Esophageal leaks repaired by a muscle onlay approach in the presence of mediastinal sepsis.

BACKGROUND: Nineteen patients were evaluated after closure of intrathoracic esophageal leaks by a pediculated muscle flap onlay repair in the presence of mediastinal and systemic sepsis. METHODS: Intrathoracic esophageal leaks with mediastinitis and systemic sepsis occurred after delayed spontaneous perforations (n = 7) or surgical and endoscopic interventions (n = 12). Six patients presented with fulminant anastomotic leaks. Seven patients had previous attempts to close the leak by surgery (n = 4) or stenting (2) or both (n = 1). The debrided defects measured up to 2 x 12 cm or involved three quarters of the anastomotic circumference and were closed either by a full thickness diaphragmatic flap (n = 13) or a pediculated intrathoracically transposed extrathoracic muscle flap (n = 6). All patients had postoperative contrast esophagography between days 7 and 10 and an endoscopic evaluation 4 to 6 months after surgery. RESULTS: There was no 30-day mortality. During follow-up (4 to 42 months), 16 patients (84%) revealed functional and morphological restoration of the esophagointestinal integrity without further interventions. One patient required serial dilatations for a stricture, and 1 underwent temporary stenting for a persistent fistula; both patients had normal control endoscopy during follow-up. A third patient requiring permanent stenting for stenosis died from gastrointestinal bleeding due to stent erosion during follow-up. CONCLUSIONS: Intrathoracic esophageal leaks may be closed efficiently by a muscle flap onlay approach in the presence of mediastinitis and where a primary repair seems risky. The same holds true for fulminant intrathoracic anastomotic leaks after esophagectomy or other surgical interventions at the gastroesophageal junction.

Should patients with esophageal atresia be submitted to esophageal substitution before they start walking?

P>Esophagocoloplasty and gastric transposition are two major methods for esophageal substitution in children with esophageal atresia, and there is broad agreement that these operations should not be performed before the children start walking. However, there are some reported advantages of performing such operations in the first months of life or in the neonatal period. In this study, we compared our experience with esophageal substitution procedures performed in walking children with esophageal atresia, with the outcomes of children who had the operation before the third month of life reported in the literature. The purpose of this study was to establish if we have to wait until the children start walking before indicating the esophageal replacement procedure. From February 1978 to October 2009, 129 children with esophageal atresia underwent esophageal replacement in our hospital (99 colonic interpositions and 30 gastric transpositions). The records of these patients were reviewed for data regarding demographics, complications (leaks, graft failures, strictures, and graft torsion), and mortality and compared with those reported in the two main articles on esophageal replacement in the neonatal period or in patients less than 3 months of age. The main complication of our casuistic was cervical anastomosis leakage, which sealed spontaneously in all except in four patients. One patient of the esophagocoloplasty group developed graft necrosis and three patients in the gastric transposition group had gastric outlet obstruction, secondary to axial torsion of the stomach placed in the retrosternal space. The long-term outcome of the patients in both groups was considered good to excellent in terms of normal weight gain, absence of dysphagia, and other gastrointestinal symptoms. The comparisons of the main complications and mortality rates in walking children with esophageal substitutions performed in the first months of life showed that the incidences of cervical anastomotic leaks and graft failures were similar, but mortality rate in the first few months of life was significantly greater than that observed in our group of patients (P = 0.001). Based on the comparison of our results with those of published series, we conclude that the recommendation of performing esophagocoloplasty or total gastric transposition in children with esophageal atresia after they start walking is still valid.

Early Complication Rates in a Single-Surgeon Series of 2500 Robotic-Assisted Radical Prostatectomies: Report Applying a Standardized Grading System

Background: Perioperative complications following robotic-assisted radical prostatectomy (RARP) have been previously reported in recent series. Few studies, however, have used standardized systems to classify surgical complications, and that inconsistency has hampered accurate comparisons between different series or surgical approaches. Objective: To assess trends in the incidence and to classify perioperative surgical complications following RARP in 2500 consecutive patients. Design, setting, and participants: We analyzed 2500 patients who underwent RARP for treatment of clinically localized prostate cancer (PCa) from August 2002 to February 2009. Data were prospectively collected in a customized database and retrospectively analyzed. Intervention: All patients underwent RARP performed by a single surgeon. Measurements: The data were collected prospectively in a customized database. Complications were classified using the Clavien grading system. To evaluate trends regarding complications and radiologic anastomotic leaks, we compared eight groups of 300 patients each, categorized according the surgeon`s experience (number of cases). Results and limitations: Our median operative time was 90 min (interquartile range [IQR]: 75-100 min). The median estimated blood loss was 100 ml (IQR: 100-150 ml). Our conversion rate was 0.08%, comprising two procedures converted to standard laparoscopy due to robot malfunction. One hundred and forty complications were observed in 127 patients (5.08%). The following percentages of patients presented graded complications: grade 1, 2.24%; grade 2, 1.8%; grade 3a, 0.08%; grade 3b, 0.48%; grade 4a, 0.40%. There were no cases of multiple organ dysfunction or death (grades 4b and 5). There were significant decreases in the overall complication rates (p = 0.0034) and in the number of anastomotic leaks (p < 0.001) as the surgeon`s experience increased. Conclusions: RARP is a safe option for treatment of clinically localized PCa, presenting low complication rates in experienced hands. Although the robotic system provides the surgeon with enhanced vision and dexterity, proficiency is only accomplished with consistent surgical volume; complication rates demonstrated a tendency to decrease as the surgeon`s experience increased. (C) 2010 European Association of Urology. Published by Elsevier B. V. All rights reserved.

Early Identification of Infectious Complications in Bariatric Surgery by the Determination of Peritoneal and Systemic Cytokines

Obesity has become a global epidemic and bariatric surgery is one of the therapeutic tools to deal with it. Postoperative complications can occur, such as staple line dehiscence and anastomotic leaks, leading to increased patient mortality. The diagnosis of these complications is frequently difficult. The objective of the present study was to determine whether peritoneal and systemic cytokines could early detect those complications. All patients who underwent open Roux-en-Y gastric bypass from February 2007 to August 2008 were prospectively evaluated. Blood and peritoneal effluent from the drain were collected for the determination of cytokine levels. We also evaluated the clinical signs and the leukograms of the patients. A total of 107 obese patients were studied. Ninety patients had no complications; 17 had at least one infectious complication which include five cases of staple line dehiscence. Until the third postoperative day, the vital signs and the leukogram did not predict the onset of infectious complications, but the cytokines (interleukin-1 beta and interleukin-6) were early markers of these complications. Cytokines are good predictors of poor postoperative evolution in bariatric surgery since peritoneal cytokines diagnose better these infectious complications even before changes in blood count and before the occurrence of clinical manifestations.

Effects of circular myotomy on the healing of esophageal suture anastomosis: an experimental study

For esophageal reconstruction in newborns with esophageal atresia, esophageal reunion with an end-to-end anastomosis is the ideal procedure, although it may result in leaks and strictures due to tension on the suture line, mainly in cases with a wide gap between the ends. Circular myotomy (Livaditis' procedure) is the best method to elongate the proximal esophageal pouch and reduce anastomotic tension. This experimental investigation in dogs was undertaken to attempt to verify that circular myotomy decreases the anastomotic leak rate in newborns with wide gap esophageal atresia, and to analyze whether the technique promotes morphologic changes in the anastomotic scar. A pilot study demonstrated that it is necessary to resect more than 8 cm (40% of the total esophageal length) in order to obtain high leak rates. In the experimental project, such resection was performed in dogs divided into two groups (control group, anastomosis only, and experimental group, anastomosis plus circular myotomy in the proximal esophageal segment). The animals were killed in the 14th postoperative day, submitted to autopsy, and were evaluated as to the presence of leaks and strictures, as well as to the features (macroscopic and microscopic aspects) of the anastomosis. Leak rates were the same in both groups. Morphometric analysis revealed that in animals in the experimental group, the anastomotic scar was thinner than the control animals, and the isolated muscular manchette distal to the site of myotomy was replaced by fibrous tissue. Correspondingly, a decreased number of newly formed small vessels were noted in the experimental animals, compared to control animals. We concluded that circular myotomy does not decrease the incidence of anastomotic leaks, and it also promotes deleterious changes in anastomotic healing.

Extreme bariatric endoscopy: stenting to reconnect the pouch to the gastrojejunostomy after a Roux-en-Y gastric bypass

Among the possible complications of bariatric surgery, fistula and partial dehiscence of the gastric suture are well known. Reoperation often is required but results in significant morbidity. Endoscopic treatment of some bariatric complications is feasible and efficient. A modified metallic stent was placed between the gastroaesophageal junction and the alimentary jejunal limb, allowing the passage of a nasoenteric feeding tube into the jejunal limb. Endoscopy showed disruption of nearly the entire staple line at the gastric pouch. The modified stent was placed and allowed wound healing. After 31 days, the stent had migrated and was removed endoscopically. Total clousure of the fistula was reported 30 days afterward. Endoscopic treatment of some bariatric surgery complications is feasible and has been reported previously. This report presents a case of a serious leakage treated by placement of a self-expandable metal stent to bridge the fistula.

Nuove strategie endoscopiche e mininvasive nel trattamento delle fistole anastomotiche pouch-anali e colorettali : endoluminal vacuum-assisted therapy (Endosponge®) first stage vs second stage treatment

La fistola anastomotica è una delle complicanze più temute nella chirurgia colo-rettale. Le anastomosi colo-rettali basse , le colo-anali e le pouch anali hanno un rischio più elevato di sviluppare una fistola anastomotica . La terapia endoluminale a pressione negativa (Endosponge®) è stata proposta come strategia di trattamento, tuttavia, la tempistica migliore in cui attuare la procedura rimane ancora poco definita. Lo scopo dello studio è confrontare i risultati ottenuti con l'Endosponge® come trattamento di prima linea rispetto a quelli in cui è stato applicato a seguito del fallimento di ulteriori trattamenti. Lo studio retrospettivo monocentrico ha incluso pazienti con fistola anastomotica trattati con Endosponge® in un periodo di tempo compreso tra novembre 2019 e novembre 2022. L'Endosponge® è stato applicato come prima linea o come salvataggio. Il dispositivo è stato applicato nella sede della deiscenza e periodicamente sostituito fino alla guarigione. La risoluzione del leak anastomotico è stata confermata con esame endoscopico. Dei 25 pazienti inclusi, 9 sono stati sottoposti a Endosponge® come trattamento di prima linea, mentre 16 sono stati sottoposti a Endosponge® di salvataggio. La deiscenza anastomotica è stata diagnosticata dopo un intervallo di tempo mediano di 14 giorni (range 10-413) nel primo gruppo e di 38 giorni (range 11-362) nel secondo (p=0,82). L'Endosponge® è stato applicato dopo 7 giorni (range 1-60) dalla diagnosi di fistola anastomotica nel primo gruppo e dopo 76 giorni (range 6-780) nel secondo gruppo (p=0,058). La risoluzione della fistola anastomotica è stata ottenuta in una percentuale di casi maggiore nel primo gruppo rispetto al secondo 88,9% vs 37,6% (p =0,033). Lo studio conferma l'efficacia dell'Endosponge® nel trattamento delle fistole anastomotiche colorettali basse quando utilizzato precocemente e come trattamento di prima linea.

Influence of Modified Posterior Reconstruction of the Rhabdosphincter on Early Recovery of Continence and Anastomotic Leakage Rates after Robot-Assisted Radical Prostatectomy

Background: Posterior reconstruction (PR) of the rhabdosphincter has been previously described during retropubic radical prostatectomy, and shorter times to return of urinary continence were reported using this technical modification. This technique has also been applied during robot-assisted radical prostatectomy (RARP); however, contradictory results have been reported. Objective: We describe here a modified technique for PR of the rhabdosphincter during RARP and report its impact on early recovery of urinary continence and on cystographic leakage rates. Design, setting, and participants: We analyzed 803 consecutive patients who underwent RARP by a single surgeon over a 12-mo period: 330 without performing PR and 473 with PR. Surgical procedure: The reconstruction was performed using two 6-in 3-0 Poliglecaprone sutures tied together. The free edge of the remaining Denonvillier`s fascia was identified after prostatectomy and approximated to the posterior aspect of the rhabdosphincter and the posterior median raphe using one arm of the continuous suture. The second layer of the reconstruction was then performed with the other arm of the suture, approximating the posterior lip of the bladder neck and vesicoprostatic muscle to the posterior urethral edge. Measurements: Continence rates were assessed with a self-administrated, validated questionnaire (Expanded Prostate Cancer Index Composite) at 1, 4, 12, and 24 wk after catheter removal. Continence was defined as the use of ""no absorbent pads."" Cystogram was performed in all patients on postoperative day 4 or 5 before catheter removal. Results and limitations: There was no significant difference between the groups with respect to patient age, body mass index, prostate-specific antigen levels, prostate weight, American Urological Association symptom score, estimated blood loss, operative time, number of nerve-sparing procedures, and days with catheter. In the PR group, the continence rates at 1, 4, 12, and 24 wk postoperatively were 22.7%, 42.7%, 91.8%, and 96.3%, respectively; in the non-PR group, the continence rates were 28.7%, 51.6%, 91.1%, and 97%, respectively. The modified PR technique resulted in significantly higher continence rates at 1 and 4 wk after catheter removal (p = 0.048 and 0.016, respectively), although the continence rates at 12 and 24 wk were not significantly affected (p = 0.908 and p = 0.741, respectively). The median interval to recovery of continence was also statistically significantly shorter in the PR group (median: 4 wk; 95% confidence interval [CI]: 3.39-4.61) when compared to the non-PR group (median: 6 wk; 95% CI: 5.18-6.82; log-rank test, p = 0.037). Finally, the incidence of cystographic leaks was lower in the PR group (0.4% vs 2.1%; p = 0.036). Although the patients` baseline characteristics were similar between the groups, the patients were not preoperatively randomized and unknown confounding factors may have influenced the results. Conclusions: Our modified PR combines the benefits of early recovery of continence reported with the original PR technique with a reinforced watertight closure of the posterior anastomotic wall. Shorter interval to recovery of continence and lower incidence of cystographic leaks were demonstrated with our PR technique when compared to RARP with no reconstruction. (C) 2010 European Association of Urology. Published by Elsevier B.V. All rights reserved.

Prospective and Randomized Comparison of Two Techniques of Staple Line Reinforcement During Open Roux-en-Y Gastric Bypass: Oversewing and Bioabsorbable Seamguard (R)

Aims: Surgical staple line dehiscence usually leads to severe complications. Several techniques and materials have been used to reinforce this stapling and thus reduce the related complications. The objective was to compare safety of two types of anastomotic reinforcement in open gastric bypass. Methods: A prospective, randomized study comparing an extraluminal suture, fibrin glue, and a nonpermanent buttressing material, Seamguard (R), for staple line reinforcement. Fibrin glue was excluded from the study and analysis after two leaks, requiring surgical reintervention, antibiotic therapy, and prolonged patient hospitalization. Results: Twenty patients were assigned to the suture and Seamguard reinforcement groups. The groups were similar in terms of preoperative characteristics. No staple line dehiscence occurred in the two groups, whereas two cases of dehiscence occurred in the fibrin glue group. No mortality occurred and surgical time was statistically similar for both techniques. Seamguard made the surgery more expensive. Conclusion: In our service, staple line reinforcement in open bariatric surgery with oversewing or Seamguard was considered to be safe. Seamguard application was considered to be easier than oversewing, but more expensive.

Long-Term Results of Percutaneous Bilioenteric Anastomotic Stricture Treatment in Liver-Transplanted Children

The purpose of this study was to evaluate the mid- and long-term results of percutaneous transhepatic cholangiography (PTC) and biliary drainage in children with isolated bilioenteric anastomotic stenosis (BAS) after pediatric liver transplantation. Sixty-four children underwent PTC from March 1993 to May 2008. Nineteen cholangiograms were normal; 10 showed intrahepatic biliary stenosis and BAS, and 35 showed isolated BAS. Cadaveric grafts were used in 19 and living donor grafts in 16 patients. Four patients received a whole liver, and 31 patients received a left lobe or left lateral segment. Roux-en-Y hepaticojejunostomy was performed in all patients. Indication for PTC was based on clinical, laboratory, and histopathologic findings. In patients with isolated BAS, dilation and biliary catheter placement, with changes every 2 months, were performed. Patients were separated into 4 groups according to number of treatment sessions required. The drainage catheter was removed if cholangiogram showed no significant residual stenosis and normal biliary emptying time after a minimum of 6 months. The relationship between risk factors (recipient`s weight < 10 kg, previous exposure to Cytomegalovirus, donor-recipient sex and weight relations, autoimmune disease as indication for transplantion, previous Kasai`s surgery, use of reduced liver grafts, chronic or acute rejection occurrence) and treatment was evaluated. Before PTC, fever was observed in 46%, biliary dilation in 23%, increased bilirubin in 57%, and increased gamma-glutamyltransferase (GGT) in 100% of patients. In the group with BAS, 24 of 35 (69%) patients had histopathologic findings of cholestasis as did 9 of 19 (47%) patients in the group with normal PTC. Of the 35 patients, 23 (65.7%) needed 1 (group I), 7 needed 2 (group II), 4 needed 3 (group III), and 1 needed 4 treatment sessions (group IV). The best results were observed after 1 treatment session, and the mean duration of catheter placement and replacement was 10 months. The primary patency rate was 61.2%, and the recurrence rate was 34.3% (group I). Seven patients (7 of 35; 20%) had their stricture treated with a second treatment session (group II). The average drainage time in group II was 24 months. During a period > 20 months, 4 patients (4 of 35; 11.4%) required 1 additional treatment session (group III), and 1 patient (1 of 35; 2.9%) had a catheter placed at the end of the study period (group IV). Drainage time in group I was significantly shorter than those in groups II, III, and IV (p < 0.05). There was no statistically significant relation between therapeutic response and the selected risk factors (p > 0.05). The majority of complications, such as catheter displacement and leakage, were classified as minor; however, 2 patients (5.7%) with hemobilia were noted. Complications increased according to the need for reintervention. In conclusion, balloon dilation and percutaneous drainage placement is safe and effective, and it has long-term patency for children with BAS after liver transplantation. Because of prolonged treatment time, reintervention may be necessary, thereby increasing the complication rate. Balloon dilation and percutaneous drainage placement should be considered as the first treatment option because of its minimally invasive nature.

Bilevel vs ICU Ventilators Providing Noninvasive Ventilation: Effect of System Leaks A COPD Lung Model Comparison

Background: Noninvasive positive-pressure ventilation (NPPV) modes are currently available on bilevel and ICU ventilators. However, little data comparing the performance of the NPPV modes on these ventilators are available. Methods: In an experimental bench study, the ability of nine ICU ventilators to function in the presence of leaks was compared with a bilevel ventilator using the IngMar ASL5000 lung simulator (IngMar Medical; Pittsburgh, PA) set at a compliance of 60 mL/cm H(2)O, an inspiratory resistance of 10 cm H(2)O/L/s, an expiratory resistance of 20 cm H(2)O/L/s, and a respiratory rate of 15 breaths/min. All of the ventilators were set at 12 cm H(2)O pressure support and 5 cm H(2)O positive end-expiratory pressure. The data were collected at baseline and at three customized leaks. Main results: At baseline, all of the ventilators were able to deliver adequate tidal volumes, to maintain airway pressure, and to synchronize with the simulator, without missed efforts or auto-triggering. As the leak was increased, all of the ventilators (except the Vision [Respironics; Murrysville, PA] and Servo I [Maquet; Solna, Sweden]) needed adjustment of sensitivity or cycling criteria to maintain adequate ventilation, and some transitioned to backup ventilation. Significant differences in triggering and cycling were observed between the Servo I and the Vision ventilators. Conclusions: The Vision and Servo I were the only ventilators that required no adjustments as they adapted to increasing leaks. There were differences in performance between these two ventilators, although the clinical significance of these differences is unclear. Clinicians should be aware that in the presence of leaks, most ICU ventilators require adjustments to maintain an adequate tidal volume. (CHEST 2009; 136:448-456)

Improved surgical technique for laparoscopic Roux-en-Y gastric bypass reduces complications at the gastrojejunostomy.

Roux-en-Y gastric bypass (RYGBP) is one of the most commonly performed bariatric procedures for morbidly obese patients. It is associated with effective long-term weight loss, but can lead to significant complications, especially at the gastrojejunostomy (GJS). All the patients undergoing laparoscopic RYGBP at one of our two institutions were included in this study. The prospectively collected data were reviewed retrospectively for the purpose of this study, in which we compared two different techniques for the construction of the GJS and their effects on the incidence of complications. In group A, anastomosis was performed on the posterior aspect of the gastric pouch. In group B, it was performed across the staple line used to form the gastric pouch. A 21-mm circular stapler was used in all patients. A total of 1,128 patients were included between June 1999 and September 2009-639 in group A and 488 in group B. Sixty patients developed a total of 65 complications at the GJS, with 14 (1.2%) leaks, 42 (3.7%) strictures, and 9 (0.8%) marginal ulcers. Leaks (0.2% versus 2%, p = 0.005) and strictures (0.8% versus 5.9%, p < 0.0001) were significantly fewer in group B than in group A. Improved surgical technique, as we propose, with the GJS across the staple line used to form the gastric pouch, significantly reduces the rate of anastomotic complications at the GJS. A circular 21-mm stapler can be used with a low complication rate, and especially a low stricture rate. Additional methods to limit complications at the GJS are probably not routinely warranted.

Anastomotic longitudinal stress due to modification of arterial longitudinal properties after anastomosis.

BACKGROUND: In our hands, in vivo segmental vessel length changes up to 5% because of blood pressure: increasing in arterial pressure is associated to decrease in segmental vessel length. METHODS AND MATERIAL: Using two piezoelectric crystals sutured on vessel wall and a high fidelity pressure probe, we recorded artery length variations as function of blood pressure, before and after an end-to-end anastomosis on four pigs carotid arteries. RESULTS: Mean arterial pressure before anastomosis = 73 mmHg (+/- 12); mean arterial pressure after anastomosis = 91 mmHg (+/- 14); mean crystals displacement before anastomosis during systole = -0.21 mm; mean crystals displacement after anastomosis during systole = +0.24 mm; mean distance between crystals before anastomosis = 12.3 mm (+/- 0.8) and after anastomosis = 11.2 mm (+/- 0.5). CONCLUSIONS: In the acute phase following an end-to-end anastomosis, an increase in blood pressure causes increasing in vessel length, with an exponential correlation. The anastomosis is constantly subjected to a longitudinal traction whose magnitude depends on blood pressure.

A modular magnetic anastomotic device for minimally invasive digestive anastomosis: proof of concept and preliminary data in the pig model.

BACKGROUND: The aim of our study was to assess the feasibility of minimally invasive digestive anastomosis using a modular flexible magnetic anastomotic device made up of a set of two flexible chains of magnetic elements. The assembly possesses a non-deployed linear configuration which allows it to be introduced through a dedicated small-sized applicator into the bowel where it takes the deployed form. A centering suture allows the mating between the two parts to be controlled in order to include the viscerotomy between the two magnetic rings and the connected viscera. METHODS AND PROCEDURES: Eight pigs were involved in a 2-week survival experimental study. In five colorectal anastomoses, the proximal device was inserted by a percutaneous endoscopic technique, and the colon was divided below the magnet. The distal magnet was delivered transanally to connect with the proximal magnet. In three jejunojejunostomies, the first magnetic chain was injected in its linear configuration through a small enterotomy. Once delivered, the device self-assembled into a ring shape. A second magnet was injected more distally through the same port. The centering sutures were tied together extracorporeally and, using a knot pusher, magnets were connected. Ex vivo strain testing to determine the compression force delivered by the magnetic device, burst pressure of the anastomosis, and histology were performed. RESULTS: Mean operative time including endoscopy was 69.2 ± 21.9 min, and average time to full patency was 5 days for colorectal anastomosis. Operative times for jejunojejunostomies were 125, 80, and 35 min, respectively. The postoperative period was uneventful. Burst pressure of all anastomoses was ≥ 110 mmHg. Mean strain force to detach the devices was 6.1 ± 0.98 and 12.88 ± 1.34 N in colorectal and jejunojejunal connections, respectively. Pathology showed a mild-to-moderate inflammation score. CONCLUSIONS: The modular magnetic system showed enormous potential to create minimally invasive digestive anastomoses, and may represent an alternative to stapled anastomoses, being easy to deliver, effective, and low cost.

Impact of preoperative embolization of the left gastric artery on the rate of anastomotic leakage after esophageal resection

Objective: Impaired blood flow of the gastric tube represents a major cause of anastomotic leakage after esophageal resection. In order to improve local vascularisation, preoperative embolization (PE) of the left gastric artery has recently been proposed. The aimof this study was to assess our initial experience of this novel approach with a particular focus on anastomotic leakage.Methods: A consecutive series of 102 patients (81 male, 21 female, median age 64 years) underwent resection (82 Ivor-Lewis procedures, 9 transhiatal resections, 11 triple incisions) for esophageal malignancies at our institution from 2000 to 2009. Since 2004, PE was used selectively in 19 patients 21 days prior to elective esophagectomy. Selection criteria were normal gastric vascular anatomy, no pre-existing vascular disease, i.e. atheromatosis of the celiac trunk or superior mesenteric artery, and resectability of the tumor. PE was performed under local anesthesia on a dedicated system in a standard fashion. Following percutaneous transfemoral visceral angiography to identify gastric vascular anatomy, embolization was performed either with 5-F or with coaxial 3-F catheters and fibered metal coils. We analyzed retrospectively patient's data, operative data, and outcome from a prospective database.Results: The overall anastomotic leakage rate was 18・6% (19/102 patients); cervical anastomosis had a leak rate of 25% compared to intrathoracic anastomosis leak rate of 18・2%. While 17 of 83 patients without PE developed anastomotic leakage (20・5%), there were only 2 of 19 patients after PE revealing an anastomotic leakage (10・5%). Otherwise, patients with PE had no more other complications. There was only one PE-related complication (i.e. partial splenic necrosis).Mean hospital stay was 25 days versus 27 days for patients with PE and without PE, respectively. The mortality rate was 7・8% (8/102 patients), whereby four deaths were related to anastomotic leakage (1 and 3 patients with PE and without PE, respectively).Conclusion: PE is an interesting novel approach to improve gastric blood flow in order to minimize anastomotic leakage. Its application is safe and technically easy. Our preliminary experience revealed a decrease of the anastomotic leakage rate of almost 50%.