfied; 10 ⫽ totally
satisfied). Patients were also asked whether they thought
they had made the correct decision to have a revision op-
eration. A 24-hour pH study, esophageal manometry, and
endoscopy were not routinely scheduled during follow-up.
They were only performed when clinically indicated.
SPSS for Windows (version 12.0.1; SPSS Inc, Chicago,
IL) was used for statistical analysis of the data. Student
t tests and nonparametric tests (Mann-Whitney and Wil-
coxon) were used to compare groups, as applicable. Differ-
ences were considered to be significant if P ⬍ .05. The
Kaplan-Meier method was used to calculate the chance of
recurrent dysphagia after reoperation over time.
Results
Between November 1997 and June 2006, 19 patients un-
derwent a reoperation for achalasia. Eight of these patients
were referred to us following primary surgery at another loca-
tion. There were 6 women and 13 men, and the mean age at
surgery was 48 years (range, 22–73). All primary operations
involved a single myotomy through either the thorax or abdo-
men (Table 1). Six of the patients underwent additional endo-
scopic interventions after their first myotomy operation (5
pneumatic dilatations, 1 botulinum toxin injection), but these
procedures failed to relieve symptoms of dysphagia.
Seven of the patients had persistent dysphagia following
their primary surgery; 3 had no improvement (all underwent
laparoscopic myotomy without anterior fundoplication),
and 4 had minor improvement in dysphagia after the first
myotomy (all laparoscopic myotomy with anterior fundo-
Table 1
Operative details
Variable Patients
(N)
Primary surgery technique
Laparoscopic myotomy ⫹ anterior fundoplication 10
Laparoscopic myotomy without anterior fundoplication 3
Thoracoscopic myotomy without anterior fundoplication 2
Myotomy via thoracotomy without anterior fundoplication 4
Reoperative surgery technique
Laparoscopic myotomy ⫹ anterior fundoplication 11
Laparoscopic myotomy without anterior fundoplication 1
Myotomy via laparotomy without anterior fundoplication 1
Thoracoscopic myotomy without anterior fundoplication 4
Myotomy via thoracotomy without anterior fundoplication 2
Interval between first and second operation
Mean 81 mo (range, 1–276 mo) 19
Median 60 mo 19
Operative time
Mean 104 min (range, 40–180 min) 19
Cause of failure (multiple causes possible)
Scar tissue at gastroesophageal junction 12
Incomplete myotomy 8
Diverticulum 1
No anatomical abnormality found 3
680 B.A. Grotenhuis et al. / The American Journal of Surgery 194 (2007) 678 – 682

plication). The other 12 patients had an initially good out- Table 3

come following myotomy but then developed recurrent Approaches in primary and secondary surgery for Heller myotomy
symptoms. These symptoms included dysphagia, chest pain, (N ⫽ 19)
heartburn, or a combination of these complaints. Recurrence Primary approach Secondary approach
occurred between 1 and 252 months (mean 66, median 24
Abdominal approach Thoracic approach Total
months) after the initial operation. There were no significant
differences between the patients with persistent versus re- Abdominal approach 7 6 13
current symptoms with regards to preoperative parameters, Thoracic approach 6 0 6
objective investigations, or outcome scores. For this reason, Total 13 6 19
the population has been considered to be one group, and no
distinctions have been made.
Seventeen patients underwent additional assessment with formed at the site of the previous cardiomyotomy. Operating
esophageal manometry before the revision operation (Table 2). time ranged from 40 to 180 minutes (mean 104 minutes). The
In 5 patients the LES pressure could not be assessed because choice of operative approach compared to the primary ap-
the manometry catheter could not be positioned across the proach is summarized in Table 3. In 12 patients, a different
gastroesophageal junction. In the other 12 patients, the mean body cavity was used for access for the second operation. The
LES resting pressure was 17 mm Hg. Two patients did not mean operating time was 90 minutes if access for the second
undergo manometry because a barium swallow x-ray had operation was via a different body cavity, versus 128 minutes
shown that the gastroesophageal junction remained narrow, if both operations were undertaken through the same body
and failed to relax sufficiently with swallowing. However, no cavity (all abdominal approach).
significant stricture was found on endoscopy in these 2 pa- In 16 of the 19 patients a cause for the failure of the original
tients. surgery was identified. An incomplete myotomy or a band of
Eighteen of the patients were investigated with a preoper- scar tissue at the gastroesophageal junction (or a combination
ative barium swallow x-ray. In 12 of them a dilated esophagus of the two) were the most common causes identified (Table 1).
was identified. The extent ranged from mild dilatation to an In the 3 patients in whom no cause could be found, a further
elongated tortuous sigmoid esophagus. All 19 patients under- myotomy was undertaken, taking care to extend this well onto
went upper gastrointestinal endoscopy. Three of them were the stomach, and as far proximally as possible. The mean
diagnosed with a mechanical stricture of the lower esophagus, length of the new myotomy was 6 cm (range, 3–11).
and 2 with an esophageal diverticulum; these were the patients Intraoperative perforation of the mucosa of the esophagus
in whom LES pressure could not be assessed by manometry or gastric cardia occurred in 3 patients. These perforations
(see above). Columnar epithelium was not seen in any of the were all recognized and repaired intraoperatively, and they had
patients. The final decision to undertake a revision operation no adverse impact on the patient’s recovery. One patient suf-
was based on a combination of the clinical presentation and the fered an intraoperative tear of the splenic capsule, but no
outcome of the preoperative investigations. splenectomy was necessary. Three of these 4 patients who had
The mean time from the first to the second myotomy was 81 an intraoperative complication had their reoperation done via
months (range, 1–256, median 60 months). Re-myotomy was the same body cavity (abdominal in these patients) as the
performed by a laparoscopic approach in 12 patients, an open primary operation. The other patient who had an intraoperative
abdominal approach in 1 patient, a thoracoscopic approach in mucosal perforation underwent a laparoscopic reoperation af-
4 patients, and an open left thoracotomy in 2 patients (Table 1). ter an initial myotomy via thoracotomy. Seventeen patients had
One of the thoracotomies was undertaken because of extensive an uncomplicated recovery following surgery. Two developed
pleural adhesions within the left thoracic cavity following an postoperative pneumonia following a laparotomy and a lapa-
initial thoracoscopic approach. In 11 patients an anterior fun- roscopic approach, respectively. These patients were success-
doplication was added to the procedure. In 1 patient an esoph- fully treated with antibiotics. None of the patients required
ageal diverticulum was also excised; the diverticulum had further surgery during the admission for re-myotomy. The
postoperative hospital stay ranged from 1 to 8 days (mean 4
days). The stay was longer in the patients who underwent open
Table 2 surgery (either laparotomy or thoracotomy) compared to min-
Esophageal manometry outcomes imal invasive surgery (7 vs 3 days, P ⬍ .05).
Variable Patients (N) One patient was lost to follow-up. The other 18 patients
LES resting pressure
were monitored for 3 months to 8 years (mean 3.6 years).
Reduced (⬍10 mm Hg) 3 Symptoms and their frequencies after revisional surgery are
Normal (10–30 mm Hg) 7 shown in Fig. 1. Fig. 2 shows the chance of recurrence of
Elevated (⬎30 mm Hg) 2 daily dysphagia over time in our study group. Ten (56%) of
Not able to assess 5 the patients were able to eat an unrestricted diet. Sixteen
LES relaxation patients experienced an improvement in the overall outcome
Normal 4 (including dysphagia, odynophagia, heartburn, and chest
Incomplete 8 pain), and 2 were not helped by further surgery. The mean
Not able to assess 5 satisfaction score was 7 (range, 2–10; Table 4). Addition of
Primary esophageal peristalsis
a fundoplication or change in surgical approach (alternate
Absent 17
Present 0
body cavity) did not affect the satisfaction score or postop-
erative dysphagia score. All patients stated that they thought
 B.A. Grotenhuis et al. / The American Journal of Surgery 194 (2007) 678 – 682
Fig. 1. Outcome of reoperations for achalasia (N ⫽ 18).
they had made the correct decision to undergo a revision
operation. Two patients underwent a third operation for
persistent dysphagia: 1 underwent a thoracoscopic myot-
omy 10 weeks after the first reoperation and 1 had an
esophagectomy 6 years after the second cardiomyotomy
operation. Both of these patients had a grossly dilated (sig-
moid) esophagus at the time of the first revision procedure.
Comments
Minimally invasive cardiomyotomy, using either a laparo-
scopic or thoracoscopic technique, is now considered to be the
treatment of choice in patients with symptomatic primary acha-
lasia [1–3]. However, even though these procedures are safe
and effective in the majority of patients, recent studies have
shown that up to 20% of patients still experience some symp-
toms following surgery [4 – 6]. Unfortunately, there is little
consensus about what is the most appropriate treatment for
persistent or recurrent symptoms after cardiomyotomy. Some
favor endoscopic pneumatic balloon dilatation, whereas others
recommend surgical re-exploration. In recent years there have
Fig. 2. Postoperative recurrence of daily dysphagia (N ⫽ 18).
681
been a few reports published on reoperative cardiomyotomy
following a failed first myotomy, but the reports were of small
numbers, as well as relatively short durations of follow-up
[6 –9]. We have described our experience in 19 patients with
persistent or recurrent symptoms of dysphagia following an
initial cardiomyotomy.
Although incomplete LES relaxation and absence of pri-
mary peristalsis in the esophageal body are considered pathog-
nomic for achalasia, the underlying manometric pattern in
patients with persistent or recurrent achalasia symptoms was
less clear in our study group. LES relaxation was not always
incomplete. Furthermore, dilatation of the esophagus was not
always present on barium swallow studies, and endoscopy did
not reveal any evidence of a mechanical obstruction to the
esophagus, or other causes of dysphagia. No relationship could
be identified between the results of the preoperative objective
tests and the subjective clinical outcome following re-myot-
omy. However, all patients did have clinically important symp-
toms of dysphagia, with or without reflux symptoms before
re-myotomy, and these symptoms improved significantly in
most patients following reoperation. Also, all 3 of the patients
in whom no cause of dysphagia was identified had a significant
improvement in their symptoms following re-myotomy.
Therefore, the selection of patients for reoperation ultimately is
clinically based, although investigations can reinforce clinical
assessments.
In 12 (63%) of our patients, the surgeon chose the alternate
body cavity for gaining access to the distal esophagus (Table
3). This approach almost certainly provided the surgeon with
easier access to the distal esophagus, as there were fewer
adhesions in the “new” cavity. This meant that it was easier to
identify tissue planes, and there were fewer intraoperative
Table 4
Satisfaction scores after reoperation (N ⫽ 18)
Satisfaction score Frequency
7–10: Good/Excellent 9
4–6: Fair 7
0–3: Poor 2
682 B.A. Grotenhuis et al. / The American Journal of Surgery 194 (2007) 678 – 682
complications. For these reasons, the choice of undertaking the
second operation via the alternate body cavity resulted in
reduced operating time.
An incomplete myotomy or a band of scar tissue at the
gastroesophageal junction were thought to be the cause of
recurrent or persistent dysphagia in most of our patients.
However, in several cases we were not able to determine the
exact cause of the problem. This was sometimes because we
undertook the revision surgery through a different body
cavity. For example, when the first procedure was under-
taken through an abdominal approach, the myotomy was
performed on the anterior wall of the esophagus. If the
revision operation was then undertaken via a left thoraco-
scopic approach, the left postero-lateral esophagus was ex-
posed, and subsequently the myotomy was performed on
this part of the esophageal wall. Therefore, at revision the
anterior wall of the esophagus was not exposed, and the
previous myotomy was not examined directly. Hence, our
strategy usually entailed a pragmatic approach that involved
extending the myotomy as far as possible both proximally
and distally, irrespective of whether the previous myotomy
was thought to have been adequate or otherwise.
In 2 reviews of achalasia, no ideal length of myotomy
has been found that maximally relieves dysphagia and min-
imizes the occurrence of reflux [1,2]. Although most reports
describe a minimal gastric myotomy of 2 cm, in our patients
a distal myotomy was performed until 5 mm below the
gastroesophageal junction. This can be justified by the fact
that our reoperation rate for achalasia (5%) is similar to that
in other studies (2% to 13%) [4,6,7,9].
In 12 patients scar tissue in the region of the gastroesoph-
ageal junction and esophageal hiatus was thought to be part
of the problem. Excessive scar tissue in this region is also
known to contribute to dysphagia following anti-reflux sur-
gery [11], and it also has been described as a cause of failure
following cardiomytomy [4 – 6].
It is known that dysphagia (secondary to an inadequate
myotomy or periesophageal scarring) after a primary Heller
myotomy recurs early, often not later than 3 years [12].
Nevertheless, the long-term results of operative treatment
for achalasia tend to worsen with time. Furthermore, abnor-
mal gastroesophageal reflux and related complications may
also occur at more than 10 years postoperatively. These
factors also justify long-term follow-up.
Our medium to long-term follow-up data (3 months to 8
years; mean 3.6 years) after redo myotomy show that about
half of the patients still experience some symptoms of dyspha-
gia and/or odynophagia. Other studies have shown lower rates
of dysphagia, but these studies present smaller populations and
shorter follow-up [6 –9]. However, the mean overall satisfac-
tion score of 7 in our population and the fact that every patient
thought he or she had made the right choice to undergo a
revision operation, suggests that patients are aware of the
chronicity of achalasia (surgery will not restore normal swal-
lowing), and they are probably satisfied with an improvement
in symptoms if all of the symptoms cannot be fully relieved. In
89% of the patients (16/18) at least some improvement was
achieved (overall clinical outcome rated as fair, good or ex-
cellent). This is comparable to the results from other studies
where between 80% and 100% of patients have benefited from
surgical revision [6 –9]. Also, this percentage is similar to the
outcome of primary cardiomyotomy in our hospitals (93%
improved, N ⫽ 167) [13].
Alternative treatments for failure after primary operations
for achalasia include pneumatic balloon dilatation or esopha-
gectomy. Balloon dilatation has been reported to achieve a
variable success rate, ranging from 33% to 80% [4,6], and it
often results in short-term improvement of symptoms [4]. A
more radical approach is an esophagectomy. This is used in
end-stage achalasia where the esophagus is grossly dilated,
causing uncorrectable dysphagia. The drawbacks of this oper-
ation are higher risks of morbidity and mortality [5]. Of the 2
patients in our series who rated their outcome as poor, 1
recently underwent esophagectomy for a mega-esophagus, af-
ter which his satisfaction score improved to excellent.
In conclusion, reoperation for persistent or recurrent
achalasia is a safe procedure, resulting in a reasonable
outcome. However, the criteria for recommending reop-
eration are often not clear, and the results of any inves-
tigations need to be interpreted with caution, with pa-
tients’ individual clinical symptoms also being very
important when deciding to recommend further surgery
(dysphagia at least similar or worse than prior to the
primary surgery). Changing from the original approach to
the alternative body cavity (ie, from abdomen to thorax
or vice versa) gives easier access to the distal esophagus
and gastric cardia, provides better exposure of the oper-
ative field, and makes the revision procedure easier to
perform.
References
[1] Abir F, Modlin I, Kidd M, et al. Surgical treatment of achalasia:
current status and controversies. Dig Surg 2004;21:165–76.
[2] Woltman T, Pellegrini C, Oelschlager B. Achalasia. Surg Clin N Am
2005;85:483–93.
[3] Ramacciato G, Mercantini P, Amodio P, et al. The laparoscopic
approach with antireflux surgery is superior to the thoracoscopic
approach for the treatment of esophageal achalasia. Surg Endosc
2002;16:1431–7.
[4] Zaninotto G, Constantini M, Portale G, et al. Etiology, diagnosis, and
treatment of failures after laparoscopic Heller myotomy for achalasia.
Ann Surg 2002;235:186 –92.
[5] Miller D, Allen M, Trastek V, et al. Esophageal resection for recur-
rent achalasia. Ann Thorac Surg 1995;60:922– 6.
[6] Patti M, Molena D, Fisichella P, et al. Laparoscopic Heller myotomy
and Dor fundoplication for achalasia: analysis of successes and fail-
ures. Arch Surg 2001;136:870 –7.
[7] Gorecki P, Hindler A, Libbey J, et al. Redo laparoscopic surgery for
achalasia. Is it feasible? Surg Endosc 2002;16:772– 6.
[8] Duffy P, Awad Z, Filipi C. The laparoscopic reoperation of failed
Heller myotomy. Surg Endosc 2003;17:1046 –9.
[9] Iqbal A, Tierney B, Haider M, et al. Laparoscopic re-operation for
failed Heller myotomy. Dis Esophagus 2006;19:193–9.
[10] Mitchell P, Watson D, Devitt P, et al. Laparoscopic cardiomyotomy
with a Dor patch for achalasia. Can J Surg 1995;38:445– 8.
[11] O’Reilly M, Mullins S, Reddick E. Laparoscopic management of
failed antireflux surgery. Surg Laparosc Endosc 1997;7:90 –3.
[12] Di Simone M, Felice V, D’Errico A, et al. Onset timing of delayed
complications and criteria of follow-up after operation for esophageal
achalasia. Ann Thorac Surg 1996;61:1106 –10.
[13] Bessell J, Lally C, Schloithe A, et al. Laparoscopic cardiomyo-
tomy for achalasia: long-term outcomes. Aust NZ J Surg 2006;76:
558 – 62.