clinical application of mucosal valve technique for anastomosis during esophagogastrostomy

ORIGINAL ARTICLE

Clinical Application of Mucosal Valve Techniquefor Anastomosis During Esophagogastrostomy

Bin Li & Yu-Min Li & Jian-Hua Zhang & Yun-Feng Su &

Cheng Wang & Zhi-Qiang Wang & Yun-Jiu Gou &

Tie-Niu Song & Jian-Bao Yang

Received: 20 July 2013 /Accepted: 1 October 2013 /Published online: 18 October 2013# 2013 The Society for Surgery of the Alimentary Tract

AbstractBackground The study aims to compare the efficacy in prevention of anastomotic complications using layer-to-layer mucosalvalve technique versus circular stapled technique for esophagogastric intrathoracic anastomosis after resection for esophageal andgastric cardiac carcinoma.Methods From January 2005 to December 2010, 136 patients received layer-to-layer mucosal valve technique (LM group), 219received circular stapled anastomosis (CS group) after curative intent resection for esophageal and gastric cardiac carcinoma. Thetechnique details were reported and the clinical results were analyzed.Results The two groups were comparable on clinical baseline characteristics. The average duration of operation was longer withLM technique by 16 min, but without statistical significance (P=0.073). There was no anastomotic leakage in the LM group,while in the CS group, leakage occurred in seven patients (3.2 %, P=0.047). Both the incidence and grade of postoperativedysphagia were significantly lower in the LM group (P <0.05). Significantly fewer patients experienced stricture after LMtechnique (3.8%) compared with CS anastomosis (18.2 %, P <0.001). CS anastomosis was associated with a significantly higherincidence of persistent stricture requiring more dilatation (P <0.001). Symptoms of reflux were better controlled by LMtechnique; 82.7 % of patients were asymptomatic with respect to reflux compared to 58.9 % in the CS group, P <0.001. Andthere was a significant reduction in the incidence of esophagitis in remnant esophagus in the LM group (P=0.001).Conclusions The layered mucosal valve anastomosis could significantly diminish the incidence of anastomotic complicationsand could be used as an alternative for esophagogastric anastomosis after resection of esophageal and gastric cardiac carcinoma.

Keywords Esophageal carcinoma . Esophagogastrostomy .

Hand-sewn anastomosis . Circular stapler . ComplicationIntroduction

Surgical resection remains the mainstay of treatment for esoph-ageal and gastric cardiac carcinoma worldwide, whetherperformed for curative intent or for palliation of dysphagia.1–3

Reconstruction of the upper alimentary tract is predominantlydone using the stomach and accomplished by means of hand-sewn or mechanical esophagogastric anastomosis. Anastomoticcomplications, such as leakage, stricture, and reflux, continue tobe the main causes of postoperative morbidity, mortality, andpoor quality of life.4

–10 Various anastomotic techniques havebeen employed in an effort to reduce the complications, butthere is insufficient evidence to recommend one technique overthe others.11

–15 This warrants further research on how to improvethe anastomotic healing process and accordingly improve

B. Li :Y.<M. Li (*)Gansu Provincial Key Laboratory of Digestive System Tumors,Lanzhou University Second Hospital, Lanzhou University SecondClinical Medical College, 82 Cuiyingmen, Chengguan District,Lanzhou 730030, People’s Republic of Chinae-mail: [email protected]

B. Li : J.<H. Zhang :Y.<F. Su :C. Wang : Z.<Q. Wang :Y.<J. Gou :T.<N. Song : J.<B. YangDepartment of Thoracic Surgery, Lanzhou University SecondHospital, Lanzhou University Second Clinical Medical College,Lanzhou 730030, People’s Republic of China

J Gastrointest Surg (2013) 17:2051–2058DOI 10.1007/s11605-013-2382-3

outcomes for patients after esophagogastrostomy. Currently,though mechanically stapled anastomosis is used commonly,some situations may warrant the use of hand-sewn technique.At our hospital, a layer-to-layer mucosal valve technique foresophagogastric anastomosis has been in routine use since2005.16 We report our technique details and clinical results inthis study, aim to determine whether this technique was safe, andwhether it was associated with a diminished incidence of anas-tomotic complications.

Materials and Methods

Patients

Between January 2005 and December 2010, a total of 355consecutive patients with esophageal or gastric cardia carci-noma underwent intrathoracic esophagogastrostomywith gas-tric conduit reconstruction at our hospital. They were operatedon by two surgical teams; all of these surgeons were similarlyexperienced in esophageal surgery. Among these patients, 136received layer-to-layer mucosal valve technique (LM group);another 219 received circular stapled anastomosis, and theyserved as control group (CS group). Preoperative clinicalevaluation was performed for all patients consisting of ultra-sonography of the neck, CT scan of the thorax and abdomen,barium swallow examination, upper gastrointestinal (UGI)endoscopy with biopsy, and Tc99m whole-body bone scan.Electrocardiograph, spirometry, hematological, and biochem-istry tests were performed routinely. No recruited patientsreceived preoperative chemotherapy or radiotherapy, or both.The ethics committee of the hospital approved this retrospec-tive study.

Operative Technique

General anesthesia was administered with a double-lumentube. The patient was placed in right lateral position, andthe chest was entered via a left posterolateral thoracotomythrough the sixth or seventh intercostal space. The intra-thoracic esophagus was isolated en bloc within an enve-lope of the adjoining tissues. The diaphragm was opened,the stomach was mobilized, the left gastric vessels weretransected at their base, and the right gastroepiploic vesselswere preserved. A 6-cm-wide gastric conduit was createdby a linear cutter stapler (Johnson & Johnson) along thegreater curvature, and the staple line was inverted withinterrupted 3–0 silk sutures. Thoracoabdominal standardtwo-field lymphadenectomy was performed according tothe lymph node mapping system.17 No pyloroplasty wasundertaken in this study.

For layer-to-layer mucosal valve anastomosis, the mainsurgical procedures included the following: (1) The esophagus

was clipped with an angle clamp at least 5 cm proximal to theupper edge of the tumor and dissected transversely below theclamp. The specimen was removed. Three stitches ofinterrupted 2–0 silk suture were placed at 4, 6, and 8 o’clockpositions between the outer layer of the esophageal muscularisand gastric seromuscular layers, 2 to 3 cm away from theintended level of anastomosis, to reinforce the posterolateralwall of the anastomotic orifice (Fig. 1a). (2) A 1-cm-widecircumferential myotomy through the esophageal muscularisapproximating above the angle clamp was made, and themuscularis was swept a further 1 cm proximally while keepingthe submucosa and mucosa intact. The gastric seromuscularlayer was incised about 3 cm in diameter at the anastomosisposition and swept a further 1 cm distally the same way as inthe esophagus (Fig. 1b). (3) After ensuring meticulous hemo-stasis, the posterior row of the esophageal muscularis and thegastric seromuscular layer was stitched with interrupted 3–0silk sutures (Fig. 1c). (4) The esophageal mucosa was thensharply transected approximating above the angle clamp tocreate a 2-cm-wide mucosal sleeve, and this resected speci-men was the proximal margin. The gastric mucosa layer wasopened transversely (the length of incision should match thediameter of the esophageal lumen) (Fig. 1d). (5) Supportive 3–0 silk suture was placed at 3 and 9 o’clock position to definethe corners of the mucosal layers and obtain optimal view. Therear mucosal layers were stitched with 4–0 absorbable Vicrylantibacterial suture (polyglyconate) in everted interruptedstitches to achieve mucosa-to-mucosa approximation. Eachbit of the stitches was evenly placed (approximately 3 mmapart at a 3 mm depth) without pulling the sutures too tight toavoid strangulation of tissue (Fig. 1e). (6) Once the rear rowwas completed, the nasogastric tube was placed across theanastomosis into the intrathoracic stomach, and a feeding tubewas placed in the duodenum for postoperative nutritionalsupport routinely (Fig. 1f). (7) The front mucosal layers werethen stitched the same way as the rear row with 4–0 Vicrylsuture. So a 3-cm-long mucosal tube was created (Fig. 1g). (8)The anterior row of the esophageal muscularis and the gastricseromuscular layer was completed in a manner similar to thatof the posterior row (Fig. 1h). (9) Three stitches for reinforcingthe anastomotic posterolateral wall were tied, and anotherthree stitches were added at 2, 10, and 12 o’clock positionsbetween the outer layer of the esophageal muscularis andgastric seromuscular layers, 3 cm away from the anastomosis,to reinforce the anterolateral wall of the anastomotic orifice.Subsequently, the gastric wall was brought up wrappingaround the anastomotic orifice; mucosal tube was invaginated,like a “valve”-shaped stump into the gastric conduit (Fig. 1i).For stapler anastomosis, a circular stapler (EES 25, Johnson &Johnson) was used, and reinforcement was routinelyperformed with five or six stitches between the esophagealmuscularis and gastric seromuscular layers for better securingof the anastomosis.

2052 J Gastrointest Surg (2013) 17:2051–2058

Liquid oral diet was usually commenced on the sixth toeighth postoperative day (POD), after both the clinical evalu-ation and the radiologic study that ruled out anastomoticleakage. If a leakage was present, sufficient drainage wasparamount and was usually achieved by drains placed duringthe operation, nasogastric decompression and appropriate an-tibiotics were used, nutrition was maintained by combinedparenteral and feeding tube, and oral diet was resumed fol-lowing closure of the leakage.

Follow-up and Assessments

Postoperative staging was classified according to the sixthedition AJCC TNM Classification 18 using resected speci-mens. Operative mortality was defined as any death duringthe first 30 days after the operation or during the same hospi-talization with the operation. The anastomosis was routinelyassessed by barium swallow examination performed on POD

5 to 7; orally ingested methylene blue and/or contrast-enhanced CT scan of the thorax were given in clinicallysuspected patients. Anastomotic leakage was defined as thevisualization of gastrointestinal contents or methylene blue inthe thoracic drainage and the extravasation of water-solublecontrast medium on the radiological investigations (includingesophagogram, CT scan) accompanied by clinical featuresindicative of leakage (including evidence of a peri-anastomotic collection, local inflammation, evacuation of airor saliva, mediastinitis, abscess, empyema, and pneumotho-rax). After discharge, patients were routinely examined every3 months during the first year, every 6 months during thefollowing 2 years, and annually thereafter at our hospital. Atthe sixth postoperative month, we used validated question-naires to assess dysphagia and reflux symptoms.19 Sleepdisturbance and modifications to sleeping arrangements werealso recorded. The presence or absence of respiratory symp-toms such as nocturnal cough, wheezing, or recurrent chest

Fig. 1 a Three stitches were placed between the outer layer of theesophageal muscularis and gastric seromuscular layers. b One-centime-ter-wide circumferential myotomy through the esophageal musculariswas made, and the muscularis was swept a further 1 cm proximally whilekeeping the submucosa andmucosa intact. The gastric seromuscular layerwas incised about 3 cm in diameter at the anastomosis position and swepta further 1 cm distally. c The posterior row of the esophageal muscularisand the gastric seromuscular layers was stitched with interrupted sutures.d The esophageal mucosa was sharply transected approximating abovethe angle clamp to create a 2-cm-wide mucosal sleeve. The gastric

mucosa layer was opened transversely. e The rear mucosal layers werestitched with 4–0 absorbable Vicryl suture. f The nasogastric tube andfeeding tube were placed. g The front mucosal layers were stitched thesame way as the rear row, and a 3-cm-long mucosal tube was created. hThe anterior row of the esophageal muscularis and the gastricseromuscular layers was completed in a manner similar to that of theposterior row. i Another three stitches were added between the outer layerof the esophageal muscularis and gastric seromuscular layer to reinforcethe anterolateral wall of the anastomotic orifice

J Gastrointest Surg (2013) 17:2051–2058 2053

infections was recorded, as was sleep disturbance due toreflux. In case of recurrence of dysphagia, barium swallowexamination was conducted instantly. UGI endoscopy wasimmediately recommended in patients with symptoms of dys-phagia or reflux and routinely recommended to be performedannually at follow-up with or without symptoms. Follow-updata was obtained by medical records, telephone interviews,or mail correspondences.

The grade of dysphagia was evaluated according toStoller’s scoring system: 0=able to eat a normal diet, nodysphagia; 1=able to swallow some solid foods; 2=able toswallow only semisolid foods; 3=able to swallow liquidsonly; 4=unable to swallow anything, total dysphagia.20 Anas-tomotic stricture was defined either as the predilation diameterof the anastomotic orifice on barium swallow image≤9 mm,measured by radiologists without knowing the research-related information, or as a stenosis that precluded passageof a 9-mm endoscope and required endoscopic dilation, in theabsence of malignant recurrence defined by endoscopic biop-sy. Strictures were dilated with a Savary bougie over a guidewire visualized under fluoroscopic guidance to a maximaldiameter of 15 mm in the same session. Dilatation sessionswere repeated if necessary until patients could swallow solidfoods. Severity of the stricture was graded by the number ofdilatations needed to relieve dysphagia,21 mild=1 to 2 andsevere≥3. The severity of esophagitis was graded using theLos Angeles Classification System: A=one (or more) mucosalbreak no longer than 5 mm that does not extend between thetops of two mucosal folds; B=one (or more) mucosal breakmore than 5 mm long that does not extend between the tops oftwo mucosal folds; C=one (or more) mucosal break that iscontinuous between the tops of two or more mucosal folds butwhich involves less than 75 % of the circumference; and D=one (or more) mucosal break which involves at least 75 % ofthe esophageal circumference.22

Statistical Analysis

Related data was considered according to the anastomoticapproach (LM group or CS group). The SPSS statisticalpackage (version 16.0, SPSS Inc., Chicago, IL, USA) wasused for analysis. Student’s t test or Mann–Whitney U testwas used for continuous data. Chi-square test or Fisher’s exacttest was used for categorical data. Two-tailed P <0.05 wasconsidered to indicate statistical significance.

Results

Perioperative Clinicopathologic Characteristics

There were 136 patients enrolled in the LM group and 219 inthe CS group. The clinicopathologic data of the two groups

were compared in Table 1. There were no significant differ-ences in age, gender, tumor localization, histological type, andtumor staging. There were also no differences in anastomoticposition, blood loss, and postoperative hospital stay. Therewas a tendency for increased duration of operation with theLM procedure by 16 min, but without statistical significance(P=0.073). Overall postoperative complications were 34.6 %in the LM group and 37.9 % in the CS group without signif-icant difference (P=0.525). No patients died intraoperativelyin our series. The overall operative mortality was 2.3 %,without significant difference between the two groups (P=1.000). Among them, three died of respiratory failure, two ofanastomotic leakages, two of myocardial infarction, and oneof pulmonary thromboembolism.

Evaluation of Anastomosis

There was no anastomotic leakage in the LM group, while inthe CS group, anastomotic leakage occurred in seven patients(3.2 %, P=0.047, Table 1). All of the leakages were treatedconservatively, and these patients were associated with alonger hospital stay.

To assess postoperative stricture and reflux, we excludedthe patients of operative mortality (n =8), and the results weresummarized in Table 2. We found that both the incidence andgrade of dysphagia were significantly lower in the LM group(P <0.05). Significantly fewer patients were observed withstricture after layer-to-layer mucosal valve technique (3.8 %)compared with circular stapled anastomosis (18.2 %, P <0.001). Overall, dilatation was performed in 7 patients in theLM group (5.3%) and 43 in the CS group (20.1 %,P <0.001).Circular stapled anastomosis was associated with a signifi-cantly higher incidence of persistent stricture requiring moredilatation (P <0.001).

The postoperative barium swallow examination of thosetreated by mucosal valve anastomosis showed no regurgita-tion of the contrast medium in 30° Trendelenburg position(Fig. 2). The incidence of reflux symptoms was significantlylower in the LM group during the follow-up period, with82.7 % patients denying all symptoms of reflux even whensupine, compared with 58.9 % in the CS group, P <0.001.There was also a significant reduction in the incidence of sleepdisturbance due to reflux in the LM group compared to the CSgroup (P=0.027).

Endoscopic Findings

A total of 183 patients underwent postoperative UGI endos-copy, including 66 in the LM group and 117 in the CS group.The mean interval between operation and final endoscopy was28.5±19.1 months (range, 2 to 79; median, 29), and 29.7±18.5 months and 27.3±15.9 months in the LM and CS group,respectively, without significant difference (P=0.536). The


final endoscopic findings were presented in Table 3. Theincidence of esophagitis in esophageal remnant was 15.2 %in the LM group and 39.3 % in the CS group; the differencewas significant (P=0.001). However, no difference was found

between the two groups when comparing the severity ofesophagitis (P=1.000). Follow-up endoscopy confirmed ma-lignant recurrence at the anastomotic orifice in nine patients,without statistical difference between the two anastomotic

Table 1 Perioperative clinico-pathologic characteristics LM group n =136 CS group n =219 P value

Age, mean±SD (years) 57.1±8.7 58.8±9.2 0.426

Gender, n (%) 0.943

Male 107 (78.7) 173(79.0)

Female 29(21.3) 46(21.0)

Tumor localization, n (%) 0.977

Middle thoracic 72 (52.9) 115 (52.5)

Lower thoracic 39 (28.7) 65 (29.7)

Gastric cardia 25 (18.4) 39 (18.8)

Anastomotic position, n (%) 0.910

Above the aortic arch 108 (79.4) 175 (79.9)

Under the aortic arch 28 (20.6) 44 (20.1)

Operative duration, mean±SD (min) 229±39 213±36 0.073

Intraoperative blood loss, mean±SD (mL) 403±178 396±183 0.474

pTNM staging, n (%) 0. 499

I 13 (9.6) 25 (11.4)

IIa 31 (22.8) 52 (23.7)

IIb 33 (24.3) 55 (25.1)

III 56 (41.2) 81 (37.0)

IV (M1a-abdominal para-aorta) 3 (2.2) 6 (2.7)

Histological type, n (%) 0.975

Squamous cell carcinoma 107 (78.7) 172 (78.5)

Adenocarcinoma 29 (21.3) 47 (21.5)

Postoperative hospital stay, mean±SD (day) 12.2±4.4 12.7±6.7 0.500

Complications, n (%) 47 (34.6) 83 (37.9) 0.525

Anastomotic leakage, n (%) 0 (0) 7 (3.2) 0.047

Operative mortality, n (%) 3 (2.2) 5 (2.3) 1.000

Table 2 Postoperative assess-ments of dysphagia, stricture, andreflux

LM group (n =133) CS group (n =214) P value

Grade of dysphagia, n (%) 0.000

0 125 (94.0) 169 (79.0)

1 4 (3.0) 13 (6.1)

2 3 (2.3) 17 (7.9)

3 1 (0.8) 13 (6.1)

4 0 (0) 2 (0.9)

Anastomotic stricture, n (%) 5 (3.8) 39 (18.2) 0.000

Anastomotic dilation, n (%) 7 (5.3) 43 (20.1) 0.000

Severity of stricture, n (%) 0.000

No dilation 126 (94.7) 171 (79.9)

Mild 6 (4.5) 34 (15.9)

Severe 1 (0.8) 9 (4.2)

Asymptomatic of reflux, n (%) 110 (82.7) 126 (58.9) 0.000

Sleep disturbance due to reflux, n (%) 8 (6.0) 29 (13.6) 0.027


approaches (P=1.000). These patients received radiotherapyor esophageal stent, respectively.

Discussion

Surgical resection for esophageal and gastric cardiac carcinomawith immediate reconstruction with gastric conduit is currentlyconsidered to be the standard treatment for curative measures orpalliation.1

–3 However, this may be jeopardized by anastomoticcomplications. Early anastomotic complications, such asleakage, are serious risks that increase not only postoperativemorbidity and mortality but also eating difficulties, includingodynophagia and dysphagia. Late complications, such asstricture or reflux, are major factors that decrease the long-termfunctional results and quality of patients’ lives. The incidences ofleakages are still substantial;23

,24 in a recent study, investigatorsobserved that global health status remains significantly reduced

in long-term survivors after esophagogastrostomy comparedwith population controls, and although some symptoms im-proved, deteriorated swallowing function and gastroesophagealreflux remained.8 As the postoperative outcomes of patients maybe extremely influenced by the quality of the anastomosis,4 itbecomes a hot topic for surgeons to ameliorate the anastomotictechnique without sacrificing the effect of the operation.

As with wound healing, the healing of anastomosis is a seriesof carefully regulated steps beginning immediately once thereconstruction occurs. The development of leakage and strictureis complicated and is often related to various factors such astechnique, tension, ischemia, and infection of the anastomoticregion.6 Esophagogastric anastomosis heals in a hyperacidicenvironment. The esophageal and gastric mucosa have goodantiacid stress and anti-infection capability, but the antiacid stressability of the submucosa and muscularis is poor. Moreover, thesubmucosa has excellent blood flow and tends to develop aconnection much more easily than the muscularis, so

Fig. 2 Postoperative bariumswallow examination of mucosalvalve anastomosis, withoutregurgitation of the contrastmedium in 30° Trendelenburgposition (anastomotic orificeshown by arrow)

Table 3 Endoscopic findings atfinal UGI endoscopy

LA Los Angeles ClassificationSystem

LM group (n =66) CS group (n =117) P value

Esophagitis, n (%) 10 (15.2) 46 (39.3) 0.001

Severity of esophagitis, n (%) 1.000

LA - A 2 (3.0) 9 (7.7)

LA - B 3 (4.5) 15 (12.8)

LA - C 4 (6.1) 17 (14.5)

LA - D 1 (1.5) 5 (4.3)

Anastomotic recurrence, n (%) 3 (4.5) 6 (5.1) 1.000


anastomotic healing relies mainly on union of the mucosa andsubmucosa.25 Animal experiments demonstrated that the single-layer mucosal anastomosis was smoother with better elasticitythan the whole-layer anastomosis; in addition, with the nature ofprimary intention wound healing, the inflammatory reaction andscar formationwas slight, whichmight be themain reason for thelow rate of anastomotic stricture formation.26 The clinical appli-cation of layered anastomosis also results in both low leakageand stricture rate after esophagogastrostomy.26

–28 Therefore, it isconsidered that anastomosis should be performed in a layeredfashion.

At our hospital, we use a layered mucosal valve anastomo-sis for reconstruction. The results from the present studysuggest that this technique is a safe procedure and is associat-ed with a markedly diminished incidence of anastomoticcomplications. No anastomotic leakage was observed in pa-tients with layered mucosal valve technique. Patientsexpressed greater satisfaction with their ability to swallowcomfortably, as reflected by the decreased formation of anas-tomotic strictures and decreased need for dilatations due todysphagia. Strictures were minimal and easily dilated even ifthey occur, with most patients only requiring a single dilation.On the other hand, the rate of leakage was 3.2 % in circularstapler anastomosis, and a proportion of patients suffered fromrefractory stricture requiring a higher number of dilations tomaintain patency of the lumen.

Dysphagia is the most frequently mentioned symptom asso-ciated with anastomotic stricture. When a stricture develops, therecurrence of dysphagia defeats one of the main aims of surgery,which is to restore normal swallowing function.29 The circularstapler anastomosis was more frequently associated with thedevelopment of anastomotic strictures.15 The reasons why stric-ture rate was more common with the circular stapled methodmay include the following: a lack of accurate mucosa-to-mucosaconjugation during anastomosis leading to muscularis explora-tion and tissue ischemia in the anastomotic portion, which leadsto inflammation, ulcer, granulation tissue formation, and exces-sive fibrosis, which have the potential to develop into strictureand even leakage during the healing process; furthermore, thecircumferentially placed unabsorbable metal staples limit thecapability of the esophageal lumen to dilate beyond the sizeobtained originally.13

,26,30–32

The layered mucosal valve anastomosis in our study adoptedmucosa-to-mucosa single-layer sutures. This ensured accurateapposition of the mucosal and muscular layers and avoidedmuscularis exploration, acid or alkali damage, reduced inflam-mation, and scar formation. At the same time, absorbable Vicrylantibacterial suture was used for mucosa suturing, which inhibitsbacterial growth and consequently prevent postoperative infec-tion, perhaps playing a role in decreasing stricture formation.33

Furthermore, with the circumferential myotomy through theesophageal muscularis, the tension of anastomosis was reduced,and the anastomotic orifice of the mucosa and muscularis was in

a different level. Takada and colleagues found that a singlecircular myotomy reduced anastomotic tension by about 50 %and that two myotomies placed 2 cm apart reduced tension asmuch as 75 %.34 Finally, vascularized tissue is a mainstay in thetreatment of esophageal perforation: placement of a well-vascularized stomach around the esophageal remnant and theanastomosis in the form of a wrap would improve the anasto-motic healing environment to prevent leakage.

Our data also revealed that the layered mucosal valvetechnique can be effective in controlling postoperative gastro-esophageal reflux. Reflux symptoms can decrease quality oflife and have been reported to be present in 60 to 80 % ofpatients.35 Reflux of acid and bile into the esophageal remnantis known to predispose the development of Barrett’s epitheli-um in the remnant esophagus; Barrett’s epithelium may even-tually lead to a new carcinoma in those who have achievedlong-term survival after esophagogastrostomy.36 In our sup-poses, by the reinforcement, the stomach tube was brought upwrapping around the esophageal remnant and the anastomo-sis, and an invagination was created, which can be applied as aone-way valve, allowing food to pass but closing the anasto-motic orifice during coughing, bending, or supine position,which may contribute to the lower incidence of gastroesoph-ageal reflux in our series, as shown in Fig. 2, without regur-gitation of the contrast medium in Trendelenburg position.

Lastly, another consideration in choosing this technique foranastomosis is the medical cost. In our experience, the layeredmucosal valve anastomosis was cheaper in terms of the consum-ables required and therefore more suitable for underdevelopedregions.

We understand the potential limitations in our study. Firstand foremost, it is a retrospective analysis and includes a risk ofsome selection biases. Moreover, although a validated ques-tionnaire was used to assess reflux symptoms, there was noobjective evaluation using esophageal manometry or a 24-hesophageal pH testing to confirm the lack of pathologic reflux;there remains a component of reporting bias from the patients.

In conclusion, the layered mucosal valve esophagogastricanastomosis is a safe and effective anastomotic technique, whichcan significantly diminish the incidence of anastomotic compli-cations, and could be used as an alternative for anastomosis afterresection of esophageal and gastric cardiac carcinoma.

Conflict of Interest The authors have declared no conflict of interest.

References

1. Swanson SJ, Batirel HF, Bueno R, JaklitschMT, Lukanich JM,Allred E,Mentzer SJ, Sugarbaker DJ. Transthoracic esophagectomy with radicalmediastinal and abdominal lymph node dissection and cervicalesophagogastrostomy for esophageal carcinoma. Ann Thorac Surg.2001; 72(6): 1918–25.


2. Enzinger PC, Mayer RJ. Esophageal cancer. N Engl J Med. 2003;349(23): 2241–52.

3. Boone J, Livestro DP, Elias SG, Borel Rinkes IH, van HillegersbergR. International survey on esophageal cancer: part I surgical tech-niques. Dis Esophagus. 2009; 22(3): 195–202.

4. Lerut T, Coosemans W, Decker G, De Leyn P, Nafteux P, vanRaemdonck D. Anastomotic complications after esophagectomy.Dig Surg. 2002; 19(2): 92–8.

5. Rizk NP, Bach PB, Schrag D, Bains MS, Turnbull AD, Karpeh M,Brennan MF, Rusch VW. The impact of complications on outcomesafter resection for esophageal and gastroesophageal junction carcino-ma. J Am Coll Surg. 2004; 198(1): 42–50.

6. Ercan S, Rice TW, Murthy SC, Rybicki LA, Blackstone EH. Doesesophagogastric anastomotic technique influence the outcome ofpatients with esophageal cancer? J Thorac Cardiovasc Surg. 2005;129(3): 623–31.

7. Aly A, Jamieson GG, Watson DI, Devitt PG, Ackroyd R, StoddardCJ. An antireflux anastomosis following esophagectomy: a random-ized controlled trial. J Gastrointest Surg. 2010; 14(3): 470–5.

8. Donohoe CL, McGillycuddy E, Reynolds JV. Long-term health-related quality of life for disease-free esophageal cancer patients.World J Surg. 2011; 35(8): 1853–60.

9. Rutegård M, Lagergren P, Rouvelas I, Lagergren J. Intrathoracicanastomotic leakage and mortality after esophageal cancer resection:a population-based study. Ann Surg Oncol. 2012; 19(1): 99–103.

10. van der Schaaf M, Lagergren J, Lagergren P. Persisting symptomsafter intrathoracic anastomotic leak following oesophagectomy forcancer. Br J Surg. 2012; 99(1): 95–9.

11. Kim RH, Takabe K. Methods of esophagogastric anastomoses fol-lowing esophagectomy for cancer: a systematic review. J Surg Oncol.2010; 101(6): 527–33.

12. Biere SS, Maas KW, Cuesta MA, van der Peet DL. Cervical orthoracic anastomosis after esophagectomy for cancer: a systematicreview and meta-analysis. Dig Surg. 2011; 28(1): 29–35.

13. Saluja SS, Ray S, Pal S, Sanyal S, Agrawal N, Dash NR, Sahni P,Chattopadhyay TK. Randomized trial comparing side-to-side stapledand hand-sewn esophagogastric anastomosis in neck. J GastrointestSurg. 2012; 16(7): 1287–95.

14. Maas KW, Biere SS, Scheepers JJ, Gisbertz SS, Turrado Rodriguez VT,van der Peet DL, Cuesta MA. Minimally invasive intrathoracic anasto-mosis after Ivor Lewis esophagectomy for cancer: a review of transoral ortransthoracic use of staplers. Surg Endosc. 2012; 26(7): 1795–802.

15. Honda M, Kuriyama A, Noma H, Nunobe S, Furukawa TA. Hand-sewn versus mechanical esophagogastric anastomosis after esopha-gectomy: a systematic review and meta-analysis. Ann Surg. 2013;257(2): 238–48.

16. Su YF. Prevention of anastomotic complications after esopha-gogastrostomy. Health Vocational Education. 2005; 23(16): 171–2.

17. Korst RJ, Rusch VW, Venkatraman E, Bains MS, Burt ME, DowneyRJ, Ginsberg RJ. Proposed revision of the staging classification foresophageal cancer. J Thorac Cardiovasc Surg. 1998; 115(3): 660–70.

18. Sobin LH, Wittekind CH. TNM classification of malignant tumors.6th edition. New York: Wiley Liss; 2002.

19. ChineseGastroesophageal RefluxDisease Study Group. The value ofreflux diagnostic questionnaires in the diagnosis of gastroesophagealreflux disease. Chin J Dig. 2003; 23(11): 651–4.

20. Stoller JL, Samer KJ, Toppin DI, Flores AD. Carcinoma of theesophagus: a new proposal for the evaluation of treatment. Can JSurg. 1977; 20(5): 454–9.

21. Briel JW, Tamhankar AP, Hagen JA, DeMeester SR, Johansson J,Choustoulakis E, Peters JH, Bremner CG, DeMeester TR. Prevalenceand risk factors for ischemia, leak, and stricture of esophageal anas-tomosis: gastric pull-up versus colon interposition. J Am Coll Surg.2004; 198(4): 536–42.

22. Lundell LR, Dent J, Bennett JR, Blum AL, Armstrong D, GalmicheJP, Johnson F, HongoM, Richter JE, Spechler SJ, Tytgat GNJ,WallinL. Endoscopic assessment of oesophagitis: clinical and functionalcorrelates and further validation of the Los Angeles classification.Gut. 1999; 45(2): 172–80.

23. Sarela AI, Tolan DJ, Harris K, Dexter SP, Sue-Ling HM.Anastomotic leakage after esophagectomy for cancer: a mortality-free experience. J Am Coll Surg. 2008; 206(3): 516–23.

24. Worrell S, Mumtaz S, Tsuboi K, Lee TH, Mittal SK. Anastomoticcomplications associated with stapled versus hand-sewn anastomo-sis. J Surg Res. 2010; 161(1): 9–12.

25. Kakegawa T. Forty years’ experience in surgical treatment for esoph-ageal cancer. Int J Clin Oncol. 2003; 8(5): 277–88.

26. Zhao YF, Xie CH, Wang Y, Kou YL, Liu LX. Clinical application andexperimental study of layered anastomosis in esophagogastrostomy.Chin J Clin Thorac Cardiovasc Surg. 2005; 12(3): 164–8.

27. Zhu ZJ, Zhao YF, Chen LQ, Hu Y, Liu LX,Wang Y, Kou YL. Clinicalapplication of layered anastomosis during esophagogastrostomy.World J Surg. 2008; 32(4): 583–8.

28. Maciver RH, Sundaresan S, DeHoyos AL, Sisco M, Blum MG.Mucosal tube technique for creation of esophageal anastomosis afteresophagectomy. Ann Thorac Surg. 2009; 87(6): 1703–7.

29. Nederlof N, Tilanus HW, Tran TC, Hop WC, Wijnhoven BP, deJonge J. End-to-end versus end-to-side esophagogastrostomy afteresophageal cancer resection: a prospective randomized study. AnnSurg. 2011; 254(2): 226–33.

30. Law S, Fok M, Chu KM, Wong J. Comparison of hand-sewn andstapled esophagogastric anastomosis after esophageal resection forcancer: a prospective randomized controlled trial. Ann Surg. 1997;226(2): 169–73.

31. Furukawa Y, Hanyu N, Hirai K, Ushigome T, Kawasaki N, ToyamaY, Nakayoshi T, Yanaga K. Usefulness of automatic triangular anas-tomosis for esophageal cancer surgery using a linear stapler (TA-30).Ann Thorac Cardiovasc Surg. 2005; 11(2): 80–6.

32. Aiko S, Ando N, Shinozawa Y, Ozawa S, Kitajima M, Kurose I,Tsuchiya M. Increased chemiluminescence and ulcer development inthe low blood flow state of the gastric tube for esophageal replace-ment. J Clin Gastroenterol. 1993; 17(Suppl 1): S161-S6.

33. Suárez Grau JM, De Toro Crespo M, Docobo Durántez F, RubioChaves C, Martín Cartes JA, Docobo Pérez F. Prevention of surgicalinfection using reabsorbable antibacterial suture (Vicryl Plus) versusreabsorbable conventional suture in hernioplasty. An experimentalstudy in animals. Cir Esp. 2007; 81(6): 324–9.

34. Takada Y, Kent G, Filler RM. Circular myotomy and esophageallength and safe esophageal anastomosis: an experimental study. JPediatr Surg. 1981; 16(3): 343–8.

35. Dresner SM, Griffin SM, Wayman J, Bennett MK, Hayes N, RaimesSA. Human model of duodenogastro-oesophageal reflux in thedevelopment of Barrett’s metaplasia. Br J Surg. 2003; 90(9): 1120–8.

36. O’Riordan JM, Tucker ON, Byrne PJ, McDonald GS, Ravi N,Keeling PW, Reynolds JV. Factors influencing the development ofBarrett’s epithelium in the esophageal remnant postesophagectomy.Am J Gastroenterol. 2004; 99(2): 205–11.


clinical application of mucosal valve technique for anastomosis during esophagogastrostomy

Documents