Bypass circuits as the source of thromboemboli during extracorporeal membrane oxygenation.

To determine the presence and extent of thrombus formation in the apparatus used for extracorporeal membrane oxygenation we studied various portions of the polyvinylchloride circuit from five infants who received extracorporeal membrane oxygenation for 70 to 330 hours. All infants had right-sided cannulation. Sections were cut from the circuit at the time of decannulation and subjected to light and scanning electron microscopy. The site that contained the most thrombus formation was the membrane oxygenator bypass circuit, which is subjected to repeated periods of unclamping and clamping to direct blood flow through the membrane oxygenator. Autopsy results from nonsurvivors showed evidence of pulmonary and renal infarcts, a left frontal lobe infarct, a thromboembolus of the left external and internal carotid arteries, and thrombi in the lungs, kidney, brain, and coronary arteries. One survivor had computed tomographic evidence of infarction of the left middle cerebral artery distribution. We suggest that the areas of the extracorporeal membrane oxygenation circuit subjected to repeated changes in flow dynamics may be the source of microemboli.

Hypertrichosis cubiti.

Hypertrichosis is an unusual but well-recognized genetic condition. Hypertrichosis may be generalized or limited to specific body areas, in which case it is usually not associated with other anomalies. Five previous cases of hypertrichosis cubiti have been reported, with short stature in 2 sibs being the only other associated abnormalities. We report on a child with hairy elbows, developmental delay, facial asymmetry, and delayed speech with normal parents. Our patient may represent severe expression of the hairy elbow syndrome or constitute a previously unrecognized syndrome.

The role of prolonged esophageal pH monitoring in the diagnosis of gastroesophageal reflux.

To reassess 24-hour esophageal monitoring and determine if shorter time periods might retain its diagnostic benefits, we studied 16 gastroesophageal reflux (GER) patients and eight controls. Esophageal pH monitoring was performed for 24 hours, with patients in an upright position during the day and supine when retiring. During the 24-hour pH monitoring period, the mean percentage time that pH was less than 4.0 in GER patients, 13.2% +/- 2.9% (SEM), was significantly higher than in normal subjects, 0.7% +/- 0.2% (SEM). Analysis of individual data indicated clear separation of GER patients from normal subjects when in the upright posture, but 25% of GER patients were within the range of the normal subjects when supine (overnight). Three-hour time periods after meals were analyzed. Postprandial pH monitoring, when compared with 24-hour pH monitoring, can identify GER with a 77% sensitivity and a 96% specificity. A 12-hour period (four hours after the dinner meal and eight hours supine) can identify GER with a 94% sensitivity and a 100% specificity. We conclude that (1) 24-hour pH monitoring of the esophagus may have a continuing role in research aspects of GER, (2) 12-hour pH monitoring is a highly accurate test that could be adapted to patients' work schedules or to outpatient telemetry, and (3) postprandial pH testing is a practical, less expensive, and accurate method of diagnosing GER that could be utilized by any gastroenterology diagnostic unit.

The association of histologic gastritis with gastroesophageal reflux and delayed gastric emptying.

The aims of this study were to: 1) investigate the incidence of histologic gastritis in patients with gastroesophageal reflux (GER); 2) ascertain if gastritis in GER patients could be correlated with impaired gastric emptying; and 3) determine if the presence of histologic antral gastritis correlated with other parameters of esophageal and gastric function. Twenty-three GER patients, mean age 53.3 years (range 28-68 years) with subjective and objective evidence for GER; and 20 normal subjects (13 males and seven females), mean age 28.7 years (range 19-46 years), underwent upper gastrointestinal endoscopy. Antral biopsies obtained from the greater curvature were graded as: 0 = normal; 1 = chronic gastritis; 2 = chronic active gastritis; and 3 = chronic atrophic gastritis. All patients underwent a gastric emptying study using an isotope-labeled semisolid meal. Eighteen of the 23 GER patients (78%) had histologic gastritis compared to two (10%) of the normals. No subject had endoscopic evidence of gastritis. Gradings of histologic gastritis were significantly (p less than 0.05) correlated with delayed gastric emptying. Twelve GER patients had severe gastritis (grades 2 or 3) and their gastric emptying, 79.7% +/- 5.8 (mean +/- S.E.M.) retention of isotope at 90 minutes after the meal, was significantly slower (p less than 0.01) than the 11 GER patients with either grade 0 or 1 gastritis, 56.1% +/- 5.9 retention, or the normal subjects, 51.8% +/- 1.7. We conclude that: 1) histologic gastritis is associated with GER disease; and 2) slowing of gastric emptying can be significantly correlated with increased severity of histologic gastritis in GER patients.

Metoclopramide in gastroesophageal reflux disease: rationale for its use and results of a double-blind trial.

We investigated the acute effect of metoclopramide on lower esophageal sphincter pressure, esophageal contraction amplitude, and gastric emptying and compared metoclopramide, 10 mg four times a day, to placebo in improving the symptoms and objective parameters of reflux esophagitis in 19 patients in a randomized, double-blind 4-wk outpatient trial. Orally administered metoclopramide, 10 mg, significantly accelerated gastric emptying of a semisolid meal in patients in whom it was delayed; lower esophageal sphincter pressure was significantly increased for up to 90 min, but there were no changes in esophageal contraction amplitude. During the treatment trial, metoclopramide resulted in an overall improvement in heartburn and regurgitation of 60%, significantly better than 32% improvement after placebo (p less than 0.05). Compared to baseline symptoms scores, metoclopramide significantly improved both daytime and nighttime heartburn and regurgitation. Compared to placebo-treated patients, the metoclopramide group had significantly fewer episodes of daytime heartburn and regurgitation (p less than 0.05), while nighttime symptoms significantly improved with both treatments. Mean antacid consumption was significantly reduced by metoclopramide, 61%, compared to placebo-treated patients, 21% (p less than 0.05), who were ingesting a mean of 1.9 oz of antacid daily. Endoscopic and histological improvement were similar in both groups, although histological healing occurred in three patients after metoclopramide compared with none in the placebo group. Our data suggest that: 1) gastric emptying and lower esophageal sphincter pressure were significantly improved by acute administration of oral metoclopramide; 2) metoclopramide therapy for 4 wk is significantly more effective than placebo (medium dose antacid therapy) in relieving the symptoms of gastroesophageal reflux without significantly altering objective parameters of esophagitis; 3) metoclopramide effectively addresses the diffuse upper gastrointestinal motor disturbances present in reflux esophagitis patients.

Effect of metoclopramide on normal and delayed gastric emptying in gastroesophageal reflux patients.

Gastric emptying has an important role in the pathophysiology of gastroesophageal reflux disease. We investigated the effect of metoclopramide, a gastric prokinetic agent, in gastroesophageal reflux patients with normal as well as delayed emptying. Twenty-six patients with subjective and objective evidence of gastroesophageal reflux ingested an egg salad sandwich meal labeled with 99mtechnetium-DTPA for a baseline study, and then again on a separate day after receiving oral metoclopramide, 10 mg, 30 min prior to the test meal. The mean percent isotope remaining in the stomach after 90 min improved significantly (P less than 0.001) from 70.3 +/- 3.9% (SEM) to 55.2 +/- 4.2% after metoclopramide. Fourteen (54%) had a basal emptying in the normal range of 34-69% retention of isotope at 90 min, (means +/- 2 SD), while it was slow in 12 (46%). For those with delayed basal gastric emptying, the mean retention of 88.9 +/- 2.9% at 90 min was significantly (P less than 0.005) decreased by metoclopramide to 68.6 +/- 6.1%. In those patients with a normal basal gastric emptying and a mean retention of 54.4 +/- 2.3% at 90 min, there was also significant improvement (P less than 0.025) to 43.6 +/- 3.6% after metoclopramide. These data indicate that metoclopramide increased gastric emptying in gastroesophageal reflux patients with normal as well as delayed gastric emptying. Therefore on a patient management level a trial of metoclopramide is warranted in patients with gastroesophageal reflux disease and is not limited by the gastric emptying status of the patient.

Endoscopic findings in chronic hemodialysis patients with upper gastrointestinal bleeding.

Over a 7-year period 20 episodes of upper gastrointestinal bleeding in 18 patients on chronic hemodialysis were evaluated endoscopically. The site, type of lesion and outcome were studied. We confirm that 1) mucosal inflammation is more commonly found in these patients than that reported in upper gastrointestinal bleeding in the general population and 2) mortality of gastrointestinal bleeding is not higher in this subset of patients than in the general population. The indications for endoscopy in this population are better defined than in other upper gastrointestinal bleeders.

Effects of metoclopramide and bethanechol on delayed gastric emptying present in gastroesophageal reflux patients.

Gastric emptying has been reported to be delayed in a significant percentage of patients with gastroesophageal reflux. The rationale for the use of metoclopramide and bethanechol in gastroesophageal reflux has been based on their ability to stimulate lower esophageal sphincter pressure and enhance acid clearance mechanisms. In this study, we investigated the comparative efficacies of metoclopramide and bethanchol in improving the rate of gastric emptying in gastroesophageal reflux patients in whom delayed emptying was present. Gastric emptying studies used an isotope-labeled mixed solid-liquid meal. Thirteen reflux patients with delayed gastric emptying received metoclopramide, 10 mg intramuscularly, and subcutaneous bethanechol, 0.07 mg/kg, in a randomized single-blind fashion. Eleven additional reflux patients with delayed gastric emptying received oral metoclopramide, 10 mg, in an open-labeled fashion. After parenteral metoclopramide, gastric emptying was significantly (p less than 0.05) faster compared with both the initial basal day and the bethanechol treatment day. Compared with the normal gastric emptying rate established in 26 control subjects, metoclopramide accelerated gastric emptying into the normal range. Bethanechol did not increase gastric emptying. Metoclopramide orally also significantly improved gastric emptying. Our study indicates that metoclopramide, both parenterally and orally, increased the rate of gastric emptying in those reflux esophagitis patients in whom it was delayed, while bethanechol did not improve the degree of gastric retention in the same patients. Our results extend the rationale for the therapeutic efficacy of metoclopramide in gastroesophageal reflux disease.

Reassessment of esophageal histology in normal subjects: a comparison of suction and endoscopic techniques.

We have examined esophageal biopsies from 18 asymptomatic volunteers. These normal subjects were also evaluated by esophageal manometry with determination of motor function of the esophagus and lower esophageal sphincter pressure, a modified Bernstein acid infusion test, and a basal pH reflux test. In 12 subjects, biopsies were obtained by suction technique; the remaining six had pinch biopsies performed during upper gastrointestinal endoscopy. Forty-nine (92.5%) of the 53 suction biopsies yielded tissue as deep as muscularis mucosa or at least ample amounts of lamina propria. In contrast, all 28 endoscopic biopsies yielded squamous epithelium only. Moreover, suction biopsies were generally well-oriented (83%), whereas endoscopic biopsies were less commonly well-oriented (35.7%). All volunteer subjects had a negative acid reflux test confirming the absence of latent or asymptomatic reflux. Only one subject (5.6%) had biopsies which met histologic criteria for gastroesophageal reflux. We conclude that: 1) more tissue, more information, and better orientation is achieved with suction than with endoscopic biopsies of the esophagus; 2) the low false-positive rate observed in volunteers in whom gastroesophageal reflux was objectively excluded emphasizes the value of esophageal biopsy as a reliable index in the evaluation of gastroesophageal reflux.

Significant hemorrhage as a complication of hydraulic suction biopsy of the esophagus.

An histological diagnosis of gastroesophageal reflux disease is more accurately made with biopsies obtained by the hydraulic suction method than by endoscopic biopsy forceps. We report significant hemorrhage as a complication of biopsying the esophagus by hydraulic suction. This bleed was explained by an aberrantly large vessel in the lamina propria of an otherwise histologically normal esophagus. This risk of hemorrhage in our extensive experience is less than 0.1%, and hydraulic suction biopsy can be regarded as a safe technique and the optimal method of obtaining histological information in the esophagus.