Pneumatic dilatation (PD) has been successfully used for treatment of achalasia in adults and children (1-10). The goal of the procedure is the improvement of esophageal clearance by forceful partial rupture of the lower esophageal sphincter (LES). A recognized complication of the procedure is esophageal perforation (EP).
The reported frequency of EP caused by the pneumatic dilatation procedure in achalasia in adults varies from 0% to 18% (4,5,11-16) but in most series does not exceed 5%. The standard of practice has been to manage these perforations with surgical closure, even though there is clinical experience to support medical management (15,17-22). We report three cases of successful medical management of esophageal perforation after PD in children with achalasia.
Between 1981 and 1996, 44 PD procedures were performed as primary treatment of achalasia in 22 children with ages ranging from 16 months to 18 years. Six additional dilatations were done for recurrent symptoms after a Heller myotomy with or without a Nissen fundoplication (4 and 2 patients, respectively). Esophageal perforation occurred in 3 of our 50 procedures (6%).
The first patient was a 3.5-year-old boy in whom achalasia was diagnosed at 18 months of age, based on a history of vomiting after coughing, which began at 11 months of age; a positive family history of achalasia (his older sister underwent a Heller myotomy with a loose fundoplication when she was 11 years old; parents are first cousins) and characteristic findings in an upper gastrointestinal series, which showed a mildly dilated esophagus with tertiary contractions and delayed esophageal emptying through a tapered distal esophagus. An upper gastrointestinal endoscopy showed no evidence of esophagitis or stricture.
At the time of diagnosis the patient underwent esophageal dilatation elsewhere using Fogarty catheters sizes 5 through 10 mm. One year later his symptoms recurred and he was referred to our hospital for pneumatic dilatation. The first PD has been done under general anesthesia using a Rigiflex (Microvasive, Milford, MA, U.S.A.) 30-mm dilator with 12 psi pressure for 1 minute, repeated twice. A small amount of blood was seen on the dilator. Endoscopy after dilatation revealed a few superficial abrasions without active bleeding. During fluoroscopy, dilute Gastrografin (Bracco Diagnostics, Princeton, NJ, U.S.A.) was injected through the esophageal tube. There was no evidence of extravasation or mucosal tear. The patient tolerated the procedure well.
His symptoms did not improve and he underwent PD again within 2 weeks of the first procedure. The second procedure was done with a Rigiflex 35-mm dilator using 10 psi for 1 minute, repeated twice.
Mild bleeding from small erosions were seen on esophagoscopy but no free mediastinal air or extravasation of Gastrografin from the distal esophagus was found during fluoroscopy or on esophagogram. The only positive finding was a small outpouching of the distal esophagus caused by either mucosal tear or supporting muscle layers (Fig. 1). During recovery, the patient complained of mild pain substernally and vomited twice, the first time with a small amount of blood. He was admitted for observation: Oral feeding was not allowed, and he received intravenous hydration and intravenous penicillin 312.5 mg every 6 hours. The patient remained afebrile but was irritable overnight. In the morning he was asymptomatic. However, a repeat chest film was interpreted as positive for a small amount of air adjacent to the esophageal wall. A few hours later he developed fever to 39.5 °C, and mediastinal air was diagnosed. An esophagogram showed significant extravasation of Gastrografin (Fig. 2).
The patient was given total parenteral nutrition for 10 days with Ranitidine to minimize acid secretion. Triple antibiotics: metronidazole 125 mg every 6 hours, gentamycin 25 mg every 8 hours, and mezlocillin 850 mg every 6 hours were given to sterilize the mediastinum. Substernal pain and fever ceased in 24 and 48 hours, respectively. Leukocytosis (12.6 × 103) reverted to normal levels on the third day of treatment. Repeat chest film on the second day of admission showed bilateral, increased density in the lung base and right pleural effusion. Three days later these findings had resolved.
Complete healing of the esophageal perforation was documented by esophagogram on the ninth day of the hospital stay (Fig. 3).
The second patient was an 18-year-old male with a 10-month history of dysphagia for solids. He described a feeling of food getting stuck in his midchest when he swallowed. His previous medical history was significant for pneumonia at 11 and 13 years of age. The family history was remarkable for esophageal carcinoma in the paternal grandfather.
His initial evaluation for dysphagia was done at an outside hospital and results showed mild narrowing of the distal esophagus above the gastroesophageal junction without endoscopic or histologic evidence of inflammation or tumor. During the same endoscopic procedure the patient underwent esophageal dilatation using a 50 French Maloney dilator. Results of a subsequent motility study showed esophageal spasm with partial relaxation of the lower esophageal sphincter with swallowing and a hypertensive LES. The patient was given one tablet of Procardia 1 10 mg, twice daily, with a good initial response. Subsequently, his symptoms returned and progressed despite increasing doses of Procardia 10 mg t.i.d.
When seen in consultation at the hospital his weight and height were at the 75th and 70th percentiles, respectively. The physical examination was unremarkable.
A repeat study of esophageal motility showed a hypertensive LES with mean resting pressure of 82 mmHg, with a peak pressure well above 150 mmHg; failure of the LES to show any relaxation on initiation or continuation of swallowing, observed in wet and dry swallows; an upper esophageal sphincter normotensive at mean resting pressure of 58 mmHg with normal relaxation on initiation of wet swallows; an esophageal body showing uncoordinated movements with multiple simultaneous contractions of upper and lower esophageal body, as well as some swallows without peristaltic activity.
A barium esophagogram revealed abnormal retention of contrast in the distal esophagus at the gastroesophageal junction. The esophagus took 15 minutes to empty in a thin stream. No focal mass was noted in the esophagus. The patient was diagnosed with achalasia and was scheduled for pneumatic dilatation.
The first dilation was done under conscious sedation, with an image intensifier used for positioning. A Rigiflex 35-mm dilator was used, with a pressure of 19 psi for 3 minutes.
Upon withdrawing the ballon and guide wire, no blood was noted on the balloon itself. A small catheter was then passed into the esophagus and 10 ml Gastrografin was infused into the distal esophagus. During fluoroscopy the contrast pooled in the distal esophagus with no evidence of extravasation from the esophageal lumen and with no passage of the contrast into the stomach in 2 minutes of viewing with fluoroscopy. The contrast was withdrawn through the catheter and the catheter was removed. An attempt to use a 40-mm dilator was unsuccessful because of lack of patient cooperation. The second pneumatic dilatation was done a week later under general anesthesia with 320 mg Propofol, 700 μg L-fentanyl, and 3 mg midazolam, using a 40-mm Rigiflex dilator and 15 psi pressure for 3 minutes.
The results of the fluoroscopic examination showed good expansion of the lower esophageal sphincter with minimal pinch seen at the midpoint of the balloon across the diaphragm. The balloon was deflated and withdrawn with the wire in place. A significant amount of bright red blood was seen on the balloon on its withdrawal from the mouth.
Gastrografin contrast solution was injected into the nasogastric catheter, confirming the position of the tube in the esophagus. The contrast did not appear to leak from the esophagus and did not pass into the stomach during 30 seconds of fluoroscopic viewing. The catheter was used to aspirate the remaining contrast and was removed. The patient tolerated the procedure well with stable oxygen saturations and vital signs.
In the recovery room the patient experienced prolonged chest pain and nausea, for which he was given meperidine and prochlorperazine. A subsequent esophagogram with Gastrografin was done 2 hours later and demonstrated extravasation of the contrast outside of the distal esophagus (Fig. 4).
The patient was admitted and treated with total restriction of oral intake; total parenteral nutrition; and intravenous antibiotics, 1 g cefazolin and 500 mg metronidazole, respectively, every 8 hours for 7 days. Results of chest radiographs on admission and in the following 24 hours showed no pleural effusion or significant mediastinal emphysema. The patient became free of pain and afebrile in 48 and 72 hours, respectively. Repeat ingestion of Gastrografin in 7 days revealed a completely healed esophagus (Fig. 5). The patient was sent home to complete a full 10-day course of oral antibiotics.
The third patient was an 18-year-old female with a long history achalasia, initially manifested with solid food intolerance at 6 months of life. Subsequent intolerance of pureed food appeared as episodes of gagging, choking, and vomiting that relieved her symptoms.
The diagnosis was established based on results of a barium esophagogram and an esophageal motility study, performed at an outside hospital, which showed dilatation of the distal esophagus with minimal emptying through the esophagogastric junction, high resting pressure (60 mmHg) of the LES, no relaxation of the LES with swallowing, and absence of peristalsis in the esophageal body.
At 2 years of age the patient underwent a Heller myotomy at our hospital without relief of dysphagia. Subsequent treatment included surgical gastrostomy and multiple retrograde dilatations with Tucker dilators ranging in size from 34 to 40 French. The gastrostomy tube was removed in 1 year. During the next 15 years she did well with rare episodes of choking but with normal weight gain and growth.
At age 17.5 years she became more symptomatic with retrosternal burning and sensation of food sticking in the esophagus, which was relieved by vomiting or drinking water. An endoscopy of the upper esophagus showed friable mucosa of the distal esophagus with loss of normal vascular pattern. Findings in an esophageal biopsy were consistent with mild esophagitis, and results of a repeat esophageal motility study revealed mean pressure in the LES of 73 mmHg, absence of relaxation of the LES with swallowing, simultaneous contractions of low and high amplitude with wet and dry swallows, and normal pressure in the upper esophageal sphincter.
The patient underwent pneumatic dilatation with a 30-mm Rigiflex dilator using 17 psi pressure for 3 minutes. Repeat esophagoscopy showed fresh blood at the level of narrowed esophagus just above the gastroesophageal junction.
There were some irregularities of the distal esophagus but no evidence of esophageal perforation was revealed, either by fluoroscopy or by barium esophagogram immediately after PD. The patient tolerated the procedure well. Four hours later she complained of lower retrosternal and right pleuritic pain. Results of an esophagogram with Gastrografin were positive for a collection of contrast in the right posterolateral aspect of the esophagus, consistent with esophageal perforation (Fig. 6).
The patient was admitted and received 10 days' treatment with bowel rest, total parenteral nutrition, H2 blocker and intravenous antibiotics gentamycin 100 mg every 8 hours, amplicillin 500 mg every 6 hours, and metronidazole 250 mg every 6 hours. Low-grade fever of 38 °C, chest pain and leukocytosis 19.1 × 103/cm resolved on the third day of treatment. A repeat barium esophagogram showed rapid esophageal emptying and no evidence of intra- or extraluminal contrast collection. The patient was discharged home on a regular diet.
Pneumatic dilatation was developed as the primary treatment of achalasia. It works by forcefully rupturing the circular muscle fibers of the the LES to improve esophageal emptying compromised by an absence of esophageal peristalsis and incomplete and discoordinated relaxation of the LES. Despite the improved technique and development of new noncompliant dilators, e.g., the Rigiflex balloon dilator, esophageal perforation still occurs.
Since the first successful closure of an EP by Barrett in 1947 (23) immediate surgical intervention remains the treatment of choice for the majority of patients. However, there are limited circumstances when nonsurgical management of EP is feasible and associated with a low mortality rate (0% to 12.5%) (15,17-21), compared with the mortality rate (10% to 17%) after surgical repair (24,25).
Medical treatment may be used for patients with small perforations caused by pneumatic dilatations that occur in controlled environment without significant contamination of the mediastinum. The key factor in using medical management is early diagnosis, which requires a high level of vigilance, including routine postprocedure esophagograms performed after recovery from sedation, after PD, and before the first feeding, or if the patient develops any symptoms such as prolonged pain associated with fever and vomiting more than 2 hours after the procedure. The most common symptom of esophageal perforation is prolonged substernal chest pain (1,14,25,26). The other symptoms and signs are fever, leukocytosis, pleural effusion, and mediastinal air. One of our three patients developed severe chest pain that persisted for more than 3 hours after the procedure. The other two had mild substernal pain and one had hematemesis caused by a mucosal tear. The subsequent course of this lesion is usually benign but may progress to complete perforation (13,27), as most likely occurred in our patient.
Ischemic changes in the esophageal mucosa develop because of rupture of the circular layer of muscle and interruption of nutrient vessels. This is hypothesized to be the probable cause of gradually developing EP and delayed radiologic diagnosis (28). It also could explain why an esophagogram with water soluble contrast after PD is not 100% sensitive (13,25,26). In our cases there was neither extravasation of contrast agent or free air in the mediastinum on films obtained immediately after PD. The irregularity of the distal esophagus, which might be the earliest sign of mucosal tear, was seen in one patient (13).
Our patients met the criteria for nonsurgical treatment of EP-namely, that it should be well contained in the mediastinum, the cavity should drain back into esophagus, there should be an absence of communication of the EP with the pleural space, and there should be no evidence of sepsis (19).
In medical management of our patients we followed the concept of nonsurgical treatment of EP originally proposed by Mengoli and Klassen (17). Their approach includes measures to prevent or stop contamination of the mediastinum by giving nothing by mouth, suctioning with nasogastric tube to prevent reflux of gastric contents into the esophagus, and administering intravenous antibiotics and parenteral nutrition to provide nutritional support while the perforation heals.
The basic therapy was later modified by addition of H2 blockers (15). The need for a nasogastric tube for decompression and removal of gastric contents becomes less meaningful if H2 blockers are used. We chose not to use a nasogastric tube in our patients to avoid stimulation of vomiting by inserting the tube and gagging induced by leaving the tube in place. This plan has been recommended by others (15,19).
Medical management consisted of no oral intake of food and fluids; total parenteral nutrition to provide for nutritional needs, in that feeding was contraindicated because of perforation; use of intravenous antibiotics to eradicate bacteria released into the mediastinum by the perforation; and use of H2 blockers to inhibit gastric secretion. Response to treatment was evaluated by absence or disappearance of fever, decrease in white blood count, and disappearance of chest pain. Our three patients resumed oral intake after an esophagogram done on days 7, 9, and 10, respectively, showed that extravasation of contrast was no longer occurring.
Our experience with the medical treatment of EP after PD for achalasia in three young patients corresponds with the results shown in published data of similar treatment in older patients with complete recovery in from 93% to 100% of cases (17-21).
1. Vantrappen G, Hellemans J. Treatment of Achalasia and related motor disorders. Gastroenterology
2. Berquist WE, Byrne WJ, Ament ME, et al. Achalasia: diagnosis, management, and clinical course in 16 children. Pediatrics
3. Nakayama DK, Shorter NA, Boyle JT, et al. Pneumatic dilatation and operative treatment of achalasia in children. J Ped Surg
4. Ciarola DA, Traube M. Short term clinical monitoring after pneumatic dilatation. Dig Dis Sci
5. Abid S, Champion G, Richter JE, et al. Treatment of achalasia: the best of both worlds. Am J Gastroenterol
6. Cox J, Bucklon GK, Bennett JR. Balloon dilatation in achalasia: a new dilator. Gut
7. Gelfand MD, Kazarack RA. An experience with polyethylene balloons for pneumatic dilatation in achalasia. Am J Gastroenterol
8. Barkin JS, Gueland M, Reirer, et al. Forceful balloon dilatation: an outpatient's procedure for achalasia. Gastrointest Endosc
9. Kadakia SC, Wong RKH. Graded pneumatic dilatation using Rigiflex achalasia dilator in patients with primary esophageal achalasia. Am J Gastroenterol
10. Azizkhan RG, Tapper D, Eraklis A. Achalasia in childhood: a 20 year experience. J Pediatr Surg
11. Mandelstam P, Sugawa C, Silvis SE, et al. Complications associated with esophagogastroduodenoscopy and with esophageal dilatation. Gastrointest Endosc
12. Dellipiani AW, Hewetson KA. Pneumatic dilatation in the management of Achalasia: experience of 45 cases. Q J Med
13. Ott DJ, Richter JE, Wu WC, et al. Radiographic evaluation of esophagus immediately after pneumatic dilatation for achalasia. Dig Dis Sci
14. Miller RE, Tiszenkel HI. Esophageal perforation due to pneumatic dilatation for achalasia. Surg Gynecol Obstet
15. Swedlung A, Traube M, Siskind BN, McCallum RW. Nonsurgical management of esophageal perforation from pneumatic dilatation in achalasia. Dig Dis Sci
16. Newcomer MK, Brazer SR. Complications of upper gastrointestinal endoscopy and their management. Gastrointest Endosc Clin N Am
17. Mengoli LR, Klassen KP. Conservative management of esophageal perforation. Arch Surg
18. Lyons WS, Seremetis MG, deGuzman VC, Peabody JW. Ruptures and perforations of the esophagus: the case for conservative supportive management. Ann Thorac Surg
19. Cameron JL, Keiffer RF, Hendrix TR, et al. Selective nonoperative management of contained intrathoracic esophageal disruptions. Ann Thorac Surg
20. Ajalat GM, Mulder DG. Esophageal perforations. Arch Surg
21. Wesdorp I, Bartelsman JFWM, Huibregtse K, et al. Treatment of instrumental esophageal perforation. Gut
22. Lo AY, Surick B. Ghazia. Non-operative management of esophageal perforation secondary to balloon dilation. Surg Endosc
23. Barrett NR. Report of case of spontaneous perforation of the esophagus successfully treated by operation. Br J Surg,
24. Pricolo VE, Park CS, Thompson WR. Surgical repair of esophageal perforation due to pneumatic dilatation for achalasia. Is myotomy really necessary? Arch Surg
25. Sarr MG, Pemberton JH, Payne WS. Management of instrumental perforations of the esophagus. J Thorac Cardiovasc Surg
26. Nair LA, Reynolds JC, Parkman HP, et al. Complications during pneumatic dilatation for achilasia or diffuse esophageal spasm. Analysis of risk factors, early clinical characteristics, and outcome. Dig Dis Sci
27. Adams H, Roberts GM, Smith PM. Esophageal tears during pneumatic balloon dilatation for the treatment of achalasia. Clin Radiol
28. Zegel HG, Kressel HY, Levine GM, Rosato EF. Delayed esophageal perforation after pneumatic dilatation for the treatment of Achalasia. Gastrointest Radiol