Prospective randomized study of Maloney esophageal dilation—blinded versus fluoroscopic guidance

0016-5107/90/3603-0272$02.00 GASTROINTESTINAL ENDOSCOPY Copyright © 1990 by the American Society for Gastrointestinal Endoscopy

Prospective randomized study of Maloney esophageal dilation-blinded versus fluoroscopic guidance S. A. McClave, MD R. A. Wright, MD P. G. Brady, MD Louisville, Kentucky and Tampa, Florida

Controversy exists over the need for fluoroscopic guidance when performing Maloney esophageal dilation. This prospective randomized single-blinded study evaluated the safety and efficacy of blinded and fluoroscopic technique. A total of 162 dilations were performed in 43 patients with benign esophageal strictures referred for maintenance dilation, 88 randomized to blinded technique, and 74 to fluoroscopic guidance. Use of fluoroscopic guidance resulted in a higher rate of successful dilations than the blinded technique (96% vs. 80%, P < 0.05) and greater operator accuracy in assessing success of dilation (98.6% vs. 85%, P < 0.05). Although adverse events (such as passage into trachea, gross hematemesis, impaction in a hiatal hernia) occurred more often with blinded technique (11.3% vs. 5.4%), this difference did not reach statistical significance. However, recognition of adverse events was significantly greater with fluoroscopic guidance than with the blinded technique (100% vs. 20%, P < 0.05). No difference existed between groups regarding speed of dilation, patient comfort, or blood on the dilator. Patients were unreliable in their assessment of dilation success. Fluoroscopic guidance is recommended when performing Maloney dilation. (Gastrointest Endosc 1990;36:272-275)

Esophageal dilation with red rubber mercury-filled Maloney bougienage is considered safe and effective therapy for benign peptic strictures of the esophagus. Although the reported incidence of serious complications from esophageal dilation is less than 1 in 400 dilations, the design of the Maloney dilator with its soft flexible rubber tip, has been demonstrated to provide a greater margin of safety than the EderPuestow wire-guided system.} The technique of Maloney dilation varies with the operator. Many physicians pass Maloney dilators blindly, often with the patient sitting in a high back chair. 2 The use of fluoroscopic guidance has been Received June 1, 1989. For revision July 5, 1989. Accepted December 28,1989. From the Departments of Medicine, University of Louisville School of Medicine, Louisville, Kentucky and University of South Florida College of Medicine, Tampa, Florida. Reprint requests: Stephen A. McClave, MD, Department of Medicine, Division of Gastroenterology, University of Louisville, Louisville, Kentucky 40292. Presented in part at the American Society for Gastrointestinal Endoscopy Meeting, May 17,1989, Washington, D.C. 272

recommended to increase the safety and accuracy of the procedure. 3 • 4 However, the addition of this modification requires use of a fluoroscopy unit (often not in the endoscopy suite), exposes the patient and physician to radiation, and adds to the cost of the procedure. 5 We designed a prospective randomized singleblinded study to examine the advantages of fluoroscopic guidance when performing Maloney dilation. With patients randomized to undergo Maloney dilation by either blinded or fluoroscopically guided technique, multiple parameters were evaluated. These parameters included duration or speed, comfort, safety, success of dilation (passage through the stricture), and relief of dysphagia. PATIENTS AND METHODS

Patients with benign esophageal strictures referred for maintenance Maloney dilation were entered into the study and prospectively randomized to undergo dilation in blinded fashion or with fluoroscopic guidance. Patients were exGASTROINTESTINAL ENDOSCOPY

cluded if they had malignant esophageal strictures, were uncooperative, or had benign but difficult strictures requiring fluoroscopically guided or wire-guided dilation. Patients received topical pharyngeal anesthesia without sedation. No patients were excluded from the study because of intolerance to dilation with topical anesthesia only. Patients were randomized to one technique at each dilation session, with all dilations done on that day in the same fashion. Patients were placed in the supine position, inclined at a 30- to 45-degree angle. Multiple dilations were performed at each session, with no more than three dilations performed against resistance. Patients were dilated in French sizes up to the high 40s to low 50s (average 49.85 French). Upon subsequent sessions at a later date, patients were randomized again. Thus, a patient might receive multiple dilations at one session with one technique, and then have more dilations performed at a separate session with the opposite technique. All dilations were performed by an operator in the presence of a monitor. All dilations were timed by the monitor. Following blinded passage of the dilator by an operator, the position was checked fluoroscopically by the monitor prior to withdrawal. Successful dilation by either technique was identified by passage through the stricture documented on fluoroscopy by the monitor. Following each dilation, the monitor documented the operator and patient assessment of dilation success (passage through stricture). The patient's subjective assessment of comfort (easy, moderate, or difficult) was recorded after each dilation. Dilators were checked for the presence of blood following each dilation, and the occurrence of any adverse events was recorded. Symptomatic relief of dysphagia was evaluated with a dysphagia scale (Table 1). The degree of dysphagia was assessed prior to dilation and again 1 to 4 weeks postdilation. The statistical tests were done by chi-square analysis, with statistical significance reached at p < 0.05. RESULTS

One-hundred sixty-two dilations were performed at 54 sessions, in a total of 43 patients. Of the 162 total dilations, 88 were randomized to blinded technique (30 sessions), and 74 to fluoroscopic guidance (24 sessions). Etiologies of the benign esophageal strictures are included in Table 2. Fluoroscopic guidance was shown to have a statistically significant higher success rate of dilation than the blinded technique, 96% (71 of 74) vs. 80% (71 of 88), respectively (p < 0.05). Three of the fluoroscopically guided dilations were unsuccessful: one entering

Table 1. Dysphagia scale

o= 1= 2= 3= 4= 5=

No dysphagia Occasional dysphagia, at least once per week Occasional dysphagia, at least once per day Dysphagia at every meal for solids only Dysphagia for solids and liquids Cannot handle secretions

VOLUME 36, NO.3, 1990

Table 2. Etiology of benign strictures Peptic Schatzki's ring Post-radiation Anastomotic Post-endoscopic variceal sclerotherapy

29 8 2 2 1

VVeb

1

Table 3. Success/complications

Successful dilations Adverse events Operator aware adverse events Operator assessment correct

Blind (88)

Fluoroscopic (74)

(%)

(%)

80.0 11.3 20.0 85.0

96.0" 5.4

100.0" 98.6"

"p < 0.05.

the trachea, one curling in the esophagus, and one curling in the hypopharynx. Seventeen of the blinded dilations were unsuccessful: four entering the trachea, four curling in the esophagus, five meeting resistance which required withdrawal, and four failing to pass through the stricture (inadequate dilation). As would be expected, use of fluoroscopy led to significantly greater accuracy in operator assessment of dilation success. The operator correctly assessed the success of dilation in 98.6% of the fluoroscopically guided dilations, but only 85.0% of the blinded dilations. In 13 of the 88 blinded dilations and only 1 of the 74 fluoroscopic dilations, the operator claimed successful dilation while the monitor documented failure of dilator passage through the stricture. Patient's determination of dilation success was poor, being correct only 84.5% overall. In almost all cases, any chest discomfort was interpreted by the patient as a successful dilation. No attempt was made to train patients in their assessment during this study. Although adverse events occurred twice as often in the blinded group compared with fluoroscopy (11.3% vs. 5.4%), this trend did not reach statistical significance (p = 0.287). Adverse events included passage into the trachea (one in fluoroscopic, four blinded), gross hematemesis (two in blinded), and curling in a hiatal hernia pouch causing overt pain (three in fluoroscopic, four blinded). Of concern was the fact that the operator was aware of only 2 of the 10 adverse events (the two cases of gross hematemesis) that occurred with the blinded technique. The operator was aware of all four adverse events that occurred with the fluoroscopic dilations. Of the four blinded dilations that entered the trachea, neither the patient nor the operator was aware, and the patient failed to demonstrate any associated symptoms (such as coughing or 273

dyspnea). This difference in operator awareness of adverse events was statistically significant (Table 3). No statistically significant difference existed between groups regarding speed of dilation, patient comfort, or the presence of blood post-dilation. Despite a trend suggesting greater improvement in dysphagia with fluoroscopic technique (61.1% vs. 41.6%), this difference did not reach statistical significance (p = 0.349). Improvement in dysphagia did not correlate well with success of dilation (Table 4). DISCUSSION

Mechanical peroral esophageal dilation has become the first line of therapy in the management of esophageal stricture. The overall morbidity for esophageal dilation ranges from 0.1 to 2.2% and is related mainly to perforation and bleeding. I, 6 The use of mercuryfilled bougies is associated with a lower morbidity rate (0.4%) than the use of wire-guided metal olive system (0.6%), as reflected by their different rates of perforation (0.09% vs. 0.3%, respectively).1 The technique of blinded passage of a dilator is thought to be associated with a greater risk of perforation than string-, wire-, or fluoroscopically guided systems. 3, 6-8 The overall mortality for esophageal dilation ranges from 0.05 to 0.5%.6,8 Studies have shown a pattern of dilator use in this country, with over 55% of cases performed with mercury-filled (predominantly Maloney) dilators. I ,6 Another 40% utilized a wire-guided system such as the Eder- Peustow or Savary or were evenly split between Table 4. Clinical parameters

Blood on dilator Time per dilation (sec) Patient comfort Easy Moderate Difficult Dysphagia Average score (pre-dilation) Average score (post-dilation) Dysphagia improved

Blind

Fluoroscopic

(88)

(74)

17.0% 12.0

27.0% 12.5

34.0% 42.0% 23.0%

49.0% 29.0% 21.0%

1.34 0.75 41.6%

1.00 0.44 61.1%

Table 5. Pattern use of dilator1,6 Dilator Mercury-filled Eder-Puestow/Savary Other (balloon) a b

Mandelstam a

Boyce a

Webb b

(%)

(%)

(%)

55 40 5

58 40 2

56 20 24

Benign and malignant strictures. Benign strictures only.

274

the wire-guided systems and the modern balloon dilators (Table 5). Differences in pattern of use in the latter group may be explained by the fact that Webb's series 6 involved only benign strictures and, thus, may have required less guide wire dilations. Overall, fluoroscopy is used during esophageal di1ation in 23 to 32% of cases. 6 Although fluoroscopic guidance has been recommended for wire-guided systems,3, 7, 8 its specific use for Maloney dilation is controversial. An advantage of Maloney dilation is attributed to the belief that fluoroscopy is not needed,6 or at least not needed in simple uncomplicated previously dilated strictures. 3 Webb6 has suggested that without fluoroscopy, tactile sensation may allow the dilating physician to determine whether the dilator has curled proximal to a stricture or is impacted in a hiatal hernia pouch below the stricture. Numerous reports have recommended the use of fluoroscopy with Maloney dilation for complicated tight strictures5 ,9 or initial treatment sessions 4 and in cases where difficulty is encountered or dysphagia is not relieved following dilation. 6 The use of fluoroscopy has been criticized for its potential to add time, cost, and risk of radiation to the dilating procedure. 5 Results of this study show that the use of fluoroscopy with Maloney dilation not only increases the success rate for the procedure, but may contribute significantly to increased safety. Ninety-six percent of the fluoroscopically guided dilations were successful in passing through the stricture, which was significantly greater than the 80% success rate with the blinded technique. The reasons for unsuccessful dilation in the blinded group (entering the trachea, curling above a stricture, and meeting resistance) could be avoided by the use of fluoroscopy which would enable the operator to uncurl and redirect the dilator and push against resistance while maintaining the luminal aXIS.

Assessment of dilation success by the dilating physician is poor with the blinded technique and was significantly less accurate than the assessment with fluoroscopy (85.0% vs. 98.6%). Combining patient assessment with that of the operator in the blinded technique did not improve accuracy, as the patient's assessment overall was poor (84.5% correct). The pattern of occurrence of adverse effects in this study suggested an increase in safety with the use of fluoroscopic guidance. Although not statistically significant, adverse events occurred twice as often in the blinded group as in the fluoroscopically guided dilations (11.3% vs. 5.4%). Of concern was the fact that adverse events which occurred in the blinded group failed to elicit symptoms which would alert the dilating physician to the occurrence of such an event. In the cases where the dilator passed into the trachea, the use of topical pharyngeal anesthesia may have interGASTROINTESTINAL ENDOSCOPY

fered with the gag reflex. As in most dilations, there was some degree of agitation, discomfort, and resistance, but no specific symptoms of cough, cyanosis, dyspnea, or stridor occurred to distinguish these dilations from uncomplicated ones. The use of fluoroscopy enabled the prompt recognition of adverse events early in their development. With regard to the presence of blood on the dilator following dilation, there was no difference in incidence between the two groups. Furthermore, there seems to be no correlation between the presence of blood on the dilator and either adverse events or unsuccessful dilations. Although it was anticipated that the use of tactile sensation in the blinded technique might slow the duration of dilation, the study showed that there was no difference between groups in the time required per dilation. Likewise, there was no difference in patient comfort between groups. Although the study suggests greater improvement of dysphagia with fluoroscopic technique, assessment of symptoms in an individual patient was difficult. As a group, these patients with benign strictures undergoing maintenance dilation had minimal symptoms, and low pre-dilation dysphagia scores made it difficult to show significant change post-dilation. On an individual basis, symptoms of dysphagia showed poor correlation with dilation success, a trend noted in other studies as well. lO, 11 Developing our own dysphagia scale (modified from those in the literature) 10, 12 and shortening post-dilation assessment from 4 weeks to 1 week failed to improve the sensitivity of our scale. A better technique to determine dilation success is evaluation of luminal patency by means of increasing sizes of barium spheres or capsules,lO, 11 a technique which was not used in this study. The results of this study could be interpreted to conclude that an 85% success rate for blind Maloney bougienage is acceptable, that the significance of these adverse events was overstated, and that use of fluoroscopy should be reserved for those patients who have a complicated stricture, who get poor relief of dysphagia post-dilation, or who experience a previous adverse event. Multiple factors would suggest that this interpretation is incorrect. First, a number of parameters reached statistical significance favoring use of fluoroscopic guidance. Blinded dilation, with a lower success rate, might result in a greater number of repeat (otherwise unnecessary) dilations. Because symptoms correlate so poorly, fluoroscopy would provide objec-

VOLUME 36, NO.3, 1990

tive evidence of dilation success. Furthermore, the patients in this study did not represent a difficult subgroup, but instead, had very easy strictures, had no previous problems with dilation, and were undergoing chronic maintenance dilation. These are the patients who would usually be chosen to undergo blinded dilation. No serious complications occurred in these patients. While the use of fluoroscopy might not prevent the occurrence of adverse effects, long-term blinded dilation would allow a certain degree of risk to go unrecognized. Early identification of adverse events by fluoroscopy would enable a switch to a wireguided system and greater potential patient safety. On the basis of the results of this study, fluoroscopic guidance is recommended when performing Maloney dilation, even in those patients with simple benign strictures undergoing maintenance dilation. Fluoroscopic guidance improves success rate, allows early recognition of adverse events, and enables greater accuracy in assessment of dilation success.

REFERENCES 1. Mandelstam P, Sugawa C, Silvis SE, Nebel OT, Rogers BHG. Complications associated with esophagogastroduodenoscopy and with esophageal dilation. Gastrointest Endosc 1976;23:169. 2. Boyce HW, Palmer ED. Techniques of clinical gastroenterology. Springfield, IL: Charles C Thomas, 1975:237. 3. Tulman AB, Boyce HW. Complications of esophageal dilation and guidelines for their prevention. Gastrointest Endosc 1981;27:221-34. 4. Boyce HW. Precepts of safe esophageal dilation (Editorial). Gastrointest Endosc 1977;23:215. 5. Mangla JC, Kothari T. Esophageal dilation with metal olives under fiberoptic endoscopic control. Am J Gasotroenterol 1980;73:260-4. 6. Webb WA. Esophageal dilation: personal experience with current instruments and techniques. Am J Gastroenterol 1988;83:471-5. 7. Bradpiece HA, Galland RB, Murray J, Spencer J. Esophageal dilatation as an outpatient procedure. Surg Gynecol Obstet 1988;167:45-8. 8. Croker JR. Nonsurgical management of benign esophageal stricture in the elderly. Geriat Med Today 1982;1:27-32. 9. Kozarek RA, Phelps JE, Partyka EK, Sanowski RA. Intraluminal pressures generated during esophageal bougienage. Gastroenterology 1981;81:833-7. 10. Cox JGC, Winter RK, Maslin SC, et al. Balloon or bougie for dilatation of benign esophageal stricture? An interim report of a randomized controlled trial. Gut 1988;29:1741-7. 11. Goldschmid S, Boyce HW, Brown JI, Brady PG, Nord HJ, Lyman GH. A new objective measurement of the severity of dysphagia. Gastrointest Endosc 1987;33:171. 12. Earlam R, Cunha Melo JR. Benign esophageal strictures: historical and technical aspects of dilatation. Br J Surg 1981;68:829-36.

275