Introduction
Achalasia is a rare esophageal motility disorder characterized by aperistalsis in
the esophageal body and incomplete relaxation of the lower esophageal sphincter (LES)
in response to swallowing [1 ]
[2 ]
[3 ]
[4 ]
[5 ]
[6 ]
[7 ]
[8 ], with progressive dysphagia from solids to liquids being the most important clinical
manifestation [9 ]
[10 ]
[11 ]
[12 ]. Currently, laparoscopic Heller myotomy (LHM) with fundoplication is considered
the standard first-line therapy for achalasia [1 ]
[2 ]
[3 ]
[4 ]
[5 ]
[6 ]
[13 ]
[14 ].
Inoue et al. [15 ] introduced the concept of natural orifice transluminal endoscopic surgery (NOTES)
[16 ]
[17 ] associated with submucosal dissection [18 ] and named this technique peroral endoscopic myotomy (POEM). Studies have shown that
the short- to mid-term efficacy of POEM is greater than 90 % and that the procedure
is associated with less postoperative pain [19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
[25 ]. However, this procedure is frequently associated with some common adverse effects,
such as pneumoperitoneum and surgical emphysema [26 ].
Although this new treatment could revolutionize the treatment of achalasia, the lack
of knowledge regarding its costs and learning curve are barriers to its adoption in
clinical practice. The aim of this study was to conduct a comparative cost-utility
analysis of POEM and LHM with fundoplication.
Materials and methods
This economic pilot study consisted of a prospective economic cost-utility evaluation
from a randomized controlled trial comparing POEM and LHM with fundoplication. Data
were collected from a public tertiary university hospital for a period of 1 year.
This study was approved by the ethics committee of the Faculty of Medicine of the
University of São Paulo.
Randomized controlled trial
The randomized controlled trial was prospective, randomized, and conducted in a single
center. This study was approved by the ethics committee of the Faculty of Medicine
of the University of São Paulo and it was register in Clinical Trial by the number
NCT02138643.
A simple randomization was carried out of patients diagnosed with symptomatic esophageal
achalasia, eligible for the study and admitted on Gastrointestinal Department of the
institution.
Patients
The patient inclusion criteria were: diagnosis of chagasic or idiopathic achalasia;
age between 18 and 60 years; dysphagia score ≥ II; and Eckhardt score > 3.
The exclusion criteria were: prior treatment for achalasia; prior esophageal, mediastinal
or gastric surgery; pseudoachalasia; hepatic cirrhosis; esophageal varices; Barrett’s
esophagus; esophageal stenosis; malignant esophageal lesions; esophageal distal diverticulum;
and coagulopathy.
Interventions
Peroral endoscopic myotomy
POEM was performed while the patient was under general anesthesia with positive pressure
ventilation and CO2 insufflation. A submucosal tunnel was created approximately 2 or 3 cm below the gastroesophageal
transition (GET), and 10 mL of saline with methylene blue was injected into the posterior
wall of the mid-esophagus.
ENDO CUT Q configuration of ERBE VIO (ERBE, Elektromedizin GmbH, US) equipment was
used to create the submucosal tunnel. This configuration was selected considering
that it is an alternating, monopolar and high frequency cutting mode that is characterized
by alternating cutting and coagulation cycles.
After creating an adequate submucosal tunnel and performing haemostatic revision,
myotomy of the distal (2 to 3 cm below the GET) to proximal (up to 8 cm above the
GET and approximately 2 cm below the mucosal access) region was performed. SPRAY COAG
(ERBE, Elektromedizin GmbH) configuration was also used to ensure effective surface
coagulation without contact with depth of thermal penetration, being suitable for
tissue devitalization and to stop diffuse bleeding. In addition, the submucosal tunnel
was performed using T-type hybrid knives (ERBE, Elektromedizin GmbH) and ERBEJet2
pump cartridges (ERBE, Elektromedizin GmbH).
Total myotomy, which involved sectioning of the longitudinal muscle fibers, allowed
endoscopic passage without resistance through the GET. After revision of the mucosal
aspect of the esophagus and stomach with thermal or surgical probing in the submucosal
tunnel, the mucosal access was closed with clips.
Resolution Clip (Resolution Clips, Boston Scientific, Natick, Massachusetts, United
States) was used to close the endoscopic myotomy. The choice of this device took into
account the high retention rate, greater penetration into the tissue, possibility
of repositioning, and ease of handling.
Laparoscopic Heller MYOTOMY WITH FUNDOPLICATION
LHM was performed with the patient under general anesthesia. Five to six incisions
were made in the abdominal region to introduce the trocars. After the introduction
of five or six trocars, the gastric vessels were sectioned, and the gastric fundus
was mobilized. Then, the GET was dissected, and the hepatogastric ligament membrane
was sectioned.
The diaphragmatic pillars were dissected, and the esophagus was isolated from the
gap; the adhesions in the mediastinum were disconnected. Then, myotomy was performed
by sectioning the muscle fibers in the LES and advancing 3 cm into the stomach and
at least 6 cm into the esophagus. Hiatoplasty (approximation of the diaphragm abutments)
was performed at separate points.
Fundoplication was performed by means of three suture lines (Pinottiʼs technique)
covering the entire myotomy as follows: the first line was between the posterior wall
of the esophagus and the posterior wall of the stomach; the second line was between
the left lateral wall of the esophageal myotomy and the gastric fundus between the
transition from the anterior to the posterior wall; and the third line was between
the right lateral wall of the esophageal myotomy and the anterior surface of the gastric
fundus.
Economic analysis
Direct medical costs were obtained via prospective data collection from the local
database of the institution and estimated using microcosting. The costs were obtained
in Brazilian Real (R$) in January 2017 and adjusted using the Broad National Consumer
Price Index from the period (2.61897 %), followed by conversion to United States dollars
(US$) in August 2018 (US$ 1.00 = R$ 3.77).
The study included all direct medical costs associated with the treatment in the time
horizon of the study and represented the institutionʼs direct monetary expenses for
the treatment and clinical follow-up of the patient. [Table 1 ] shows how the monetary data were obtained. The costs were grouped into three categories:
intraoperative, postoperative and clinical follow-up.
Table 1
Description of the methodology for obtaining cost data.
Department
Component
Costs
Human Resources
Doctors, nurses, physiotherapists, and administrative professionals
Based on the standard cost indicated in the hospital budget per profession per month
divided by the time required for treatment
Pharmacy sector
Medicine and hospital material
Based on the unit values paid in the last public bidding
Medicinal gases
Oxygen, nitrous oxide, carbon dioxide, and nitrogen
Based on the cost of medical gas per liter per minute
Laboratory sector
Laboratory examinations
Based on the unit price of each laboratory examination
Imaging sector
Esophageal manometry, upper digestive endoscopy, abdominal ultrasound, and electrocardiogram
Based on the unit price of each imaging examination
The first category included monetary data associated with the procedure and anesthetic
recovery. The second category comprised the costs from the immediate postoperative
period until up to 30 days after the procedure, including costs due to complications.
Finally, the third category covered all of the costs of clinical follow-up, such as
medical interviews and conventional and high-resolution esophageal manometry and endoscopy.
Medical interviews and endoscopic examinations were performed at one, six and 12 months.
Conventional and high-resolution esophageal manometry were performed at 6 and 12 months.
Utility
Quality-adjusted life years (QALYs) were used to measure efficacy, being defined as
a year of life with remission of achalasia symptoms. The QALYs is a generic measure
of disease burden, including the quality and the quantity of life lived. The QALYs
of each patient were estimated for one year, respecting the time horizon. The participants
used this instrument to self-record their health status in the following five areas:
mobility, self-care, usual activities, pain/discomfort and anxiety/depression.
The QALY instrumental was applied at two points during a clinical consultation: immediately
after randomization and in the last cycle of the time horizon. The purpose of the
first point was self-assessment of health status before the procedure, while the goal
of the second point was to identify the degree of improvement in self-assessment 1
year after treatment. The difference in the first and second QALYs was assumed to
be directly related to the intervention for achalasia.
Statistical analysis
The results are presented as the average and standard deviation. The monetary results
are described as the average and standard deviation. Student’s t-test was used to
compare averages.
To test the hypothesis that both treatments are equally effective, a non-parametric
repeated-measures analysis of variance (ANOVA) model was used [27 ]
[28 ]. The 95 % confidence intervals (CIs) were calculated and a two-tailed alpha of 0.05
was used in the hypothesis test. The tests were performed using R Software and Excel
2017.
Cost-utility analysis
A cost-utility analysis was performed in which the incremental cost-utility ratio
(ICUR) was estimated using the final costs and the differences in the pre-procedure
and post-procedure efficacy derived from the patient interviews to determine the incremental
cost per QALY gained. The ICUR was calculated as the result between in the total cost
of POEM and LHM divided by the difference in the utility between the therapeutic alternatives.
Uncertainty regarding the ICUR was determined by a sensitivity analysis, which considered
the costs and QALYs associated with minor and major complications in each group. Minor
complications were defined as bronchoaspiration, subcutaneous emphysema and intraoperative
bleeding during myotomy; major complications included esophageal or gastric iatrogenic
perforation, pneumothorax, hypertension, pneumonia, haemothorax and pleural effusion.
These results were compared with the reference ICUR.
Results
In total, 40 patients with achalasia were recruited; 20 patients were treated with
POEM, and 20 patients were treated with LHM. The demographic data and the details
of the POEM and LHM procedures are summarized in [Table 2 ] and [Table 3 ]. POEM and LHM were successfully completed in all patients (technical success rates,
100 %, P = 1.000), and the operative duration of the endoscopic procedure was half of that
for the surgical procedure (P = 0.001). The total myotomy extent in the POEM group was on average 1.5 cm greater
than that in the LHM group (P = 0.001). The patients who underwent POEM had shorter hospital stays than did the
patients who underwent LHM (P = 0.009).
Table 2
Baseline demographics.
POEM (n = 20)
LHM (n = 20)
P value
Male, n (%)
12 (60)
14 (70)
0.507
Age (years), average ± SD (range)
44.9 ± 14.6 (21–72)
44.2 ± 13.2 (21–64)
0.875
Aetiology of achalasia, n (%)
Chagasic
5 (25)
4 (20)
Idiopathic
15 (75)
16 (80)
Eckardt score, average (range)
8.0 ± 1.7 (6–9.00)
8.5 ± 1.9 (7.25–9.75)
0.558
Tolerated diet, n (%)
0
0
4 (20)
4 (20)
5 (25)
4 (20)
10 (50)
12 (60)
1 (5)
0
LES pressure (mmHg), average ± SD (range)
34.6 ± 11.3 (16.1–55.5)
29.4 ± 9.8 (20.4–59)
0.177
LES, lower oesophageal sphincter; LHM, laparoscopic Heller myotomy with fundoplication;
POEM, peroral endoscopic myotomy; SD, standard deviation.
Table 3
Short-term outcomes of POEM and LHM.
POEM (n = 20)
LHM (n = 20)
P value
Technical success rate, n (%)
20 (100)
20 (100)
–
Length of myotomy (cm), average ± SD
0.001
Total
10.9 ± 1.1
8.3 ± 1.5
Esophageal myotomy
7.3 ± 1.2
5.5 ± 1.1
Gastric myotomy
3.6 ± 0.7
2.8 ± 0.7
Operative duration (minutes), average ± SD
95.7 ± 30.5
218.7 ± 50.9
0.001
Length of hospital stay (days), average
3.7
5.4
0.009
Recurrence of dysphagia, n (%)
1 (5)
–
0.001
Complication rate, n (%)
9 (64,3)
5 (35,7)
0.016
Low-risk complications, n
7
–
Subcutaneous emphysema
3
–
Bleeding
3
–
Bronchoaspiration
1
–
High-risk complications, n
2
5
Pleural effusion
–
1
Pneumonia
–
1
Esophageal or gastric iatrogenic perforation
2
–
Pneumothorax
–
2
Hemothorax
–
1
LES pressure (mmHg), average ± SD
15.8 ± 7.4
14.4 ± 6.8
0.448
19.2 ± 7.1
13.4 ± 3.3
0.073
Eckardt score, average ± SD
1.0 ± 1.4
0.2 ± 0.5
0.186
0.6 ± 1.2
0.4 ± 0.7
0.380
LES, lower oesophageal sphincter; LHM, laparoscopic Heller myotomy with fundoplication;
POEM, peroral endoscopic myotomy; SD, standard deviation.
Fourteen adverse events (AEs) (35 %) were observed in 12 of 40 patients, including
three patients in the LHM group and nine in the POEM group (P = 0.016). AEs were classified as low risk, including subcutaneous emphysema, bleeding,
and bronchoaspiration, and high risk, including pleural effusion, pneumonia, esophageal
or gastric iatrogenic perforation, pneumothorax, and hemothorax.
All five complications in LHM patients were high risk: one case of pleural effusion,
one case of pneumonia, two cases of pneumothorax, and one case of hemothorax. In the
POEM group, seven complications were low risk: three cases of subcutaneous emphysema,
three cases of bleeding, and one case of bronchoaspiration. Two patients had high-risk
complications, including two cases of esophageal or gastric iatrogenic perforation.
Of the 40 patients involved in the study, 37 underwent upper digestive endoscopy in
1 month, 34 in 6 months and 35 in 1 year. The reflux rate was 64.6 % in POEM and 11.1 %
in LHM (P < 0.002). The esophagitis rate was significantly higher in POEM in all periods (P = 0.014; P < 0.001; P = 0.002). After 1 month, 6 months, and 1 year, esophagitis rates were 0 %, 5.6 %
and 11.1 % in LHM and 29.4 %, 62.5 % and 64.6 % in POEM, respectively.
At 3 months 57 % of patients who had POEM and 20 % who had LHM had reflux esophagitis.
In the POEM group, esophagitis was observed in five patients in 1 month (Grade A:
1; Grade B: 3 and Grade C: 1); 10 patients in 6 months (Grade A: 3; Grade B: 2 and
Grade C: 5) and 11 patients in 1 year (Grade A: 5; Grade B: 4 and Grade C: 2). These
patients were treated conservatively with oral use of proton pump inhibitors. In the
LHM group, one patient had esophagitis at 6 month (Grade A) and two patients at 1
year (Grade B: 1; Grade C: 1).
The Eckardt score and LES pressure were significantly reduced in both groups during
all cycles of the time horizon, although they were similar in both groups independent
of the cycle analyzed ([Table 3 ]).
Costs
The total cost, including the costs incurred intraoperatively, postoperatively and
during the follow-up period, was US$ 86,312.63, and POEM was responsible for 60.7 %
(US$ 52,383.85) of this value. The average cost per patient in the POEM group was
significantly higher than that in the LHM group (US$ 2,619.19 ± 399.53 vs 1,696.44 ± 412.21,
P < 0.001).
Costs incurred intraoperatively accounted for 56 % (US$ 48,335.07) of the total cost.
Although this period represented the component with the highest financial cost in
both groups, the cost in the LHM group was half of the cost in the POEM group (P < 0.001) ([Fig. 1 ]).
Fig. 1 Comparison of the average cost of POEM and LHM for each cost component.
The component with the second highest financial cost (25 %, US$ 21,620.51) was the
postoperative period; on average, the cost in the LHM group was 22 % less than that
in the POEM group (P < 0.001) ([Fig. 1 ]).
Overall, the average cost of complications was higher in the POEM group than in the
LHM group. For POEM patients without complications, the average cost was US$ 2,477.60 ± 158.01,
while for POEM patients with minor and major complications, the average cost was US$ 2,527.99 ± 163.53
and US$ 3,666.78 ± 375.22, respectively. The average cost in the LHM group was US$ 1,621.67 ± 246.22
and US$ 2,120.13 ± 901.33 for patients without and with major surgical complications,
respectively.
QALYs
QALYs did not significantly differ between the groups before the procedure (POEM = 0.506 ± 0.181
vs LHM = 0.653 ± 0.083, P > 0.05). Over the time horizon, the QALYs increased in both groups (POEM = 0.940 ± 0.200
vs LHM = 0.985 ± 0.069, P = 0.385). The difference between the final and preoperative QALYs for POEM and LHM
was 0.434 ± 0.164 and 0.332 ± 0.100, respectively; however, this difference was not
significant (P = 0.397).
Cost-utility analysis
The difference in the final cost of the interventions was US$ 922.75, and the difference
in the final QALYs was 0.102. Therefore, the ICUR was US$ 9,046.41/QALY.
Sensitivity analysis
The sensitivity analysis revealed variation in the ICUR when we analyzed the costs
of major and minor complications and the clinical success in both groups ([Fig. 2 ]). The ICUR was directly proportional to the cost of POEM and, consequently, inversely
proportional to LHM. Major endoscopic complications associated with the treatment
of mucosal perforation increased the ICUR by 114 % compared to the reference ICUR
because the cost of POEM in this situation increased 40 % (US$ 3,666.78 ± 375.22 vs
US$ 2,619.19 ± 399.53).
Fig. 2 Sensitivity analysis of POEM and LHM comparing the reference ICUR with ICUR from major,
minor, and clinical success of each group. The ICUR was calculated as the difference
in cost of POEM and LHM divided by the difference in QALYs of POEM and LHM of according
with each condition analyzed. The arrow indicates the percentage of impact of the
ICUR result comparing reference ICUR with ICUR of the condition analyzed.
In contrast, the clinical success of POEM decreased the ICUR by 17 % because the cost
of POEM was 5.5 % lower than the reference cost (US$ 2,477.60 ± 158.01 vs US$ 2,619.19 ± 399.53).
Similarly, when a major surgical complication occurred, the cost of LHM increased
by 25 % (US$ 2,120.13 ± 901.33 vs U$$ 1,696.44 ± 412.21) and, consequently, the ICUR
decreased by 46 % (US$ 4.892,74 vs US$ 9,046.41/QALY).
Discussion
Until recently, the only options available for treatment of achalasia were pharmacological
treatments, Botox, pneumatic dilation, and surgery [29 ]. However, technological progress and the development of minimally invasive surgery
have allowed the creation of NOTES [16 ]
[17 ] associated with submucosal dissection [18 ], and the called POEM, revolutionizing the treatment of achalasia [30 ]
[31 ]
[32 ]
[33 ].
This study is the first economic evaluation of achalasia management strategies and
was performed using data from a prospective randomized controlled study based on real
public costs. These costs were analyzed using a bottom-up analysis, which is an economic
methodology that guarantees a high degree of detail and precision in determining costs
for health institutions [34 ]
[35 ], as the monetary values exactly reflect the reality of health system costs.
This economic methodology differs from previous economic studies [19 ]
[20 ] based on results extracted retrospectively, in which main disadvantage is the lack
of precision in collecting cost data. The total cost of POEM and LHM was US$ 2,619.19 ± 399.53
and US$ 1,696.44 ± 412.21, respectively (P < 0.001). In addition, the QALYs associated with POEM were slightly higher than those
associated with LHM (0.434 ± 0.164 vs 0.332 ± 0.100, P = 0.397).
Thus, the ICUR was US$ 9,046.41/QALY, which is consistent with previous reports [19 ]
[20 ]. The component with the highest financial cost in our study was the intraoperative
period, with POEM costing twice as much as LHM (P < 0.001). This difference could be explained by: (1) the high cost of disposable
endoscopic materials; and (2) the use of depreciated equipment in public hospitals.
Disposable endoscopic materials, including metal clips (Resolution Clips, Boston Scientific, Natick, Massachusetts, United States), T-type hybrid knives (ERBE,
Elektromedizin GmbH, GA, US) and ERBEJet2 pump cartridges (ERBE, Elektromedizin GmbH),
accounted for 75 % (US$ 23,933.23, average US$ 1,196.66 ± 285.39 per patient) of the
intraoperative cost of POEM. In Brazil, these materials are imported, representing
an important part of the total cost of the treatment of complex patients in public
hospitals [36 ].
Negotiations between public managers and medical companies are necessary to ensure
the attainment of good-quality disposable endoscopic materials at lower costs. If
the cost of disposable endoscopic materials decreased by 10 %, there would be a savings
of US$ 1,196.66 ± 285.39, reducing the ICUR by 17 % relative to the reference ICUR.
Thus, this result represents an area of potential cost reduction for POEM.
Regarding the second reason, public hospitals are maintained by the System Unit Health
(Sistema Único de Saúde–SUS). This system is financed by the government, making it
a reference for private systems in the country. The goal of the SUS is to ensure comprehensive
care for patients, prioritizing strategies for public health conditions. In this context,
surgical instruments are considered as items of lower importance, and the purchase
of new products is a long-term process. Thus, the vast majority of these instruments
have depreciated, and their cost is incorporated throughout the process, justifying
the lower cost of LHM in the intraoperative period.
A total of 14 AEs (35 %) were observed, which is higher than the rate reported in
the literature. Of these, 64.3 % were in the POEM group and 35.7 % were in the LHM
group. Iatrogenic perforations of the mucosa were responsible for 14 % of complications
in the POEM group, thus increasing the cost of this procedure, especially due to the
greater number of clips used to close the lesion. Pneumothorax was the main major
surgical complication in LHM, which impacted the increase in the cost of this group
too.
The institutional experience with LHM is more consolidated than with POEM, with a
protocol with the treatment guidelines for the main adverse surgical events. The lack
of expertise with management of POEM complications had a direct impact on postoperative
costs (POEM US$823.99 vs LHM US$474.75, P < 0.001),because extra resources were used, including imaging exams to assess contrast
leakage.
Therefore, hospital costs are inversely proportional to technical and clinical experience:
the lower the level of expertise, the higher the institutional cost. Our sensitivity
analysis ([Fig. 2 ]) reinforces this condition, because the clinical success rate for POEM decreased
the ICUR by 17 %, given that the overall cost of treatment was 5.5 % lower. This finding
was also confirmed by an economic evaluation, which indicated that high performance
levels for POEM will decrease complication rates, directly impacting the reduction
in hospital costs [20 ].
Cost-effectiveness is difficult to define but is generally understood as good value
or good results not costing much money. Brazil has not officially adopted a willingness-to-pay
threshold; however, one study [37 ] has reported that there is an implied threshold of one to three times the Brazilian
gross domestic product. Therefore, we can consider POEM a cost-effective technology
in a time horizon of 1 year, as reported in the literature [19 ]
[20 ].
Conclusion
POEM is more cost-effective than LHM over a 1-year time horizon. POEM costs can be
minimized by reducing the cost of the endoscopic materials and improving experience
with this procedure.