April 1998

anesthesia, the stomach mounted in an ex-vivo chamber was exposed to ASA or NO-ASA (80 mM) with or without 100 mM HC1 for 30 rain, perfused with saline before and after the exposure, and changes in potential difference (PD), mucosal blood flow (GMBF) and luminal pH were measured simultaneously. Gastric lesions were induced by hypothermic stress (lowering body temperature to 28°C for 4 h) in anesthetized rats or by oral administration of ethanol (60% ethanol in 150 mM HCI) in unanesthetized rats. ASA (20 mg/kg) or NO-ASA (33 mg/kg) was given SC 30 min before the onset of hypothermia or ethanol treatment. Mucosal PGE~ contents and luminal NOx levels were determined after administration of NO-ASA. RESULTS: Mucosal application of ASA (80 mM) markedly decreased the PD and slightly increased luminal pH (acid back-diffusion) with minimal effect on mucosal blood flow, while NO-ASA (80 mM) caused a marked increase in mucosal blood flow with no effect on PD and pH. ASA itself was ulcerogenic, causing damage in the mucosa when administered orally, and markedly potentiated gastric ulcerogenic response to hypothermic stress with no effect on acid secretion when given SC. NO-ASA, however, was not ulcerogenic by itself, did not modify the ulcerogenic response to stress, and even showed a dose-dependent protection against HCl/ethanol-induced gastric lesions. When NO-ASA was given intragastrically to pylorus-ligated rats, a large amount of NOx was detected in both the gastric contents and serum. NO-ASA administered either PO or SC produced an equipotent inhibition of mucosal PGE 2 generation in the stomach, as compared to ASA. In addition, NO-ASA significantly suppressed carrageenan-induced rat paw edema, similar to ASA. CONCLUSION: These results suggest that unlike ASA, NO-ASA had neither a topical irritating action on the stomach nor worsening effect on gastric ulcerogenic response to stress, but rather provided gastric protection against ethanol, despite inhibiting COX activity and showing antiinflammatory action, similar to ASA. NO-ASA, probably by releasing NO, exerted protective effects that counteracted the potential damaging effects of COX inhibition.

• G1288 ESOPHAGO-PHARYNGEAL DISTRIBUTION OF REFLUXED GASTRIC ACID IN PATIENTS WITH VOCAL CORD NODULE AND CHRONIC SINUSITIS. S, Ulualp, R. Toohill, B. Massey, R.C. Arndorfer, W.J. Hogan, R. Shaker. MCW Dysphagia Institute, Division of Gastroenterology & Hepatology and Department of Otolaryngology & Human Communication, Medical College of Wisconsin and VAMC, Milwaukee, WI.

It has been suggested that gastroesophago-pharyngeal reflux (GEPR) contributes to the pathogenesis of chronic sinusitis and vocal cord (VC) nodule formation. However, the esophago-pharyngeal distribution of refluxed gastric acid in these patient groups has not been systematically studied. We evaluated the prevalence of GEPR among 11 CT confirmed chronic sinusitis patients (51-+4 yrs) who had not responded to conventional therapy, 11 patients with vocal cord nodule (43 -+6 yrs) and 21 normal healthy controls (50-+4 yrs). A 3-site ambulatory esophago-pharyngeal pH monitoring technique (probe location: 2cm proximal (PH), 3-4 cm distal to UES (PE) and 5cm proximal to LES (DE) high pressure zones) was used. A pharyngeal pH drop below 4 was accepted as a true reflux event only if it was coincident with or preceded by esophageal pH declines of similar or larger magnitude. Studies were performed while subjects were on a uniform 2500 calorie diet (provided). Results: GEPR episodes were documented in 7 of 11 patients (1-12 episodes) with sinusitis, 8 of 11 patients with vocal cord nodules (1-9 episodes) and 5 of 21 normal volunteers (1-3 episodes) (P < 0.05). None of these events were associated with coughing or belching. In all groups pharyngeal acid events occurred in upright position. As a group the average number of pharyngeal and proximal esophageal acid reflux events as well as proximal esophageal acid exposure time in sinusitis patients were significantly higher than that of normal controls (p< 0.05) (Table).

Number O f Episodes Acid Exposure Time PH PE DE PH PE DE

Sinusitis 3±1" 24±5* 49±9 0.1±0 3.2±0.9* 7.7±2.1 VCNodules 2.3±1 9±2 42±9 0.06±0.02 0.5±0.1 5.1±2.7 Normals 0.3±0.1 10±2 27±3 0.02±0.01 1.0±0.2 4.6±1.2

Condusions: 1) Compared to normal controls, prevalence of pharyngeal reflux of gastric acid is significantly higher in patients with chronic sinusitis unresponsive to conventional therapy and patients with vocal cord nodules. This suggests a different esophago-pharyngeal distribution pattern of gastric refluxate in these patient groups; 2) although not conclusive, these findings suggest that gastroesophago-pharyngeal reflux may contribute to the pathogenesis of chronic sinusitis and vocal cord nodules in some adult patients.

• G1289

PHARYNGO-UES CONTRACTILE REFLEX IN PATIENTS WITH POSTERIOR LARYNGITIS. S. Ulualp, R. Toohill, M. Kern, R. Shaker, MCW Dysphagia Institute, Depts of Medicine and Otolaryngology and Human Communications, Medical College of Wisconsin, and VAMC, Milwaukee, WI.

Earlier studies have shown that pharyngeal stimulation in humans by injection of minute amounts of water results in the augmentation of the upper esophageal sphincter (UES) resting pressure (pharyngo-UES contractile reflex). It has been postulated that stimulation of this reflex by contact of

Esophageal, Gastric, and Duodenal Disorders A315

refluxate with the pharyngeal mucosa during pharyngeal reflux events may augment UES pressure and enhance the defense barrier against further entry of refluxate into the pharynx. However, this reflex has not been evaluated in patients with supraesophageal injuries induced by gastroesophagopharyngeal reflux. Our aim was to evaluate the pharyngo-UES contractile reflex in a group of patients with objective findings of posterior laryngitis clinically compatible with reflux injury. Methods: We studied 14 patients (48 _+ 6 yrs) and 13 healthy age matched controls (53-+6 yrs). All patients had stroboscopic evidence of posterior laryngitis and complained of chronic hoarseness. None of the controls had any symptoms attributable to aerodigestive tract disorders. Eleven of 13 controls also had transnasal laryngoscopy and exhibited normal laryngeal and pharyngeal structures. UES pressure was monitored by a sleeve assembly (Dentsleeve, Parkside SA, Australia) which in addition to a 6x0.4x0.3 cm sleeve device incorporated perfusion side holes at proximal and distal ends of the sleeve. The proximal site was positioned 2 cm above the UES high pressure zone facing posteriorly and was used for pharyngeal water injection. Pharyngo-UES contractile reflex was evoked by injecting incrementally increasing volumes (0.1 ml) of room temperature water into the pharynx directed posteriorly. Each volume was tested x3. The minimum volume of water that elicited response to all three trials was considered the threshold volume for triggering the reflex. Results: The threshold volume required to evoke the pharyngo-UES contractile reflex in the laryngitis group (0.4 + 0.05 ml) was significantly higher than that of the control (0.2 -+ 0.04 ml) (P < 0.05). The preinjection UES pressure in the patient group (42 _ 4mmHg SE) was significantly lower than that of controls (53 -+ 4mmHg SE). Following stimulation by injection of threshold volume, the maximum post-injection pressure in patients (75 +-_ 6mmHg) was similar to that of the controls (78 -+ 6mmHg). Percent increase in UES pressure following stimulation of the pharyngo-UES contractile reflex in the laryngitis group (99-+ 15%) was also significantly higher than that of controls (55 -+ 11%). Conclusions: The pharyngo-UES contractile reflex is present in patients with posterior laryngitis. Compared to normal controls a significantly larger volume of liquid is required to trigger the pharyngo-UES reflex in posterior laryngitis patients but when triggered the maximum UES pressure induced by this reflex is similar between the two groups. These findings suggest an intact efferent motor limb but an altered afferent sensory limb of the reflex in posterior laryngitis patients.

G1290

REVELATIONS ABOUT 24-HOUR AMBULATORY PHARYNGEAL PH MONITORING. S. Ulualp, R. Toohill, R. Hoffmann, R. C. Arudorfer, R. Shaker. MCW Dysphagia Institute, Depts of Med. and Otolaryngol., Medical College of Wisconsin and VAMC Milwaukee, WI.

Ambulatory 24 hr pharyngeal pH monitoring is increasingly used for evaluation of patients with supraesophageal complications of reflux disease such as posterior laryngitis (PL). However, the discriminatory power of this test has not been systematically studied. Aims: To determine the reproducibility, sensitivity, and specificity, optimum pH threshold, and recording technique of 24 hr ambulatory pharyngeal pH monitoring. Method: We did a total of 112 ambulatory 3-site pharyngoesophageal pH monitoring studies in 36 patients with PL documented by stroboscopy and 33 normal healthy age matched controls (NC). In addition, thirteen PL and 14 NC were studied twice and 8 NC were studied 4 times, twice on diet ad libitum and twice on a standard 2500 calorie meal. A pharyngeal pH drop below 4 (threshold pH) was accepted as true reflux event only if it was coincident with or preceded by esophageal pH declines of similar or larger magnitude. Recordings were also analyzed for threshold pH of 5. Results: Prevalence of pharyngeal acid reflux in PL (78%) was significantly higher than NC (21%) (p < 0.01). Sensitivity with threshold pH 4 of 77 -+ 8 was similar to that of pH 5, however, the specificity decreased significantly from 76 _+ 9 (For pH 4) to 57 -+ 11 (For pH 5). Reproducibility of the test was 69% and 79% for PL and NC respectively (Table). Diet did not alter the prevalence of pharyngeal acid reflux events. As a group, PL patients had significantly more pharyngeal reflux events and acid exposure time compared to NC (p < 0.05).

pH Threshold of 4 pH Threshold of 5 Patients Controls Patients Controls

Positive (both studies) 38% 14% 69% 21% Positive (one study) 31% 21% 15 % 57 % Negative (both studies) 31% 65% 16% 22% Reproducibility 69% 79% 85% 43%

Discriminate analysis showed that pH threshold of 5 was not a better discriminator than the pH of 4, for any parameters. However, it did increase the reproducibility of the test in PL but not in NC (Table). Conclusions: Gastroesophagopharyngeal acid reflux is more prevalent in patients with PL than NC. A substantial minority of patients with PL do not exhibit pharyngeal acid reflux of either pH 4 or 5 in repeat studies and may not have acid related disease. The relatively low reproducibility of the 24 hr pH monitoring may reflect heterogeneity of etiology of PL and episodic and infrequent nature of pharyngeal reflux events. Pharyngeal pH threshold of 5 increases the reproducibility of the test among patients without decreasing its sensitivity and needs to be applied only in patients with high suspicion for the supraesophageal complications of reflux.