Surg Endosc DOI 10.1007/s00464-014-3781-6
and Other Interventional Techniques
A prospective randomized trial of intralesional triamcinolone injections after endoscopic dilation for complex esophagogastric anastomotic strictures Steroid injection after endoscopic dilation Ju´lio C. Pereira-Lima • Michele Lemos Bonotto • Gustavo Dru¨gg Hahn • Guilherme Watte • Ce´sar Vivian Lopes Carlos Eduardo Oliveira dos Santos • Cla´udio R. Teixeira
•
Received: 10 April 2014 / Accepted: 23 July 2014 Ó Springer Science+Business Media New York 2014
Abstract Background and aims Postoperative esophageal strictures frequently recur. We assessed the efficacy of injecting triamcinolone after bougie dilation in comparison to dilation alone. Methods In a double-blind randomized fashion, 19 patients (68 % male with mean age of 53-years old) with non-dilated esophagogastric complex strictures after esophagectomy with gastric pull-up were assigned to receive dilation alone (control) or 40 mg of triamcinolone at the borders of the wall lacerations caused by the bougienage in each dilation session during the study follow-up. Dysphagia and complications were assessed at 1, 2, and 6 months. Primary end-point was to be dysphagia-free. Results After 1 month of the beginning of therapy, 4 patients in the steroid group were without dysphagia, in comparison to 0 patient in the control group (P = 0.021).
J. C. Pereira-Lima M. Lemos Bonotto G. D. Hahn (&) C. V. Lopes Department of Gastroenterology and Hepatology, Santa Casa Hospital/Federal University of Health Sciences of Porto Alegre (UFCSPA), Porto Alegre, Brazil e-mail: [email protected] J. C. Pereira-Lima e-mail: [email protected] J. C. Pereira-Lima C. R. Teixeira Rio Grande do Sul Foundation of Gastroenterology (FUGAST), Porto Alegre, Rio Grande do Sul, Brazil G. Watte The Post-Graduation Program in Chest Medicine Sciences, Santa Casa Hospital, Porto Alegre, Brazil C. E. O. dos Santos Santa Casa de Bage´ Hospital, Bage´, Brazil
Six months after endoscopic therapy, 62 % of the cases in the triamcinolone group versus none in the control group were dysphagia-free (P = 0.009). There was no perforation nor hemorrhage. Conclusions Injecting triamcinolone after every dilation session next to or at the borders of the lacerations caused by the dilators, results in a significant improvement or resolution of dysphagia. Keywords Benign esophageal strictures Postoperative Endoscopic dilation Triamcinolone Anastomotic strictures after esophagectomy with gastric pull-up are presently the main cause of benign esophageal strictures in the Western hemisphere, due to the widespread use of proton pump inhibitors (PPI) in the last two decades, which reduced the frequency of esophageal peptic strictures after treatment [1–4]. Endoscopic dilation with or without fluoroscopy is considered the initial treatment of choice for both simple and complex strictures [2, 5, 6]. Complex strictures (tortuous, longer than 2 cm, asymmetrical and/or with a luminal diameter less than 9 mm) are usually caused by caustic ingestion, radiation injury, aggressive endotherapy for Barrett’s esophagus or early esophageal cancer or are postoperative. These strictures are more difficult to dilate and are associated with a higher recurrence rate. Complex postoperative strictures usually need 4–8 dilations sittings either with conventional bougienage or balloon dilation technique in a 3–18 months follow-up [2, 5, 7]. Although additional endoscopic techniques combined with dilation such as steroid injections, topical mitomycinC application, incisional therapy, and stent placement are recommended for the management of refractory strictures,
123
Surg Endosc
few trials have addressed this issue in a randomized fashion. Moreover, postoperative stenosis frequently recurs and demands many dilation sessions, turning into refractory strictures [5, 7, 8]. In this study, we aimed to compare the ‘‘ab initio’’ use of triamcinolone injections in conjunction with endoscopic dilation with bougies with dilation alone in the treatment of complex anastomotic strictures.
Materials and methods Patients with dysphagia to solids (dysphagia grade 2) after esophagectomy with gastric pull-up and cervical handsewn esophagogastrostomy, with a confirmed anastomotic stricture that were not passable with a 0.98 cm endoscope were included. Exclusion criteria were suspicion of recurrent malignancy at the diagnosis, previous dilations or radiotherapy, active anastomotic leak, and refusal to enter in the protocol. All patients gave signed informed consent to the procedures and to the study before randomization. The study protocol was approved by the research Ethics Commission of our Institution. Endoscopic procedures The randomization was computer-generated by the www. researchrandomizer.org site. Randomization was performed by study personnel not involved in patients’ followup immediately before endoscopy. Patients were randomized to receive a total of 40 mg triamcinolone acetonide (Ophtaac 40, Laborato´rio Ophtalmos, Sa˜o Paulo, Brasil) injections around and at the borders of the tears caused by the bougie dilation after each therapeutic endoscopy session or to sham injections after bougie dilation. The endoscopists were not blinded to the therapy; however, the patients and the research fellows responsible for the follow-up were blinded to patient’s therapy. Upper endoscopies were performed under conscious sedation with midazolam and fentanyl by the Unit endoscopic staff or gastroenterology fellows, who were not involved in patient clinical care follow-up. Patients were progressively dilated (rule of 3 bougies of increasing size per session) at 14-day intervals until the 51F (17 mm) dilator was introduced. Savary-Gilliard bougies(WilsonCook Medical Inc., Winston Salem, USA) under guidance of a biliary guide-wire (Jagwire, Boston Scientific, Natick, USA) were used. At baseline, data on age, sex, indication for esophagectomy, administration of neoadjuvant or adjuvant
123
chemotherapy, dysphagia grade, and first inserted bougie size were recorded. Since the etiology of postoperative stenosis is ischemic, no PPIs were recommended. At baseline, 1, 2, and 6 months after the beginning of therapy, patients were interviewed by phone or personally by the research fellow, who asked standardized questions on dysphagia and complications. Dysphagia was graded as previously published: 0, taking normal diet, no dysphagia; 1, unable to swallow certain solid foods (e.g., grilled beef); 2, able to swallow only semisolid soft foods; 3, able to swallow just liquids; and 4, unable to swallow liquids in adequate amounts or difficulty in swallowing the own saliva [1]. In case recurrent dysphagia was reported during a follow-up contact or a patient visit to the Outpatient Unit of our hospital, an endoscopy with dilation according to the initial randomization was performed within a week during the 6-month follow-up period. The primary study end-point was the number of patients without dysphagia at follow-up. Secondary end-points were total number of dilation sessions per patient, complications and dysphagia grade at the follow-up. Statistical analysis Data were presented as mean ± SD or frequency and percentage. We performed associations between variables with the v2 tests (Pearson’s or Linear-by-Linear Association) and within-patient analysis with McNemar’s test. For comparing continuous variables, a Student t test or an unequal variance t test was used. Statistical significance was accepted at P \ 0.05. All statistical analyses were performed using the Statistical Package for the Social Sciences (PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc).
Results Between December, 2012, and February, 2014, 42 patients with postoperative esophageal stenosis were assessed for study eligibility. 5 patients were excluded due to prior dilation, 4 due to the fact that the stricture was passable with the endoscope, 7 due to concomitant radiotherapy, 6 to postoperative stenosis due to another surgical technique (esophagocoloplasty or esophagojejunostomy) and 1 case has not accepted to participate in the study. 19 patients with postoperative stricture due to cervical esophagogastrostomy who did not undergo previous radiotherapy or endoscopic dilation were randomized to receive triamcinolone (N = 10) or sham therapy (N = 9). Between study randomization and follow-up end (6 months), 1 patient
Surg Endosc Table 1 Clinical characteristics of the study baseline Characteristic
Total
Control
Triamcinolone
Male
13 (68)
4 (21)
9 (47)
0.06
Age (year)
54 ± 12
52 ± 15
56 ± 8
0.46
Dysphagia
P Value
0.06
2
13 (69)
8 (42)
5 (27)
3
4 (21)
1 (5)
3 (16)
4
2 (10)
–
2 (10)
Simple size = 19 patients. Data are presented as mean ± SD or No. (%). Dysphagia was graded as previously published: 1, unable to swallow certain solid foods (e.g. grilled beef); 2, able to swallow only semisolid soft foods; 3, able to swallow just liquids; 4, unable to swallow liquids in adequate amounts or difficulty in swallowing the own saliva [1]
and 2). There was no difference in dysphagia improvement when comparing the first and the second month visits. On the 6th month follow-up evaluation, 62 % of the patients who received steroid injections (5 out of 8 patients) were dysphagia-free versus 0 % (0 of 9 patients) of the control group (P = 0.009) (Fig. 2). In addition, 2 patients in the control group were with grade 3 dysphagia 6 months after randomization. The difference between groups was already statistically significant after 1 month of therapy (4 out of 10 patients in the steroid group—40 %— versus none in the dilation alone group, P = 0.021). The mean number of dilation sessions needed in the steroid group was 3 ± 2 versus 4 ± 2 in the dilation alone group (P = 0.388). Major complications
died and 1 patient presented with anastomotic cancer recurrence. Indication for esophagectomy in the randomized patients was squamous cell carcinoma in 18 and adenocarcinoma in 1. The mean time frame between the day of surgery and the day of the first endoscopic dilation was 76 (range 31–182) days in the steroid group and 68 days in the dilation alone group (range 29–205), (P = 0.6). Other patients’ characteristics at baseline are at Table 1. Response to treatment Both groups had improvement in dysphagia grade when comparing baseline with the 6-month follow-up visit (Fig. 1). However, not only the improvement, but also the abolishment of dysphagia, were significantly greater in the triamcinolone group in comparison to the controls (Figs. 1
Discussion In this study, the injection of intralesional triamcinolone in the wounds provoked by the shearing effect of the bougies in the dilated stenotic esophageal segment provided significantly more dysphagia-free patients in the follow-up, than performing dilation with Savary-Gilliard bougies alone. More patients were also significantly dysphagia-free in the first month, as well as in the end of the study followup (6 months). Although we have not addressed quality of life scores, to be dysphagia-free since the very beginning of the therapy, as well as to require less frequent dilation
12
*
10
8
Patient
Fig. 1 Comparison between the groups in relation to dysphagia in the baseline and after intervention (N = 19). *Triamcinolone vs Control, P \ .001. **Triamcinolone in the 6th month versus Triamcinolone in the baseline, P \ .001
There were no perforations, clinically overt bleeding or esophageal candidiasis in the studied population.
6
**
4
2
0 Baseline
1st month
Triamcinolone without dysphagia
2nd month
Triamcinolone with dysphagia
6th month
Control with dysphagia
123
Surg Endosc 6
P. 009
5
Patient
4 3 2 1 0 Triamcinolone
Control
Dysphagia 0
1
2
3
Fig. 2 Comparison between the groups in relation to dysphagia in the 6th month after the intervention (N = 16). Dysphagia was graded as previously published: 0, taking normal diet, no dysphagia; 1, unable to swallow certain solid foods (e.g., grilled beef); 2, able to swallow only semisolid soft foods; 3, able to swallow just liquids [1]. Linearby-Linear Association v2 test
sessions, certainly means a high improvement in someone’s quality of life. Our study has important strengths: (1) the adequate method of steroid injection—after performing dilation and aiming at the zone that will be contracted by the scaring of the wound, instead of injecting randomly before the bougienage; (2) the enrollment of an uniform sample of only true complex strictures-all postoperative, all not passable with the endoscope and all treatment naı¨ve; (3) the intense follow-up for 6 months, a time frame, which after that, dysphagia-free dilated patients will probably not have dysphagia recurrence. On the other hand, the major drawback of our study is the small sample size. However, we decided to interrupt the trial due to the notable difference in treatment response between the two study arms. During the past decade, there has been growing interest in the use of steroid injections for the treatment of refractory benign esophageal stenosis. Kochar et al. [6] retrospectively analyzed 71 stricture cases (19 of which with postoperative strictures) that were being managed with a program of intermittent endoscopic dilation using polyvinyl bougies, it was observed that after triamcinolone injections, the number of monthly dilations needed to keep the patients without dysphagia dropped from 1.24 to 0.5 after the introduction of steroid injections. The decrease in this dilation index was significantly observed in benign strictures from all etiologies. Different steroids such as betamethasone or dexamethasone have been used intralesionally along with endoscopic
123
dilation in other retrospective series with encouraging results [3, 9]. The technique of dilation (either bougie or balloon) or injections (before or after dilation), as well as the performance of injections varies from only in the first session to as many sessions as the number of dilations [4, 5, 9–11]. In one randomized comparative study involving 21 patients with strictures of different causes and variable number of previous dilation sittings, the additional use of triamcinolone (10 mg per injection, 4 injections, 1 in each quadrant) after esophageal bougie dilation resulted in an increase in the dysphagia-free period, a decrease in the periodic dilation index and in the mean number of dilation sessions per patient [10]. A decrease in the need for repeated balloon dilation and in the average time for new balloon dilation in patients with peptic strictures submitted to triamcinolone injections compared to sham injections was observed in a randomized comparative trial involving 30 patients [11]. These authors included stenosis which were passable by the endoscope and patients with the socalled ‘‘recalcitrant’’ strictures. The investigators considered recalcitrant (a synonym for refractory), strictures that recurred after only one dilation session instead of five, as usually considered in the literature [12]. On the other hand, the study was blinded and the groups were stratified to dysphagia grade before treatment, presence of esophagitis and prior therapies [11]. Finally, a recently published randomized trial from Netherlands, which compared triamcinolone injections before bougie dilation with dilation alone, is the only clearly negative study in the literature [4]. In this Dutch study, involving 60 patients, 45 % of the patients in the steroid group were dysphagia-free at 6-month versus 36 % of the controls. Median time to repeat dilations was 108 days for the treated group and 42 days for the controls (NS). Candida esophagitis developed in four cases of the corticosteroid group, but in none of the controls. We have not observed any case of candidiasis in the remnant esophagus, and this minor side effect was also not reported in other studies [3, 6, 10, 11]. The investigators performed 10 mg injections of triamcinolone in each quadrant of the stenotic segment before the passage of the dilator. Since bougie dilators provoke lacerations in one or two parts of the stenotic segment (usually in just one), 2 or 3 of the triamcinolone injections were lost with this technique. Our approach was to specifically target the injections next to or at the border of the wall lacerations, in order to reduce the inflammatory and recurrent cicatrix response at the anastomotic stricture. This way, the dilated esophageal segment would contract less and the lumen would remain wider. During gastric pull-up, the left epiploic artery is ligated, which makes the cervical esophagogastrostomy susceptible to ischemia. Although the exact mechanism is unknown,
Surg Endosc
scaring resulting from chronic ischemic inflammation causes the anastomotic stenosis. Intralesional steroids have been shown to interfere with collagen synthesis, fibrosis, and chronic scaring processes [5]. Triamcinolone inhibits the transcription of matrix protein genes, such as fibronectin and procollagen, and diminishes the synthesis of collagenase activity inhibitors, like alfa-2-macroglobulin [5]. Thus, corticosteroid injection prevents the cross-linking of collagen that results in scar contracture. If the scar is stretched by a dilator, the contracture will presumably not occur or will be reduced after intralesional triamcinolone injection. Reducing dilation sessions is of paramount significance, not only due to patient‘s quality of life or cost reductions, but, most importantly, to diminish the perforation risk. Since perforation rates are calculated per session and a perforation can occur with equal frequency notwithstanding whichever session is being performed [2], the reduction in the number of dilation sessions in order to mitigate patient’s dysphagia would be of great importance. Another criticism that could be raised against our study is the use of Savary-Gilliardbougies instead of balloon dilators. This is the first single-center prospective comparative study, which analyzed the use of steroid injections and dilation exclusively in naı¨ve postoperative strictures. Dilations were not only performed by experts, but also by gastroenterology fellows with less than a year of training in endoscopy. This means that our method of injecting next to or at the wall tear caused by the dilator, as bougie dilation itself, is widely safe and applicable. Balloon dilation is undoubtedly more ‘‘fashionable’’ nowadays. Theoretically, balloons would have a lesser perforation risk than bougies, since the latter dilators exert not only radial forces as they are passed, but also longitudinal forces as the result of a shearing effect. The former dilators, in contrast, deliver exclusively a radial force that is transmitted simultaneously over the entire length of the stricture rather than progressively from its proximal to distal extent. Nevertheless, an advantage, in terms of efficacy or safety of one method over the other, has not been shown in clinical studies. On the other hand, balloon dilators are much more expensive, since they are single-use [13]. In conclusion, this study showed that adding triamcinolone after every dilation session next to or at the borders of the lacerations caused by the dilators, results in a significant improvement or resolution of dysphagia.
Watte ([email protected]), Ce´sar Vivian Lopes (drcvlopes@ gmail.com) and Carlos Eduardo Oliveira dos Santos (ddendo@ uol.com.br) have no conflicts of interest or financial ties to disclose. Dr. Ju´lio C. Pereira-Lima ([email protected]) is on the speakers bureau of Takeda Pharmaceutical and receives honoraria as consultant of Boston Scientific. Dr. Cla´udio R. Teixeira ([email protected]) is consultant on the speakers bureau of Fujinon Endoscopy.
References 1. Pereira-Lima JC, Ramires RP, Zamin I Jr et al (1999) Endoscopic dilation of benign esophageal strictures: report on 1043 procedures. Am J Gastroenterol 94:1497 2. Raymondi R, Pereira-Lima JC, Alves AV et al (2008) Endoscopic dilation of benign esophageal strictures without fluoroscopy: experience of 2750 procedures. Hepatogastroenterology 55:1342–1348 3. Gonc¸alves C, Almeida N, Gomes D et al (2006) Injec¸a˜o Intralesional de Betametasona nas Estenoses Benignas do Esoˆfago. J Port Gastroenterol 13:22–25 4. Hirdes MM, van Hooft JE, Koornstra JJ et al (2013) Endoscopic corticosteroid injections do not reduce dysphagia after endoscopic dilatation therapy in patients with benign esophagogastric anastomotic strictures. Clin Gastroenterol Hepatol 11:795–801 5. Kochhar R, Poornachandra KS (2010) Intralesional steroid injection therapy in the management of resistant gastrointestinal strictures. World J Gastrointest Endosc 2:61–68 6. KochharR Makharia GK (2002) Usefulness of intralesional triamcinolone in treatment of benign esophageal strictures. Gastrointest Endosc 56:829–834 7. Wijkerslooth LR, Vleggaar FP, Siersema PD (2011) Endoscopic management of difficult or recurrent esophageal strictures. Am J Gastroenterol 106:2080–2091 8. Miyashita M, Onda M, Okawa K et al (1997) Endoscopic dexamethasone injection following balloon dilatation of anastomotic stricture after esophagogastrostomy. Am J Surg 174:442–444 9. Orive-Calzada A, Bernal-Martinez A, Navajas-Laboa M et al (2012) Efficacy of intralesional corticosteroid injection in endoscopic treatment of esophageal strictures. Surg Laparosc Percutan Tech 22:518–522 10. Altintas E, KacarS TuncB et al (2004) Intralesional steroid injection in benign esophageal strictures resistant to bougie dilation. J Gastroenterol Hepatol 19:1388–1391 11. Ramage JI Jr, Rumalla A, Baron TH et al (2005) A prospective, randomized, double-blind, placebo-controlled trial of endoscopic steroid injection therapy for recalcitrant esophageal peptic strictures. Am J Gastroenterol 100:2419–2425 12. Kochman ML, McClave SA, Boyce HW (2005) The refractory and the recurrent esophageal stricture: a definition. Gastrointest Endosc 62:474–475 13. Mo¨nkemu¨ller K, Kalauz M, Fry L (2010) Endoscopic dilation of benign and malignant esophageal strictures. In: Mo¨nkemu¨ller K, Wilcox CM, Mun˜oz-Navas M (eds.) Interventional and therapeutic gastrointestinal endoscopy. Front Gastrointest Res. Basel, Karger, vol 27; pp. 91–105
Disclosures Drs. Michele Lemos Bonotto (michele.lb@hotmail. com), Gustavo Dru¨gg Hahn ([email protected]), Guilherme
123
and Other Interventional Techniques
A prospective randomized trial of intralesional triamcinolone injections after endoscopic dilation for complex esophagogastric anastomotic strictures Steroid injection after endoscopic dilation Ju´lio C. Pereira-Lima • Michele Lemos Bonotto • Gustavo Dru¨gg Hahn • Guilherme Watte • Ce´sar Vivian Lopes Carlos Eduardo Oliveira dos Santos • Cla´udio R. Teixeira
•
Received: 10 April 2014 / Accepted: 23 July 2014 Ó Springer Science+Business Media New York 2014
Abstract Background and aims Postoperative esophageal strictures frequently recur. We assessed the efficacy of injecting triamcinolone after bougie dilation in comparison to dilation alone. Methods In a double-blind randomized fashion, 19 patients (68 % male with mean age of 53-years old) with non-dilated esophagogastric complex strictures after esophagectomy with gastric pull-up were assigned to receive dilation alone (control) or 40 mg of triamcinolone at the borders of the wall lacerations caused by the bougienage in each dilation session during the study follow-up. Dysphagia and complications were assessed at 1, 2, and 6 months. Primary end-point was to be dysphagia-free. Results After 1 month of the beginning of therapy, 4 patients in the steroid group were without dysphagia, in comparison to 0 patient in the control group (P = 0.021).
J. C. Pereira-Lima M. Lemos Bonotto G. D. Hahn (&) C. V. Lopes Department of Gastroenterology and Hepatology, Santa Casa Hospital/Federal University of Health Sciences of Porto Alegre (UFCSPA), Porto Alegre, Brazil e-mail: [email protected] J. C. Pereira-Lima e-mail: [email protected] J. C. Pereira-Lima C. R. Teixeira Rio Grande do Sul Foundation of Gastroenterology (FUGAST), Porto Alegre, Rio Grande do Sul, Brazil G. Watte The Post-Graduation Program in Chest Medicine Sciences, Santa Casa Hospital, Porto Alegre, Brazil C. E. O. dos Santos Santa Casa de Bage´ Hospital, Bage´, Brazil
Six months after endoscopic therapy, 62 % of the cases in the triamcinolone group versus none in the control group were dysphagia-free (P = 0.009). There was no perforation nor hemorrhage. Conclusions Injecting triamcinolone after every dilation session next to or at the borders of the lacerations caused by the dilators, results in a significant improvement or resolution of dysphagia. Keywords Benign esophageal strictures Postoperative Endoscopic dilation Triamcinolone Anastomotic strictures after esophagectomy with gastric pull-up are presently the main cause of benign esophageal strictures in the Western hemisphere, due to the widespread use of proton pump inhibitors (PPI) in the last two decades, which reduced the frequency of esophageal peptic strictures after treatment [1–4]. Endoscopic dilation with or without fluoroscopy is considered the initial treatment of choice for both simple and complex strictures [2, 5, 6]. Complex strictures (tortuous, longer than 2 cm, asymmetrical and/or with a luminal diameter less than 9 mm) are usually caused by caustic ingestion, radiation injury, aggressive endotherapy for Barrett’s esophagus or early esophageal cancer or are postoperative. These strictures are more difficult to dilate and are associated with a higher recurrence rate. Complex postoperative strictures usually need 4–8 dilations sittings either with conventional bougienage or balloon dilation technique in a 3–18 months follow-up [2, 5, 7]. Although additional endoscopic techniques combined with dilation such as steroid injections, topical mitomycinC application, incisional therapy, and stent placement are recommended for the management of refractory strictures,
123
Surg Endosc
few trials have addressed this issue in a randomized fashion. Moreover, postoperative stenosis frequently recurs and demands many dilation sessions, turning into refractory strictures [5, 7, 8]. In this study, we aimed to compare the ‘‘ab initio’’ use of triamcinolone injections in conjunction with endoscopic dilation with bougies with dilation alone in the treatment of complex anastomotic strictures.
Materials and methods Patients with dysphagia to solids (dysphagia grade 2) after esophagectomy with gastric pull-up and cervical handsewn esophagogastrostomy, with a confirmed anastomotic stricture that were not passable with a 0.98 cm endoscope were included. Exclusion criteria were suspicion of recurrent malignancy at the diagnosis, previous dilations or radiotherapy, active anastomotic leak, and refusal to enter in the protocol. All patients gave signed informed consent to the procedures and to the study before randomization. The study protocol was approved by the research Ethics Commission of our Institution. Endoscopic procedures The randomization was computer-generated by the www. researchrandomizer.org site. Randomization was performed by study personnel not involved in patients’ followup immediately before endoscopy. Patients were randomized to receive a total of 40 mg triamcinolone acetonide (Ophtaac 40, Laborato´rio Ophtalmos, Sa˜o Paulo, Brasil) injections around and at the borders of the tears caused by the bougie dilation after each therapeutic endoscopy session or to sham injections after bougie dilation. The endoscopists were not blinded to the therapy; however, the patients and the research fellows responsible for the follow-up were blinded to patient’s therapy. Upper endoscopies were performed under conscious sedation with midazolam and fentanyl by the Unit endoscopic staff or gastroenterology fellows, who were not involved in patient clinical care follow-up. Patients were progressively dilated (rule of 3 bougies of increasing size per session) at 14-day intervals until the 51F (17 mm) dilator was introduced. Savary-Gilliard bougies(WilsonCook Medical Inc., Winston Salem, USA) under guidance of a biliary guide-wire (Jagwire, Boston Scientific, Natick, USA) were used. At baseline, data on age, sex, indication for esophagectomy, administration of neoadjuvant or adjuvant
123
chemotherapy, dysphagia grade, and first inserted bougie size were recorded. Since the etiology of postoperative stenosis is ischemic, no PPIs were recommended. At baseline, 1, 2, and 6 months after the beginning of therapy, patients were interviewed by phone or personally by the research fellow, who asked standardized questions on dysphagia and complications. Dysphagia was graded as previously published: 0, taking normal diet, no dysphagia; 1, unable to swallow certain solid foods (e.g., grilled beef); 2, able to swallow only semisolid soft foods; 3, able to swallow just liquids; and 4, unable to swallow liquids in adequate amounts or difficulty in swallowing the own saliva [1]. In case recurrent dysphagia was reported during a follow-up contact or a patient visit to the Outpatient Unit of our hospital, an endoscopy with dilation according to the initial randomization was performed within a week during the 6-month follow-up period. The primary study end-point was the number of patients without dysphagia at follow-up. Secondary end-points were total number of dilation sessions per patient, complications and dysphagia grade at the follow-up. Statistical analysis Data were presented as mean ± SD or frequency and percentage. We performed associations between variables with the v2 tests (Pearson’s or Linear-by-Linear Association) and within-patient analysis with McNemar’s test. For comparing continuous variables, a Student t test or an unequal variance t test was used. Statistical significance was accepted at P \ 0.05. All statistical analyses were performed using the Statistical Package for the Social Sciences (PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc).
Results Between December, 2012, and February, 2014, 42 patients with postoperative esophageal stenosis were assessed for study eligibility. 5 patients were excluded due to prior dilation, 4 due to the fact that the stricture was passable with the endoscope, 7 due to concomitant radiotherapy, 6 to postoperative stenosis due to another surgical technique (esophagocoloplasty or esophagojejunostomy) and 1 case has not accepted to participate in the study. 19 patients with postoperative stricture due to cervical esophagogastrostomy who did not undergo previous radiotherapy or endoscopic dilation were randomized to receive triamcinolone (N = 10) or sham therapy (N = 9). Between study randomization and follow-up end (6 months), 1 patient
Surg Endosc Table 1 Clinical characteristics of the study baseline Characteristic
Total
Control
Triamcinolone
Male
13 (68)
4 (21)
9 (47)
0.06
Age (year)
54 ± 12
52 ± 15
56 ± 8
0.46
Dysphagia
P Value
0.06
2
13 (69)
8 (42)
5 (27)
3
4 (21)
1 (5)
3 (16)
4
2 (10)
–
2 (10)
Simple size = 19 patients. Data are presented as mean ± SD or No. (%). Dysphagia was graded as previously published: 1, unable to swallow certain solid foods (e.g. grilled beef); 2, able to swallow only semisolid soft foods; 3, able to swallow just liquids; 4, unable to swallow liquids in adequate amounts or difficulty in swallowing the own saliva [1]
and 2). There was no difference in dysphagia improvement when comparing the first and the second month visits. On the 6th month follow-up evaluation, 62 % of the patients who received steroid injections (5 out of 8 patients) were dysphagia-free versus 0 % (0 of 9 patients) of the control group (P = 0.009) (Fig. 2). In addition, 2 patients in the control group were with grade 3 dysphagia 6 months after randomization. The difference between groups was already statistically significant after 1 month of therapy (4 out of 10 patients in the steroid group—40 %— versus none in the dilation alone group, P = 0.021). The mean number of dilation sessions needed in the steroid group was 3 ± 2 versus 4 ± 2 in the dilation alone group (P = 0.388). Major complications
died and 1 patient presented with anastomotic cancer recurrence. Indication for esophagectomy in the randomized patients was squamous cell carcinoma in 18 and adenocarcinoma in 1. The mean time frame between the day of surgery and the day of the first endoscopic dilation was 76 (range 31–182) days in the steroid group and 68 days in the dilation alone group (range 29–205), (P = 0.6). Other patients’ characteristics at baseline are at Table 1. Response to treatment Both groups had improvement in dysphagia grade when comparing baseline with the 6-month follow-up visit (Fig. 1). However, not only the improvement, but also the abolishment of dysphagia, were significantly greater in the triamcinolone group in comparison to the controls (Figs. 1
Discussion In this study, the injection of intralesional triamcinolone in the wounds provoked by the shearing effect of the bougies in the dilated stenotic esophageal segment provided significantly more dysphagia-free patients in the follow-up, than performing dilation with Savary-Gilliard bougies alone. More patients were also significantly dysphagia-free in the first month, as well as in the end of the study followup (6 months). Although we have not addressed quality of life scores, to be dysphagia-free since the very beginning of the therapy, as well as to require less frequent dilation
12
*
10
8
Patient
Fig. 1 Comparison between the groups in relation to dysphagia in the baseline and after intervention (N = 19). *Triamcinolone vs Control, P \ .001. **Triamcinolone in the 6th month versus Triamcinolone in the baseline, P \ .001
There were no perforations, clinically overt bleeding or esophageal candidiasis in the studied population.
6
**
4
2
0 Baseline
1st month
Triamcinolone without dysphagia
2nd month
Triamcinolone with dysphagia
6th month
Control with dysphagia
123
Surg Endosc 6
P. 009
5
Patient
4 3 2 1 0 Triamcinolone
Control
Dysphagia 0
1
2
3
Fig. 2 Comparison between the groups in relation to dysphagia in the 6th month after the intervention (N = 16). Dysphagia was graded as previously published: 0, taking normal diet, no dysphagia; 1, unable to swallow certain solid foods (e.g., grilled beef); 2, able to swallow only semisolid soft foods; 3, able to swallow just liquids [1]. Linearby-Linear Association v2 test
sessions, certainly means a high improvement in someone’s quality of life. Our study has important strengths: (1) the adequate method of steroid injection—after performing dilation and aiming at the zone that will be contracted by the scaring of the wound, instead of injecting randomly before the bougienage; (2) the enrollment of an uniform sample of only true complex strictures-all postoperative, all not passable with the endoscope and all treatment naı¨ve; (3) the intense follow-up for 6 months, a time frame, which after that, dysphagia-free dilated patients will probably not have dysphagia recurrence. On the other hand, the major drawback of our study is the small sample size. However, we decided to interrupt the trial due to the notable difference in treatment response between the two study arms. During the past decade, there has been growing interest in the use of steroid injections for the treatment of refractory benign esophageal stenosis. Kochar et al. [6] retrospectively analyzed 71 stricture cases (19 of which with postoperative strictures) that were being managed with a program of intermittent endoscopic dilation using polyvinyl bougies, it was observed that after triamcinolone injections, the number of monthly dilations needed to keep the patients without dysphagia dropped from 1.24 to 0.5 after the introduction of steroid injections. The decrease in this dilation index was significantly observed in benign strictures from all etiologies. Different steroids such as betamethasone or dexamethasone have been used intralesionally along with endoscopic
123
dilation in other retrospective series with encouraging results [3, 9]. The technique of dilation (either bougie or balloon) or injections (before or after dilation), as well as the performance of injections varies from only in the first session to as many sessions as the number of dilations [4, 5, 9–11]. In one randomized comparative study involving 21 patients with strictures of different causes and variable number of previous dilation sittings, the additional use of triamcinolone (10 mg per injection, 4 injections, 1 in each quadrant) after esophageal bougie dilation resulted in an increase in the dysphagia-free period, a decrease in the periodic dilation index and in the mean number of dilation sessions per patient [10]. A decrease in the need for repeated balloon dilation and in the average time for new balloon dilation in patients with peptic strictures submitted to triamcinolone injections compared to sham injections was observed in a randomized comparative trial involving 30 patients [11]. These authors included stenosis which were passable by the endoscope and patients with the socalled ‘‘recalcitrant’’ strictures. The investigators considered recalcitrant (a synonym for refractory), strictures that recurred after only one dilation session instead of five, as usually considered in the literature [12]. On the other hand, the study was blinded and the groups were stratified to dysphagia grade before treatment, presence of esophagitis and prior therapies [11]. Finally, a recently published randomized trial from Netherlands, which compared triamcinolone injections before bougie dilation with dilation alone, is the only clearly negative study in the literature [4]. In this Dutch study, involving 60 patients, 45 % of the patients in the steroid group were dysphagia-free at 6-month versus 36 % of the controls. Median time to repeat dilations was 108 days for the treated group and 42 days for the controls (NS). Candida esophagitis developed in four cases of the corticosteroid group, but in none of the controls. We have not observed any case of candidiasis in the remnant esophagus, and this minor side effect was also not reported in other studies [3, 6, 10, 11]. The investigators performed 10 mg injections of triamcinolone in each quadrant of the stenotic segment before the passage of the dilator. Since bougie dilators provoke lacerations in one or two parts of the stenotic segment (usually in just one), 2 or 3 of the triamcinolone injections were lost with this technique. Our approach was to specifically target the injections next to or at the border of the wall lacerations, in order to reduce the inflammatory and recurrent cicatrix response at the anastomotic stricture. This way, the dilated esophageal segment would contract less and the lumen would remain wider. During gastric pull-up, the left epiploic artery is ligated, which makes the cervical esophagogastrostomy susceptible to ischemia. Although the exact mechanism is unknown,
Surg Endosc
scaring resulting from chronic ischemic inflammation causes the anastomotic stenosis. Intralesional steroids have been shown to interfere with collagen synthesis, fibrosis, and chronic scaring processes [5]. Triamcinolone inhibits the transcription of matrix protein genes, such as fibronectin and procollagen, and diminishes the synthesis of collagenase activity inhibitors, like alfa-2-macroglobulin [5]. Thus, corticosteroid injection prevents the cross-linking of collagen that results in scar contracture. If the scar is stretched by a dilator, the contracture will presumably not occur or will be reduced after intralesional triamcinolone injection. Reducing dilation sessions is of paramount significance, not only due to patient‘s quality of life or cost reductions, but, most importantly, to diminish the perforation risk. Since perforation rates are calculated per session and a perforation can occur with equal frequency notwithstanding whichever session is being performed [2], the reduction in the number of dilation sessions in order to mitigate patient’s dysphagia would be of great importance. Another criticism that could be raised against our study is the use of Savary-Gilliardbougies instead of balloon dilators. This is the first single-center prospective comparative study, which analyzed the use of steroid injections and dilation exclusively in naı¨ve postoperative strictures. Dilations were not only performed by experts, but also by gastroenterology fellows with less than a year of training in endoscopy. This means that our method of injecting next to or at the wall tear caused by the dilator, as bougie dilation itself, is widely safe and applicable. Balloon dilation is undoubtedly more ‘‘fashionable’’ nowadays. Theoretically, balloons would have a lesser perforation risk than bougies, since the latter dilators exert not only radial forces as they are passed, but also longitudinal forces as the result of a shearing effect. The former dilators, in contrast, deliver exclusively a radial force that is transmitted simultaneously over the entire length of the stricture rather than progressively from its proximal to distal extent. Nevertheless, an advantage, in terms of efficacy or safety of one method over the other, has not been shown in clinical studies. On the other hand, balloon dilators are much more expensive, since they are single-use [13]. In conclusion, this study showed that adding triamcinolone after every dilation session next to or at the borders of the lacerations caused by the dilators, results in a significant improvement or resolution of dysphagia.
Watte ([email protected]), Ce´sar Vivian Lopes (drcvlopes@ gmail.com) and Carlos Eduardo Oliveira dos Santos (ddendo@ uol.com.br) have no conflicts of interest or financial ties to disclose. Dr. Ju´lio C. Pereira-Lima ([email protected]) is on the speakers bureau of Takeda Pharmaceutical and receives honoraria as consultant of Boston Scientific. Dr. Cla´udio R. Teixeira ([email protected]) is consultant on the speakers bureau of Fujinon Endoscopy.
References 1. Pereira-Lima JC, Ramires RP, Zamin I Jr et al (1999) Endoscopic dilation of benign esophageal strictures: report on 1043 procedures. Am J Gastroenterol 94:1497 2. Raymondi R, Pereira-Lima JC, Alves AV et al (2008) Endoscopic dilation of benign esophageal strictures without fluoroscopy: experience of 2750 procedures. Hepatogastroenterology 55:1342–1348 3. Gonc¸alves C, Almeida N, Gomes D et al (2006) Injec¸a˜o Intralesional de Betametasona nas Estenoses Benignas do Esoˆfago. J Port Gastroenterol 13:22–25 4. Hirdes MM, van Hooft JE, Koornstra JJ et al (2013) Endoscopic corticosteroid injections do not reduce dysphagia after endoscopic dilatation therapy in patients with benign esophagogastric anastomotic strictures. Clin Gastroenterol Hepatol 11:795–801 5. Kochhar R, Poornachandra KS (2010) Intralesional steroid injection therapy in the management of resistant gastrointestinal strictures. World J Gastrointest Endosc 2:61–68 6. KochharR Makharia GK (2002) Usefulness of intralesional triamcinolone in treatment of benign esophageal strictures. Gastrointest Endosc 56:829–834 7. Wijkerslooth LR, Vleggaar FP, Siersema PD (2011) Endoscopic management of difficult or recurrent esophageal strictures. Am J Gastroenterol 106:2080–2091 8. Miyashita M, Onda M, Okawa K et al (1997) Endoscopic dexamethasone injection following balloon dilatation of anastomotic stricture after esophagogastrostomy. Am J Surg 174:442–444 9. Orive-Calzada A, Bernal-Martinez A, Navajas-Laboa M et al (2012) Efficacy of intralesional corticosteroid injection in endoscopic treatment of esophageal strictures. Surg Laparosc Percutan Tech 22:518–522 10. Altintas E, KacarS TuncB et al (2004) Intralesional steroid injection in benign esophageal strictures resistant to bougie dilation. J Gastroenterol Hepatol 19:1388–1391 11. Ramage JI Jr, Rumalla A, Baron TH et al (2005) A prospective, randomized, double-blind, placebo-controlled trial of endoscopic steroid injection therapy for recalcitrant esophageal peptic strictures. Am J Gastroenterol 100:2419–2425 12. Kochman ML, McClave SA, Boyce HW (2005) The refractory and the recurrent esophageal stricture: a definition. Gastrointest Endosc 62:474–475 13. Mo¨nkemu¨ller K, Kalauz M, Fry L (2010) Endoscopic dilation of benign and malignant esophageal strictures. In: Mo¨nkemu¨ller K, Wilcox CM, Mun˜oz-Navas M (eds.) Interventional and therapeutic gastrointestinal endoscopy. Front Gastrointest Res. Basel, Karger, vol 27; pp. 91–105
Disclosures Drs. Michele Lemos Bonotto (michele.lb@hotmail. com), Gustavo Dru¨gg Hahn ([email protected]), Guilherme
123
