Effects of esophageal shortening on the gastroesophageal barrier: An experimental study on the causes of reflux in esophageal atresia

Effects of Esophageal Shortening on the Gastroesophageal Barrier: An Experimental Study on the Causes of Reflux in Esophageal Atresia By Sandra

Montedonico,

Juan A. Diez-Pardo, Madrid,

Background/Purpose: Gastroesophageal reflux (GER) is frequently recognized after surgical repair of esophageal atresia. The aim of this study was to test the hypothesis that one or more components of the gastroesophageal pressure barrier are weakened by esophageal anastomosis undertension. Methods: Lower esophageal sphincter pressure (LESP), crural sling pressure (CSP), and the length of the intraabdominal segment of the esophagus (LIAE) were measured by pullthrough perfusion manometry in 20 rats before and after resection of 15 mm of the cervical esophagus, and in eight rats before and after esophageal transection (control group).

Anne

K. Possdgel,

and Juan A. Tovar

Spain

not significantly significantly.

change

in controls.

CSP

did

not

change

Conc/usions:Anastomosis of the esophagus under tension in this model decreases significantly the lower esophageal sphincter tone and length of the intraabdominal esophagus, but it does not change the crural sling pressure. Postoperative reflux in patients operated on for esophageal atresia might be in part, caused by this mechanism. J Pediatr Surg 34:300-303. Copyright o 7999 by W.B. Saunders Company.

Resu/ts:This manouver decreased the LESPfrom 44.9 ? 17.4 to 30.9 i 12.3 mm Hg and the LIAE from 17.9 i 2.8 to 15.8 ? 2.4 mm (PC .05) in experimental animals, whereas they did

INDEX WORDS: ter, esophagus, flux.

ASTROESOPHAGEAL REFLUX (GER) is so common after surgical treatment of esophageal atresia (EA) that it is not unreasonable to consider it as part of the disease.l Extensive esophageal mobilization and anastomosis under tension are generally accepted as good explanations for gastroesophageal barrier insufficiency because partial denervation and traction on the cardia may weaken the lower esophageal sphincter.* Both preoperative3 and postoperative manometric assessment’ show that peristaltic dysfunction of the body of the esophagus probably related to neuromuscular disorganization4m6contributes to GER disease in these patients. The purpose of the current study is to examine whether pure anastomotic tension accounts for gastroesophageal barrier failure in rats and, if so, by which mechanism.

From the Department of Surgery and Experimental Laboratory, Hospital Infantil Vniversitario “La Paz, “Madrid, Spain.. Presented at the 45th Annual International Congress of the British Association of Pnediatric Surgeons, Bristol, England, July 21-24, 1998. Supported by FIS Grants #95/0498 and 9710287. Address reprint requests to Dr Juan A. Tovar; Department of Surgery, Hospital Infantil Vniversitario ‘?.a Paz,” I? de la Castellana 261, 28046, Madrid, Spain. Copyright o 1999 by WB. Saunders Company 0022.3468/99/3402-0015$03.00/0

according to previously described techniques.7-11 Briefly, a tip-occluded single-lumen catheter (outer diameter, 1 mm; internal diameter, 0.5 mm) with a distal side hole (1.0 X 0.5 mm) connected to an external transducer (HP 1280; Hewlett Packard, Palo Alto, CA) and continuously perfused with bubble-free distilled water (0.4 mL/min) using a high-pressure, low-compliance pneumohydraulic pump (Mui Scientific, Mississauga, Ontario, Canada), was advanced into the stomach within which the pressure was registered by a monitor (Schiller, Switzerland) with a screen display and on-line printout at a paper speed of 2.5 mm/s. The atmospheric pressure at the level of the atrium served at “0” reference. The recording orifice of the catheter was oriented toward the lesser curvature of the stomach-the plane at which the lower esophageal sphincter shows its highest pressure value.” To measure the pressures in the gastroesophageal barrier, the tube subsequently was withdrawn into the esophagus at a constant speed (1 mm/s) by a specially made mechanical device. The values recorded are the average of three successive pull-throughs. The variables analyzed were the lower esophageal sphincter pressure or difference between intragastric pressure and the peak of the more distal component of the pressure profile (LESP), the cmral sling pressure or difference between baseline pressure and the peak of the more proximal component of the pressure profile (CSP), and the length of the intraabdominal segment of the esophagus or distance between the onset of the first component and the respiratory reversal point at which the esophagus becomes intrathoracic (LIAE). After the control values were measured, the esophagus was exposed on the left side of the trachea in 20 rats through a midline cervicotomy taking care of preserving both vagi, and a length of 15 mm was resected, reestablishing the continuity by end-to-end, one-layer anastomosis with interrupted 8-O polypropylene sutures, using the surgical microscope (Wild M-650, Herrbrugg, Switzerland). The remaining eight rats underwent complete esophageal transection and reanastomosis to serve as the control group (Fig 1). Oral feeds were immediately resumed, and 1 week later the manometric measurements were repeated with identical techniques in both groups. Values (in millimeters and millimeters of mercury) are presented as

300

Journal

G

MATERIALS

AND

METHODS

Adult male Wistar rats (n = 28) weighing 300 to 500 g (Criffa; Barcelona, Spain) were submitted to manometric studies in the supine position under intraperitoneal anesthesia (6.25 mg/lOO g ketamine hydrochloride and 0.5 mg/lOO g diazepam) and spontaneous breathing

of Pediatric

Esophageal anastomotic

Surgery,

atresia, manometry, rat, sphinctension, gastroesophageal re-

Vol34,

No 2 (February),

1999: pp 300-303

ESOPHAGEAL

SHORTENING

AND

REFLUX

301

IN ATRESIA

ments, the LESP and the LIAE were decreased significantly 1 week after esophageal resection but did not change after esophageal transection. CSP did not change significantly. DISCUSSION

Fig 1. (Top) Procedure on the control group: schematic drawing shows the esophageal transection in the neck followed by end-to-end reconstruction with 8-O polypropylene interrupted stitches. (Bottom) Procedure on the experimental group: schematic drawing shows e esophagus and cervical esophageal resection of 15 mm esophagus. anastomosis under tension. T, trach

means +SD. After ascertaining the normality of their distribution within groups by comparison against theoretical values for the same means (Kolmogorov-Smimov tests) in both control and experimental groups, the results of the baseline and postoperative conditions were compared with paired parametric tests (Student’s t). P values of less than .05 were accepted as significant to reject the null hypothesis that there were no differences between repeated measurements. The experimental protocol was approved by the local institutional Research Committee and met the requirements of the European Union for animal care and research (EC86/L609).

RESULTS

Five rats of the experimental group died of anastomotic leaks within 3 days of the operation, but there were no other complications, and in particular no strictures. The tracings obtained in the baseline conditions were similar to those previously described by our group in this model.7-11 Upon withdrawal of the catheter from the stomach, a first pressure plateau corresponding to the esophageal sphincter was seen clearly followed by a return of pressure to intraabdominal levels. Finally, a series of phasic inspiratory contractions corresponding to the diaphragmatic crural sling appeared until the recording orifice was above the level of the diaphragm where inspiratory oscillations became negative. Table 1 and Fig 2 show that in comparison with the baseline measure-

Gastroesophageal reflux (GER) is determined mainly by a failure of the gastroesophageal barrier that consists of a set of mechanical components including the permanent tone of the lower esophageal sphincter, the intermittent compression of the esophagus by the diaphragmatic crural sling, and the effect of permanently positive pressures on the intraabdominal esophagus. This failure may be caused by insufficient tone or nondeglutory relaxations of the lower esophageal sphincter, to an abnormal sling function, or to an insufficient length of the infradiaphragmatic esophagus. If the barrier fails, there is a second level of defense, the esophageal peristalsis, that may compensate for this insufficiency by clearing the refluxate. The conditions met in survivors of EA repair largely facilitate GER, which is so frequent that it affects 54% to 80% of patients depending on the diagnostic methods used.‘J3 First, the esophageal body component of the antireflux barrier is damaged in these individuals who have perianastomotic aperistalsis and predominant lowamplitude, nonpropulsive contractions in the lower esophagus for life. W* Vagal injury caused by extensive esophageal dissection during repair may contribute to such dysfunction3J3 but there also is preoperative evidence of intrinsic structural and functional anomalies of the esophageal wa11.4-6Moreover, Romeo et al3 demonstrated manometrically esophageal dysmotility in both the proximal and distal ends of the esophagus in 20 newborns before EA repair. Second, the antireflux barrier components also are compromised in this particular group of patients, although the mechanisms involved are not well known. The shortening of the infradiaphragmatic segment of the esophagus might well play a role, and it is very likely that nondeglutory relaxations, a well-known mechanism of reflux that has not been addressed in this particular clinical situation, also contribute. Sphincteric dysfuncTable 1. Components and After

of the Gastroesophageal

Esophageal

Transection

Esophageal Transection In = 8)

LESP (mm Hg) CSP (mm Hg) LIAE (mm) NOTE. Values *P<

.05 against

Barrier

and Resection

Before

in the Rat

Esophageal

Resection

(n = 15) Postresection

Baseline

Postresection

Baseline

32.7 ril 13.8 7.9 2 2.9

31.3 ? 12.5 6.9 k 4.9

44.9 t 17.4 9.4 I 9.9

30.9 ir 112.3~ 6.4 2 6.8

16.3 i 1.9

16.9 + 1.7

17.9 I 2.8

15.8 + 2.4’

expressed baseline

as mean within

i: SD. the same

group.

MONTEDONICO

302

ET AL

Fig 2. (A) Pressure profile of the gastroesophageal barrier in control group before esophageal transection. (6) Postoperative image. (C) Pressure profile of the gastroesophageal junction before esophageal shortening. (D) Postoperative image. After esophageal shortening the LESP and the LIAE in the same animals were visibly reduced. The baseline before and after LESP corresponds to gastric and infradiaphragmatic esophageal resting pressures respectively, whereas that above CSP shows the intrathoracic situation of the esophagus because of the negative inspiratory oscillations.

tion caused by either structural malformation or denervation might be reasonably expected, but there is little information on whether it is actually congenital or acquired. A few studies reported normal preoperative lower esophageal sphincteric pressures,3*‘3whereas most investigators found these pressures lowered in the postoperative period.* The finding of normal postoperative sphincteric pressures in some children in whom the anastomotic tension was reduced by Livaditis myotomy2,‘* reinforces the interpretation that anastomotic tension plays a role in postoperative sphincteric dysfunction leading to GER. Reflux recently has been demonstrated after esophageal shortening in the rabbit by pH-probe assessmentI but the mechanisms involved have not been explored in

that model. Our investigation-that did not aim at demonstrating GER in the rat after esophageal shortening but rather at examine the behaviour of the gastroesophageal barrier before and after it-clearly shows that this manipulation weakened the barrier by directly decreasing the sphincteric pressure and the length of the intraabdominal esophagus. The use of an animal model in which intrinsic structural anomalies are absent authorizes the attribution of all postoperative changes to the effects of esophageal traction on the components of the barrier. The current study supports the conception that anastomosis of the esophagus under tension-a common situation in the clinical setting-facilitates GER by significantly weakening the gastroesophageal barrier.

REFERENCES 1. Tovar J, Diez Pardo J, Murcia J, et al: Ambulatory 24-hour manometric and pH metric evidence of permanent impairment of clearance capacity in patients with esophageal atresia. J Pediatr Surg 30:1224-1231, 1995 2. Takano K, Iwafuchi M, Ucbiyama M, et al: Evaluation of lower esophageal sphincter function in infants and children following esophageal surgery. J Pediati Surg 23:410-414, 1988 3. Romeo G, Zuccarello B, Proietto F, et al: Disorders of the esophageal motor activity in atresia of the esophagus. J Pediatr Surg 22:120-124, 1987 4. Nakazato Y, Landing B, Wells T: Abnormal Auerbach plexus in the esophagus and stomach of patients with esophageal atresia and tracheoesophageal fistula. J Pediatr Surg 21:831-837,1986

5. Hokama A, Myers N, Kent M, et al: Esophageal atresia with tracheoesophageal fistula; A histopathologic study. Pediatr Surg Int 1:117-121, 1986 6. Zuccarello B, Nicotina P, Centorrino A, et al: Immunohistochemical study on muscle actinin content of atresic esophageal upper pouch. Int J Pediatr Surg Sci 2:75-78, 1988 7. Wang WL, Tovar JA, Eizaguirre I, et al: Airway obstruction gastroesophageal reflux. An experimental study of the pathogenesis this association. J Pediatr Surg 28:995-998, 1993

and of

8. Qi B, Diez-Pardo JA, Soto C, et al: Transdiaphragmatic pressure gradients and lower esophageal sphincter after tight abdominal wall plication in the rat. J Pediatr Surg 31:1666-11X9, 1996

ESOPHAGEAL

SHORTENING

AND

REFLUX

IN ATRESIA

9. Qi B, Soto C, Diez-Pardo JA, et al: An experimental study on the pathogenesis of gastroesophageal reflux after repair of diaphragmatic hernia. JPediatr Surg 32:1310-1313, 1997 10. Soto C, Qi B, Diez-Pardo JA, et al: Identification of the diaphragmatic crural component of the gastro-esophageal barrier in the rat. Dig Dis Sci 42:2420-2425, 1997 11. Montedonico S, Godoy .I, Mate A, et al: Muscular architecture and manometric image of the gastroesophageal barrier in the rat. Dig Dis Sci (in press)

303 12. Schneeberger A, Scott R, Rubin, et al: Esophageal function following Livaditis repair of long-gap esophageal atresia. J Pediatr Surg 22:779-783, 1987 13. Shono T, Suita S, Arima T, et al: Motility function of the esophagus before primary anastomosis in esophageal atresia. J Pediatr Surg 28673-676, 1993 14. Guo W, Fonkalsmd E, Swaniker F, et al: Relationship of anastomotic tension to the development of gastroesophageal reflux. J Pediatr Surg 32: 1337-1340, 1997