Baylor University Medical Center Proceedings October 2017 - 422

Table 2. Outcomes of patients undergoing POEM vs LHM for
management of achalasia
POEM (N = 12)

LHM (N = 11)

P value

Mucosal injury during
surgery

Mucosal necrosis

Postoperative leak

Return to operating room

30-day readmission

Complications (n)

Eckardt scores (mean ± SD)
Preoperative

8.3 ± 1.8

6.9 ± 3.2

0.20

Immediate postoperative

1.3 ± 1.0

0.7 ± 1.2

0.39

Late postoperative

1.2 ± 1.6

3.0 ± 0.7

0.08

Follow-up days, median (IQR)
Success after 6 months

385 (274-500) 285 (217-473)

0.59†

82% (9/11)

66% (6/9)

0.62*

In primary surgery group

80% (8/10)

83% (5/6)

0.99*

In redo group

100% (1/1)

33% (1/3)

0.99*

P value based on *Fisher's exact test; †Wilcoxon two-sample test.
IQR indicates interquartile range; LHM, laparoscopic Heller myotomy; POEM, per-oral
endoscopic myotomy; SD, standard deviation.

noticed intraoperatively immediately distal to the LES on the
gastric mucosa. The three mucosal injuries in the LHM group
were noticed at the time of surgery and were repaired with a
ﬁgure-of-eight 4-vicryl suture. There was one area of necrosis of
the mucosa in the POEM group. The creation of the submucosal
tunnel compromised the perfusion of the overlying mucosa. No
perforation was seen, and nothing was done at the time of surgery.
Follow-up esophagram did not show a leak. One contained leak
was seen on the postoperative esophagram in the POEM group.
The mucosal approximation with hemoclips was not adequate,
and a small amount of contrast leaked into the submucosal tunnel. The patient was immediately taken to the endoscopy suite
for an additional hemoclip to be placed. Once the hemoclip was
placed, a follow-up esophagram showed resolution of the contained leak. The patient never experienced tachycardia or fever.
No clinical diﬀerence was seen from the rest of the cohort. No
patients needed to return to the operating room. No patients
were readmitted within 30 days of their procedure.
Postoperative outcomes were evaluated with Eckardt scores
to indicate the success of the procedure (Table 2). In the immediate postoperative perod, Eckardt scores signiﬁcantly improved
to a mean of 1.3 ± 1.0 (P < 0.0001) in 7 of 12 patients in the
POEM group and 0.7 ± 1.2 (P = 0.03) in 3 of 11 patients in
the LHM group. The other patients failed to follow up in the
immediate phase due to distance from the facility. In follow-up
at least 6 months after surgery, patients were called to obtain
Eckardt scores. The mean score in the POEM group was 1.2 ±
1.6 for 11 of 12 patients at a mean of 483 days, and 3.0 ± 0.7
for 9 of 11 patients in the LHM group with a mean of 273
days. The remaining patients did not have 6 months elapse
422

from the operation to report a score. Success of the procedure
with Eckardt scores ≤3 after 6 months was 82% in the POEM
group and 66% in the LHM group (P = 0.62).
A subset analysis was done for patients who had the procedure as a primary surgery versus a redo surgery. One patient in
the POEM group had a prior LHM and reported an Eckardt
score of 2, indicating success. Three patients in the LHM group
had prior LHM, and only one reported a successful score of 3.
The other 2 patients reported scores of 5 and 6. No signiﬁcant
diﬀerences were observed in the success rate between the two
groups (Table 2).
DISCUSSION
These results indicate a favorable success rate in our institution's early experience with POEM and LHM. The success rate
was 82% for POEM and 66% for LHM at a follow-up of at
least 6 months. There were no complications related to either
procedure that altered the patient's postoperative course, except
for one patient with a contained leak in the POEM group. This
was identiﬁed with an immediate postoperative esophagram
without further morbidity.
POEM has become more widespread, and a signiﬁcant
amount of data has been reported on its eﬃcacy and outcomes.
Inoue et al reported their 500-patient experience with an adverse
event rate of 3.2% (14), and Sharata et al showed a 6% morbidity
rate in 100 patients (15). We show a 25% adverse event incidence
in our 12 patients with no Clavien Dindo grade IV or V and
only one grade III event. This is a much higher rate but related
to our early experience and low patient numbers. This rate is
much more comparable to other reports of early experiences,
such as Hungness et al (20), who reported an approximate 18%
adverse event rate. We anticipate this rate will decrease once the
number of procedures increases. Despite the high adverse event
rate, there was very little alteration in the patients' postoperative
clinical courses. The hospital length of stay was 1.6 days in our
data, which is similar to other reports noting a length of stay of
1 to 3 days (15, 24, 25) in high-volume analysis.
A large meta-analysis by Marano et al compared POEM
and LHM (26). It revealed that POEM has a slightly better
maintenance of a lower Eckardt score than LHM, but the difference didn't reach statistical signiﬁcance. Our data reﬂect the
same trend (Figure 2). There was also no signiﬁcant diﬀerence
between operative times in the meta-analysis. Our data reﬂect
the same ﬁnding, with 136 minutes for POEM and 153 minutes for LHM. We did see a diﬀerence in our complication
rates for POEM and LHM, where the meta-analysis showed no
diﬀerence. Finally, POEM has produced Eckardt scores of ≤3
in 90% to 98% of patients in several studies (18, 24, 27, 28).
These success rates seem to surpass the success of LHM and
pneumatic dilation (86% and 76%, respectively) (7). Overall,
our data, showing success rates of 82% for POEM and 66% for
LHM, follow the trend reported in the literature.
A major limitation to this study is statistical power due to
the low numbers of patients for each procedure. Even though
the cohort numbers were low, they could adequately illustrate
our early experience with POEM and LHM for evaluation.

Baylor University Medical Center Proceedings

Volume 30, Number 4

Table of Contents for the Digital Edition of Baylor University Medical Center Proceedings October 2017