Introduction
Gastric outlet obstruction (GOO), characterized by blockage in the antropyloric area
or duodenum, can stem from both malignant and benign causes. While malignancy is more
common in adults, benign GOO, often caused by peptic ulcer disease (PUD), caustic
ingestion, opioid use, and nonsteroidal anti-inflammatory drug (NSAID) use, remains
a significant clinical challenge.[1]
[2] The incidence of GOO due to PUD has decreased since the discovery of Helicobacter pylori and its association with PUD.[3]
[4]
[5] Complications of GOO include malnutrition, aspiration, pain, and respiratory failure.[6]
[7]
[8] In India, caustic ingestion, both accidental and suicidal, is a major cause, with
studies showing a high percentage of patients developing GOO after such ingestion.
The chances of developing GOO is directly proportional to mucosal injury.[9]
[10]
[11] Opioid use, particularly in India, has also emerged as a frequent cause of benign
GOO, possibly due to prolonged mucosal contact with acidic contents and inhibition
of enteric reflexes.[12]
[13]
[14] Chronic NSAID use contribute to ulceration by inhibiting prostaglandin production,
necessitating concomitant use of H2 receptor antagonists or proton-pump inhibitors
(PPIs).[15] Misoprostol, a synthetic prostaglandin E1 analog, protects the gastric lining by
stimulating mucus and bicarbonate secretion, mucosal blood flow, and cellular repair.[16]
[17] Studies have shown its efficacy in healing peptic ulcers and reducing complications
like GOO in chronic NSAID users. For maintenance phase PPIs are more effective.[18]
[19]
[20]
[21] Endoscopic balloon dilatation (EBD) is the preferred treatment for benign GOO, with
surgery reserved for refractory cases.[22] We hypothesized that the addition of misoprostol therapy along with standard endoscopic
therapy and concomitant PPI may result in higher and rapid ulcer healing rates and,
therefore, may lead to better outcomes; moreover, misoprostol is more useful in acute
healing phase, hence, we used concomitant misoprostol therapy with PPIs for a duration
of 2 weeks after cessation of etiological agents, and PPI therapy was continued for
14 weeks versus PPI alone and standard endoscopic therapy (EBD) and conducted this
pilot study.
Materials and Methods
This randomized controlled trial was conducted at SMS Medical College and Associated
Hospital, Jaipur, Rajasthan, India, from August 2020 to December 2023 (CTRI/2020/07/026501)
([Fig. 1]). The study included patients over 18 years old with benign GOO caused by NSAIDs,
opioids, caustics, or PUD. Chronic NSAID use was defined as use > 3 times/week for
≥ 3 months, and chronic opioid use as daily consumption for ≥ 4 months.[23]
[24]
Fig. 1 Flow chart of the study.
Exclusion Criteria
-
Contraindications to misoprostol (pregnancy, history of cerebrovascular accident,
coronary artery disease, uncontrolled hypertension, atypical uterine bleeding),
-
GOO associated with chronic pancreatitis, Crohn's disease, tuberculosis, or fibrotic/healed/nonactive
strictures,
-
continued use of NSAIDs/opioids/smoking,
-
H. pylori infection,
-
malignant etiology, and
-
refusal to consent.
Diagnosis of GOO was based on clinical, radiological, and endoscopic findings. Four-quadrant
biopsies were taken to rule out malignancy, tuberculosis, Crohn's disease, and H. pylori, and rapid urease tests was performed in all. Patients were randomized to either
sequential EBD plus PPI (pantoprazole 40 mg twice a day) for 14 weeks (control group)
or misoprostol (400 mcg three times a day for 2 weeks) plus EBD plus PPI (pantoprazole
40 mg twice a day) for 14 weeks (intervention group). H. pylori-positive patients received standard triple therapy and were excluded. Randomization
was done using a computer-generated random numbers done by Research Randomizer app.
Patients were followed for 1 year. The sample size of 40 patients was empirically
determined due to limited existing data.
EBD was performed by a single endoscopist (P.A.) using controlled radial expansion
(CRE) balloon dilators (6–20 mm) Boston Scientific. Patients were kept in fasting
for 8 to 12 hours, and gastric decompression was performed if needed. Pharyngeal anesthesia
and conscious sedation were administered. The stricture was visualized endoscopically
and a guidewire was advanced through it. The balloon was positioned across the stricture
under fluoroscopy and inflated to a predetermined pressure for 60 seconds (up to 2 minutes
if needed) as shown in [Fig. 2]. Dilation was repeated every 2 to 3 weeks until a 15-mm lumen was achieved. Gastric
residue assessment was also used to guide dilatation.[25] Esophageal strictures, when present, were also dilated using CRE balloons to 15 mm.
The endpoint of esophageal stricture dilatation was 15 mm and once the diameter of
more than 12 mm was achieved, then GOO was addressed endoscopically.[26]
[27] Adjunctive therapies like corticosteroids, sucralfate, triamcinolone, antacids,
and stenting were not used. Use of sucralfate in the acute or subacute phase of corrosive
ingestion is recommended.[28] Since all the patients of caustic GOO presented after 3 weeks of ingestion, hence
it was not used. PPIs were used per institutional practice for 14 weeks and throughout
the study for idiopathic PUD-related GOO.
Severity and nutritional status was addressed using the Zargar classification for
caustic injury and GOO score.[9]
[29] Patients with corrosive GOO (Zargar IIB/IIIA) underwent feeding jejunostomy (FJ)
for nutritional support. Dilatation was started after 3 weeks postingestion. In noncorrosive
GOO, patients with a GOO score < 2 underwent FJ. Prior to endoscopic intervention,
oral intake was gradually introduced as luminal diameter increased as per protocol.[25] Misoprostol was administered sublingually or rectally if oral intake was not possible.
Parenteral nutrition was used for patients with dehydration or poor nutritional status
until FJ became functional. There is some evidence that patients with caustic esophageal/pyloric
stricture should be dilated at least after a period of 3 weeks of conservative management.[30] In our study, all patients presented to us after 3 weeks of ingestion, hence we
started dilatation as soon as technically feasible.
Outcome Analysis
The primary outcome was obviation of need of surgery.
The secondary outcomes included:
-
symptom resolution (heartburn, vomiting, abdominal pain, postprandial fullness),
-
number of EBD sessions, and
-
improvement in nutritional status (hemoglobin, serum albumin).
Procedural success was defined as achieving 15 mm dilation and no gastric residue.[25] Clinical success was defined as sustained symptom resolution (GOO score of 3), sustained
till the study's end. Primary failure was the inability to achieve procedural success,
secondary failure was the failure to achieve clinical success despite procedural success,
and recurrence was symptom relapse and stricture redevelopment after achieving clinical
success.
Follow-up visits were conducted every 2 to 3 weeks or as needed for 1 year after achievement
of procedural success. Patients' telephone number and contact address were recorded
and contact number of corresponding author (P.A.) and coauthors (M.S., G.D.) were
also provided so that patients may contact in case of any actual or perceived emergency.
Standing instructions were provided to emergency medical officers to admit them if
they presented with any significant problem, which may be related or unrelated to
the primary disease.
Statistical Analysis
It was performed using PSPP 2.0.1. Continuous variables were compared using t-tests, and categorical variables using Fisher's exact tests. p-Values of < 0.05 were considered significant. Per protocol analysis was performed.
Due to the coronavirus disease 2019 pandemic-induced travel restrictions and limitation
of endoscopic access, 28 patients (33.3%) were lost to follow-up and 16 patients were
excluded due to malignancy or other causes.
Ethical Approval
Approval was obtained from the institute review board and ethics committee prior to
the study. CTRI No. CTRI/ 2020/07/026501.
Results
Eighty-four patients were initially evaluated, and 40 completed the study (20 per
arm). Males comprised 57.5% of the study population. The most common etiology was
corrosive ingestion (50%), followed by NSAIDs (25%), peptic stricture (17.5%), and
opioid use (7.5%). The most common symptoms were abdominal pain (85%) and vomiting
(80%), and dysphagia in patients with esophageal involvement. Esophageal stricture
prevalence was similar in both groups. Most of the patients had middle esophageal
strictures (4/7 in the intervention arm and 5/8 in the control arm). One in each group
had upper esophageal stricture and two patients in both groups had multiple strictures
involving both mid and upper esophagus ([Table 1]).
Table 1
Baseline characteristics in subjects with gastroduodenal obstruction enrolled in our
study (n = 40)
|
Characteristics
|
Intervention arm (misoprostol + PPI)
|
Control arm (PPI)
|
p-Value
|
|
Age (y)
|
34.35 ± 10.649
|
37.75 ± 12.624
|
0.363
|
|
Sex – Male
|
11
|
12
|
1.000
|
|
Female
|
9
|
8
|
|
|
BMI (kg/m2)
|
20.78 ± 3.83
|
20.48 ± 2.81
|
0.780
|
|
Smoker – Male
|
6
|
4
|
0.41
|
|
Female
|
0
|
0
|
|
|
Native
|
|
|
|
|
Rajasthan
|
5
|
7
|
0.731
|
|
Other states
|
15
|
13
|
|
|
Hemoglobin (g/dL)
|
9.78 ± 2.17
|
10.55 ± 1.57
|
0.206
|
|
Serum albumin (g/dL)
|
2.99 ± 0.595
|
3.23 ± 0.525
|
0.185
|
|
Etiology
|
|
|
|
|
Corrosive
|
10
|
10
|
0.515
|
|
NSAIDs
|
4
|
6
|
|
|
Peptic
|
5
|
2
|
|
|
Opium/Opioids
|
1
|
2
|
|
|
Symptoms
|
|
|
|
|
Dysphagia
|
7
|
8
|
1.000
|
|
Abdominal pain
|
17
|
17
|
1.000
|
|
Vomiting
|
16
|
16
|
1.000
|
|
Loss of weight (≥ 5% in last 1 month)
|
15
|
11
|
0.320
|
|
Associated esophageal stricture
|
7
|
8
|
1.000
|
|
GOO score
|
1.35 ± 0.745
|
1.15 ± 0.813
|
0.422
|
|
Duration between insult and presentation as GOO (in weeks) corrosives
|
7.30 ± 2.058
|
6.90 ± 2.132
|
0.674
|
|
Symptomatic period before inclusion into study of noncorrosives (In weeks)
|
5.60 ± 1.174
|
5.30 ± 1.059
|
0.556
|
Abbreviations: GOO, gastric outlet obstruction; NSAID, nonsteroidal anti-inflammatory
drug; PPI, proton-pump inhibitor.
Clinical success was significantly higher in the intervention group (80%) compared
to the control group (35%) (p = 0.010). Procedural success was also significantly higher in the intervention group
(90%) compared to the control group (45%) (p = 0.006). Primary failure was significantly lower in the intervention group (10%)
compared to the control group (55%) (p = 0.006). Secondary failure was not significantly different between the groups ([Tables 2] and [3]).
Table 2
Final clinical success in subjects of gastroduodenal obstruction in the intervention
and control arm at the end of study period
|
Clinical success
|
Requirement of surgery
|
|
|
Control arm (PPI)
|
7
|
12 + 1 (stricturoplasty/pyloroplasty)
|
20
|
|
Intervention arm (misoprostol + PPI)
|
16
|
3 + 1 (stricturoplasty/pyloroplasty)
|
20
|
|
20
|
20
|
p < 0.010
|
Abbreviation: PPI, proton-pump inhibitor.
Table 3
Procedural success in subjects of gastroduodenal obstruction in the intervention and
control arm
|
Procedural success
|
|
Yes
|
No
|
|
|
Control arm (PPI)
|
9
|
11
|
20
|
|
Intervention arm (misoprostol + PPI)
|
18
|
2
|
20
|
|
|
|
p = 0.006
|
Abbreviation: PPI, proton-pump inhibitor.
Gastrojejunostomy was performed in majority of patients but stricturoplasty/pyloroplasty
was performed in 2 patients (1 in each arm) who did not achieve clinical/procedural
success. The number of dilations required was significantly lower in the intervention
group (3.35 ± 0.813) compared to the control group (4.35 ± 1.040) (p = 0.004). Hemoglobin and albumin levels also significantly improved in the intervention
group (p = 0.015 and p = 0.032, respectively). Diarrhea was the most common adverse reaction, with no significant
difference between the groups ([Tables 4] and [5]).
Table 4
Improvement in nutritional status and number of dilatation sessions in both arms
|
Control arm (PPI)
|
Intervention arm (misoprostol + PPI)
|
p-Value
|
|
ΔHb (absolute increment in hemoglobin levels in g/dL)
|
0.870 ± 0.801
|
1.540 ± 0.859
|
0.015
|
|
ΔAlbumin (absolute increment in albumin levels in g/dL)
|
0.400 ± 0.229
|
0.675 ± 0.493
|
0.032
|
|
Number of dilatations
|
4.35 ± 1.040
|
3.35 ± 0.813
|
0.004
|
Abbreviations: Hb, hemoglobin; PPI, proton-pump inhibitor.
Table 5
Adverse reactions in both the groups
|
Intervention arm (misoprostol + PPI)
|
Control arm (PPI)
|
p-Value
|
|
Diarrhea
|
9
|
5
|
0.320
|
|
Abdominal pain
|
5
|
4
|
1.000
|
|
Bloating
|
3
|
4
|
1.000
|
|
Headache
|
2
|
1
|
1.000
|
|
Menstrual irregularity
|
0
|
0
|
–
|
Abbreviation: PPI, proton-pump inhibitor.
Discussion
The basis of conducting this study and including misoprostol as an add-on therapy
in all etiologies of GOO was based on evidence that misoprostol increases bicarbonate
secretion, promotes mucosal barrier and integrity, and possibly also enhances motility,
thus decreasing acid contact with mucosa and promote mucosal healing, but in the maintenance
phase of ulcer healing PPIs are supposed to be superior.[16]
[18] These properties can benefit any etiology of ulcerative GOO.
The optimum duration and timing of misoprostol therapy in patients with corrosive
intake or with a history of chronic opioid/NSAID use is unclear. So, we empirically
gave misoprostol for 2 weeks after exposure cessation, previous studies used misoprostol
therapy concomitant with long-term justified NSAIDs use.[19]
[20]
[21] Since we included only patients in whom the exposure to etiological agent has ceased
like NSAID/opium/smoking, we gave misoprostol only in the healing phase and continued
PPI for the healing phase of 14 weeks.
In our study, the most common etiology of benign GOO was corrosive injury (50%). There
is heterogeneity in various studies from India, as Kochhar et al[25] reported caustic ingestion (53.8%) as the most common etiology, while Maharshi et
al[31] reported opioid abuse (29%), whereas Sharma et al[32] reported PUD (42.6%). This contrast could represent either a true epidemiological
difference or reflect skewed access to health care or referral bias.
The main corrosive found in our study was hydrochloric acid (11/20) used as a toilet
cleaner and followed by sodium hydroxide (6/20) used in various industries. Corrosive
ingestion can lead to esophageal strictures, complicating GOO management. The prevalence
of opioid use in our country is 2.06% with heterogeneous distribution and is more
prevalent in our adjoining states like Haryana and Punjab.[33] In our study, 7.5% of patients used opioids (mostly tramadol) and 25% used NSAIDs
(mostly ibuprofen/paracetamol). Including excluded patients with opioid addiction,
the total opioid user breakdown was tramadol (5/9), heroin (3/9), and opium (1/9).
Opioid-induced GOO is primarily reported in India, often linked to recreational use.
NSAID users typically took them for musculoskeletal pain for over a year, but only
experienced GOO symptoms for 5 to 6 weeks before presentation. These patients were
referred to our deaddiction center for counseling and management.
Idiopathic peptic strictures (no significant NSAID use/H. pylori infection) constituted 17.5%, whereas in other countries it is believed that PUD
still constitutes around 90% of GOO.[34]
Most studies have used 15 mm endpoint for esophageal dilation, others have dilated
up to 16 to 18 mm, similarly the endpoint of pyloric dilation has been 15 mm in majority.[25] Therefore, this was defined as procedural success as dilatation of more than 15 mm
can increase the risk of perforation.[34]
[35]
We started with low diameter for balloon dilatation (usually 8 mm), dilated only till
15 mm, and as all patients of corrosive ingestion presented after the acute phase
was over (> 3 weeks), fortunately we did not encounter any postprocedure perforation.
Two patients in each study arm had multiple strictures and all these strictures were
dilated with the same diameter in the same session, sometimes colonoscope was required
in negotiating these strictures as due to gastric dilatation the upper gastrointestinal
endoscope was totally consumed and could not reach the distal stricture, especially
if the stricture was beyond the first part of the duodenum. Four patients in the control
arm while three patients in the intervention arm had complex pyloric stricture (length > 2.5 cm/tortuous/severely
angulated or edematous) [36]
[37] Difference between the two groups was not significant (p = 1.000).
We defined clinical success as symptom resolution without surgical intervention till
the completion of follow-up (1 year), which is a more relevant endpoint, as many patients
even after procedural success achieved either partial or no clinical success (vide
[Tables 2] and [3]). The possible explanation for this could be altered anatomy, persistent enteric
nervous dysfunction, and cicatrization/fibrosis. Clinical success at the end of study
in the intervention arm and control arm was 16/20 (80%) and 7/20 (35%), respectively
(p = 0.010). Moreover, the mean number of dilations required were significantly less
(p = 0.004) in the intervention group versus control (3.35 ± 0.813 vs. 4.35 ± 1.040).
In all the patients in whom we achieved stable clinical success, routine esophagogastroduodenoscopy,
which was done after the study, did not show any appreciable stricture. Primary failure
was seen in 2/20 (10%) in the intervention arm and in 11/20 (55%) in the control arm,
which was significant (p = 0.006). Secondary failure was seen in 2/18(11.11%) in the intervention arm and
in 2/9 (22.22%) in the control arm (p = 0.578).
A significant proportion of patients in the intervention arm versus the control arm
did not require surgery (p = 0.010) and this is relevant, because the majority of our patients were severely
anemic and nutritionally deprived ([Table 4]), making them high-risk candidates for surgery and for various postoperative complications.
Nutritional recovery (improvement in hemoglobin and serum albumin) was also more significant
in the intervention arm (p = 0.015 and p = 0.032, respectively), hence, effective endoscopic therapy and medical therapy is
of great value to such patients as it prevents surgery with possible high risk of
complications and poor outcomes.
Since in our study population, corrosive stricture constituted around 50%, therefore
comparison with previously published data would be more useful if we compare it with
studies that have included a sizeable proportion of corrosive strictures, as it is
obvious that a corrosive stricture would pose different set of problems as compared
to an opioid stricture or a benign PUD-related stricture and long-term prognosis is
dependent upon the etiology of the stricture as well.[38]
A few high-quality studies have been done on benign GOO from a single center from
India by Kochhar et al. In a study on 264 patients of benign GOO (> 50% corrosives),
procedural success and clinical success was 75.7 and 92.04%, respectively, requiring
a mean number of dilations of 2.55 and 5.37, respectively. While in the same study,
procedural success and clinical success in 142 patients of corrosive-induced GOO were
66.2 and 90.1% requiring a mean number of dilatations of 3.45 and 7.25, respectively,
of the 264 patients, 9 (3.4%) had perforations during EBD.[25] In another study from the same center, 31 patients of caustic GOO underwent EBD,
30 patients achieved procedural success (96.7%); with mean number of dilatations around
9, one patient developed perforation (3.2%),[39] while in another study 111 patients of caustic-induced GOO underwent EBD and procedural
and clinical success was achieved in 85.6 and 97.3%, respectively, requiring a mean
number of dilatation sessions of 5.3 and 7.21. The initial balloon diameter at the
start of dilatation and the last balloon diameter were 9.6 ± 2.06 and 14.5 ± 1.6 mm,
respectively. Perforation was encountered in 2 (1.80%) patients. There were no recurrences
in a follow-up period of 98 months.[36]
In a study by Chiu et al,[40] out of 18 patients of caustic-induced GOO, clinical success was achieved in 44.4%
only, requiring a mean number of dilations of 5.5 in the GOO-only group and 13.7 in
the esophageal stricture and GOO group. Whereas Solt at al[41] achieved a success rate of 35% in 17 patients of corrosive-induced GOO after EBD.
In another study by Rana et al,[27] 25 patients of benign GOO underwent EBD, and it was successful in 21/25 (84%). Patients
required one to six sessions of endoscopic dilatation (mean = 2.2 ± 1.2). Corrosive-induced
GOO required more dilatation sessions (3.83 ± 0.75) compared to peptic GOO (2.1 ± 0.56),
whereas in a study by Hamzaoui et al[42] that included 45 patients of only PUD-related GOO for EBD, 13 (29%) patients had
to undergo surgery despite EBD.
These studies demonstrate disparate and divergent results of EBD in pyloric strictures,
we could only achieve a procedural response rate of 45% in the control arm, which
agrees with the finding of Solt et al and Chiu et al,[40]
[41] whereas Kochhar et al[25]
[36]
[39] reported a procedural and clinical success varying from 75.7 to 96.7% and 92.04
to 97.3%, respectively; however, in misoprostol procedural and clinical success rates
were 90% (18/20) and 80% (16/20), respectively. The difference was highly significant
across the two study arms (p = 0.006 and 0.010), highlighting the efficacy of misoprostol as an adjunctive therapy.
Probable explanations for divergent result from our study (particularly the control
arm) include the fact that Kochhar et al[25] have used adjunctive therapy in the form of intralesional triamcinolone injection
in a subset of patients, besides in some studies, the mean number of dilations in
corrosives was much more (5.3–9).[25]
[36]
[39] While we have not dilated beyond five sessions, moreover in our study, most of the
patients had GOO score of < 2, which cannot be compared to other study populations
as they have not utilized this score, a low GOO score would imply a more severe stricturing
disease; furthermore, all patients of caustic ingestion had a Zargar score of IIB/IIIA,
whereas no such data is available from their study, so the comparability of study
population is uncertain, additionally as both the studies were conducted at different
tertiary care centers, there is a strong possibility of referral bias. In our study,
there are sizeable number of patients who have complex strictures (3 in the intervention
arm and 4 in the control arm) and two patients in each arm had multiple strictures,
but complex strictures were excluded by Kochhar et al,[25]
[36] hence their findings may not be comparable with us as complex strictures are usually
refractory to EBD.[37]
The number of dilatation sessions alone may not be the sole determinant of the outcome
as Chiu et al[40] reported a low clinical success rate of 44.4% despite greater number of dilatation
sessions (mean 13.7 in esophageal + pyloric stricture), closely correlating with our
observations, this reiterates that other factors such as severity of disease process,
concomitant esophageal involvement, and presence of complex/multiple stricture may
also influence outcome and these variables may be a poor prognostic marker for successful
EBD.[37]
[43]
[44]
[45] It is of interest to note that in the misoprostol treatment arm, patients with complex
and/or multiple strictures also showed markedly improved response to therapy highlighting
its efficacy as an adjunctive therapy to EBD.
We restricted our dilatation sessions to five per patients at the most, as this is
the criteria for refractory stricture.[25] Additionally, there were other reasons, a large number of patients were referral
patients from adjoining states, belonged to resource-constrained socioeconomic strata
with low education level, this in turn reflected in inadequate dietary management
through FJ in some and these patients also needed to go back to work as soon as possible,
moreover, multiple/prolonged follow-up visits were also difficult, hence, after five
sessions we opted for surgical management as there is evidence that timely surgery
leads to better quality of life and recovery.[46]
[47]
[48]
Misoprostol is well tolerated as the adverse effect profile was not significantly
different across the study arms ([Table 5]). The intervention group, which received misoprostol in addition to EBD and PPI,
exhibited significantly higher rates of clinical and procedural success and required
fewer dilations with improvement in nutritional status versus the control group.
There are some limitations of our study. First, the effective dosage, timing, and
duration of misoprostol therapy is unknown. Second, the study population had heterogeneous
etiology, so effectiveness of misoprostol might vary based on the etiology, which
is unexplored, as subgroup analysis was not performed. Moreover, it was a single-center
pilot study with a small sample size and with significant dropout rate.
