Curative endoscopic closure of a chronic esophagopleural fistula after Boerhaave syndrome using an over-the-scope clip: A case report

Abstract

BACKGROUND

Spontaneous esophageal rupture (Boerhaave syndrome) remains a life-threatening condition associated with high mortality rates. In recent years, various treatment strategies for it, including conservative and endoscopic approaches, have been increasingly reported. However, chronic esophagopleural fistula developing after conservative management is often difficult to treat, and no standard therapeutic strategy for it has been established. We report a case of a refractory chronic esophagopleural fistula successfully treated using an over-the-scope clip (OTSC).

CASE SUMMARY

A 65-year-old man underwent conservative management of spontaneous esophageal rupture and was initially discharged after clinical improvement. Subsequently, he experienced recurrent left-sided pneumonia and empyema. Approximately 1 year after onset, he was diagnosed with a refractory esophagopleural fistula and referred to our department. Because of severe bilateral emphysema, surgical intervention was considered high-risk, and endoscopic treatment was selected. Endoscopy revealed a 2 mm fistulous opening on the left wall just above the esophagogastric junction. The fistula was successfully closed using a single OTSC. Contrast esophagography performed on postoperative day 14 confirmed complete closure, and no recurrence of empyema was observed during 1-year follow-up.

CONCLUSION

OTSC can effectively treat refractory chronic fistulas developing after spontaneous esophageal rupture in patients for whom surgical intervention is high-risk.

Key Words: Boerhaave syndrome; Esophagopleural fistula; Over-the-scope clip; Chronic fistula; Endoscopic closure; Conservative management; Case report

Core Tip: The development of a chronic esophagopleural fistula following conservative management of Boerhaave syndrome is rare, such fistulas are difficult to treat, particularly in patients who are poor candidates for surgical intervention. We report a case in which a small chronic esophagopleural fistula was successfully closed using an over-the-scope clip combined with adequate pleural drainage and infection control. This case suggests that with appropriate patient selection and comprehensive management, over-the-scope clip can provide a minimally invasive and effective treatment option for chronic esophageal fistulas in high-risk surgical candidates.

INTRODUCTION

Spontaneous esophageal rupture (Boerhaave syndrome) is a full-thickness perforation of the esophagus caused by a sudden increase in intraesophageal pressure, typically associated with forceful vomiting or severe retching. Although rare, with an estimated incidence of approximately only three cases per million people per year, it is considered one of the most lethal gastrointestinal perforations. Mortality rates range from 20% to 40%, largely depending on the timing of diagnosis and therapeutic intervention, and untreated cases are reported to be almost uniformly fatal[1-3]. Contrast-enhanced chest computed tomography (CT) is the most useful imaging modality for the diagnosis of Boerhaave syndrome because it enables the detection of mediastinal emphysema, pleural effusion, and extraluminal contrast leakage. The fundamental principles of management include: (1) Closure of the perforation; (2) Control of mediastinal and thoracic infections, and (3) Adequate nutritional support. Traditionally, surgical repair with primary suture closure and thoracic drainage via thoracotomy has been regarded as the standard treatment. However, in cases accompanied by intrathoracic contamination, surgical intervention can be highly invasive and is associated with a substantial risk of postoperative respiratory complications.

Accordingly, recent treatment strategies have increasingly emphasized individualized management based on patient background and disease severity, combining conservative management, minimally invasive surgery, and endoscopic interventions as appropriate[3-10]. Conservative management may be effective in selected patients with contained perforation and a stable systemic status, but chronic complications, particularly chronic esophagopleural fistulas, remain difficult to treat. Such fistulas may lead to recurrent pneumonia and empyema and often lack an established standard therapeutic strategy.

With advances in endoscopic techniques, minimally invasive approaches using self-expandable metallic stents or the over-the-scope clip (OTSC) system have gained attention as alternative therapeutic options. OTSC provides strong full-thickness tissue approximation and has demonstrated efficacy in the closure of perforations and fistulas. However, its role in the management of chronic esophagopleural fistulous lesions developing after conservative treatment of Boerhaave syndrome has yet to be fully defined, particularly in patients who are poor candidates for surgery[11,12]. We report such a case, which was successfully treated with OTSC.

CASE PRESENTATION

Chief complaints

Recurrent left-sided empyema approximately 1 year after spontaneous esophageal rupture.

History of present illness

A 65-year-old man with a history of chronic obstructive pulmonary disease was referred to our department for treatment of a refractory chronic esophagopleural fistula associated with recurrent left-sided empyema approximately 1 year after an episode of spontaneous esophageal rupture (Boerhaave syndrome). At the time of the initial event, he presented with sudden chest and back pain and was diagnosed with spontaneous esophageal rupture based on chest radiography, contrast esophagography, and CT (Figure 1). Because the patient developed respiratory failure and shock due to massive intrathoracic contamination and was considered a high-risk candidate for general anesthesia, conservative management was selected. He was treated with mechanical ventilation, esophageal decompression via a nasogastric tube, left pleural drainage, and intravenous antibiotic therapy. The esophageal perforation closed on day 59 after onset, and the patient was discharged on day 149 after symptom onset following clinical improvement.

Figure 1 Contrast-enhanced computed tomography findings at the onset of spontaneous esophageal rupture. A: A large amount of fluid was collected in the left thoracic cavity with left-lung collapse and rightward mediastinal shift, and a nasogastric tube for decompression was in place; B: Continuity between the lower esophageal lumen and the left pleural cavity suggestive of esophageal perforation; the ruptured esophageal wall is indicated by an arrow. No mediastinal or subcutaneous emphysema was observed; C: Extensive fluid was collected in the left thoracic cavity, with compression of the left lung; D: Suspected fistulous communication between the lower esophagus and the left pleural cavity; the ruptured esophageal wall is indicated by an arrow. Fluid accumulation with an air–fluid level was present in the left thoracic cavity.

After discharge, the patient developed recurrent left-sided pneumonia and empyema and required three-time hospitalizations at another institution. Further evaluation revealed persistent communication between the esophagus and the left pleural cavity, leading to the diagnosis of a refractory chronic esophagopleural fistula. The patient was referred to our department for surgical repair, but because of the presence of severe bilateral emphysema, thoracotomy was considered high-risk by the anesthesiology team, and endoscopic treatment was chosen.

History of past illness

The patient had a history of chronic obstructive pulmonary disease.

Personal and family history

The patient had no remarkable personal or family history relevant to esophageal disease.

Physical examination

Upon admission to our department, the patient’s height was 172.0 cm, the body weight was 46.0 kg, the body temperature was 36.1 °C, the blood pressure was 17.2/10.8 KPa, and the heart rate was 70 beats/minute. Physical examination revealed no conjunctival pallor or jaundice. Cardiac auscultation demonstrated a regular rhythm without murmurs. The lung sounds were clear. The abdomen was soft and non-tender without distension. No peripheral edema was observed. Approximately 100 mL/day of purulent discharge mixed with saliva-like mucus was drained from the left pleural tube placed at the previous hospital.

Laboratory examinations

The laboratory tests conducted showed a 7030/μL white blood cell count (normal range: 3300-8600 /μL), 11.8 g/dL hemoglobin level (normal range: 13.7-16.8 g/dL), 650000/μL platelet count (normal range: 158000-348000 /μL), 0.89 mg/dL serum creatinine level (normal range: 0.65-1.07 mg/dL), 9.8 mg/dL blood urea nitrogen level (normal range: 8.0-20.0 mg/dL), 14 IU/L aspartate aminotransferase level (normal range: 13-30 IU/L), 11 IU/L alanine aminotransferase level (normal range: 10-42 IU/L), and 0.10 mg/dL C-reactive protein level (normal range: 0.01-0.14 mg/dL). The culture of the pleural fluid yielded Pseudomonas aeruginosa.

Imaging examinations

Plain chest CT was performed at the time of transfer using a multidetector CT scanner. Axial images were acquired, and multiplanar reconstructions, including coronal views, were generated to assess the extent of the pleural abscess, the position of the drainage catheter, and the anatomical relationship between the lower esophagus and the left thoracic cavity. Because the purpose of this examination was to evaluate the presence of empyema and structural changes rather than active contrast leakage, contrast-enhanced CT was not performed at this stage. The CT images demonstrated an encapsulated fluid collection with an air-fluid level in the left thoracic cavity, consistent with empyema. Severe bilateral emphysematous changes were also observed (Figure 2A). The previously placed drainage tube was located within the abscess cavity in the left lower thorax (Figure 2B).

Figure 2 Plain computed tomography findings at the time of transfer to our hospital, 1 year after onset. A: Recurrent fluid was collected with an air-fluid level in the left thoracic cavity and left-lung collapse; severe bilateral emphysematous changes were evident; B: Encapsulated fluid was collected with relatively high attenuation and intralesional free air in the left thoracic cavity, consistent with a pleural abscess; the abscess cavity in the thoracic cavity is indicated by an arrow. A chest tube was inserted into the abscess cavity for drainage, yielding saliva-like clear mucus and purulent discharge.

Upper gastrointestinal endoscopy revealed a well-demarcated fistulous opening approximately 2 mm in diameter on the left wall, just above the esophagogastric junction. The surrounding mucosa showed only mild erythema, without marked edema, necrosis, or severe fibrosis. The fistula margins were clearly identifiable, and tissue mobility was preserved, suggesting suitability for OTSC closure (Figure 3A).

Figure 3 Endoscopic and contrast findings at the time of transfer to our hospital (1 year after onset). A: Upper gastrointestinal endoscopy revealed a well-demarcated fistulous opening approximately 2 mm in diameter on the left wall just above the esophagogastric junction; the surrounding mucosa showed only mild erythema, without marked edema or necrotic changes; the esophagopleural fistula is indicated by an arrow; B: Contrast injection through a catheter inserted into the fistula demonstrated immediate flow of the contrast medium into the left pleural abscess cavity, confirming persistent communication of the esophagopleural fistula.

Contrast esophagography was performed using a water-soluble contrast agent under fluoroscopic guidance. The examination was conducted in the standing position, with additional oblique views to optimize the visualization of the lower esophagus. Leakage was assessed in real time by observing the passage of contrast medium from the esophageal lumen into the left pleural cavity. A small amount of contrast leakage from the left wall just above the esophagogastric junction into the left thoracic cavity was observed, confirming persistent communication of the esophagopleural fistula (Figure 3B). The leakage was localized, and no massive extravasation or mediastinal spread was detected. No aspiration or bronchial communication was observed during contrast administration.

Based on the clinical, radiological, and laboratory findings, the severity of empyema was assessed using the RAPID score[13], which incorporates renal function, age, purulence, infection source, and dietary factors, and a moderate- to high-risk category was suggested. In addition, the patient’s chronic obstructive pulmonary disease was severe according to the Global Initiative for Chronic Obstructive Lung Disease classification system[14], in conjunction with the marked bilateral emphysematous changes observed on CT. The presence of persistent pleural infection and reduced pulmonary reserve was considered to pose a high operative risk, particularly with regard to perioperative respiratory complications. Based on these pretreatment assessments, surgical intervention was deemed high-risk, and a less invasive endoscopic approach was selected.

FINAL DIAGNOSIS

Chronic esophagopleural fistula following conservative management of Boerhaave syndrome.

TREATMENT

Because the fistula measured approximately 2 mm and the surrounding inflammation was minimal, full-thickness capture using OTSC was considered feasible. OTSC was selected over alternative endoscopic therapies (e.g., stenting, endoscopic vacuum therapy, or fibrin-based closure) because the defect was very small, the margins were viable and mobile, and secure full-thickness closure was considered achievable without the need for luminal diversion.

Under conscious sedation, upper gastrointestinal endoscopy was performed. An OTSC system (Ovesco Endoscopy, Tübingen, Germany) was mounted on the tip of the endoscope. To ensure secure closure of the fistula, a t type (trauma clip) with an 11 mm clip width and a 21 mm cap diameter was selected. The fistula was carefully positioned in the center of the endoscopic view with adequate insufflation to optimize visualization (Figure 4A). The fistulous opening and surrounding mucosa were suctioned into the OTSC cap, ensuring sufficient inversion of the tissue (Figure 4B). Because the fistula margins were clearly identifiable and tissue mobility was preserved, ancillary devices (e.g., a grasper or anchor) were not required. After confirmation of adequate tissue capture, the OTSC was deployed to achieve full-thickness closure (Figure 4C). Immediate post-deployment inspection demonstrated that the fistula was centered within the clip and that the surrounding mucosa was securely approximated. No immediate leakage was observed, and additional clips were not required (Figure 4D). Definitive closure was achieved using a single OTSC.

Figure 4 Endoscopic closure using an over-the-scope clip. A: Endoscopic view of the 2-mm fistula located on the left wall just above the esophagogastric junction; B: The fistula and surrounding mucosa were suctioned into the over-the-scope clip (OTSC) cap to achieve adequate tissue capture; C: The OTSC was deployed after confirmation of sufficient tissue inversion into the cap, achieving full-thickness closure; D: Immediate water-soluble esophagography after OTSC placement demonstrated no visualization of the fistula and no leakage into the left pleural cavity; additional clips were not required; the esophagopleural fistula is indicated by an arrow.

To ensure adequate infection control, the left pleural drain was exchanged, and continuous suction was maintained. Pleural lavage with 200 mL of saline twice daily was performed. The lavage protocol was selected to maintain drain patency and reduce the burden of pleural contamination while avoiding excessive intrathoracic fluid retention. Because pleural fluid culture yielded Pseudomonas aeruginosa, intravenous tazobactam/piperacillin (13.5 g/day) was administered from the day of transfer until postoperative day 14. Infection control was achieved in parallel with fistula closure rather than as a completely separate step. Because chronic esophagopleural fistulas carry a high risk of recurrence if infection persists, a comprehensive strategy combining mechanical closure and continuous pleural drainage and lavage was adopted.

Early enteral nutrition was resumed the day after OTSC placement. The contrast esophagography performed on postoperative day 14 demonstrated complete closure without leakage (Figure 5A). Oral intake was resumed on postoperative day 21. The OTSC was spontaneously dislodged on postoperative day 27, and the clip was naturally expelled via the rectum (Figure 5B). Follow-up endoscopy on postoperative day 35 confirmed complete healing of the fistula, with progressive mucosal reepithelialization (Figure 5C). Chest radiography also showed improvement of the pulmonary opacities, corresponding to the resolution of empyema (Figure 5D). The patient was discharged on postoperative day 39 (Figure 6).

Figure 5 Follow-up imaging after over-the-scope clip placement. A: Water-soluble contrast esophagography performed 14 days after over-the-scope clip (OTSC) placement showed no leakage into the thoracic cavity; B: Water-soluble contrast esophagography performed 27 days after OTSC placement showed no leakage despite spontaneous OTSC dislodgement; the previously existing fistula site, now closed, is indicated by an arrow; C: Follow-up endoscopic view confirming clip dislodgement and progressive mucosal healing at the previous fistula site without recurrence; D: Chest radiograph showing improved radiolucency in the left lower-lung field, corresponding to the resolution of empyema.

Figure 6 Clinical course. Timeline of the patient’s clinical course from onset of spontaneous esophageal rupture to discharge after over-the-scope clip (OTSC) treatment. The upper axis indicates the days after onset, while the lower axis indicates the days after OTSC placement. Mechanical ventilation was required from day 0 to day 53. Fistula closure was confirmed on day 59, and the patient was discharged on day 149. Approximately 1 year later, an esophagopleural fistula was diagnosed and treated with OTSC and continuous drainage. Contrast esophagography confirmed closure on day 14 after OTSC placement. Spontaneous clip dislodgement occurred on day 27 after OTSC placement without recurrence, and the patient was discharged on day 39 after OTSC placement. OTSC: Over-the-scope clip.

OUTCOME AND FOLLOW-UP

No procedure-related complications or adverse events were observed during the clinical course. During the 1-year of follow-up, contrast-enhanced CT was performed every 3 months to evaluate for recurrence of empyema, in addition to periodic chest radiography as clinically indicated. No recurrence of empyema or fistula reopening was observed, and no recurrent dysphagia, chest pain, fever, or respiratory symptoms suggestive of fistula recurrence were noted. The patient was satisfied with the minimally invasive treatment and the favorable clinical outcome without recurrence.

DISCUSSION

Spontaneous esophageal rupture (Boerhaave syndrome) remains a highly lethal clinical entity. The treatment strategy must be individualized according to the time from onset, the extent of perforation, the degree of contamination, and the patient’s overall condition. With advances in minimally invasive approaches, various therapeutic options, including conservative and endoscopic treatments, have been increasingly reported, in addition to surgical repair[3]. Although conservative management is appropriate in selected cases, late complications remain problematic[4].

Chronic esophagopleural fistulas are frequently accompanied by fibrotic changes around the fistulous tract and persistent thoracic infection. Surgical repair in such cases is often highly invasive, and in patients with severe comorbidities, especially advanced pulmonary disease, the risk of postoperative respiratory complications may be prohibitive. In the present case, the patient had severe bilateral emphysema, and surgical intervention was considered high-risk. Thus, a minimally invasive strategy emphasizing infection control and endoscopic closure was selected.

The OTSC system provides strong full-thickness tissue approximation and has demonstrated high clinical success rates for acute gastrointestinal perforations and leaks[11]. The reported clinical success rate in acute perforations is approximately 85%, whereas the outcomes in chronic fistulas are less favorable (around 50%)[12], reflecting the challenges posed by fibrosis and reduced tissue mobility. Hagel et al[15] reported that 11 of the 17 patients (64.7%) with gastrointestinal perforations requiring surgery in their study were successfully treated with OTSC, thereby avoiding surgical intervention. Successful cases were characterized by significantly smaller defect sizes (mean length 5.5 ± 1.9 mm; mean area 21.1 ± 9.1 mm²), viable tissue edges, and fewer clips required, whereas failures were associated with larger defects, necrotic or inflamed margins, and the need for multiple clips[15]. These findings suggest that a smaller fistula size, preserved tissue elasticity, and viable margins are critical determinants of successful OTSC closure. The distinctive aspects of the present case are the long interval from initial rupture to fistula treatment (approximately 1 year), the very small fistulous opening, the preserved tissue viability despite chronicity, and durable healing despite spontaneous clip dislodgement. These features distinguish this case from previously reported acute perforation or early fistula closures and support its educational value.

In the present case, the fistula measured approximately 2 mm in diameter, and endoscopic examination revealed only mild mucosal inflammation, without marked fibrosis or necrosis. These characteristics likely facilitated effective full-thickness capture and closure using a single OTSC. Nevertheless, OTSC was not used as a stand-alone therapy. Adequate infection control through continuous pleural drainage and lavage was maintained before and after clip deployment. In chronic esophagopleural fistulas, closure of the luminal defect alone may not prevent recurrence if thoracic contamination persists. The combined strategy of mechanical closure and sustained drainage likely promoted stable tissue remodeling and prevented the re-accumulation of infected fluid. Previous reports have also suggested that comprehensive approaches, including additional techniques such as argon plasma coagulation for margin freshening, may enhance outcomes in chronic fistulas[16-18], further supporting the importance of multimodal management rather than clip application alone. This case supports the hypothesis that effective infection control combined with early mechanical closure may promote durable healing even in chronic esophageal fistulas.

Alternative endoscopic options for esophageal fistulas include self-expandable metallic stents, endoscopic vacuum therapy, and tissue sealants, such as fibrin glue. However, stent therapy may be less suitable for a very small fistula because migration, patient discomfort, and the need for prolonged luminal coverage may outweigh the expected benefit[19,20]. Endoscopic vacuum therapy is effective in larger contaminated cavities but usually requires repeated endoscopic sessions and prolonged treatment[21,22]. Fibrin-based closure may be useful in selected low-output fistulas but may provide less reliable closure in the presence of persistent pleural contamination[23,24]. In the present case, OTSC was considered the most appropriate option because the defect was small, the surrounding tissue was viable and mobile, and full-thickness closure could be achieved in a single session while maintaining simultaneous pleural drainage.

The Pittsburgh esophageal perforation severity score (PSS) has been proposed as a useful tool for risk stratification in acute esophageal perforation[25]. At the initial presentation, the patient in the present case would have been classified as high-risk (PSS 6) based on non-contained perforation, respiratory distress, tachycardia, and pleural effusion on imaging. According to the treatment algorithms proposed by Schweigert et al[25], the initial choice of nonoperative management was reasonable in this context. The subsequent development of a chronic esophagopleural fistula is consistent with the increased risk of late complications in severe cases. However, PSS primarily applies to acute management and does not directly guide treatment decisions in the chronic phase. In late-presenting fistulas, careful evaluation of local tissue characteristics, infection status, and patient comorbidities is more clinically relevant than initial severity scoring[26,27].

In the present case, spontaneous dislodgement of the OTSC occurred on day 27 after placement, but no recurrence of leakage or empyema was observed. Contrast esophagography performed on day 14 had already confirmed complete closure of the fistula, suggesting that durable full-thickness healing had been achieved prior to clip detachment. In addition, continuous pleural drainage and lavage provided adequate infection control, thereby promoting stable tissue remodeling and mucosal reepithelialization. Therefore, the absence of recurrence despite clip dislodgement can be reasonably explained by definitive fistula healing before device detachment. This clinical course indicates that when sufficient infection control and drainage are ensured, OTSC can facilitate durable closure even in chronic esophagopleural fistulas, and long-term clip retention itself may not be mandatory once tissue healing has been established.

This report has several limitations. First, it describes a single case, and the generalizability of the findings is limited. Second, long-term outcomes beyond 1 year remain unknown. Further studies are required to validate the effectiveness of OTSC in chronic esophagopleural fistulas.

In summary, this case suggests three practical conditions for successful OTSC closure in chronic post-Boerhaave fistulas: A small fistulous opening, viable and mobile tissue margins, and effective pleural drainage with adequate infection control. Conversely, OTSC may be less suitable in the presence of a large defect, severely fibrotic or necrotic margins, uncontrolled sepsis, or poor endoscopic access. In such situations, stent therapy, endoscopic vacuum therapy, or surgical repair may be more appropriate.

From a practical perspective, the treatment strategy for chronic esophagopleural fistulas may be summarized as follows: First, define the fistula size and tissue quality endoscopically; second, assess thoracic contamination and drainage adequacy; and third, select OTSC when the defect is small and tissue viability is preserved, while continuing drainage and infection control throughout the periprocedural period. The present case highlights that even in chronic esophagopleural fistulas developing after the conservative treatment of spontaneous esophageal rupture, curative management may be achieved with a minimally invasive approach. Careful patient selection, assessment of fistula characteristics, and strict infection control are essential prerequisites. In high-risk surgical candidates, the combination of OTSC and adequate drainage may be an effective and safe alternative to operative repair. These findings may contribute to clinical decision-making algorithms by supporting the use of endoscopic closure combined with adequate drainage as a viable alternative to surgery in selected high-risk patients.

CONCLUSION

The development of a chronic esophagopleural fistula after the conservative management of spontaneous esophageal rupture (Boerhaave syndrome) is a challenging clinical condition, particularly in patients with severe comorbidities in whom surgical intervention is high-risk. In the present case, careful assessment of local fistula characteristics combined with adequate infection control through continuous pleural drainage and lavage, together with OTSC placement, enabled curative treatment, even in the chronic phase. In chronic post-Boerhaave fistulas, OTSC may be considered when the defect is small, the margins are viable and mobile, and pleural infection is adequately controlled with drainage.