Nonsurgical treatment of esophageal perforations after endoscopic palliation in advanced esophageal cancer

A Comparison of Thermal Plasma Energy Versus Argon Beam Coagulator-Induced Intestinal Injury After Vaporization in a Porcine Model

OBJECTIVES

Complete cytoreduction of ovarian cancer often requires excision or ablation of bowel serosa implants. Both argon beam coagulator (ABC) and thermal plasma energy (TPE) (PlasmaJet; PlasmaSurgical, Roswell, Ga) have been used to ablate bowel serosa implants. Our objective was to identify comparable power settings as well as determine the rate of bowel perforation, depth of thermal injury, and extent of inflammatory response with ABC versus TPE in a porcine model.

MATERIALS AND METHODS

Nine pigs underwent vaporization of small bowel and colon serosa according to assigned treatment group (TPE vs ABC) and settings (ABC: 30, 50, and 70 W; TPE: Cut 10U, 20U, and 30U and Coagulation 10U, 20U, and 30U). Animals underwent necropsy with blinded histomorphologic evaluation on days 0, 3, and 10 postprocedure to assess for presence of bowel perforation, depth of thermal injury, and extent of inflammatory response.

RESULTS

At necropsy, bowel perforation was not identified in any animals. Depth of treatment with ABC in the porcine colon was variable and unrelated to power settings whereas TPE was associated with a consistent treatment depth of 1.0 mm regardless of location or power. Treatment with ABC resulted in greater tissue coagulation and desiccation as well as increased rates of mucosal necrosis, especially at higher settings (>50 W). Treatment with TPE primarily resulted in tissue ablation and minimal mucosal necrosis at low settings (Coag 10U-20U). The inflammatory response associated with TPE treatments was interpreted as biologically benign, and less than that observed with the ABC regardless of treatment settings.

CONCLUSIONS

Both ABC and TPE effectively ablate bowel serosa in a porcine model. The TPE seems to result in a more predictable tissue effect with less inflammatory response, especially when used at low power settings such as Coag 10U or 20U. These characteristics are appealing for ablation of bowel serosa implants during ovarian cancer surgery and warrant further investigation.

A successful strategy for surgical treatment of Boerhaave's syndrome

BACKGROUND

This retrospective single-institution study presents a successful treatment strategy for Boerhaave's syndrome.

METHODS

During 1995-2008, 15 patients with spontaneous esophageal perforation were treated. Patients were grouped according to time from symptoms to referral (early, <24 h; late, >24 h). In group I (early, n = 8 patients) treatment comprised primary surgical esophageal repair in seven cases and endoscopic clipping in one case. In group II (late, n = 7 patients) treatment comprised esophagectomy without primary reconstruction (4 cases) or controlled esophagocutaneous fistula (3 cases). Measures of outcome included age (years), delay to diagnosis (h), severe sepsis on admission, mortality, and hospital and intensive care unit (ICU) stay.

RESULTS

The overall hospital mortality rate was 6.6% (1/15), being 0% (0/8) in group I and 14.2% (1/7) in group II. Patient age (49.6 vs. 68.6 years, P < 0.0001), delay to diagnosis (17.75 vs. 69 h, P < 0.0001), severe sepsis on admission (0 vs. 4, P = 0.0256), and ICU stay (4 vs. 14 days, P = 0.006) were all greater in group II.

CONCLUSIONS

Early diagnosis and carefully selected therapeutic tactics can reduce the mortality rate of Boerhaave's syndrome to an acceptably low level. Methods of organ preservation and minimally invasive techniques can be applied successfully in the treatment.

A successful strategy for surgical treatment of Boerhaave's syndrome

BACKGROUND

This retrospective single-institution study presents a successful treatment strategy for Boerhaave's syndrome.

METHODS

During 1995-2008, 15 patients with spontaneous esophageal perforation were treated. Patients were grouped according to time from symptoms to referral (early, <24 h; late, >24 h). In group I (early, n = 8 patients) treatment comprised primary surgical esophageal repair in seven cases and endoscopic clipping in one case. In group II (late, n = 7 patients) treatment comprised esophagectomy without primary reconstruction (4 cases) or controlled esophagocutaneous fistula (3 cases). Measures of outcome included age (years), delay to diagnosis (h), severe sepsis on admission, mortality, and hospital and intensive care unit (ICU) stay.

RESULTS

The overall hospital mortality rate was 6.6% (1/15), being 0% (0/8) in group I and 14.2% (1/7) in group II. Patient age (49.6 vs. 68.6 years, P < 0.0001), delay to diagnosis (17.75 vs. 69 h, P < 0.0001), severe sepsis on admission (0 vs. 4, P = 0.0256), and ICU stay (4 vs. 14 days, P = 0.006) were all greater in group II.

CONCLUSIONS

Early diagnosis and carefully selected therapeutic tactics can reduce the mortality rate of Boerhaave's syndrome to an acceptably low level. Methods of organ preservation and minimally invasive techniques can be applied successfully in the treatment.

Nonoperative treatment of four esophageal perforations with hemostatic clips

Spontaneous or iatrogenic esophageal perforations are despite advances of modern surgery and intensive care medicine still potentially life-threatening events with a considerable mortality rate. Recently, encouraging results on the sealing of esophageal perforations by placement of endoluminal prostheses were reported. However, if the perforation is very proximal (close to the larynx) or very distal (involving the cardia), the situation is to our experience unsuitable for stent therapy. In these special cases non-operative treatment is still possible by application of hemostatic metal clips. We present four cases unsuitable for stent therapy where the perforation was sealed by endoscopic clip application. All patients had an uneventful recovery. Non-operative treatment of esophageal perforations with hemostatic metal clips is feasible and safe in cases not treatable with self-expanding metal stents.

Nonoperative treatment of 15 benign esophageal perforations with self-expandable covered metal stents

BACKGROUND

Spontaneous or iatrogenic esophageal perforations after endoscopic procedures are potentially life-threatening events with a considerable mortality rate. The aim of this study was to demonstrate that a nonoperative endoscopic treatment with self-expanding metal stents may have a lower morbidity and mortality rate compared with surgical treatment.

METHODS

A nonrandomized observational study was conducted with 15 consecutive patients between January 1997 and June 2004. Benign spontaneous and iatrogenic esophageal perforations after endoscopic procedures were treated with self-expandable metal stents.

RESULTS

Seven patients (group 1) underwent stent insertion with an average time delay of 45 minutes. In 8 patients (group 2), the median delay was 123 hours. All patients in group 1 had an uneventful recovery and left hospital 5 days (range, 3 to 9) after stent insertion. One patient in group 2 (1 of 8) died of pneumonia after 6 days. In any other cases, perforations healed successfully after stent placement, but the clinical course was generally complicated with sepsis and multiple organ failure. The average hospital stay was 44 days (range, 15 to 70).

CONCLUSIONS

Immediate insertion of a self-expandable metal stent enables an excellent outcome with minimal mortality and morbidity without the need for operation. Even in cases of old esophageal perforations, sealing with self-expandable metal stents is still a good option although the clinical course is much less impressive than in early treated perforations.

In vivo characterization of colonic thermal injury caused by argon plasma coagulation

BACKGROUND

Argon plasma coagulation has been rapidly accepted for endoscopic obliteration of vascular lesions and superficial tumors. Depth of injury is thought to be limited through preferential discharge of energy to nondesiccated tissue. However, argon plasma coagulation-induced injury has not been well characterized. The aim of this study was to characterize argon plasma coagulation-induced colonic injury by using a porcine model.

METHODS

Laparotomy was performed in 6 female swine and the colon exteriorized with the subjects under general anesthesia. Lesions were made with an argon plasma coagulation probe held perpendicular and 2 mm from the mucosa. Variables studied were as follows: power (45 W, 60 W, and 75 W) and duration (1, 2, or 3 seconds; n = 11 for each power/duration combination). Injury was graded as either superficial or deep, involving the muscularis propria.

RESULTS

Circular muscle layer injury correlated closely with power (p = 0.02), duration (p = 0.001), and total energy delivered (r = 0.977). Longitudinal muscle damage was associated with duration of burn (p = 0.001) and total energy delivered (r = 0.855), but correlated poorly with power (p = 0.40). No perforations occurred. Submucosal injection of saline solution had a protective effect with reductions in circular (90% to 10%, p = 0.002) and longitudinal muscle injury (50% to 0%, p = 0.1).

CONCLUSIONS

Injury to the muscularis propria occurs at recommended settings for argon plasma coagulation. Injury correlates with power setting, duration of burn, and total energy delivery. Protective arcing to nondesiccated tissue does not appear to be significant in vivo. Submucosal injection of saline solution protects against deep injury.

Endoscopic treatment of clinically symptomatic leaks of thoracic esophageal anastomoses

BACKGROUND

The mortality of thoracic anastomotic leakage following esophageal reconstruction has been reported to be as high as 60%. Early septic fulminant suture line leaks require rethoracotomy. In addition, however, clinically symptomatic leaks may also occur 2 to 7 days after resection of the esophagus.

METHODS

Among 80 esophageal reconstructions performed between January 1994 and July 1998, a total of 7 (8.75%) clinically apparent leaks of thoracic anastomoses were observed. The standard treatment consisted of endoscopic lavage, drainage and subsequent closure of the defect by repeated intraluminal and submucosal applications of fibrin glue. In 2 patients a novel approach permitting rapid closure by plugging the fistula with a Vicryl-cylinder was tried. In 4 patients the effect of endoscopic treatment on the HLA-DR expression on monocytes was investigated and compared to 6 patients with intact anastomoses.

RESULTS

All 7 patients were successfully treated via endoscopy. The cylinder plug achieved immediate closure of the leak. The measured change in HLA-DR expression reflected the improvement in the inflammatory response and thus documented the success of endoscopic treatment.

CONCLUSIONS

Endoscopic management of thoracic leakages represents a safe and relatively noninvasive therapeutic option.

The spectrum of spontaneous and iatrogenic esophageal injury: perforations, Mallory-Weiss tears, and hematomas

Esophageal perforations, Mallory-Weiss tears, and esophageal hematoma involve traumatic injury to the esophagus. These can be iatrogenic, in particular due to esophageal instrumentation, but can also occur spontaneously. The remarkable increase in diagnostic and therapeutic endoscopy as well as esophageal surgery has made instrumentation the most common cause of esophageal perforation. In many instances, spontaneous perforations are associated with retching and vomiting, which causes a sudden increase in intraesophageal pressure. A high index of suspicion leading to rapid diagnosis and appropriate therapy are needed to optimize clinical outcomes. This article focuses on esophageal perforations, Mallory-Weiss tears, and esophageal hematomas, with emphasis on etiology, pathogenesis, clinical presentation, diagnosis, management, and prevention.