Fibrin sealant for esophageal anastomosis: A phase II study

Yao-Bin Lin, Jian-Hua Fu, Yan Huang, Yi-Huai Hu, Kong-Jia Luo, Ke-Xi Wang, Amos éla Bella,Dong-Rong Situ, Ji-Yang Chen, Ting Lin, Xavier B D’Journo, Nuria M Novoa, Alessandro Brunelli,Hiran C Fernando, Robert J Cerfolio, Mahmoud Ismail, Hong Yang,the AME Thoracic Surgery Collaborative Group

Abstract BACKGROUND Esophagectomy is a pivotal curative modality for localized esophageal or esophagogastric junction cancer (EC or EJC). Postoperative anastomotic leakage(AL) remains problematic. The use of fibrin sealant (FS) may improve the strength of esophageal anastomosis and reduce the incidence of AL.AIM To assess the efficacy and safety of applying FS to prevent AL in patients with EC or EJC.METHODS In this single-arm, phase II trial (Clinicaltrial.gov identifier: NCT03529266), we recruited patients aged 18-80 years with resectable EC or EJC clinically staged as T1-4aN0-3M0. An open or minimally invasive McKeown esophagectomy was performed with a circular stapled anastomosis. After performing the anastomosis, 2.5 mL of porcine FS was applied circumferentially. The primary endpoint was the proportion of patients with AL within 3 mo.RESULTS From June 4, 2018, to December 29, 2018, 57 patients were enrolled. At the data cutoff date (June 30, 2019), three (5.3%) of the 57 patients had developed AL,including two (3.5%) with esophagogastric AL and one (1.8%) with gastric fistula.The incidence of anastomotic stricture and other major postoperative complications was 1.8% and 17.5%, respectively. The median time needed to resume oral feeding after operation was 8 d (Interquartile range: 7.0-9.0 d). No adverse events related to FS were recorded. No deaths occurred within 90 d after surgery.CONCLUSION Perioperative sealing with porcine FS appears safe and may prevent AL after esophagectomy in patients with resectable EC or EJC. Further phase III studies are warranted.

Key words: Esophageal cancer; McKeown esophagectomy; Fibrin sealant; Anastomotic leakage; Postoperative complications; Prevention

INTRODUCTION

Esophageal cancer (EC) is the seventh most common malignancy worldwide[1]. In China, approximately 477900 new diagnoses and 375000 deaths in 2015 were estimated to be due to EC, accounting for more than half of the global morbidity and mortality[1,2]. Currently, surgery remains a routine component of treatment for esophageal or esophagogastric junction cancer (EJC). McKeown esophagectomy,consisting of a cervical anastomosis, is one of the most commonly used procedures, as it provides the potential advantages of adequate resection and extensive lymphadenectomy[3]. However, compared with intrathoracic anastomosis after esophagectomy, the cervical anastomosis is associated with a higher risk of postoperative complications, especially anastomotic leakage (AL)[4,5].

AL is one of the most severe postoperative complications and results in a prolonged hospital stay, increased cost, and considerable morbidity and mortality[5,6].The incidence of esophageal AL remains high at a range of 10.6%-26.1%worldwide[7-10]. A variety of anastomotic methods have been developed to prevent AL after esophagectomy, including the application of fibrin sealant (FS) to the anastomosis. FS could strengthen the anastomosis and promote anastomotic healing by enabling polymerization and forming tight approximation of the anastomosis[11,12].However, the results from clinical trials exploring the use of FS as an anastomotic sealant in esophageal surgery have been inconsistent[13-17]. To date, only two pilot studies have been conducted to investigate the use of FS in routine esophagectomy for EC[15,16]. The current phase II trial enrolled patients undergoing McKeown surgery for EC or EJC, and aimed to evaluate the efficacy and safety of applying porcine FS to prevent AL.

MATERIALS AND METHODS

Study design

This study was a single-center, single-arm, open-label, phase II clinical trial conducted at Sun Yat-sen University Cancer Center in Guangzhou, China. The study obtained approval from the ethics committee of Sun Yat-sen University Cancer Center and was conducted according to the provisions of the Declaration of Helsinki and Good Clinical Practice guidelines. All included patients provided written informed consent.

This trial is registered at ClinicalTrials.gov, number NCT03529266.

Participants

Eligible patients were aged from 18 to 80 years with histologically proven squamous cell carcinoma or adenocarcinoma of the thoracic esophagus or esophagogastric junction staged as T1-4aN0-3M0 (according to the International Union Against Cancer and American Joint Committee on Cancer Staging System for Esophagus and Esophagogastric Junction, eighth edition). An Eastern Cooperative Oncology Group performance status of 0 to 1, normal bone marrow function, and adequate hepatic and renal function were also required. The exclusion criteria included salvage esophagectomy after failed definitive chemoradiotherapy; a history of diabetes spanning more than a decade and uncontrolled blood glucose level; allergy to any ingredients of FS.

Treatment and procedures

All patients underwent the following workup for diagnosis and pretreatment staging:Contrast esophagography; plain and enhanced computed tomography (CT) of the chest and abdomen; cervical ultrasonography; esophagogastroduodenoscopy (EGD)with endoscopic ultrasound; electronic bronchoscopy or endobronchial ultrasound to confirm involvement of the trachea and/or bronchus, if indicated; positron emission tomography–CT to rule out distant metastasis, if indicated.

An open or minimally invasive radical resection of the primary tumor was performed by one surgical team led by one of the authors (Yang H) using the McKeown esophagectomy, combined with a two-field lymphadenectomy. The dissected lymph nodes included bilateral recurrent laryngeal nerve nodes, left lower paratracheal nodes, upper/middle/lower thoracic paraesophageal nodes, subcarinal nodes, posterior mediastinal nodes, paracardiac nodes, lesser curvature nodes, left gastric nodes, common hepatic nodes, splenic nodes, and celiac nodes. After esophageal resection, anastomosis was performed in an end-to-side fashion between the esophagus and tubularized stomach or colonviaa neck incision using a circular stapler (EEA 21 or 25, Covidien, USA), with the staple line inverted by a whole-layer suture (3-0 Vicryl, Ethicon, USA). The blind end of the gastric conduit or colon was closed using a 60-mm linear stapler (Endo GIA, Covidien, USA; or Echelon Flex,Ethicon, USA) and oversewn with a seromuscular-layer suture (3-0 Vicryl, Ethicon,USA).

After complete irrigation of the surgical field, 2.5 ml of FS [Bioseal?, Guangzhou Bioseal Biotech—Subsidiary of Johnson & Johnson (China) Investment Ltd, China]was applied to the anastomosis circumferentially. A nasogastric tube, a subcutaneous drainage tube in the neck, and two chest tubes, as well as a feeding tube in the jejunum by jejunostomy, were placed intraoperatively. Details of the anastomotic technique with intraoperative application of FS to the cervical anastomosis are shown in Figure 1.

Figure 1 Illustration of the circular stapled anastomosis with application of fibrin sealant to esophagogastric anastomosis after esophageal resection. A:The stapler head was placed into cervical esophagus stump and tied carefully; B: Incision of the gastric wall and insert of the stapler; C: Esophagogastric end-to-side anastomosis with a circular stapler; D: Closure of the fundus of gastric conduit with a linear stapler; E: Anastomosis and gastric stump after circular stapling; F: The staple line was inverted with a whole-layer suture; G: Fibrin sealant was applied to the esophagogastric anastomosis circumferentially; H: Completion of the application of fibrin sealant to cervical anastomosis.

After surgery, all patients were treated in the intensive care unit on the day of operation and re-admitted to the general ward the next day for routine postoperative rehabilitation. The nasogastric tube was removed one day later if there was no occurrence of anastomotic hemorrhage. Evaluations of AL were routinely performed by contrast esophagography 7 d postoperatively, unless the examination was clinically contraindicated, such as for patients who were intubated due to respiratory failure. Adequate nutrition through enteral or parenteral nutritional support was ensured before oral intake. Modified oral feeding with recommended diet was allowed after confirming the absence of AL.

Outcome measurement

The primary endpoint was the proportion of patients with AL within the first 3 mo after operation. Patients with saliva, gastrointestinal content, or drainage stained by oral methylene blue dye collected from the neck or chest tubes were suspected to have AL, and then EGD or contrast esophagography was performed for further confirmation[18]. Diagnosis of AL was mainly established by the independent image committee depending on the visualization of total parietal defects involving the esophagus, anastomosis, staple line, or gastric tube during EGD carried out by an experienced endoscopist, or extravasation of barium during contrast esophagography[19].

The secondary endpoints were time to first oral feeding, postoperative morbidity rate, especially the postoperative anastomotic stricture rate, and postoperative mortality rate. We defined grades I and II complications as minor complications, and grades III and IV as major complications, according to the Clavien-Dindo classification of surgical complications[20].

Follow-up

Routine follow-up incl uded outpatient clinic visits once every 3 mo in the first year and then every 6 mo thereafter until the end of the study or death. The follow-up assessments performed at every visit were as follows: Physical examinations;laboratory tests including hematology, chemistry, and tumor markers; chest radiography; contrast swallow study; and ultrasound scans of the neck and abdomen.Patients were also required to undergo enhanced neck/chest/abdomen CT scans and EGD once every year.

Statistical analysis

The study was designed based on Simon’s optimal two-stage method with a onesided α of 0.05 and power of 80%. An anastomotic leakage rate of 5% or less was expected as evidence for the efficacy of FS application (alternative hypothesis),whereas a rate of 15% or more indicated by previous studies[9,10]was considered insufficient for the assessment to continue (null hypothesis). According to these assumptions, 39 evaluable patients had to be enrolled in the first stage, of whom at least 35 were required to be free from AL. An additional 18 patients would be included in the second stage, leading to a total of 57 patients for the final analyses.Overall, the advantage of FS would be sufficiently proven if 52 or more patients were protected from AL.

We performed analyses based on the intention-to-treat population composed of all enrolled patients. Frequency and percentage were presented for the description of postoperative AL, along with other categorical variables. Medians and interquartile ranges (IQRs), or mean and SD were calculated for the description of continuous variables. Time-to-event data were summarized by Kaplan-Meier methods. We did statistical analyses using SPSS 22.0 (IBM Corporation, USA).

RESULTS

Patient characteristics

Between June 4, 2018, and December 29, 2018, 64 patients were screened, of whom 57 were enrolled in the study and underwent McKeown esophagectomy with an intraoperative application of FS at the Sun Yat-sen University Cancer Center (Figure 2). The median duration of follow-up at the time of data analysis (data cutoff on June 30, 2019) was 10.2 mo (IQR: 8.3-11.5 mo).

Baseline characteristics are summarized in Table 1. The median age was 61 years(range 41-75). Forty-nine (86.0%) of the 57 patients were men. Most tumors were located in the thoracic esophagus [in 47 (82.5%) of 57 patients], while five (8.8%)patients were diagnosed with EJC, and the other five (8.8%) with multiple primary esophageal carcinomas. Most patients [55 of 57 (96.5%)] had squamous cell carcinoma,while two (3.5%) had adenocarcinoma. Of the 57 patients undergoing surgery, 22(38.6%) received neoadjuvant therapy, including five (8.8%) treated with preoperative chemotherapy and 17 (29.8%) with preoperative chemoradiotherapy. The remaining 35 (61.4%) patients underwent surgery alone.

Surgery

Figure 2 Trial profile.

A total of 56 (98.2%) patients received combined thoracoscopic and laparoscopic esophagectomy (McKeown’s procedure) with esophageal reconstruction using the gastric tube, and one (1.8%) underwent McKeown esophagectomy with colonic interposition by thoracoscopy plus open laparotomy. Most patients [56 (98.2%)]received R0 resection, while one (1.8%) underwent R2 resection because of tumor invasion of the thoracic aorta. The average (SD) operating time was 240.0 (± 35.9) min.Blood loss was generally minimal with a mean of 78.9 (± 44.3) mL. An average of 32.7(± 11.6) lymph nodes were dissected. With respect to the distribution of pathologic stage grouping, pathological complete response was achieved in eight (36.4%) of the 22 patients receiving neoadjuvant therapy, and more details are shown in Table 2.

Outcomes

Three (5.3%) of the 57 patients experienced AL within the first 3 mo after operation(Table 3). Among them, one (1.8%) patient was diagnosed with gastric fistula confirmed by EGD 8 d after surgery (Supplementary Figure 1), and recovered within 18 d with conservative treatment. The other two (3.5%) were suspected of having AL with visible loss of saliva through the cervical wound after oral intake for a few days,and were finally confirmed with esophagogastric AL. In the first case, AL was detected and confirmed by EGD 10 d after surgery, while the extravasation of watersoluble contrast during a swallow study 21 d after surgery confirmed AL in the second. Both patients recovered after conservative treatment within 28 and 14 d,respectively. The remaining 54 patients were confirmed to be free from AL by contrast esophagography or EGD, and resumed oral intake without abnormal symptoms or signs. At the data cutoff, no more patients were diagnosed with AL. Additionally,none of the three patients with esophagogastric AL or gastric fistula had received neoadjuvant treatment before surgery.

Postoperative complication rate was 49.1% (28 of 57). The incidence of major and minor complications was 17.5% and 31.6% (10 and 18 of 57), respectively (Supplementary Table 1 and Table 2). The most prevalent complications were arrhythmia [11(19.3%)] and recurrent nerve injury [10 (17.5%)] (Table 3). One (1.8%) patient was diagnosed with mild anastomotic stricture through EGD 55 d after surgery, and recovered after taking endoscopic dilatation. No adverse events related to FS were recorded, and no deaths occurred within 90 d after surgery.

The median duration of postoperative nasogastric tube placement was 1 d (IQR:1.0-2.0) (Figure 3), and the median time to resume oral feeding after operation was 8 d(IQR: 7.0-9.0). The majority of patients began oral intake on postoperative day 7 or 8 routinely after confirming the absence of anastomotic leakage by contrast esophagography. However, 17 (29.8%) of the 57 patients delayed undergoing contrast esophagography and postponed oral feeding for different reasons. Among them, five(8.8%) patients delayed oral feeding because the testing facilities stopped working on weekends, while the remaining 12 were diagnosed with postoperative contraindications to oral feeding, including six with thoracic lymphorrhea cured by medical treatment, four with respiratory failure, one with gastric fistula, and one with chylothorax cured through surgical treatment. Furthermore, the median postoperative hospital stay duration was 11 d (IQR: 9.0-12.0).

Table 1 Baseline patient characteristics, n (%)

At the data cutoff point, eight (14.3%) of 56 patients who achieved R0 resection had tumor recurrence or had died, among whom two were found with supraclavicular lymph node metastases, two with mediastinal lymph node metastases, and four with distant metastases. Of the 57 patients who underwent resection, one (1.8%) died as a result of disease progression 189 d after surgery. The disease-free survival rate in patients with R0 resection was 79.5% at one year. The 1-year overall survival rate of the whole group was 98.2%. At present, both the median disease-free survival and overall survival have not yet been reached.

Table 2 Distribution of pathologic stage groups after surgery, n (%)

DISCUSSION

To our knowledge, this is the first phase II clinical study to assess the efficacy of FS in preventing AL post-esophagectomy in patients with resectable EC or EJC. Only three(5.3%) patients developed AL within the first 3 mo after operation. The incidences of anastomotic stricture and other major postoperative complications were 1.8% and 17.5%, respectively. No death occurred within 90 d after surgery. These results indicated that the intraoperative application of FS could probably prevent AL effectively with satisfactory safety.

This study was carried out based on the results of previous clinical studies regarding the prophylactic role of tissue adhesives in preventing AL in esophageal surgery[13-16]. In those studies, the intraoperative use of FS after esophagectomy appeared safe and feasible[13-17]. Thus, we carried out this phase II clinical trial in patients with resectable EC or EJC to further assess the efficacy of FS. We have previously reported an AL rate of 14.9% prior to using FS in our institution[10]. The current study was designed to detect a 10% reduction in the incidence of postoperative AL in favor of the use of FS after esophagectomy, as compared with the expected AL rate based on our previous study (5%vs15%)[10]. The final results showed that the rate of postoperative AL was further reduced in the current study, and the outcome was better than that of previously reported series with esophagectomy alone(10.6%-26.1%)[7-10].

A variety of modified anastomotic techniques for esophagectomy have been described with the intention of minimizing the risk of postoperative AL[21-25].However, the results are inconsistent, and the optimal technique remains a matter of debate[21-25]. Sunet al[23]reported and compared a novel embedded three-layer cervical esophagogastric anastomosis with traditional cervical anastomosis with respect to the incidence of postoperative complications in 339 EC patients. The rate of esophageal AL was significantly lower with the newly reported anastomotic technique [2.4% (4 of 166)] than with the conventional two-layer anastomotic technique [7.5% (13 of 173),P= 0.031]. Chenet al[22]reported that omentoplasty might contribute to the reduction of AL after esophagectomy. The researchers conducted a meta-analysis of three randomized controlled trials with a total of 633 patients. A significantly lower rate of postoperative AL was shown in the omentoplasty group (OR = 0.26, 95%CI: 0.14-0.52,P< 0.0001) than in the non-omentoplasty group. However, large variations were present[21-25]when using modified anastomotic techniques, since the surgical procedures were complicated and technically demanding, and were therefore difficult to replicate in other centers. In the current study, FS was used to strengthen the esophagogastric or esophago-colic anastomosis in patients with EC or EJC who received standard McKeown esophagectomy with a circular stapled anastomosis.Compared with other modified techniques, the application of FS on the anastomosis was easily administrable, resulting in less variation in technique between surgeons with little added operative time.

There have been conflicting results from previous studies regarding the role of FS for both the prevention and treatment of AL[11,12,17,26-28], especially in esophageal surgery. To date, four clinical studies have reported the role of tissue adhesives in sealing esophageal anastomosis[13-16]. Upadhyayaet al[13]conducted a prospective randomized controlled trial enrolling 45 infants who underwent esophagectomy forcongenital esophageal atresia with tracheoesophageal fistula. FS (Tisseel?) was used as reinforcement on a primary end-to-end esophageal anastomosis in the study group.The AL rate was only 9.1% (2/22), compared to 43% (10/23) in the control group (P=0.017). Saldana-Corteset al[14]investigated the use of FS (Quixil?) in children after caustic esophageal injury. A colon interposition was used as the reconstruction substitute for the esophagus. Postoperative cervical anastomotic dehiscence and leakage were observed in 28.5% of patients who received FS (study group) and 50% in the control group, but the difference was not significant (P= 0.17). However, these studies were carried out in children with esophageal atresia or caustic esophageal injury. The benefit of FS as an anastomotic sealant to prevent AL after esophagectomy for EC has been questioned. So, in the present study, patients with EC or EJC were recruited, which differed from the design of Upadhyaya’s and Saldana-Cortes’s studies, in which patients with rare esophageal diseases were enrolled. On that point,only two feasibility studies have reported the use of FS to prevent AL in esophageal surgery for EC[15,16]. Haverkampet al[15]published a pilot study to evaluate the feasibility of applying fibrin coated collagen patches (Tachosil?) to esophageal anastomosis. A total of 11 patches were successfully applied to the cervical esophagogastric anastomoses. AL occurred in two (18.2%) out of 11 patients. It was concluded that the application of Tachosil?to esophageal anastomosis was technically feasible. Platet al[16]also reported a similar feasibility study focusing on the application of autologous fibrin sealant to esophageal anastomoses. That study enrolled 15 patients with EC and reached a similar conclusion. Thus, the application of FS to esophageal anastomosis for EC is safe and technically feasible according to these previous studies. Despite this, both were pilot studies, and the sample size was relatively small. Moreover, the protective effects of FS in the prevention of AL after esophagectomy still needed to be verified. In this phase II clinical trial, we prospectively enrolled 57 patients, all of whom received standard McKeown esophagectomy and perioperative sealing with FS, to investigate the use of FS as an anastomotic sealant in patients with EC or EJC. The rate of AL was significantly decreased compared with that previously reported[13-16]. Thus, safety and efficacy were favorable.

Table 3 Postoperative complications, n (%)

Figure 3 Postoperative nasogastric tube placement time.

There were several factors that might have contributed to the promoted anastomotic healing due to FS observed in this study. First, FS mainly consists of thrombin and fibrinogen, which form a stable fibrin clot by stimulating the final pathway of the coagulation cascade. The fibrin clot will form a gel-like material on the surface of surgical wounds, which can not only directly block the tissue defect, but also promote wound healing by using the fibrin network as the matrix where fibroblasts and capillary endothelial cells can proliferate to form granulation[11,29].Second, FS could provide additional benefits for anastomotic healing by increasing the maturation process of granulation tissue during the first few days following surgery[12,30]. Indeed, in Holmer’s study[30], higher mRNA levels of collagen types I and III, along with augmentation of granulation, were found in a rat model of anastomosis sutured with collagen fleece. Furthermore, since AL usually results from a preceding microleakage, a watertight esophageal anastomosis may be achieved by additional sealing with FS, leading to a lower occurrence of AL after esophageal surgery[14,17].

This study had some limitations. Patients with poorer performance status or a long history of diabetes and older patients were excluded, so the applicability of this modified technique in these patients requires additional study. Moreover, most tumors included in our study were squamous cell carcinoma of the thoracic esophagus, which differs from the histological prevalence of EC in Western countries.

Of note, this trial is a single-arm study that compared the results against the AL rates from large population studies. Hence, the generalizability of our results should be considered with caution. A randomized, phase III clinical trial comparing the application of FS after esophagectomy with esophageal surgery without sealant application in patients with resectable EC or EJC is underway (NCT03847857). In addition, due to the large variability of FS regimens, including in physicochemical characteristics, the strength of adhesive joint,etc., new types of adhesives with special properties that are especially suitable for esophageal anastomotic sealing need to be synthesized and investigated in the future.

In conclusion, our study showed that perioperative sealing with FS appears to be safe and may prevent AL after McKeown esophagectomy in patients with resectable EC or EJC. Further phase III trials are warranted to confirm our findings.

ARTICLE HIGHLIGHTS

Research background

Esophagectomy is the primary treatment for localized esophageal or esophagogastric junction cancer (EC or EJC). Postoperative anastomotic leakage (AL) remains problematic. The use of fibrin sealant (FS) may improve the strength of esophageal anastomosis and reduce the incidence of AL.

Research motivation

Previous pilot studies showed that the application of FS to esophageal anastomosis for EC is safe and technically feasible.

Research objectives

This study aimed to assess the efficacy and safety of applying FS to prevent AL in patients with EC or EJC.

Research methods

In this phase II study, the efficacy and safety of FS application were evaluated in 57 patients with resectable EC or EJC undergoing McKeown esophagectomy.

Research results

All patients received McKeown esophagectomy with intraoperative application of FS. Three(5.3%) of the 57 patients developed AL within the first 3 mo after operation, including two (3.5%)with esophagogastric AL and one (1.8%) with gastric fistula. The use of FS was associated with a low incidence of postoperative morbidity and mortality, and led to an improved outcome with rapid recovery and shortened hospital stay duration.

Research conclusions

Perioperative sealing with porcine FS appears safe and may prevent AL after esophagectomy in patients with resectable EC or EJC.

Research perspectives

The favorable benefits of fibrin sealant support further investigation in phase III trials and implementation in clinical practice.

ACKNOWLEDGEMENTS

The authors thank all the patients and their families who participated in this study.The authors thank Yi Hu, Guang-Yu Luo, and Jing-Xian Shen from the Independent Data Monitoring Committee, and Xiao-Dong Su, Ying Guo, and Zhi-Qiang Wang,who compose the Independent Image Committee, for their support. The authors thank Ji-Bin Li for his help with the statistical analysis and interpretation.