J Gastroenterol DOI 10.1007/s00535-015-1085-9
ORIGINAL ARTICLE—ALIMENTARY TRACT
Proton pump inhibitor after endoscopic resection for esophageal squamous cell cancer: multicenter prospective randomized controlled trial N. Kakushima1 • K. Hori2 • H. Ono1 • T. Horimatsu2 • N. Uedo3 • K. Ohata4 • H. Doyama5 • K. Kaneko6 • I. Oda7 • T. Hikichi8 • Y. Kawahara9 • K. Niimi10 Y. Takaki11 • M. Mizuno12 • S. Yazumi13 • A. Hosokawa14 • A. Imagawa15 • M. Niimi2 • K. Yoshimura16 • M. Muto2
•
Received: 2 March 2015 / Accepted: 20 April 2015 Ó Springer Japan 2015
Abstract Background Whether proton pump inhibitors (PPIs) relieve heartburn or precordial pain after endoscopic resection (ER) for esophageal squamous cell carcinoma (ESCC) remains unclear. The aim of this study was to investigate the efficacy of PPI therapy for these symptoms after ER for ESCC. Methods We conducted a multicenter prospective randomized controlled trial among 15 hospitals in Japan. In total, 229 patients with cT1a ESCC were randomly assigned to receive PPI therapy for 5 weeks after ER (the PPI group, n = 115) or follow-up without PPI therapy (the non-PPI group, n = 114). The primary end point was the incidence of gastroesophageal reflux disease (GERD)-like symptoms after ER from a self-reported questionnaire
This study was presented at Digestive Disease Week 2014 held in Chicago, IL, USA.
(Frequency Scale for Symptoms of GERD). Secondary end points were ulcer healing rate at 5 weeks, incidence of pain, improvement rate of symptoms in those who started PPI therapy because of GERD-like symptoms in the nonPPI group, and adverse events. Results No significant difference was observed in the incidence of GERD-like symptoms after ER between the non-PPI and PPI groups (30 % vs 34 %, respectively). No significant differences were observed in the ulcer healing rate at 5 weeks (84 % vs 85 %) and incidence of pain within 1 week (36 % vs 45 %). In nine of ten patients (90 %) who started PPI therapy because of GERD-like symptoms in the non-PPI group, PPI administration relieved GERD-like symptoms. No adverse events related to PPI administration were observed. Conclusion PPI therapy is not efficacious in reducing symptoms and did not promote healing of ulcers in patients undergoing ER for ESCC.
Electronic supplementary material The online version of this article (doi:10.1007/s00535-015-1085-9) contains supplementary material, which is available to authorized users. & N. Kakushima [email protected]
7
National Cancer Center, Tokyo, Japan
8
Fukushima Medical University, Fukushima, Japan
1
9
Okayama University Hospital, Okayama, Japan
10
The University of Tokyo Graduate School of Medicine, Tokyo, Japan
Division of Endoscopy, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi, Suntougun, Shizuoka 4118777, Japan
2
Kyoto University Graduate School of Medicine, Kyoto, Japan
11
3
Fukuoka University Chikushi Hospital, Fukuoka, Japan
Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan
12
Hiroshima City Hospital, Hiroshima, Japan
13
Kitano General Hospital, Osaka, Japan
14
Ishikawa Prefectural Central Hospital, Kanazawa, Ishikawa, Japan
Toyama University Hospital, Toyama, Japan
15
Mitoyo General Hospital, Kanonji, Kagawa, Japan
National Cancer Center East, Chiba, Japan
16
Kobe University School of Medicine, Kobe, Japan
4 5
6
NTT Medical Center Tokyo, Tokyo, Japan
123
J Gastroenterol
Keywords Esophageal cancer Squamous cell cancer Endoscopic resection Gastroesophageal reflux disease Proton pump inhibitor
Introduction Endoscopic resection (ER), including endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD), is accepted as a safe and effective treatment for superficial neoplasia in the gastrointestinal tract [1, 2]. In the esophagus, squamous cell carcinoma (SCC) with invasion limited to the lamina propria is a good candidate for ER on the basis of the low risk of lymph node metastasis [2–7]. With recent advances in diagnostic modalities and the permeation of screening endoscopy, the disease has been diagnosed at an earlier stage in more patients, resulting in an increase in the number eligible for ER [8]. After ER, an artificial ulcer is created in the esophagus. Symptoms of heartburn or precordial pain similar to those of gastroesophageal reflux disease (GERD) are occasionally observed [2]. These symptoms may be related simply to heat coagulation effects from esophageal ER, or may be due to the esophageal ulcer after ER. However, there have been no studies focusing on these postoperative symptoms. Therefore, the incidence and severity of GERD-like symptoms or precordial pain after esophageal ER remain unknown, and optimal management has not yet been completely documented. The administration of proton pump inhibitors (PPIs) is a standard treatment for esophageal erosions and ulcers related to GERD as well as for endoscopy-negative reflux disease [9–12]. The administration of PPIs promotes mucosal healing and decrease acid-related dyspepsia [9–12]. PPI administration also promotes the healing of artificial ulcers after gastric ESD and reduces the risk of postoperative bleeding [13, 14]. Moreover, PPI treatment following esophageal variceal ligation (EVL) was reported to be effective in healing post-EVL ulcers [15]. Therefore, PPI administration has been used for the treatment of esophageal ulcers after ER [2, 4, 5]. However, whether PPI administration relieves postoperative GERD-like symptoms or promotes the healing of ulcers after esophageal ER remains unclear. The present study aimed to investigate the efficacy of PPI therapy for the management of esophageal ulcers after ER for esophageal SCC (ESCC). To clarify whether PPI therapy would prevent or relieve GERD-like symptoms after esophageal ER, a randomized controlled trial was conducted to compare the incidence of GERD-like symptoms after ER between patients treated with or without a PPI after ER.
123
Methods We conducted an open-label, multicenter prospective randomized controlled study at 15 referral hospitals in Japan. Trial patients Patients who fulfilled all the following eligibility criteria were enrolled: 1.
Inclusion criteria: (a) (b)
(c) (d) (e) (f) (g) (h) (i) (j) 2.
ESCC with a clinical depth of T1a (invasion limited to the muscularis mucosa) Patients with a relatively low risk of esophageal stricture after ER—for example, after ER an estimated ulcer diameter of 5 cm or less and circumferential extent of three quarters or less [7] No prior chemotherapy for, radiotherapy for, and surgical treatment of ESCC Total Frequency Scale for Symptoms of GERD (FSSG) score of 7 points or less [16] (Fig. S1) No endoscopic findings of reflux esophagitis of Los Angeles grade A or higher [17] No use of PPIs and/or histamine H2 receptor antagonist 1 week before ER No history of radiotherapy to the neck, chest, or upper abdominal region Eastern Cooperative Oncology Group performance status of 0 or 1 Age of 20 years or more Ability to complete the FSSG and the 10-cm visual analog scale (VAS) (Fig. S2).
Exclusion criteria: (a) Active peptic ulcer in the stomach or duodenum (b) Requirement of systemic steroid administration (c) Requirement of systemic use of nonsteroidal antiinflammatory drugs (d) Inability to discontinue use of anticoagulant or antiplatelet drugs before ER (e) History of gastrectomy.
Ethical considerations and registration This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the Ethical Guidelines for Clinical Research of the Ministry of Health, Labor and Welfare of Japan. Written informed consent was provided by all patients before enrollment. The study protocol was approved by the ethics committee of Kyoto University on November 18, 2009 and the ethics committees of all participating hospitals thereafter. This study was
J Gastroenterol
registered with the University Hospital Medical Information Network Clinical Trial Registry (UMIN000002544). Randomization and treatment allocation Patients were randomly assigned to receive a PPI (rabeprazole, 20 mg) after ER (the PPI group) or follow-up without PPI administration (the non-PPI group) in a ratio of 1:1. Computer-generated randomization was performed by minimization methods, stratified by the tumor diameter (2.0 cm or greater vs less than 2.0 cm), age (20–59 years vs 60 years or older), sex, and hospital via a secure Webbased randomization system. Measured outcomes The primary end point was the proportion of patients with GERD-like symptoms after esophageal ER from a selfreported questionnaire (FSSG). Secondary end points were as follows: (1) the ulcer healing rate, as evaluated by endoscopy at 5 weeks, (2) GERD-like symptoms related to the seven items of the FSSG—Reflux Symptom (FSSGRS), (3) incidence of pain after ER, as measured with the VAS, (4) improvement rate of GERD-like symptoms in those who started taking the PPI because of GERD-like symptoms in the non-PPI group, and (5) adverse events. We used the FSSG to assess GERD-like symptoms at a cutoff score of 8 points [16, 18]. Among the 12 question of the FSSG, seven items are related to reflux symptoms (FSSG-RS). These seven items are as follows: ‘‘Do you get heartburn?’’ (question 1). ‘‘Do you sometimes subconsciously rub your chest with your hand?’’ (question 4). ‘‘Do you get heartburn after meals?’’ (question 6). ‘‘Do you have an unusual (e.g., burning) sensation in your throat)’’ (question 7). ‘‘Do some things get stuck when you swallow?’’ (question 9). ‘‘Do you get bitter liquid (acid) coming up into your throat?’’ (question 10). ‘‘Do you get heartburn if you bend over?’’ (question 12). Therefore, we also assessed these seven items included in the FSSG-RS to clarify the frequency of acid-reflux-related symptoms after ER. Pain was assessed according to the VAS score [19–21]. The proportion of patients with a VAS score of 4 points or more, which indicated pain as a therapeutic objective, was investigated. Adverse events were evaluated according to the Common Terminology Criteria for Adverse Events, version 3.0 [22]. Treatments The treatment protocol is shown in Fig. 1. Randomization was performed before ER. Patients were treated by EMR or ESD. Rabeprazole (20 mg) was administered orally daily
for 5 weeks beginning on the day after ER in the PPI group. A self-completed questionnaire notebook including the FSSG and the VAS was checked daily for 8 days after ER. Esophagogastroduodenoscopy was performed 5 weeks after ER. During the treatment protocol, the use of other PPIs, histamine H2 receptor blockers, mucosal protectives, and dopaminergic receptor antagonists was prohibited. The treatment protocol was terminated when GERD-like symptoms with an FSSG score of 10 points or more occurred during the first 8 days. We considered that an FSSG score of 10 points or more would correspond to a severe condition for the patient. Therefore, the treatment protocol was terminated to allow alternative treatment such as changing the PPI or adding mucosal protectives. In the non-PPI group, if symptoms with a FSSG score of 8 points or more occurred during the first 8 days, 20 mg of rabeprazole was administered. The improvement rate of symptoms in those in the non-PPI group who started PPI therapy because of GERD-like symptoms was investigated. Sample size and statistical analysis This study was based on the hypothesis that the administration of PPIs would reduce or prevent GERD-like symptoms after esophageal ER. Because there was no previous report regarding the incidence of symptoms after esophageal ER, a consensus meeting was held among the participating physicians before this study was started. The sample size was calculated on the assumption that the rate of GERD-like symptoms would be 70 % for the non-PPI group and 50 % for the PPI group. For a statistical power of 80 % at a one-sided significance level of 5 % more than 91 patients were required per group. On the assumption that few patients may drop out because of adverse events after ER, we decided to enroll 110 patients per group. Data were analyzed on an intention-to-treat basis. The patient population allocated to either group was considered to be the full analytical set. Fisher’s exact test was used to compare primary end points between the two groups with a one-sided significance level of 5 %. Statistical analysis was performed by a statistician (K.Y.) at an independent academic research organization.
Results Baseline characteristics From November 2009 to December 2012, 229 patients (197 males and 32 females) were enrolled and randomly assigned to the PPI group (n = 115) or the non-PPI group (n = 114) (Fig. 1). Baseline characteristics of the patients and lesions were well balanced between the groups
123
J Gastroenterol
Fig. 1 Flow diagram of participants in this study. ER endoscopic resection, FSSG Frequency Scale for Symptoms of Gastroesophageal Reflux Disease, PPI proton pump inhibitor, RPZ rabeprazole
(Table 1). Most patients (86 %, 198/229) had a single lesion. The median tumor diameter was 18 mm (range 2–50 mm) for the non-PPI group and 20 mm (range 3–50 mm) for the PPI group. The median length from the incisor teeth to the lesion was 30 cm for both the non-PPI group (range 18–41 cm) and the PPI group (range 17–40 cm). One patient in the non-PPI group did not undergo ER because no lesion was observed by endoscopy on the day of treatment. Nine patients in the non-PPI group and 11 patients in the PPI group did not finish the treatment protocol, as shown in Fig. 1.
with symptoms related to the seven items on the FSSG-RS was 21 % for each group (Table 3). Among the non-PPI group, the F-scale was 8 points or more for 28 patients during the first 5 days, during which ten patients started rabeprazole therapy. After rabeprazole therapy had started, symptom improvement was observed in nine patients within 8 days after ER. Therefore, the improvement rate of GERD-like symptoms was 90 % (95 % confidence interval 55.5–99.8 %). The other 18 patients did not take rabeprazole for unknown reasons and were discharged uneventfully.
Proportion of patients with GERD-like symptoms after esophageal ER (primary outcome)
Incidence of pain after esophageal ER
A total of 62 patients had GERD-like symptoms after ER. The median distance from the incisor teeth to the tumor among these patients was 30 cm (range 18–39 cm), which was not significantly different from that of those who did not have GERD-like symptoms. There was no statistically significant difference in the proportion of patients with GERD-like symptoms within 1 week after ER between the groups (30 and 34 % for the non-PPI group and the PPI group, respectively; Table 2). The proportion of patients
123
Pain as indicated by a VAS score of 4 points or more within 1 week after ER was observed in 36 and 45 % of patients in the non-PPI group and the PPI group, respectively, although this difference was not significant (Table 4). Ulcer healing after esophageal ER Among 228 patients who underwent ER for ESCC, complete ulcer healing at 5 weeks after ER was confirmed in 84
J Gastroenterol Table 1 Patient and lesion baseline characteristics
Table 1 continued
Non-PPI group (n = 114)
PPI group (n = 115)
Median
65
69
Median
30
30
Range
39–84
41–85
Range
10–70
10–70
IQR
11
13.5
IQR
20
15
Age (years)
Female PS at randomization
PPI group (n = 115)
Estimated diameter of mucosal defect (mm)
Sex Male
Non-PPI group (n = 114)
Estimated circumferential extent of mucosal defect 97 (85 %)
100 (87 %)
\1/4
33 (29 %)
25 (22 %)
17 (15 %)
15 (13 %)
C1/4, \1/2
51 (45 %)
57 (49 %)
C1/2, \3/4
29 (25 %)
31 (27 %)
C3/4, Bwhole
1 (1 %)
2 (2 %)
0
111 (97 %)
110 (96 %)
1
3 (3 %)
4 (3 %)
Unknown
0 (0 %)
1 (1 %)
Median
22
23
Range
14–35
14–31
IQR
4.8
4.1
Smoking habit
92 (81 %)
94 (82 %)
Alcohol drinking habit
106 (93 %)
100 (87 %)
History of gastric ulcer
17 (15 %)
11 (10 %)
History of duodenal ulcer
5 (4 %)
7 (6 %)
Multiple lesions
19 (17 %)
12 (10 %)
Median Range
30 18–41
30 17–40
IQR
5
6
Body mass index (kg/m2)
Distance from incisor (cm)
Location Cervical
1 (1 %)
4 (3 %)
Upper
15 (13 %)
10 (9 %)
Middle
66 (58 %)
62 (54 %)
Lower
28 (24.5 %)
38 (33 %)
Abdominal
4 (3.5 %)
1 (1 %)
Median
18
20
Range
2–50
3–50
IQR
5
5
Tumor diameter (mm)
Gross type of main tumor Elevated
5 (4 %)
7 (6 %)
Flat
27 (24 %)
24 (21 %)
Depressed Unknown
81 (71 %) 1 (1 %)
84 (73 %) 0 (0 %)
EP
51 (45 %)
53 (46 %)
LPM
44 (38 %)
51 (44 %)
MM
19 (17 %)
11 (10 %)
\1/4
71 (62 %)
72 (63 %)
C1/4, \1/2
29 (25 %)
28 (24 %)
C1/2, \3/4
14 (12 %)
15 (13 %)
Estimated tumor depth
Circumferential extent of tumor
EP epithelium, IQR interquartile range, LPM lamina propria mucosae, MM muscularis mucosae, PPI proton pump inhibitor, PS performance status
and 85 % of patients in the non-PPI group and the PPI group, respectively (Table 5). There was no statistically significant difference between the groups. All-cause adverse events The commonest adverse event of any grade was esophageal pain (58 %), with a similar incidence between the groups. Adverse events of grade 3 or grade 4 were observed in ten patients (4 %) because of esophageal or pharyngeal pain (Table 6).
Discussion To the best of our knowledge, this is the first randomized controlled trial to investigate the role of PPIs for the treatment of esophageal ulcers after ER for ESCC. The results of this study revealed that GERD-like symptoms occurred in approximately 30 % of patients after esophageal ER, and that the administration of PPIs did not affect the incidence of symptoms or promote healing of the ulcer after esophageal ER. The proportion of patients with GERD-like symptoms observed after esophageal ER, as assessed by the FSSG, was approximately 30 %, and the proportion of those assessed by the FSSG-RS was 21 %. In contrast, esophageal pain of all grades according to´National Cancer Institute Common Terminology Criteria for Adverse Events was observed in 58 % (133/229) of the patients. The rate of GERD-like symptoms after ER was thus overestimated. However, this study revealed an important finding that the symptoms after esophageal ER were mainly due to pain
123
J Gastroenterol Table 2 Proportion of patients with gastroesophageal reflux disease (GERD)-like symptoms within 1 week after esophageal endoscopic resection Number
GERD-like symptoms Yes
No
NE
Proportion (%)a
Fisher’s exact test P value
1-sided P = 0.79 (2-sided P = 0.57)
Non-PPI group
114
28 (25 %)
80 (70 %)
6 (5 %)
30 (22–39)
PPI group
115
34 (30 %)
76 (66 %)
5 (4 %)
34 (25–43)
NE not evaluated, PPI proton pump inhibitor, GERD gastroesophageal reflux disease, NE not evaluated, CI confidence interval
Table 3 Proportion of patients with symptoms related to the seven items of the Frequency Scale for Symptoms of Gastroesophageal Reflux Disease—Reflux Symptom (FSSG-RS) within 1 week after esophageal endoscopic resection Number
Symptoms related to the FSSG-RS Yes
No
NE
Proportion (%)a
Fisher’s exact test P value
1-sided P = 0.55 (2-sided P = 1.00)
Non-PPI group
114
18 (16 %)
90 (79 %)
6 (5 %)
21 (14–30)
PPI group
115
19 (17 %)
91 (79 %)
5 (4 %)
21 (14–29)
FSSG-RS Frequency Scale for Symptoms of GERD-Reflux Symptom, PPI proton pump inhibitor, N number, GERD gastroesophageal reflux disease, NE not evaluated, CI confidence interval
Table 4 Incidence of pain (visual analog scale score of more than 4 points) within 1 week after esophageal endoscopic resection (ER)
Number
Pain after ER Yes
Fisher’s exact test P value No
NE
Non-PPI group
114
41 (36 %)
62 (54 %)
11 (10 %)
PPI group
115
52 (45 %)
56 (49 %)
7 (6 %)
2-sided P = 0.36
PPI proton pump inhibitor, N number, ER endoscopic resection, NE not evaluated, CI confidence interval
Table 5 Complete ulcer healing at 5 weeks after esophageal endoscopic resection
N
Complete healing at 5 weeks (%) Yes
No
NE
Non-PPI group
114
96 (84 %)
12 (11 %)
6 (5 %)
PPI group
115
98 (85 %)
8 (7 %)
9 (8 %)
Fisher’s exact test P value 2-sided P = 1.00
PPI proton pump inhibitor, N number, NE not evaluated, CI confidence interval
and not to GERD-like symptoms. Only ten patients (4 %; five in each group) had severe pharyngeal or esophageal pain of grade 3. The pain was adequately relieved by mild analgesics, except in one patient, who required buprenorphine. Although the cause of severe pain was unknown, it was not related to adverse events of ER such as perforation or bleeding. The difference in the proportion of patients with GERD-like symptoms assessed by the FSSG or the FSSGRS was nearly 10 %. The five items of the FSSG that are not included in the FSSG-RS are related to motility disorder. A previous study reported that esophageal ESD may cause esophageal dysmotility in some patients, and might also have an influence on dysphagia, although without esophageal stricture [23]. The heat coagulation effect of ER or the creation of esophageal ulcer may have caused temporary dysmotility, but that is a matter of speculation.
123
The incidence of GERD-like symptoms and pain after esophageal ER was similar between the PPI group and the non-PPI group. Namely, PPI administration was not effective for prevention of these symptoms after esophageal ER. However, among the ten patients who started PPI therapy in the non-PPI group, the GERD-like symptoms rapidly improved. We did not enroll patients with apparent GERD symptoms or those with endoscopically confirmed esophagitis to avoid confusion with GERD-like symptoms at the baseline. Therefore, the temporary use of PPI therapy for those with GERD-like symptoms after ER was considered appropriate. With regard to the healing of ulcers after esophageal ER, complete healing was observed at 5 weeks in 85 % of patients regardless of PPI administration. Previous studies showed that complete healing of ulcers required 3–5 weeks
J Gastroenterol Table 6 Commonest all-cause adverse events according to the National Cancer Institute Common Terminology Criteria for Adverse Events All grades
Grade 3–4
Non-PPI group (n = 114)
PPI group (n = 115)
Non-PPI group (n = 114)
PPI group (n = 115)
Pain—esophagus
66 (59 %)
67 (58 %)
5 (4 %)
2 (2 %)
Pain—pharynx
36 (32 %)
32 (28 %)
0 (0 %)
3 (3 %)
3 (3 %)
0 (0 %)
0 (0 %)
0 (0 %)
Perforation—esophagus
4 (3.5 %)
1 (1 %)
0 (0 %)
0 (0 %)
Vomiting
1 (1 %)
2 (2 %)
0 (0 %)
0 (0 %)
Diarrhea
1 (1 %)
1 (1 %)
0 (0 %)
0 (0 %)
Fatigue Stomatitis
2 (2 %) 1 (0 %)
2 (2 %) 0 (0 %)
0 (0 %) 0 (0 %)
0 (0 %) 0 (0 %)
Stricture—esophagus
0 (0 %)
1 (1 %)
0 (0 %)
0 (0 %)
Hemorrhage—esophagus
PPI proton pump inhibitor, NCI-CTCAE National Cancer Institute Common Terminology Criteria for Adverse Events
after esophageal ER [24–26]. Ota et al. [26] reported that the type of ER (EMR or ESD), the length of the resected specimen, and the circumference of the ulcer did not affect ulcer healing in relation to PPI treatment. Although their study included only 21 patients, the results were in accordance with those of the present study, which included a much larger cohort. Our results indicate that PPI administration was not necessary to promote rapid ulcer healing after esophageal ER. No serious adverse events were observed in either group in this study. In the esophagus, the incidence of bleeding after EMR or ESD is generally low [2–7], In our study, three patients had postoperative bleeding (all in the nonPPI group). An endoscopic hemostasis procedure was used to control the bleeding in all patients. Our results showed that ER for ESCC could be safely performed without the use of a PPI. The limitation of this study is that we did not include patients with apparent GERD-like symptoms or those with esophageal adenocarcinoma. Among Japanese patients with esophageal carcinoma, more than 90 % have SCC and only 4 % have adenocarcinoma [8]. Esophageal adenocarcinoma mainly develops in the abdominal esophagus or esophagogastric junction, which is exposed to gastric reflux. We did not include these patients to avoid confusion with the baseline parameters of those who were expected to experience substantial effects of gastric reflux. Therefore, the efficacy of PPIs for patients who undergo ER for lesions at the esophagogastric junction or esophageal adenocarcinoma should be evaluated in the future. Another limitation is that we did not collect data regarding the presence of gastric atrophy. Considering that the median age of the study patients was more than 60 years and that more than 85 % of the patients were male with ESCC, we speculate that our patients represent a group of reduced gastric secretion. The background of the patients with gastric atrophy may have influenced the low incidence of
GERD-like symptoms of 30 % in the non-PPI group. However, we believe that our study patients are representative of most patients undergoing esophageal ER in Japan. In conclusion, PPI administration is not efficacious in reducing symptoms and does not improve ulcer healing after ER for ESCC. Only for patients with GERD-like symptoms after esophageal ER can on-demand PPI administration be considered.
Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Ono H, Kondo H, Gotoda T, et al. Endoscopic mucosal resection for treatment of early gastric cancer. Gut. 2001;48:225–9. 2. Kakushima N, Fujishiro M. Endoscopic submucosal dissection for gastrointestinal neoplasms. World J Gastroenterol. 2008;14:2962–7. 3. Fujishiro M, Yahagi N, Kakushima N, et al. Endoscopic submucosal dissection of esophageal squamous cell neoplasms. Clin Gastroenterol Hepatol. 2006;4:688–94. 4. Fujishiro M, Kodashima S, Goto O, et al. Endoscopic submucosal dissection for esophageal squamous cell neoplasms. Dig Endosc. 2009;21:109–15. 5. Repici A, Hassan C, Carlino A, et al. Endoscopic submucosal dissection in patients with early esophageal squamous cell carcinoma: results from a prospective Western series. Gastrointest Endosc. 2010;71:715–21. 6. Takahashi H, Arimura Y, Masao H, et al. Endoscopic submucosal dissection is superior to conventional endoscopic resection as a curative treatment for early squamous cell carcinoma of the esophagus (with video). Gastrointest Endosc. 2010;72:255–64. 7. Ono S, Fujishiro M, Niimi K, et al. Predictors of postoperative stricture after esophageal endoscopic submucosal dissection for superficial squamous cell neoplasms. Endoscopy. 2009;41:661–5. 8. Tachimori Y, Ozawa S, Fujishiro M, et al. Comprehensive registry of esophageal cancer in Japan, 2006. Esophagus. 2014;11:21–47.
123
J Gastroenterol 9. Talley NJ, Vakil N. Guidelines for the management of dyspepsia. Am J Gastroenterol. 2005;100:2324–37. 10. van Pinxteren B, Numans ME, et al. Short-term treatment with proton pump inhibitors, H2-receptor antagonists and prokinetics for gastro-oesophageal reflux disease-like symptoms and endoscopy negative reflux disease. Cochrane Database Syst Rev 2006;3:CD002095. 11. Jones RH, Baxter G. Lansoprazole 30 mg daily versus ranitidine 150 mg b.d. in the treatment of acid-related dyspepsia in general practice. Aliment Pharmacol Ther. 1997;11:541–6. 12. Kusano M, Shimoyama Y, Kawamura O, et al. Proton pump inhibitors improve acid-related dyspepsia in gastroesophageal reflux disease patients. Dig Dis Sci. 2007;52:1673–7. 13. Kakushima N, Yahagi N, Fujishiro M, et al. The healing of gastric artificial ulcers after endoscopic submucosal dissection. Dig Endosc. 2004;16:327–31. 14. Uedo N, Takeuchi Y, Yamada T, et al. Effect of a proton pump inhibitor or an H2-receptor antagonist on prevention of bleeding from ulcer after endoscopic submucosal dissection of early gastric cancer: a prospective randomized controlled trial. Am J Gastroenterol. 2007;102:1610–6. 15. Boo GB, Oh JC, Lee BJ, et al. The effect of proton pump inhibitor on healing of post-esophageal variceal ligation ulcers. Korean J Gastroenterol. 2008;41:232–40. 16. Kusano M, Shimoyama Y, Sugimoto S, et al. Development and evaluation of FSSG: frequency scale for the symptoms of GERD. J Gastroenterol. 2004;39:888–91. 17. Armstrong D, Bennett JR, Blum AL, et al. The endoscopic assessment of oesophagits: a progress report on observer agreement. Gastroenterology. 1996;111:85–92.
123
18. Oridate N, Takeda H, Mesuda Y, et al. Evaluation of upper abdominal symptoms using the frequency scale for the symptoms of gastroesophageal reflux disease in patients with laryngopharyngeal reflux symptoms. J Gastroenterol. 2008;43:519–23. 19. Bengtsson M, Ohlsson B, Ulander K. Development and psychometric testing of the visual analogue scale for irritable bowel syndrome (VAS-IBS). BMC Gastroenterol. 2007;7:16. 20. Chang FY, Lu CL, Chen CY, et al. Efficacy of dioctahedral smectite in treating patients of diarrhea-predominant irritable bowel syndrome. J Gastroenterol Hepatol. 2007;22:2266–72. 21. Watanabe H, Watanabe N, Ogura R, et al. A randomized prospective trial comparing unsedated endoscopy via transnasal and transoral routes using 5.5 mm video endoscopy. Dig Dis Sci. 2009;54:2155–60. 22. National Cancer Institute. Common Terminology Criteria for Adverse Events, version 3.0. Rockville: National Cancer Institute; 2006. 23. Bu BD, Linghy EQ, Li HK, et al. Influence of endoscopic submucosal dissection on esophageal motility. World J Gastroenterol. 2013;19:4781–5. 24. Inoue H, Takeshita K, Nagahama T, et al. Endoscopic mucosal resection of early esophageal cancer: EMRT and EMRC techniques. Stomach Intestine. 1993;28:161–9. In Japanese with English abstract. 25. Honda M, Nakamura T, Hori Y, et al. Process of healing of mucosal defects in the esophagus after endoscopic mucosal resection: histological evaluation in a dog model. Endoscopy. 2010;42:1092–5. 26. Ota M, Ohki T, Nakamura T, et al. Investigation of ulcer healing after esophageal EMR/ESD. Esophagus. 2010;7:101–4.
ORIGINAL ARTICLE—ALIMENTARY TRACT
Proton pump inhibitor after endoscopic resection for esophageal squamous cell cancer: multicenter prospective randomized controlled trial N. Kakushima1 • K. Hori2 • H. Ono1 • T. Horimatsu2 • N. Uedo3 • K. Ohata4 • H. Doyama5 • K. Kaneko6 • I. Oda7 • T. Hikichi8 • Y. Kawahara9 • K. Niimi10 Y. Takaki11 • M. Mizuno12 • S. Yazumi13 • A. Hosokawa14 • A. Imagawa15 • M. Niimi2 • K. Yoshimura16 • M. Muto2
•
Received: 2 March 2015 / Accepted: 20 April 2015 Ó Springer Japan 2015
Abstract Background Whether proton pump inhibitors (PPIs) relieve heartburn or precordial pain after endoscopic resection (ER) for esophageal squamous cell carcinoma (ESCC) remains unclear. The aim of this study was to investigate the efficacy of PPI therapy for these symptoms after ER for ESCC. Methods We conducted a multicenter prospective randomized controlled trial among 15 hospitals in Japan. In total, 229 patients with cT1a ESCC were randomly assigned to receive PPI therapy for 5 weeks after ER (the PPI group, n = 115) or follow-up without PPI therapy (the non-PPI group, n = 114). The primary end point was the incidence of gastroesophageal reflux disease (GERD)-like symptoms after ER from a self-reported questionnaire
This study was presented at Digestive Disease Week 2014 held in Chicago, IL, USA.
(Frequency Scale for Symptoms of GERD). Secondary end points were ulcer healing rate at 5 weeks, incidence of pain, improvement rate of symptoms in those who started PPI therapy because of GERD-like symptoms in the nonPPI group, and adverse events. Results No significant difference was observed in the incidence of GERD-like symptoms after ER between the non-PPI and PPI groups (30 % vs 34 %, respectively). No significant differences were observed in the ulcer healing rate at 5 weeks (84 % vs 85 %) and incidence of pain within 1 week (36 % vs 45 %). In nine of ten patients (90 %) who started PPI therapy because of GERD-like symptoms in the non-PPI group, PPI administration relieved GERD-like symptoms. No adverse events related to PPI administration were observed. Conclusion PPI therapy is not efficacious in reducing symptoms and did not promote healing of ulcers in patients undergoing ER for ESCC.
Electronic supplementary material The online version of this article (doi:10.1007/s00535-015-1085-9) contains supplementary material, which is available to authorized users. & N. Kakushima [email protected]
7
National Cancer Center, Tokyo, Japan
8
Fukushima Medical University, Fukushima, Japan
1
9
Okayama University Hospital, Okayama, Japan
10
The University of Tokyo Graduate School of Medicine, Tokyo, Japan
Division of Endoscopy, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi, Suntougun, Shizuoka 4118777, Japan
2
Kyoto University Graduate School of Medicine, Kyoto, Japan
11
3
Fukuoka University Chikushi Hospital, Fukuoka, Japan
Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan
12
Hiroshima City Hospital, Hiroshima, Japan
13
Kitano General Hospital, Osaka, Japan
14
Ishikawa Prefectural Central Hospital, Kanazawa, Ishikawa, Japan
Toyama University Hospital, Toyama, Japan
15
Mitoyo General Hospital, Kanonji, Kagawa, Japan
National Cancer Center East, Chiba, Japan
16
Kobe University School of Medicine, Kobe, Japan
4 5
6
NTT Medical Center Tokyo, Tokyo, Japan
123
J Gastroenterol
Keywords Esophageal cancer Squamous cell cancer Endoscopic resection Gastroesophageal reflux disease Proton pump inhibitor
Introduction Endoscopic resection (ER), including endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD), is accepted as a safe and effective treatment for superficial neoplasia in the gastrointestinal tract [1, 2]. In the esophagus, squamous cell carcinoma (SCC) with invasion limited to the lamina propria is a good candidate for ER on the basis of the low risk of lymph node metastasis [2–7]. With recent advances in diagnostic modalities and the permeation of screening endoscopy, the disease has been diagnosed at an earlier stage in more patients, resulting in an increase in the number eligible for ER [8]. After ER, an artificial ulcer is created in the esophagus. Symptoms of heartburn or precordial pain similar to those of gastroesophageal reflux disease (GERD) are occasionally observed [2]. These symptoms may be related simply to heat coagulation effects from esophageal ER, or may be due to the esophageal ulcer after ER. However, there have been no studies focusing on these postoperative symptoms. Therefore, the incidence and severity of GERD-like symptoms or precordial pain after esophageal ER remain unknown, and optimal management has not yet been completely documented. The administration of proton pump inhibitors (PPIs) is a standard treatment for esophageal erosions and ulcers related to GERD as well as for endoscopy-negative reflux disease [9–12]. The administration of PPIs promotes mucosal healing and decrease acid-related dyspepsia [9–12]. PPI administration also promotes the healing of artificial ulcers after gastric ESD and reduces the risk of postoperative bleeding [13, 14]. Moreover, PPI treatment following esophageal variceal ligation (EVL) was reported to be effective in healing post-EVL ulcers [15]. Therefore, PPI administration has been used for the treatment of esophageal ulcers after ER [2, 4, 5]. However, whether PPI administration relieves postoperative GERD-like symptoms or promotes the healing of ulcers after esophageal ER remains unclear. The present study aimed to investigate the efficacy of PPI therapy for the management of esophageal ulcers after ER for esophageal SCC (ESCC). To clarify whether PPI therapy would prevent or relieve GERD-like symptoms after esophageal ER, a randomized controlled trial was conducted to compare the incidence of GERD-like symptoms after ER between patients treated with or without a PPI after ER.
123
Methods We conducted an open-label, multicenter prospective randomized controlled study at 15 referral hospitals in Japan. Trial patients Patients who fulfilled all the following eligibility criteria were enrolled: 1.
Inclusion criteria: (a) (b)
(c) (d) (e) (f) (g) (h) (i) (j) 2.
ESCC with a clinical depth of T1a (invasion limited to the muscularis mucosa) Patients with a relatively low risk of esophageal stricture after ER—for example, after ER an estimated ulcer diameter of 5 cm or less and circumferential extent of three quarters or less [7] No prior chemotherapy for, radiotherapy for, and surgical treatment of ESCC Total Frequency Scale for Symptoms of GERD (FSSG) score of 7 points or less [16] (Fig. S1) No endoscopic findings of reflux esophagitis of Los Angeles grade A or higher [17] No use of PPIs and/or histamine H2 receptor antagonist 1 week before ER No history of radiotherapy to the neck, chest, or upper abdominal region Eastern Cooperative Oncology Group performance status of 0 or 1 Age of 20 years or more Ability to complete the FSSG and the 10-cm visual analog scale (VAS) (Fig. S2).
Exclusion criteria: (a) Active peptic ulcer in the stomach or duodenum (b) Requirement of systemic steroid administration (c) Requirement of systemic use of nonsteroidal antiinflammatory drugs (d) Inability to discontinue use of anticoagulant or antiplatelet drugs before ER (e) History of gastrectomy.
Ethical considerations and registration This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the Ethical Guidelines for Clinical Research of the Ministry of Health, Labor and Welfare of Japan. Written informed consent was provided by all patients before enrollment. The study protocol was approved by the ethics committee of Kyoto University on November 18, 2009 and the ethics committees of all participating hospitals thereafter. This study was
J Gastroenterol
registered with the University Hospital Medical Information Network Clinical Trial Registry (UMIN000002544). Randomization and treatment allocation Patients were randomly assigned to receive a PPI (rabeprazole, 20 mg) after ER (the PPI group) or follow-up without PPI administration (the non-PPI group) in a ratio of 1:1. Computer-generated randomization was performed by minimization methods, stratified by the tumor diameter (2.0 cm or greater vs less than 2.0 cm), age (20–59 years vs 60 years or older), sex, and hospital via a secure Webbased randomization system. Measured outcomes The primary end point was the proportion of patients with GERD-like symptoms after esophageal ER from a selfreported questionnaire (FSSG). Secondary end points were as follows: (1) the ulcer healing rate, as evaluated by endoscopy at 5 weeks, (2) GERD-like symptoms related to the seven items of the FSSG—Reflux Symptom (FSSGRS), (3) incidence of pain after ER, as measured with the VAS, (4) improvement rate of GERD-like symptoms in those who started taking the PPI because of GERD-like symptoms in the non-PPI group, and (5) adverse events. We used the FSSG to assess GERD-like symptoms at a cutoff score of 8 points [16, 18]. Among the 12 question of the FSSG, seven items are related to reflux symptoms (FSSG-RS). These seven items are as follows: ‘‘Do you get heartburn?’’ (question 1). ‘‘Do you sometimes subconsciously rub your chest with your hand?’’ (question 4). ‘‘Do you get heartburn after meals?’’ (question 6). ‘‘Do you have an unusual (e.g., burning) sensation in your throat)’’ (question 7). ‘‘Do some things get stuck when you swallow?’’ (question 9). ‘‘Do you get bitter liquid (acid) coming up into your throat?’’ (question 10). ‘‘Do you get heartburn if you bend over?’’ (question 12). Therefore, we also assessed these seven items included in the FSSG-RS to clarify the frequency of acid-reflux-related symptoms after ER. Pain was assessed according to the VAS score [19–21]. The proportion of patients with a VAS score of 4 points or more, which indicated pain as a therapeutic objective, was investigated. Adverse events were evaluated according to the Common Terminology Criteria for Adverse Events, version 3.0 [22]. Treatments The treatment protocol is shown in Fig. 1. Randomization was performed before ER. Patients were treated by EMR or ESD. Rabeprazole (20 mg) was administered orally daily
for 5 weeks beginning on the day after ER in the PPI group. A self-completed questionnaire notebook including the FSSG and the VAS was checked daily for 8 days after ER. Esophagogastroduodenoscopy was performed 5 weeks after ER. During the treatment protocol, the use of other PPIs, histamine H2 receptor blockers, mucosal protectives, and dopaminergic receptor antagonists was prohibited. The treatment protocol was terminated when GERD-like symptoms with an FSSG score of 10 points or more occurred during the first 8 days. We considered that an FSSG score of 10 points or more would correspond to a severe condition for the patient. Therefore, the treatment protocol was terminated to allow alternative treatment such as changing the PPI or adding mucosal protectives. In the non-PPI group, if symptoms with a FSSG score of 8 points or more occurred during the first 8 days, 20 mg of rabeprazole was administered. The improvement rate of symptoms in those in the non-PPI group who started PPI therapy because of GERD-like symptoms was investigated. Sample size and statistical analysis This study was based on the hypothesis that the administration of PPIs would reduce or prevent GERD-like symptoms after esophageal ER. Because there was no previous report regarding the incidence of symptoms after esophageal ER, a consensus meeting was held among the participating physicians before this study was started. The sample size was calculated on the assumption that the rate of GERD-like symptoms would be 70 % for the non-PPI group and 50 % for the PPI group. For a statistical power of 80 % at a one-sided significance level of 5 % more than 91 patients were required per group. On the assumption that few patients may drop out because of adverse events after ER, we decided to enroll 110 patients per group. Data were analyzed on an intention-to-treat basis. The patient population allocated to either group was considered to be the full analytical set. Fisher’s exact test was used to compare primary end points between the two groups with a one-sided significance level of 5 %. Statistical analysis was performed by a statistician (K.Y.) at an independent academic research organization.
Results Baseline characteristics From November 2009 to December 2012, 229 patients (197 males and 32 females) were enrolled and randomly assigned to the PPI group (n = 115) or the non-PPI group (n = 114) (Fig. 1). Baseline characteristics of the patients and lesions were well balanced between the groups
123
J Gastroenterol
Fig. 1 Flow diagram of participants in this study. ER endoscopic resection, FSSG Frequency Scale for Symptoms of Gastroesophageal Reflux Disease, PPI proton pump inhibitor, RPZ rabeprazole
(Table 1). Most patients (86 %, 198/229) had a single lesion. The median tumor diameter was 18 mm (range 2–50 mm) for the non-PPI group and 20 mm (range 3–50 mm) for the PPI group. The median length from the incisor teeth to the lesion was 30 cm for both the non-PPI group (range 18–41 cm) and the PPI group (range 17–40 cm). One patient in the non-PPI group did not undergo ER because no lesion was observed by endoscopy on the day of treatment. Nine patients in the non-PPI group and 11 patients in the PPI group did not finish the treatment protocol, as shown in Fig. 1.
with symptoms related to the seven items on the FSSG-RS was 21 % for each group (Table 3). Among the non-PPI group, the F-scale was 8 points or more for 28 patients during the first 5 days, during which ten patients started rabeprazole therapy. After rabeprazole therapy had started, symptom improvement was observed in nine patients within 8 days after ER. Therefore, the improvement rate of GERD-like symptoms was 90 % (95 % confidence interval 55.5–99.8 %). The other 18 patients did not take rabeprazole for unknown reasons and were discharged uneventfully.
Proportion of patients with GERD-like symptoms after esophageal ER (primary outcome)
Incidence of pain after esophageal ER
A total of 62 patients had GERD-like symptoms after ER. The median distance from the incisor teeth to the tumor among these patients was 30 cm (range 18–39 cm), which was not significantly different from that of those who did not have GERD-like symptoms. There was no statistically significant difference in the proportion of patients with GERD-like symptoms within 1 week after ER between the groups (30 and 34 % for the non-PPI group and the PPI group, respectively; Table 2). The proportion of patients
123
Pain as indicated by a VAS score of 4 points or more within 1 week after ER was observed in 36 and 45 % of patients in the non-PPI group and the PPI group, respectively, although this difference was not significant (Table 4). Ulcer healing after esophageal ER Among 228 patients who underwent ER for ESCC, complete ulcer healing at 5 weeks after ER was confirmed in 84
J Gastroenterol Table 1 Patient and lesion baseline characteristics
Table 1 continued
Non-PPI group (n = 114)
PPI group (n = 115)
Median
65
69
Median
30
30
Range
39–84
41–85
Range
10–70
10–70
IQR
11
13.5
IQR
20
15
Age (years)
Female PS at randomization
PPI group (n = 115)
Estimated diameter of mucosal defect (mm)
Sex Male
Non-PPI group (n = 114)
Estimated circumferential extent of mucosal defect 97 (85 %)
100 (87 %)
\1/4
33 (29 %)
25 (22 %)
17 (15 %)
15 (13 %)
C1/4, \1/2
51 (45 %)
57 (49 %)
C1/2, \3/4
29 (25 %)
31 (27 %)
C3/4, Bwhole
1 (1 %)
2 (2 %)
0
111 (97 %)
110 (96 %)
1
3 (3 %)
4 (3 %)
Unknown
0 (0 %)
1 (1 %)
Median
22
23
Range
14–35
14–31
IQR
4.8
4.1
Smoking habit
92 (81 %)
94 (82 %)
Alcohol drinking habit
106 (93 %)
100 (87 %)
History of gastric ulcer
17 (15 %)
11 (10 %)
History of duodenal ulcer
5 (4 %)
7 (6 %)
Multiple lesions
19 (17 %)
12 (10 %)
Median Range
30 18–41
30 17–40
IQR
5
6
Body mass index (kg/m2)
Distance from incisor (cm)
Location Cervical
1 (1 %)
4 (3 %)
Upper
15 (13 %)
10 (9 %)
Middle
66 (58 %)
62 (54 %)
Lower
28 (24.5 %)
38 (33 %)
Abdominal
4 (3.5 %)
1 (1 %)
Median
18
20
Range
2–50
3–50
IQR
5
5
Tumor diameter (mm)
Gross type of main tumor Elevated
5 (4 %)
7 (6 %)
Flat
27 (24 %)
24 (21 %)
Depressed Unknown
81 (71 %) 1 (1 %)
84 (73 %) 0 (0 %)
EP
51 (45 %)
53 (46 %)
LPM
44 (38 %)
51 (44 %)
MM
19 (17 %)
11 (10 %)
\1/4
71 (62 %)
72 (63 %)
C1/4, \1/2
29 (25 %)
28 (24 %)
C1/2, \3/4
14 (12 %)
15 (13 %)
Estimated tumor depth
Circumferential extent of tumor
EP epithelium, IQR interquartile range, LPM lamina propria mucosae, MM muscularis mucosae, PPI proton pump inhibitor, PS performance status
and 85 % of patients in the non-PPI group and the PPI group, respectively (Table 5). There was no statistically significant difference between the groups. All-cause adverse events The commonest adverse event of any grade was esophageal pain (58 %), with a similar incidence between the groups. Adverse events of grade 3 or grade 4 were observed in ten patients (4 %) because of esophageal or pharyngeal pain (Table 6).
Discussion To the best of our knowledge, this is the first randomized controlled trial to investigate the role of PPIs for the treatment of esophageal ulcers after ER for ESCC. The results of this study revealed that GERD-like symptoms occurred in approximately 30 % of patients after esophageal ER, and that the administration of PPIs did not affect the incidence of symptoms or promote healing of the ulcer after esophageal ER. The proportion of patients with GERD-like symptoms observed after esophageal ER, as assessed by the FSSG, was approximately 30 %, and the proportion of those assessed by the FSSG-RS was 21 %. In contrast, esophageal pain of all grades according to´National Cancer Institute Common Terminology Criteria for Adverse Events was observed in 58 % (133/229) of the patients. The rate of GERD-like symptoms after ER was thus overestimated. However, this study revealed an important finding that the symptoms after esophageal ER were mainly due to pain
123
J Gastroenterol Table 2 Proportion of patients with gastroesophageal reflux disease (GERD)-like symptoms within 1 week after esophageal endoscopic resection Number
GERD-like symptoms Yes
No
NE
Proportion (%)a
Fisher’s exact test P value
1-sided P = 0.79 (2-sided P = 0.57)
Non-PPI group
114
28 (25 %)
80 (70 %)
6 (5 %)
30 (22–39)
PPI group
115
34 (30 %)
76 (66 %)
5 (4 %)
34 (25–43)
NE not evaluated, PPI proton pump inhibitor, GERD gastroesophageal reflux disease, NE not evaluated, CI confidence interval
Table 3 Proportion of patients with symptoms related to the seven items of the Frequency Scale for Symptoms of Gastroesophageal Reflux Disease—Reflux Symptom (FSSG-RS) within 1 week after esophageal endoscopic resection Number
Symptoms related to the FSSG-RS Yes
No
NE
Proportion (%)a
Fisher’s exact test P value
1-sided P = 0.55 (2-sided P = 1.00)
Non-PPI group
114
18 (16 %)
90 (79 %)
6 (5 %)
21 (14–30)
PPI group
115
19 (17 %)
91 (79 %)
5 (4 %)
21 (14–29)
FSSG-RS Frequency Scale for Symptoms of GERD-Reflux Symptom, PPI proton pump inhibitor, N number, GERD gastroesophageal reflux disease, NE not evaluated, CI confidence interval
Table 4 Incidence of pain (visual analog scale score of more than 4 points) within 1 week after esophageal endoscopic resection (ER)
Number
Pain after ER Yes
Fisher’s exact test P value No
NE
Non-PPI group
114
41 (36 %)
62 (54 %)
11 (10 %)
PPI group
115
52 (45 %)
56 (49 %)
7 (6 %)
2-sided P = 0.36
PPI proton pump inhibitor, N number, ER endoscopic resection, NE not evaluated, CI confidence interval
Table 5 Complete ulcer healing at 5 weeks after esophageal endoscopic resection
N
Complete healing at 5 weeks (%) Yes
No
NE
Non-PPI group
114
96 (84 %)
12 (11 %)
6 (5 %)
PPI group
115
98 (85 %)
8 (7 %)
9 (8 %)
Fisher’s exact test P value 2-sided P = 1.00
PPI proton pump inhibitor, N number, NE not evaluated, CI confidence interval
and not to GERD-like symptoms. Only ten patients (4 %; five in each group) had severe pharyngeal or esophageal pain of grade 3. The pain was adequately relieved by mild analgesics, except in one patient, who required buprenorphine. Although the cause of severe pain was unknown, it was not related to adverse events of ER such as perforation or bleeding. The difference in the proportion of patients with GERD-like symptoms assessed by the FSSG or the FSSGRS was nearly 10 %. The five items of the FSSG that are not included in the FSSG-RS are related to motility disorder. A previous study reported that esophageal ESD may cause esophageal dysmotility in some patients, and might also have an influence on dysphagia, although without esophageal stricture [23]. The heat coagulation effect of ER or the creation of esophageal ulcer may have caused temporary dysmotility, but that is a matter of speculation.
123
The incidence of GERD-like symptoms and pain after esophageal ER was similar between the PPI group and the non-PPI group. Namely, PPI administration was not effective for prevention of these symptoms after esophageal ER. However, among the ten patients who started PPI therapy in the non-PPI group, the GERD-like symptoms rapidly improved. We did not enroll patients with apparent GERD symptoms or those with endoscopically confirmed esophagitis to avoid confusion with GERD-like symptoms at the baseline. Therefore, the temporary use of PPI therapy for those with GERD-like symptoms after ER was considered appropriate. With regard to the healing of ulcers after esophageal ER, complete healing was observed at 5 weeks in 85 % of patients regardless of PPI administration. Previous studies showed that complete healing of ulcers required 3–5 weeks
J Gastroenterol Table 6 Commonest all-cause adverse events according to the National Cancer Institute Common Terminology Criteria for Adverse Events All grades
Grade 3–4
Non-PPI group (n = 114)
PPI group (n = 115)
Non-PPI group (n = 114)
PPI group (n = 115)
Pain—esophagus
66 (59 %)
67 (58 %)
5 (4 %)
2 (2 %)
Pain—pharynx
36 (32 %)
32 (28 %)
0 (0 %)
3 (3 %)
3 (3 %)
0 (0 %)
0 (0 %)
0 (0 %)
Perforation—esophagus
4 (3.5 %)
1 (1 %)
0 (0 %)
0 (0 %)
Vomiting
1 (1 %)
2 (2 %)
0 (0 %)
0 (0 %)
Diarrhea
1 (1 %)
1 (1 %)
0 (0 %)
0 (0 %)
Fatigue Stomatitis
2 (2 %) 1 (0 %)
2 (2 %) 0 (0 %)
0 (0 %) 0 (0 %)
0 (0 %) 0 (0 %)
Stricture—esophagus
0 (0 %)
1 (1 %)
0 (0 %)
0 (0 %)
Hemorrhage—esophagus
PPI proton pump inhibitor, NCI-CTCAE National Cancer Institute Common Terminology Criteria for Adverse Events
after esophageal ER [24–26]. Ota et al. [26] reported that the type of ER (EMR or ESD), the length of the resected specimen, and the circumference of the ulcer did not affect ulcer healing in relation to PPI treatment. Although their study included only 21 patients, the results were in accordance with those of the present study, which included a much larger cohort. Our results indicate that PPI administration was not necessary to promote rapid ulcer healing after esophageal ER. No serious adverse events were observed in either group in this study. In the esophagus, the incidence of bleeding after EMR or ESD is generally low [2–7], In our study, three patients had postoperative bleeding (all in the nonPPI group). An endoscopic hemostasis procedure was used to control the bleeding in all patients. Our results showed that ER for ESCC could be safely performed without the use of a PPI. The limitation of this study is that we did not include patients with apparent GERD-like symptoms or those with esophageal adenocarcinoma. Among Japanese patients with esophageal carcinoma, more than 90 % have SCC and only 4 % have adenocarcinoma [8]. Esophageal adenocarcinoma mainly develops in the abdominal esophagus or esophagogastric junction, which is exposed to gastric reflux. We did not include these patients to avoid confusion with the baseline parameters of those who were expected to experience substantial effects of gastric reflux. Therefore, the efficacy of PPIs for patients who undergo ER for lesions at the esophagogastric junction or esophageal adenocarcinoma should be evaluated in the future. Another limitation is that we did not collect data regarding the presence of gastric atrophy. Considering that the median age of the study patients was more than 60 years and that more than 85 % of the patients were male with ESCC, we speculate that our patients represent a group of reduced gastric secretion. The background of the patients with gastric atrophy may have influenced the low incidence of
GERD-like symptoms of 30 % in the non-PPI group. However, we believe that our study patients are representative of most patients undergoing esophageal ER in Japan. In conclusion, PPI administration is not efficacious in reducing symptoms and does not improve ulcer healing after ER for ESCC. Only for patients with GERD-like symptoms after esophageal ER can on-demand PPI administration be considered.
Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Ono H, Kondo H, Gotoda T, et al. Endoscopic mucosal resection for treatment of early gastric cancer. Gut. 2001;48:225–9. 2. Kakushima N, Fujishiro M. Endoscopic submucosal dissection for gastrointestinal neoplasms. World J Gastroenterol. 2008;14:2962–7. 3. Fujishiro M, Yahagi N, Kakushima N, et al. Endoscopic submucosal dissection of esophageal squamous cell neoplasms. Clin Gastroenterol Hepatol. 2006;4:688–94. 4. Fujishiro M, Kodashima S, Goto O, et al. Endoscopic submucosal dissection for esophageal squamous cell neoplasms. Dig Endosc. 2009;21:109–15. 5. Repici A, Hassan C, Carlino A, et al. Endoscopic submucosal dissection in patients with early esophageal squamous cell carcinoma: results from a prospective Western series. Gastrointest Endosc. 2010;71:715–21. 6. Takahashi H, Arimura Y, Masao H, et al. Endoscopic submucosal dissection is superior to conventional endoscopic resection as a curative treatment for early squamous cell carcinoma of the esophagus (with video). Gastrointest Endosc. 2010;72:255–64. 7. Ono S, Fujishiro M, Niimi K, et al. Predictors of postoperative stricture after esophageal endoscopic submucosal dissection for superficial squamous cell neoplasms. Endoscopy. 2009;41:661–5. 8. Tachimori Y, Ozawa S, Fujishiro M, et al. Comprehensive registry of esophageal cancer in Japan, 2006. Esophagus. 2014;11:21–47.
123
J Gastroenterol 9. Talley NJ, Vakil N. Guidelines for the management of dyspepsia. Am J Gastroenterol. 2005;100:2324–37. 10. van Pinxteren B, Numans ME, et al. Short-term treatment with proton pump inhibitors, H2-receptor antagonists and prokinetics for gastro-oesophageal reflux disease-like symptoms and endoscopy negative reflux disease. Cochrane Database Syst Rev 2006;3:CD002095. 11. Jones RH, Baxter G. Lansoprazole 30 mg daily versus ranitidine 150 mg b.d. in the treatment of acid-related dyspepsia in general practice. Aliment Pharmacol Ther. 1997;11:541–6. 12. Kusano M, Shimoyama Y, Kawamura O, et al. Proton pump inhibitors improve acid-related dyspepsia in gastroesophageal reflux disease patients. Dig Dis Sci. 2007;52:1673–7. 13. Kakushima N, Yahagi N, Fujishiro M, et al. The healing of gastric artificial ulcers after endoscopic submucosal dissection. Dig Endosc. 2004;16:327–31. 14. Uedo N, Takeuchi Y, Yamada T, et al. Effect of a proton pump inhibitor or an H2-receptor antagonist on prevention of bleeding from ulcer after endoscopic submucosal dissection of early gastric cancer: a prospective randomized controlled trial. Am J Gastroenterol. 2007;102:1610–6. 15. Boo GB, Oh JC, Lee BJ, et al. The effect of proton pump inhibitor on healing of post-esophageal variceal ligation ulcers. Korean J Gastroenterol. 2008;41:232–40. 16. Kusano M, Shimoyama Y, Sugimoto S, et al. Development and evaluation of FSSG: frequency scale for the symptoms of GERD. J Gastroenterol. 2004;39:888–91. 17. Armstrong D, Bennett JR, Blum AL, et al. The endoscopic assessment of oesophagits: a progress report on observer agreement. Gastroenterology. 1996;111:85–92.
123
18. Oridate N, Takeda H, Mesuda Y, et al. Evaluation of upper abdominal symptoms using the frequency scale for the symptoms of gastroesophageal reflux disease in patients with laryngopharyngeal reflux symptoms. J Gastroenterol. 2008;43:519–23. 19. Bengtsson M, Ohlsson B, Ulander K. Development and psychometric testing of the visual analogue scale for irritable bowel syndrome (VAS-IBS). BMC Gastroenterol. 2007;7:16. 20. Chang FY, Lu CL, Chen CY, et al. Efficacy of dioctahedral smectite in treating patients of diarrhea-predominant irritable bowel syndrome. J Gastroenterol Hepatol. 2007;22:2266–72. 21. Watanabe H, Watanabe N, Ogura R, et al. A randomized prospective trial comparing unsedated endoscopy via transnasal and transoral routes using 5.5 mm video endoscopy. Dig Dis Sci. 2009;54:2155–60. 22. National Cancer Institute. Common Terminology Criteria for Adverse Events, version 3.0. Rockville: National Cancer Institute; 2006. 23. Bu BD, Linghy EQ, Li HK, et al. Influence of endoscopic submucosal dissection on esophageal motility. World J Gastroenterol. 2013;19:4781–5. 24. Inoue H, Takeshita K, Nagahama T, et al. Endoscopic mucosal resection of early esophageal cancer: EMRT and EMRC techniques. Stomach Intestine. 1993;28:161–9. In Japanese with English abstract. 25. Honda M, Nakamura T, Hori Y, et al. Process of healing of mucosal defects in the esophagus after endoscopic mucosal resection: histological evaluation in a dog model. Endoscopy. 2010;42:1092–5. 26. Ota M, Ohki T, Nakamura T, et al. Investigation of ulcer healing after esophageal EMR/ESD. Esophagus. 2010;7:101–4.
