Non-surgical treatment of esophageal perforation after pneumatic dilation for achalasia: a case series

Abstract

Esophageal perforation is the most serious complication of pneumatic dilation for achalasia and is traditionally managed by conservative therapy or surgical repair. We present four achalasia patients who underwent pneumatic dilatation, complicated by an esophageal perforation. All patients were treated successfully with endoscopic treatment: two patients with Eso-SPONGE^® vacuum therapy, in the other two patients, esophageal defects were closed endoscopically using Endoclips. The time between dilatation and detection of the perforation was less than 24 h in all cases. Non-surgical treatment resulted in a relatively short hospital stay, ranging from 5 to 10 days, and an uneventful recovery in all patients. Based on our experience, endoscopic clipping and/or vacuum therapy are relatively new, valuable, minimally invasive techniques in the management of patients with small, well-defined esophageal tears with contained leakage and should be considered as primary therapeutic option for iatrogenic perforation in achalasia.

Keywords:

• Achalasia
• esophageal
• perforation pneumatic dilation
• endoscopic treatment

Introduction

Esophageal perforation is the most serious complication of pneumatic dilation for achalasia and is traditionally managed by conservative therapy or surgical repair [1–3]. With the introduction of new endoscopic techniques in the past decade, endoscopic management as primary therapy of iatrogenic perforation has been gaining ground [4–7]. We report four achalasia patients that underwent pneumatic dilatation complicated by an esophageal perforation. Two patients were treated successfully with Eso-SPONGE^® vacuum treatment, in the other two patients the esophageal defects were closed endoscopically using endoclips. An overview of the clinical characteristics of the cases is displayed in Table 1.

Table 1. Clinical characteristics and course of treatment of esophageal perforation in four cases.

Case	Sex/ age	BMI	Weight loss > 10 kg	Symptom duration (years)	Achalasia type	IRP-4 (mmHg)	Barium Column height at 5 min (cm)	Esophageal width (cm)	Balloon size (mm)	Dilation pressure (PSI)	Site of perforation	Distance to EGJ (cm)	Length of perforation (cm)	Time until diagnosis (hours)	Extraluminal contrast on CT	Type of treatment	Hospital stay (days)	Treatment success
I	F/39	22	+	<1	1	22	3.7	3.0	40	8	Distal, L	5	3	12	Contained	EsoSPONGE	10	+
II	F/55	21	+	2	2	37	6.7	3.2	40	8	Distal, V	3	3	0	Contained	EndoClips  + esoSPONGE	7	+
III	F/70	20	+	<1	1	24	10.0	1.5	40	8	Distal, L	1	5	3	No extraluminal contrast*	EndoClips	5	+
IV	F/80	32	–	10	2	¥	2.6	4.3	35	8	Distal, R	1	4	0	Contained	EndoClips	6	+

EGJ: esophagogastric junction; I: Integrated relaxation pressure; L: left; R: right; V: ventral. *After endoscopic closure with Endoclips. ^¥Catheter did not pass the LES; therefore, IRP could not be calculated.

Case presentation

Case I

A 39-year-old woman was referred to our department with symptoms of dysphagia. High-resolution manometry and barium esophagogram revealed achalasia type I. The patient was informed of the different treatment options and shared decision was made to perform pneumatic dilatation. Because of persistent symptoms after a 30 and 35-mm Rigiflex balloon dilation, she was re-dilated with a 40-mm balloon 4 weeks after the previous session. After an uneventful procedure, the patient was discharged, but after 12 hours she presented to the emergency room with symptoms of chest pain, vomiting and dyspnea. A CT scan of the chest was performed and showed a perforation of the distal esophagus with a para-esophageal fluid collection. The patient was kept nil per mouth and intravenous benzylpenicilline, gentamicin and metronidazole were started. An urgent upper endoscopy was performed and showed a 3-cm tear, located five centimeters above the diaphragm impression on the left lateral side. An overtube was introduced and placed proximal to the defect. Subsequently, the Eso-SPONGE^® (B Braun Medical AG) was unfolded in the esophageal lumen. Since the chest CT scan revealed persistent mediastinal pneumatosis, the endosponge was retracted the next day, to make sure that it completely covered the perforation. The sponge was connected to a vacuum pump system with a negative pressure of 75 mmHg and flushed three times daily. Parenteral nutrition was started. Over the course of the following days, the patient improved clinically and the sponge could be removed endoscopically after 7 days with a follow-up CT scan showing no more leakage. Endoscopic inspection showed a closed esophageal wall with the presence of some granulation tissue. The patient was discharged on a soft diet and proton pump inhibitor therapy and recovered quickly without further symptoms.

Case II

A 55-year old female patient previously diagnosed with achalasia type II for which she already had been dilated twice up to 35 mm, was admitted with an esophageal perforation after a 40-mm pneumatic dilatation. The perforation was detected after deflating the dilation balloon, revealing a 3-cm linear perforation just proximal to the Z-line (Figure 1(a)). An endoscope with cap was introduced and the perforation was closed with seven Endoclips (Olympus). The patient was kept nil per mouth and was started on the same antibiotic therapeutic protocol as case I. As the post-treatment chest CT scan showed extraluminal air and localized contrast spill in the mediastinal region, she was re-treated the next day by placement of the Eso-SPONGE^®. Parenteral nutrition was started. Symptoms resolved soon after start of the vacuum therapy. Five days later, a CT scan confirmed absence of any esophageal leakage, and the sponge could be removed endoscopically after which a healed esophageal wall with granulation tissue became apparent (Figure 1(b,c)). The patient was discharged 2 days later. At 1-month follow-up, the patient reported improvement of dysphagia and was tolerating a normal diet without regurgitation or chest pain.

Figure 1. (a–c) Endoscopic view of the esophageal defect in case II. Immediately after pneumatic dilatation, a laceration 3 cm above the Z-line was observed (a). The patient was treated with Eso-SPONGE^® vacuum therapy (b). After 5 days, endoscopic inspection showed healing of the esophageal wall defect with granulation tissue (c).

Case III

The third patient, a 70-year-old female with achalasia type I, was treated with a 40-mm diameter Rigiflex balloon because of lack of symptom relief after two previous (30 and 35 mm) dilatations. Four hours after the procedure, the patient developed severe chest pain whereupon endoscopic re-evaluation took place. A 5-cm long perforation became apparent (Figure 2(a)), which was closed with 15 endoclips (Figure 2(b,c)). An esophagogram and CT scan performed the next day did not show any esophageal leakage. The patient was started on intravenous antibiotics and a clear liquid diet along with duodenal tube feedings. The patients could be discharged 4 days later in good clinical condition.

Figure 2. (a–c) Endoscopic images of esophageal perforation in case III. The defect was located just above the esophagogastric junction (a) and was successfully closed with 15 endoclips (b and c).

Case IV

An 80-year-old female was referred to our clinic because of long-standing dysphagia. The diagnosis of achalasia type II was established by barium esophagogram and manometry. Two weeks after a first 30-mm pneumatic dilation, the patient underwent a second dilation with a 35-mm diameter balloon. Endoscopic inspection immediately after dilation showed a 4-cm linear esophageal perforation just proximal to the gastroesophageal junction. The perforation was closed endoscopically with eight endoclips. The patient was kept nil per mouth and tube-fed. The CT study performed the next day showed localized extraluminal contrast spill just proximal of the esophagogastric junction (Figure 3). Under intravenous antibiotics, symptoms resolved swiftly. A second chest CT scan 3 days later confirmed absence of extraluminal contrast and the patient was discharged 6 days post-procedure. After 12 weeks she visited our out-patient clinic and reported complete symptom remission.

Figure 3. Axial (a) and coronal (b) CT scan of the chest showing evidence of contained esophageal perforation on the right side of the distal esophagus in case IV, including extraluminal air and contrast (arrows).

Discussion

Although endoscopic pneumatic dilatation is regarded an effective and safe, nonsurgical therapeutic option for achalasia [8,9], the procedure may be complicated by esophageal perforation. Typically, this occurs at the distal part of the esophagus, just above the level of the lower esophageal sphincter where the balloon is placed [10]. With an estimated occurrence of 0.5–5% of dilations in larger series, it is the most common and serious adverse event of balloon dilatation [9,11]. In this case series, we report four female patients who all suffered a perforation in a second or third dilation session. Other demographic, manometric or radiologic characteristics however, differed widely among our patients. Naturally, we cannot draw any conclusions in terms of predictive factors for esophageal perforation based on this case series. In literature, several risk factors have been suggested to increase the risk of perforation, including age, balloon size, a more stringent protocol with multiple graded dilatations, weight loss, long-standing symptoms, and higher dilation pressures [12–16].

In patients with suspected esophageal perforation after pneumatic dilatation, a swift and decisive diagnostic approach is of utmost importance to limit diagnostic delay. Once the diagnosis is confirmed, nil per mouth, intravenous fluids, and broad-spectrum antibiotics should be initiated. Further management ranges from watchful waiting to surgical drainage or even repair. Traditionally, most esophageal perforations were managed surgically [1–3]. Primarily in patients with larger tears and free mediastinal spill, surgical repair is still considered the strategy of choice. However, surgery involves considerable risks and therefore, previous studies have suggested that a less invasive treatment approach is more appropriate in clinically stable patients with contained perforation (e.g., well-defined tears with localized extraluminal spill) [16,17]. In search of less-invasive methods to close esophageal perforations, endoscopic techniques, such as over the scope clips and esophageal stenting, have been explored over the past years. However, these techniques also come with additional risks, while results are not overwhelming [4,6,7]. Stents may dislocate in the dilated esophagus of the achalasia patient or may not close off the wall at the proximal end. Therefore, current guidelines mainly advice the use of endoclips in the endoscopic management of iatrogenic perforation [5].

In our series, esophageal defects were closed successfully in all patients; in two patients the perforation healed and symptoms resolved after closure with endoclips; two other patients were effectively treated with the relatively new method of endoscopic vacuum therapy. Although both methods have previously been reported in patients with iatrogenic esophageal perforations [18,19], this is the first series reporting these techniques in esophageal perforation caused by an endoscopic balloon dilatation in achalasia. One of the advantages of sponge vacuum therapy is the ability to clean the perforation cavity of debris using a minimally invasive technique. It enhances esophageal healing and formation of granulation tissue by reducing edema and bacterial contamination [20,21]. Of note, there are currently no comparative studies looking at the optimal endoscopic technique for iatrogenic perforations. The choice of endoscopic closure should depend on the duration of diagnostic delay, the size and location of the perforation and the endoscopic expertise available at the center. Clipping and/or vacuum therapy resulted in relatively short duration of hospitalization and an uneventful recovery in all our patients. In our center it is standard protocol to start patients on a liquid diet 3 days before the procedure. Hence, a clean esophagus without stasis and retention during the procedure might have contributed to treatment success in our series. Nonetheless, endoscopic therapy, combined with supportive treatment and careful observation, is a valuable option for treatment of patients with small (<5cm), well-defined esophageal tears with contained leakage and no signs of systemic infection. Based upon our experience, use of endoclips or sponge vacuum therapy should be considered in the treatment of iatrogenic perforation in achalasia. Prospective series and possibly comparative studies could assist to determine the definite role of these endoscopic techniques as non-surgical options of treating post-dilation perforations.

Informed consent

Informed consent was obtained for publication of patient information.

Author contributions

Guarantor of the article: Renske A.B. Oude Nijhuis, MD. Specific author contributions: R.O.N. was involved in the acquisition of patient data. R.O.N, and A.B. had a role in collecting and/or interpreting data, and drafting the manuscript. A.B, B.T, P.F and J.B. had a role in reviewing and revising the manuscript for important intellectual content. All authors approved the final draft submitted.

Disclosure statement

R.O.N., J.B., and B.T. have no competing interests. A.B. received research funding from Nutricia, Norgine and Bayer and received speaker and/or consulting fees from Laborie, EsoCap, Diversatek, Medtronic, Dr. Falk Pharma, Calypso Biotech, Thelial, Robarts, Reckett Benkiser, Regeneron, Celgene, Bayer, Norgine, AstraZeneca, Almirall and, Allergan. P.F. received research funding from Boston Scientific and received speaker and/or consulting fees fom Olympus, Cook, and Fujifilm.