We describe a novel intrathoracic esophagogastric anastomotic technique designed to create a large diameter anastomosis while simultaneously maintaining conduit blood supply to minimize the incidence of anastomotic leaks and strictures. Construction and orientation of the stomach conduit designed to optimize upper gastrointestinal tract function is also described.
We describe a novel esophagogastric anastomotic technique (“side-to-side: staple line-on-staple line”, STS) for intrathoracic anastomoses designed to create a large diameter anastomosis while simultaneously maintaining conduit blood supply. This technique aims to minimize the incidence of anastomotic leaks and strictures, which is a frequent source of morbidity and occasional mortality after esophagectomy. We analyze the results of this STS technique on 368 patients and compared outcomes to 112 patients who underwent esophagogastric anastomoses using an end-to-end stapler (EEA) over an 8-year time interval at our institution.
The STS technique involves aligning the remaining intrathoracic esophagus over the tip of the lesser curve staple line of a stomach tube, created as a replacement conduit for the esophagus. A linear stapling device cuts through and restaples the conduit staple line to the lateral wall of the esophagus in a side-to-side fashion. The open common lumen is then closed in two layers of sutures.
There was a total of 12 (3.8%) anastomotic leaks in patients who underwent STS esophagogastric anastomosis. Two of eight patients (25%) had anastomotic leaks after esophagectomy for end-stage achalasia as compared to a 2.8% leak rate (10/336) after esophagectomy for other conditions. Eighteen (5.2%) patients required a median of 2 dilatations for anastomotic stricture after STS anastomosis. Supplemental jejunostomy feedings were required in only 11.1% of patients undergoing STS anastomoses following hospital discharge. In contrast, patients undergoing EEA anastomoses demonstrated anastomotic leak and stricture rates of 16.1% and 14.3% respectively (p<0.01). Time analysis of postoperative contrast studies following the STS technique typically demonstrated a straight/uniform diameter conduit with essentially complete contrast emptying into the small bowel within 3 minutes in 88.4% of patients.
The incidence of esophagogastric anastomotic leaks and strictures were extremely low using this novel anastomotic technique. Additionally we believe that based on time and qualitative analyses of postoperative contrast studies, this technique appears to optimize postoperative upper gastrointestinal tract function; however, further comparative studies are needed.
Esophagogastric anastomotic leaks represent a not uncommon complication after esophagectomy1. Additionally, an anastomotic leak has been associated to unfavorable longer-term outcomes including hospital readmission, early mortality, and occasionally poor quality of life2,3,4. An anastomotic stricture represents a long-term complication, which can also be a consequence of an anastomotic leak5. Anastomotic strictures both negatively impact quality of life as well as escalate the cost of care.
As more esophageal cancer patients are living longer as a result of endoscopic detection of early stage adenocarcinoma and induction chemoradiation therapy for more locally advanced cancers, optimizing stomach conduit function also becomes important. Gastric conduit “function” mainly, however, relies on gravity for drainage. Gastric conduit construction and orientation can affect upper gastrointestinal tract function and therefore poor conduit "function" can be a result of technical issues.
We have used a novel "side-to-side: staple line-on-staple line" (STS) technique for intrathoracic esophagogastric anastomosis since 2009. This technique is designed to create a larger diameter anastomosis as compared to anastomoses made with end-to-end staplers (EEA) while simultaneously maintaining conduit blood supply to reduce the incidence of anastomotic leaks and strictures. We describe this novel anastomotic technique and additionally described gastric conduit construction and orientation to optimize function. We compared the results of this STS technique to anastomoses performed with EEA staplers over an 8-year time interval at our institution.
This study was approved by Indiana University-Purdue University in Indianapolis institutional review board (1109006832). This procedure was performed on all patients requiring resection and resection of the distal intrathoracic esophagus for malignant or benign diseases since 2009 at Indiana University Hospital.
1. Pre-anesthesia phase
2. Initial laparotomy phase
3. Thoracic phase
4. Re-laparotomy: “double flip” phase
From 2009 to 2017, a total of 368 patients were identified who underwent an STS intrathoracic esophagogastric anastomosis and of these 12 (3.8%) had anastomotic leaks. Five of these patients demonstrated grade I/II leaks and required no intervention. Six patients and one patient respectively experienced grade III and grade IV leaks requiring endoscopic stenting and/or surgical intervention2. A leak rate of 25% (2/8) was observed after esophagectomy for end-stage achalasia as compared to a 2.8% leak rate (10/336) where esophagectomy was performed for other conditions. There were 4 (1.1%) patient who died postoperatively, none of which had an anastomotic complication. Of STS patients, 18 (5.0%) required a median of 2 dilatations for symptomatic anastomotic strictures. Supplemental jejunostomy feedings were required in only 11.1% of these patients following hospital discharge. In contrast, of the 112 patients identified who underwent thoracoscopic end-to-end mechanical stapler (EEA) anastomosis over this same time interval, 16.1% and 14.3% demonstrated anastomotic leaks and symptomatic strictures respectively (p<0.01) despite all (100%) of these patients being maintained on a limited diet with supplemental jejunostomy tube feedings for at least one month following surgery (Table 1). Demographic and comorbidities of both STS and EEA groups were statistically similar; however, there was a trend towards more cardiac disease in the STS cohort (Table 2). Time analysis of postoperative contrast studies in first 208 patients who underwent an STS anastomosis typically demonstrated a straight/uniform diameter conduit with essentially complete (>95%) contrast emptying into the small bowel within 3 min in 184 (88.4%) patients. The remaining 11.6% of patients studied, had near complete (>95%) contrast emptying into the small bowel within 5 minutes. No patient studied had contrast transit time in excess of 5 minutes.
Figure 1: Creation of the stomach conduit. The stomach is secured at three points, which are retracted outward. A relatively uniform diameter conduit is created with staplers aiming just inferior to the previous staple line on the lesser curve. (Taken from with permission11.) Please click here to view a larger version of this figure.
Figure 2: Alignment of the esophagus over the stomach conduit. The mid left lateral aspect of the upper third of the intrathoracic esophagus is aligned over the lesser curve staple line at the tip of the conduit with 4 initial tacking sutures approximately 2 to 3 cm apart. (Taken from with permission11.) Please click here to view a larger version of this figure.
Figure 3: Preparation for a side-to-side communication between the esophagus and stomach conduit. A 1 to 2 cm rent is made in the stomach conduit across the lesser curve staple line. Tacking sutures are placed between the adjacent conduit and esophageal walls beginning in the middle through the lesser curve staple line then two sutures on either side. (Taken from with permission11.) Please click here to view a larger version of this figure.
Figure 4: Creation of a side-to-side communication between the esophagus and stomach conduit. A 45 mm endoscopic stapler is fired cutting through and restapling the lesser curve staple line. Typically only 2/3rds of the stapler length is used. An ellipse of conduit is removed over the lesser curve staple line (dotted line) rather than extending the rent laterally to equalize the length of the cut end of the esophagus and gastrotomy. (inset) (Taken from with permission11.) Please click here to view a larger version of this figure.
Figure 5: Closure of the open common lumen. The open common lumen is hand-closed in two layers of sutures beginning with inverted interrupted absorbable sutures. The first suture layer is imbricated by a second layer of Lembert silk sutures placing the stomach suture a few mm inferior to the first layer of sutures. (inset) (Taken from with permission11.) Please click here to view a larger version of this figure.
Figure 6: Over sew of the lesser curve staple line and pleural buttress of the anastomosis. The superior aspect of the lesser curve conduit staple line can be over sewn at this point with interrupted silk sutures inferiorly until the right gastric vessels are encountered. For anastomoses established in near the thoracic inlet, a flap of mobilized pleura is tacked to the stomach conduit to contain small anastomotic leaks. (Taken from with permission11.) Please click here to view a larger version of this figure.
Figure 7: Vascularized soft tissue buttress of the anastomosis. Pericardial fat is loosely wrapped around anastomoses created in the middle aspect of the posterior mediastinum to contain any small areas of anastomotic dehiscence. Please click here to view a larger version of this figure.
Figure 8: Final (“double flip”) laparotomy phase. The laparotomy incision is reopened. The right gastroepiploic fat and conduit carefully tacked to the crus with 2-0 silk sutures. (Taken from with permission11.) Please click here to view a larger version of this figure.
STS (N=368) | EEA (N=112) | |
Anastomotic Leaks | 3.8%* | 16.1% |
Anastomotic Stricture | 5.2%* | 14.3% |
Post Op J-Tube Usage | 11.0%* | 100.0% |
Table 1. Anastomotic leaks/strictures and use of postoperative feeding jejunostomy (Post Op J-Tube Usage) following hospital discharge comparing patients undergoing open STS and thoracoscopic EEA esophagogastric anastomoses using an Ivor Lewis approach performed at Indiana University Simon Cancer Center from 2009 to 2017. (*p value < 0.01, chi-square)
STS (n=278) | EEA (n=82) | P-Value | |
Age at Diagnosis (yrs) | 60.3 ± 11.4 (23-84) | 60.6 ± 9.0 (38-80) | 0.80 |
Gender | |||
Male | 228 (82.0%) | 69 (84.1%) | 0.66 |
Female | 50 (18.0%) | 13 (15.9%) | |
Cormorbidity | |||
Cardiac | 114 (41.0%) | 24 (29.3%) | 0.06 |
Diabetes Mellitus | 70 (25.2%) | 19 (23.2%) | 0.71 |
COPD | 32 (11.5%) | 12 (14.6% | 0.45 |
Histology | |||
Adenocarcinoma | 237 (85.3%) | 74 (90.2%) | 0.25 |
Other Diagnoses | 41 (14.7%) | 8 (9.8%) | |
Neoadjuvant Therapy | 200 (71.9%) | 59 (71.9%) | 0.99 |
Table 2: Demographic and comorbidity comparison open STS and thoracoscopic EEA anastomoses using an Ivor Lewis approach performed at Indiana University Simon Cancer Center from 2009 to 2015. Mean and standard deviation with range given for continuous variables. P values for continuous variables generated by Student’s t-test and chi-square for discrete variables.
Esophagectomy represents a very extensive surgical procedure. Adverse long-term quality of life has been linked to patients experiencing postoperative complications including anastomotic leaks3. Risk factors for an anastomotic leak primarily include creation of an anastomosis with poor blood supply. An anastomotic leak not only represents a significant source of postoperative morbidity, but also can commonly result in stricture. A stricture can also be a result of technical issues including performing a small diameter anastomosis. Besides impacting quality of life, strictures add to overall medical costs when dilatation is needed5. Accurately performing several steps is of utmost importance to minimize complications as well as achieve good oncologic and upper gastrointestinal tract functional outcomes.
Esophagogastric anastomosis can be accomplished by several methods, including hand sewn, EEA, and linear stapler techniques representing the majority. A report from the Society of Thoracic Surgeons General Thoracic Database cited an overall leak rate of 9.3% in patients undergoing intrathoracic esophagogastric anastomoses6. While postoperative mortality due to anastomotic leak seems to be decreasing, subsequent stricture rates remain high, ranging between 10 and 56%7. Collard and Orringer described a linear stapler technique to create a side-to-side cervical esophagogastric anastomosis8,9. The posterior triangulated opening formed by the linear stapler was demonstrated to result in a low leak rate as well as resistance to stricture. A retrospective study from Mayo Clinic reported a 5.6% incident of intrathoracic anastomotic leak in 177 patients where a linear stapled technique was used versus an 8.3% leak rate in 48 patients who underwent an EEA stapled anastomosis. While this difference did not reach statistical difference, the one-year probability for stricture was 32% after EEA anastomosis as compared to only 8.6% with linear stapled techniques, which was significant5. Wang and colleagues performed a prospective clinical trial involving 155 patients who were randomized into one of three esophagogastric anastomotic methods10. Impressively, no postoperative strictures developed in patients undergoing a linear stapler technique as compared to 9.6% and 19.1% in the hand sewn and circular stapled cases respectively, which was statistically significant. Prior studies involving linear stapler anastomoses have utilized the anterior wall of the stomach conduit for the anastomotic site. This approach may lead to an ischemic strip of conduit between the lesser curve staple line and the anastomosis predisposing to leak. Novel to our technique, collateral blood supply to the stomach conduit tip is preserved by cutting through and restapling the lesser curve staple line.
There are limitations to the study. First, this represents a retrospective analysis. Despite the retrospective nature however, we utilized this STS approach uniformly in all patients undergoing surgery for mid-esophageal to proximal stomach pathology over the study interval as an “intent to treat” including stable patients who sustained esophageal perforations during dilatation for stricture where repair was not possible. Common to any intrathoracic side-to-side anastomotic technique is the need to dissect an additional 3 to 4 cm of proximal esophagus, potentially decreasing the length of surgical esophageal margin in cases of malignancy and also potentially resulting in some degree of esophageal devascularization. To avoid devascularization, a critical point is not to dissect the intrathoracic esophagus any further superiorly toward the neck than the tip of the stomach conduit will reach without tension.We speculate the higher leak rate observed using this technique for patients with end-stage achalasia may be related to further devascularization of a thicken esophageal wall after mobilization to perform an STS anastomosis where the preexisting blood supply may be poor. Esophageal dilatation frequently seen in achalasia cases makes hand sewing the open common lumen very difficult, which may also be a factor.Based on this experience, we now believe that achalasia is a contraindication to perform STS intrathoracic anastomoses.Of note, for long or more central cancers where an esophagogastric anastomosis needs to be created near the thoracic inlet to achieve an adequate proximal esophageal margin, we have utilized a somewhat shorter initial side-to-side communication not utilizing the entire length of the 45 mm GIA which however has potential to be more prone to stricture formation.
Unlike the esophagus, the stomach is a passive conduit, gravity dependent for drainage. Several variables including conduit diameter and length as well as conduit orientation can, therefore, significantly impact upper gastrointestinal tract function. Ingested food has the potential to hang up in three locations: the esophagogastric anastomosis, the stomach body, and gastric outlet. Poor conduit function can be a result of technical issues in any of these three areas. Poor conduit emptying paradoxically can cause more “reflux,” not only negatively impacting quality of life but also occasionally resulting in aspiration. “Minimally invasive” (laparoscopic/thoracoscopic) approaches, which utilize an EEA stapler for esophagogastric anastomoses, although still representing the minority of esophagectomy cases performed, have become increasingly popular. We believe however our open technique as described not only allows creations of precise STS esophagogastric anastomosis to reduce the stricture rates but additionally optimizes conduit construction and orientation with a straight non-redundant stomach conduit including pyloroplasty with minimal tendency for ingested food materials to hang up in these areas as compared to thoracoscopic approaches. The measured contrast transit times from mouth to small bowel on routine postoperative studies would support excellent conduit function with our technique, however comparative studies using other techniques are needed. Our observations would support minimal and self-limiting “dumping” symptoms in the vast majority of STS patients however specific quality of life assessments are currently underway. Finally, with the thoracotomy approach described, we have observed little difference with respect to acute and long-term postoperative discomfort as compared to patients undergoing a thoracoscopic approach at our institution.
In summary, we believe this novel STS technique can significantly reduce the morbidity and occasional mortality of esophagogastric anastomotic complications following esophagectomy. Conduit construction and orientation as described additionally optimizes upper gastrointestinal tract function. Finally, this technique is easily adapted and reproducible.
The authors have nothing to disclose.
None.
100 mm Linear Stapler (ILA Autosuture, “green” cartridge, 4.8 mm staple height) | Covidien | 3973 | Surgical Stapler |
3-0 silk (Perma hand black, 8×18", SH needle, 1/2 circle 26 mm, C013) | Ethicon | C013D | Suture Material |
3-0 silk (Perma hand black, 8×30", SH needle, 1/2 circle 26 mm, C017) | Ethicon | C017D | Suture Material |
3-0 vicryl (Coated vicryl violet, 8×18", SH needle, 1/2 circle 26 mm, J774) | Ethicon | VCP774D | Suture Material |
3-0 vicryl (Coated vicryl violet, 8×27", SH needle, 1/2 circle 26 mm, J784) | Ethicon | VCP784D | Suture Material |
45 mm Endoscopic Stapler (Flex “green” cartridge, 4.1 mm staple height) | Ethicon | SC45A | Surgical Stapler |
60 mm Endoscopic Tristapler | Ethicon | SC60A | Surgical Stapler |
Flex “green” cartridge, 4.1 mm staple height | Ethicon | GST45G | Surgical Stapler |
Flex 60, “black” cartridge (for 60 mm Endoscopic Tristapler) | Ethicon | GST60T | Surgical Stapler |
Foceps Debakey 7.75 inch | Jarit | 320-101 | Surgical Instrument |
Forceps Debakey 12 inch | Jarit | 320-103 | Surgical Instrument |
Forceps Debakey 9.5 inch | Jarit | 320-102 | Surgical Instrument |
Needle Holder Mayo-Hegar 10 inch | Codman | 36-2019 | Surgical Instrument |
Needle Holder Mayo-Hegar 7 inch | Codman | 36-2017 | Surgical Instrument |
Needle Holder Mayo-Hegar 8 inch | Codman | 36-2018 | Surgical Instrument |
Needle Holder Ryder 10 inch | Codman | 36-3005 | Surgical Instrument |
Needle Holder Ryder 9 inch | Jarit | 121-164 | Surgical Instrument |