Gastroesophageal reflux (GER) is common in anesthetized dogs and can cause esophagitis, esophageal stricture, and aspiration pneumonia.
Gastroesophageal reflux (GER) is common in anesthetized dogs and can cause esophagitis, esophageal stricture, and aspiration pneumonia.
To determine whether preanesthetic IV administration of esomeprazole alone or esomeprazole and cisapride increases esophageal pH and decreases the frequency of GER in anesthetized dogs using combined multichannel impedance and pH monitoring.
Sixty-one healthy dogs undergoing elective orthopedic surgery procedures.
Prospective, randomized, placebo-controlled study. Dogs were randomized to receive IV saline (0.9% NaCl), esomeprazole (1 mg/kg) alone, or a combination of esomeprazole (1 mg/kg) and cisapride (1 mg/kg) 12–18 hours and 1–1.5 hours before anesthetic induction. An esophageal pH/impedance probe was utilized to measure esophageal pH and detect GER.
Eight of 21 dogs in the placebo group (38.1%), 8 of 22 dogs in the esomeprazole group (36%), and 2 of 18 dogs in the combined esomeprazole and cisapride group (11%) had ≥1 episode of GER on impedance testing during anesthesia (P < .05). Esomeprazole was associated with a significant increase in gastric and esophageal pH (P = .001), but the drug did not significantly decrease the frequency of GER (P = .955). Concurrent administration of cisapride was associated with a significant decrease in the number of reflux events (RE) compared to the placebo and esomeprazole groups (P < .05).
Preanesthetic administration of cisapride and esomeprazole decreases the number of RE in anesthetized dogs, but administration of esomeprazole alone was associated with nonacid and weakly acidic reflux in all but 1 dog.
5-hydroxytryptamine receptor 4
area under the curve
lower esophageal sphincter
multichannel intraluminal impedance monitoring
nonsteroidal anti-inflammatory drug
proton pump inhibitor
Gastroesophageal reflux (GER) and its associated sequelae are well-documented causes of morbidity and potential mortality in a variety of species, including people, dogs, and cats.[1-5] Gastroesophageal reflux during anesthesia is associated with 46–65% of cases of benign esophageal stricture in dogs, and represents the most common cause of high-grade esophagitis and stricture formation in dogs.[4, 5] Lower esophageal sphincter (LES) tone is modulated by excitatory and inhibitory vagal pathways, and LES basal tone contributes substantially to increased pressure at the esophagogastric junction. In addition, contraction of the crural diaphragm around the LES further increases pressure, particularly during inspiration and increased abdominal pressure. Relaxation of the LES is mediated by nonadrenergic noncholinergic pathways, and has been shown to occur with administration of commonly utilized injectable preanesthetic and inhalant anesthetic agents.[8-12]
The diagnosis of GER in dogs can be challenging given the intermittent nature of the problem, and the need for diagnostic equipment such as videofluoroscopy, esophagoscopy, and esophageal pH probes that are not readily available at many veterinary hospitals. Esophageal pH monitoring was considered the gold standard for documenting GER, but esophageal pH probes are unable to detect subtle changes in pH and nonacid reflux, resulting in an underestimation of reflux events (RE).[13-16] The benefits of combined multichannel intraluminal impedance (MII)/pH-metry allows determination of reflux independent of its acidity,[16, 17] and facilitates detection of nonacid (pH ≥ 7.5) and weakly acidic reflux (4.0 < pH < 7.5), both of which have been associated with coughing and laryngitis in people.[13, 14, 17]
Previous studies evaluating GER in anesthetized dogs have focused on the use of metoclopramide and proton pump inhibitors (PPIs) to prevent reflux.[18, 19] All studies, however, utilized esophageal pH probes, potentially underestimating the frequency of GER, and no studies have evaluated the effects of cisapride, a 5-hydroxytryptamine receptor 4 (5-HT4) agonist more potent than metoclopramide. We hypothesized that preanesthetic administration of esomeprazole in dogs undergoing elective orthopedic surgery would increase gastric and esophageal pH, and that the administration of cisapride with esomeprazole would decrease the frequency of GER.
Sixty-one client-owned dogs admitted to the University of California, Davis, VMTH for elective orthopedic surgery procedures were recruited for inclusion into this prospective, randomized, double blind, placebo-controlled study. A sample size of 18 dogs in each group was determined to have adequate power (> 80%) to detect a difference of 1 RE, given a probability of type I error of 5% and a SD of 1.
All dogs underwent a comprehensive physical examination, followed by minimum database consisting of measurement of hematocrit and plasma protein concentration and a serum biochemistry panel in dogs < 5 years old. In addition, a CBC was performed on all dogs ≥5 years old. Dogs with a history of GER, regurgitation, vomiting, esophagitis, or coughing were excluded from the study. In addition, any dog treated with gastric antacids or prokinetics within 1 month of surgery was excluded. Dogs receiving treatment with nonsteroidal anti-inflammatory medications (NSAIDs) were included in the study, but this was recorded and factored into analysis. The study protocol was evaluated and approved by the University of California, Davis, School of Veterinary Medicine Animal Care and Use Committee (IACUC), and owners of all dogs signed an informed consent form before enrollment of their pet in the study.
Each dog was randomized by an Excel software program random number generator into 1 of 3 treatment groups to receive either (1) esomeprazole sodium1(1 mg/kg IV); (2) esomeprazole and cisapride2 (both at 1 mg/kg IV); or (3) saline placebo (0.9%) administered IV at 10 mL over 3 minutes. Esomeprazole was administered as an IV infusion over 3 minutes according to an established protocol. Cisapride was diluted with sterile saline (0.9%) to a total volume of 100 mL, and administered over a 15-minute infusion. All treatments were administered by IV catheter 12–18 hours and 1–1.5 hours before induction.
All dogs were fasted for at least 12 hours before induction. The anesthetic protocol was identical in each dog, consisting of premedication with hydromorphone3 (0.05 mg/kg SC) and atropine4 (0.02 mg/kg SC), followed by induction of anesthesia with propofol5 (4 mg/kg IV) and diazepam6 (0.3 mg/kg IV). Anesthesia was maintained in all dogs with isoflurane7 (1–3%), titrated to maintain appropriate surgical anesthetic depth. To ensure adequate analgesia, each dog received a loading dose of fentanyl8 (5 μg/kg IV) followed by a constant rate infusion (0.4 μg/kg/min IV). Intermittent positive pressure ventilation was administered in all dogs after induction of anesthesia. Lactated Ringer's solution (10 ml/kg/h IV) was administered throughout anesthesia. Esophageal temperature probes and stethoscopes were avoided during anesthetic monitoring to minimize artifact during recording. The anesthetist responsible for administering and monitoring anesthesia was blinded to the treatment group assignment.
Immediately after induction, a 6.4-French (2.13-mm) esophageal multi-use pH/impedance probe9 was attached to an electrical external reference pad that was placed in the axillary region of the dog. All of the polyurethane esophageal probes had 6 impedance rings, each spaced 2 cm apart, and 1 pH sensor (Fig 1). The pH electrode of the MII-pH probe was calibrated within 10 minutes of use in buffer solutions of pH 4.0 and 7.0. The esophageal probe then was introduced into the esophagus via the oral cavity by use of a loop snare10 passed through the biopsy channel of a fiberoptic endoscope.11 Probe placement was performed in all dogs by 1 of 2 investigators (SLM or KLD) skilled in endoscope handling to ensure consistency in the positioning of the probe. The probe was advanced into the greater curvature region of the stomach in 41 dogs to record gastric pH for 2 minutes before probe placement in the distal esophagus. After recording of gastric pH, the pH sensor on the esophageal probe was positioned 6 cm proximal to the gastroesophageal junction (GEJ) in all dogs, and no portion of the probe traversed the LES during the recording period. Deionized water was used to rinse any residual gastric acid from the pH probe, and air introduced into the stomach and esophagus during probe placement was carefully suctioned. The probe then was secured in place using tape wrapped around the maxilla, and the probe was attached to a recording device12 from which data were uploaded onto a computer by proprietary software from the manufacturer. Esophageal pH and impedance were recorded throughout the surgical procedure until the probe was removed immediately before extubation of the dog.
A reflux episode was defined as a 50% decrement in ohms seen in 2 consecutive impedance channels in the distal esophagus for >2 seconds from the pre-episodic esophageal baseline recording. The pH of the refluxate was classified as strongly acidic (pH < 4.0), weakly acidic (4.0 < pH < 7.0), or nonacidic (pH ≥ 7.0).
The mean acid clearance time was defined as the duration of each acid RE, beginning at the moment the pH decreased to < 4.0 and ending when the pH increased to ≥ 4.1.
The potential effects of esomeprazole and cisapride on fecal consistency were assessed by evaluating the dogs’ feces before and after drug administration. Fecal consistency was evaluated at least 3 times daily, and recorded by veterinary technicians and senior students who were blinded to the treatment group of each animal.
All data were coded and recorded into SPSS 17.0 software.13 A one-way analysis of variance (ANOVA) and the nonparametric Mann–Whitney U-test were performed to evaluate statistical differences among treatment groups. A multiway ANOVA was utilized to evaluate the effect of multiple covariates on the development of loose feces. A probability of type I error < .05 was deemed significant.
Sixty-one dogs were randomized into 3 groups consisting of a placebo group (n = 21), esomeprazole group (n = 22), and a group receiving a combination of esomeprazole and cisapride (n = 18). All animals were systemically healthy based upon the history and normal physical examination and minimum database results. The age of the dogs ranged from 0.5 to 12.0 years (mean ± SD, 4.9 ± 3.4 years). Body weight ranged from 3.6 to 52.0 kg (mean ± SD, 25.0 ± 12.9 kg). Body condition score (BCS) ranged from 4 to 9 (mean ± SD, 5.9 ± 1.29). There were no significant differences in age, body weight, or BCS among the treatment groups (P > .05). Labrador retrievers represented the most common pure breed (n = 13/61; 21%) and 19/61 dogs (31%) were mixed-breed. The total esophageal pH/impedance analysis time ranged from 1.8 to 7.2 hours (mean ± SD, 3.2 ± 1.2 hours). The most common surgical procedures were extracapsular cruciate ligament stabilization (n = 19/61; 31%), tibial plateau leveling osteotomy (n = 10/61; 16%), and elbow arthroscopy (n = 6/61; 10%). Twenty-four of 61 dogs (39%) were receiving treatment with NSAIDs during enrollment. Four were in the placebo group, 9 in the esomeprazole group, and 11 in the combined esomeprazole with cisapride group. Prescribed NSAIDs included meloxicam,14 deracoxib,15 carprofen,16 aspirin,17 firocoxib,18 and naproxen.19 No morbidity or death was associated with anesthesia or the study procedures.
Gastric pH was measured in 41 dogs, and mean gastric pH in dogs that received esomeprazole (7.7 ± 1.7) and the combination of cisapride and esomeprazole (6.6 ± 1.5) was significantly higher than pH in dogs receiving the placebo (3.9 ± 1.9; P = .001) (Fig 2). No difference was found in mean gastric pH between the esomeprazole and combination groups (P = .275).
The mean esophageal pH for the esomeprazole group was 6.4 ± 0.2. This was significantly higher than the mean esophageal pH of 5.6 ± 0.3 for the placebo group (P = .028) and the same as the mean esophageal pH of 6.4 ± 0.2 for the combination esomeprazole and cisapride group. The mean percentage of procedural time that the esophageal pH was < 4.0 was 2.0 ± 6.6% for the esomeprazole group. This was significantly lower than the mean percentage of procedural time that the esophageal pH was < 4.0 for the placebo group (15.5 ± 32.8%; P = .028) and similar to the mean percentage of procedural time that the esophageal pH was <4.0 for the combination esomeprazole and cisapride group (0.0 ± 0.0%; P = .195) (Fig 3).
Eighteen of 61 dogs (29.5%) enrolled in the clinical trial had at least 1 episode of GER during anesthesia, representing 8 of 21 dogs (38.1%) in the placebo group, 8 of 22 dogs (36.4%) in the esomeprazole group, and 2 of 18 dogs (11.1%) in the combined esomeprazole and cisapride group. In total, 30 individual RE were documented in the 18 dogs. Of these, 16.7% (5/30) were strongly acidic, 43.3% (13/30) were weakly acidic, and 40.0% (12/30) were nonacid reflux (NAR) events. There were a total of 15 RE in 8 dogs within the esomeprazole group, 10 of which were nonacid events, 4 of which were weakly acidic events, and 1 that was strongly acidic. There were a total of 13 RE in 8 dogs within the placebo group, 2 of which were nonacid events, 7 of which were weakly acidic events, and 4 that were strongly acidic. One of the 8 dogs in the placebo group had both a weakly acidic RE and 2 strong acid RE. Two dogs refluxed 1 time each in the combination group, and both RE were weakly acidic. The coadministration of cisapride and esomeprazole was associated with a significantly decreased number of RE in comparison with the placebo group (P = .046) and the group that received esomeprazole alone (P = .049; Fig 4). Esomeprazole administration was not associated with a reduction in the frequency of GER (P = .955). The mean ± SD number of NAR events was 0.10 ± 0.30 for the placebo group (n = 21), 0.45 ± 0.91 for the esomeprazole group (n = 22) and 0.00 ± 0.00 for the esomeprazole with cisapride group (n = 18). Comparison of the number of NAR events between the placebo and esomeprazole group (P = .093) and placebo and esomeprazole and cisapride groups (P = .188) was not significant, but the esomeprazole group had a significantly higher number of NAR events compared with the esomeprazole and cisapride group (P = .041).
The mean acid clearance time was significantly longer in the placebo group (56.6 ± 66.0 minutes; range, 2.4–164.2 minutes) versus the esomeprazole group (23.2 ± 4.3 minutes; range, 24.2–30.2 minutes; P = .012). Dogs that received cisapride in combination with esomeprazole had no strongly acidic RE, thus mean acid clearance time cannot be reported.
Diarrhea characterized by liquid to soft-formed feces was documented in 10% of dogs in the placebo group, 10% of dogs in the esomeprazole group, and 59% of dogs in the cisapride and esomeprazole group (Table 1). The combination group was 12.8× more likely to have diarrhea than the placebo and esomeprazole groups (95% CI for OR = 2.23, 74.1, P = .002). There was a significant association between animal group and NSAID use (P = .027), but there was no association between NSAID use and the development of diarrhea (P = .375). It can be concluded that the significant association between esomeprazole and cisapride use and the development of diarrhea occurred independent of NSAID use.
|Placebo||Esomeprazole||Esomeprazole & Cisapride|
|NSAIDsb,d||4/21 (19%)||9/22 (41%)||11/18 (61%)|
|Diarrheaa||2/20 (10%)||2/20 (10%)||10/17 (59%)c|
Gastroesophageal reflux (GER) during anesthesia is a relatively common phenomenon in people and dogs, and has been documented to occur in approximately 50% of dogs that receive preanesthetic administration of morphine.[8, 11] Approximately 25% of the dogs that develop GER have evidence of refluxed gastric content reaching the pharynx, increasing the risk of aspiration pneumonia.[8, 11] Reflux of gastric acid also has been implicated in the development of postanesthetic esophagitis and esophageal stricture formation.[4, 5]
The administration of a variety of anesthetic premedications, induction agents, and maintenance drugs, including morphine, glycopyrrolate, atropine, diazepam, xylaxine, propofol, and halogenated anesthetic agents are associated with weakening of LES tone and increased risk of GER.[8, 10, 11, 21-23] Atropine, acepromazine, and xylazine administration to dogs was shown to decrease LES pressure to 13.2, 18.6, and 11.7%, respectively. In addition, isoflurane and halothane administration to dogs has been shown to decrease LES barrier pressure by 21%.
An alternative mechanism for GER involves transient LES relaxations (TLESRs) that reflect prolonged LES relaxations that are not associated with swallowing and occur as a normal physiologic response to gastric distention, allowing venting of gas from the stomach.[24, 25] Transient LES relaxations are part of a vago-vagal reflex, triggered by distension of the stomach, and are found to be the main mechanism of both acid and nonacidic RE both in healthy volunteers and patients.[24, 25]
Acidic and nonacidic RE occurred in 18/61 dogs (29.5%). These findings parallel those previously reported in the veterinary literature.[8, 18, 21, 26] The parenteral administration of esomeprazole alone in this investigation was associated with a significant increase in gastric pH with a consequent reduction in the percentage of time that the esophageal pH was < 4.0. Seven dogs in the placebo group had mean esophageal acid clearance times > 2.4 minutes, and 2 dogs had mean esophageal clearance times of 126.5 and 164 minutes. There is compelling in vivo evidence in experimental animal models of esophagitis induced in rabbits, dogs, and cats that the dripping or infusion of an acidic solution of hydrochloric acid alone (pH 1.6–2.0) on the esophagus over a period of 30–60 minutes is associated with minimal to no damage to the esophagus, but the addition of pepsin to the hydrochloric acid is associated with both macroscopic (eg, edema, hyperemia, hemorrhage) and microscopic (eg, epithelial loss, erosions, inflammation, mucosal, and submucosal hemorrhage) evidence of acute esophagitis.[27-30] These findings are paralleled by the increased esophageal mucosal H+ ion back diffusion and increased permeability in experimental acid esophagitis in rabbits after perfusion of acid (pH 2) with pepsin. Prolonged exposure (> 1–2 hours) of the esophageal mucosa to acid (pH < 4) is an important cause of esophagitis and potential stricture formation, particularly in anesthetized or sleeping animals, because the proteolytic pH range for the conversion of pepsinogen to pepsin is between 1.5 and 3.5, and esophageal acid clearance does not occur during sleep or anesthesia.[17, 32-35]
The administration of esomeprazole alone did not decrease the frequency of GER in the dogs, but RE in 7 of 8 dogs in this group all were nonacidic or weakly acidic in nature, potentially decreasing the likelihood of esophagitis and esophageal stricture formation. The pathogenesis of reflux esophagitis and stricture formation is complex, and clearly not caused by the reflux of gastric acid alone. Different components of the gastroduodenal contents, including pepsin, bile acids, and trypsin may have different sites or mechanisms of esophageal injury, and additional studies evaluating the interaction among these factors are warranted. The pH of the refluxate needs to be related to the optimal pH value of the proteolytic enzyme activity and the pKa of the bile salts to determine which bile salts at what pH will remain in solution or be un-ionized and cause severe mucosal permeability changes or esophagitis.
Our findings are in contrast to a previous study in anesthetized dogs that showed a significant reduction in the number of RE after PO administration of a single dose of omeprazole at 1 mg/kg to dogs at least 4 hours before induction. Likely reasons for the differences in outcome between the studies include differences in defining reflux, the use of an esophageal pH probe in the former study, differences in the anesthetic protocols, and differences in the patient population. Body position of the anesthetized dogs was not standardized in this study, but a previous publication did not show a significant correlation between body position and GER. The results of our study parallel those of a recent study that utilized combined multichannel intraluminal impedance (MII) technology with pH monitoring in people, in which the oral administration of omeprazole q12h for 7 days was not associated with a decrease in GER.
The addition of parenterally administered cisapride with esomeprazole did not alter gastric pH compared to dogs receiving esomeprazole alone, but this group of dogs had a significantly lower number of RE compared with the placebo group and esomeprazole group (P < .05). A previous study examined the effects of low dose and high dose administration of metoclopramide in anesthetized dogs on frequency of GER. Low dose administration (0.4 mg/kg IV, followed by CRI of 0.3 mg/kg/h) failed to decrease the frequency of GER, but high dose administration (1 mg/kg IV, followed by a CRI of 1 mg/kg/h for the study duration) was associated with a 54% reduction in the relative risk (RR) of developing GER. This unusually high dose of metoclopramide can cause extrapyramidal signs and sedation with prolonged administration. Cisapride is a far more potent 5-HT4 receptor agonist than metoclopramide, and represents a viable and effective prokinetic agent that enhances gastric emptying and increases LES sphincter pressure tone.20 Additional studies are warranted to determine whether PO cisapride is as effective as the parenterally administered form.
This study represents the first in the veterinary literature utilizing dual MII/pH-metry for the detection of acid and nonacid gastroesophageal reflux in anesthetized dogs. Previous studies on GER in the veterinary literature have relied upon the use of pH probes for detection of reflux, defining and characterizing GER by esophageal pH < 4.0 (acid reflux) or pH > 7.5 (nonacid reflux) in dogs.[8, 11, 18, 21, 26] However, the absence of impedance potentially misses reflux events characterized by subtle pH changes (eg, pH 6.0 to 5.0), and those composed of weakly acidic and nonacidic refluxate, potentially resulting in decreased documentation of reflux.[14, 15, 39] Impedance (Z) technology relies upon the principle of resistance to the passage of flow of an electrical current and is inversely related to conductivity. Impedance is influenced by the physical characteristics of intraluminal substrates, and gastric refluxate has high electrical conductivity, or low impedance, whereas intraluminal air has a low conductivity, or high impedance. Interpretation of the waveform generated by the computer can be used to determine whether the refluxate originated orally or aborally. In addition, the associated pH of the refluxate is used to classify the pH as strongly acidic (pH < 4), weakly acidic (4.0 < pH < 7.0), or nonacidic (pH ≥ 7.0). Esophagoscopy-assisted placement of the pH sensor of the dual MII/pH probe 6 cm proximal to the GEJ for all dogs was necessary for accurate and repeatable documentation of GER.
Esomeprazole1 is the s-isomer of omeprazole and is 97% bound to plasma proteins. It is metabolized more slowly than omeprazole, resulting in a larger area under the concentration-versus-time curve (AUC) after administration of the same dose. In addition, inter-patient variability in intragastric pH and AUC is less with esomeprazole than with omeprazole. The administration of esomeprazole in our study was not associated with any clinically relevant adverse effects, although 2/20 dogs on esomeprazole alone (10%) had soft feces after administration of the drug. Neither of these dogs received NSAIDs while enrolled in the study. In the dog, cisapride stimulates contraction of the LES without effect on esophageal peristalsis, because the entire esophageal body is composed of striated muscle. Although other prokinetic agents such as metoclopramide have been assessed in the prevention of GER under anesthesia, comparative studies have shown that cisapride is far superior in stimulating gastric emptying and increasing LES sphincter pressure.20 Parenteral cisapride was relatively well tolerated by the dogs in this study, but there was a significant increase in the frequency of diarrhea in the cisapride group, independent of NSAID administration (P = .002). These effects were self-limiting in all dogs affected, and resolved within 48 hours after cessation.
In summary, this study showed that cisapride significantly decreased the frequency of GER in dogs whereas esomeprazole had no significant effect on the number of RE. Despite these findings, esomeprazole was associated with a significant increase in esophageal pH, and all but one of the RE documented in the esomeprazole group were nonacidic or weakly acidic in nature. Preanesthetic administration of esomeprazole to dogs might decrease the incidence of esophagitis and esophageal stricture formation, because the percentage of time the esophageal pH was < 4.0 was significantly decreased in this group. The coadministration of cisapride with esomeprazole may further reduce the number of nonacidic RE, which are less likely to cause esophagitis but may be associated with aspiration pneumonia, chronic cough, dysphonia, globus, and laryngeal pathology in people.[42, 43] Additional studies examining the influence of PO administered esomeprazole and cisapride are warranted to determine whether these findings can be reproduced in a clinical setting.
This study was supported in part by funding from the Students Training in Advanced Research (STAR) Program at the University of California Davis, School of Veterinary Medicine. The results of this study were presented in part at the 2010 ACVIM Forum in Anaheim, CA.
The authors greatly appreciate the support of Sandhill Scientific Inc, the technical support of Marni McChesney and Michele Callahan, and the technicians in the UC Davis VMTH Anesthesiology and Small Animal Medicine Services.
Nexium, AstraZeneca, Wilmington, DE
Compounded Cisapride, North Carolina State University College of Veterinary Medicine, Veterinary Teaching Hospital Pharmacy, Raleigh, NC
Hydromorphone HCl, Hospira Inc, Lake Forest, IL
Atropine, Baxter, Deer Field, IL
Propofol, Novaplus, Schaumburg, IL
Diazepam, Hospira Inc
Isoflurane, Piramal, Bethlehem, PA
Fentanyl citrate, Hospira Inc
Esophageal pH/impedance probes, model # ZPN-BS-01E and ZPN-BS-46E, Sandhill Scientific, Inc, Denver, CO
Loop snare, model F016, Endoscopy Support Services, Brewster, NY
Olympus XP-10 and Olympus GIF P140 endoscopes, Olympus America Inc, Center Valley, PA
ZepHr and Sleuth recording devices, Sandhill Scientific, Inc
IBM SPSS 17.0 software for the Macintosh, IBM Corporation, Armonk, NY
Metacam, Boehringer Ingelheim Vetmedica, Inc, St. Joseph, MO
Deramaxx, Novartis Animal Health US, Inc, Greensboro, NC
Rimadyl, Pfizer, New York, NY
Aspirin, Bayer, Shawnee Mission, KS
Previcox, Merial, Duluth, GA
Weiser HF, Holscher A, Zimmermann T. Effects of cisapride and metoclopramide on the lower esophageal motility. A pressure on pH-metric study. Digestion 1986;34:142 (abstract)