Infectious complications of endoscopic retrograde cholangiopancreatography. A prospective assessment.

A prospective assessment of the bacterial complications of endoscopic retrograde cholangiopancreatography (ERCP) was made in 97 unselected patients undergoing 101 endoscopies. Blood cultures were taken before and five and ten minutes after completion of the procedure. In an attempt to identify a potential source of infection, aspirates were taken from the pancreaticobiliary duct (PBD) after injection of contrast material. Blood cultures from all procedures were negative. In 14 patients PBD aspirates yielded Pseudomonas aeruginosa, pyocin type 22 and serotype O6, suggesting a common source for this organism. Isolation of this strain ceased after more rigorous cleansing and disinfection of the endoscope. The occurrence of bacteremia following ERCP is low, but there is a risk of transmission of potential nosocomial pathogens with this procedure.

Infectious complications of endoscopic injection sclerotherapy.

We prospectively assessed the infectious complications of esophageal injection sclerotherapy (EIS) in 38 patients who underwent 104 procedures. Blood cultures were taken prior to and five and ten minutes after injection of the sclerosing agent in all procedures in an attempt to determine the frequency of positive blood cultures. Surveillance cultures were obtained from each patient's pharynx and from the biopsy channel of the endoscope to identify potential sources of bacteremia. The rate of blood culture positivity before injection was not significantly different from that after injection (1.9% vs 4.3%). In only one procedure was the same organism isolated five and ten minutes after sclerotherapy. The isolate in both samples was a Corynebacterium species. Endoscope surveillance cultures were positive prior to 42 of 102 procedures, although none of those organisms subsequently were isolated in the blood cultures. Since the rate of positive blood cultures following EIS is no greater than that before the procedure, the use of prophylactic antibiotics is unnecessary.

The extent and pattern of gastro-oesophageal reflux in patients with scleroderma oesophagus: the effect of low-dose omeprazole.

Ambulatory 24-hour oesophageal pH studies were obtained from 11 patients with scleroderma who expressed either dysphagia (n = 6) or heartburn (n = 5) as their predominant oesophageal symptom. No significant differences were found in the extent of pattern of reflux between these two groups. The pH data of both scleroderma groups were combined and compared to an age- and sex-matched group of control subjects (n = 11). The reflux demonstrated by scleroderma patients was significantly greater than the control group in every category (P < 0.01). The percentage of time the pH was < 4.0 was not significantly different (P > 0.05) upright (29.9 +/- 19.8%) vs. supine (44.2 +/- 28.5%) in patients with scleroderma. Eight scleroderma patients underwent repeat pH studies while taking low-dose omeprazole (20 mg daily) and reflux was reduced significantly (P < 0.01) in all patients. The authors believe that 24-hour ambulatory oesophageal pH-monitoring should be routinely conducted in scleroderma patients to provide quantitative reflux data, even when heartburn is not expressed as a symptom. Omeprazole, 20 mg daily, provides adequate protection from the H+ component of the refluxate.

The effect of nasal CPAP on nocturnal reflux in patients with aperistaltic esophagus.

It has been shown that nasal continuous positive airway pressure (nasal CPAP) significantly reduces nocturnal reflux both in patients with sleep apnea and in patients without sleep apnea but consistent abnormal nocturnal reflux. The mechanism by which CPAP is thought to reduce reflux includes the elevation of the resting lower esophageal sphincter (LES) pressure. In this study, we tested the effect of nasal CPAP in two groups of patients with aperistaltic esophagus but with different resting LES pressure. Seven patients with scleroderma esophagus and six patients treated for achalasia were tested over a 48-h period. On the first night, the patients were untreated; on the second night, both groups received applied nasal CPAP at 8 cm H2O pressure. The percentage of time the pH < 4.0, the number of reflux events > 5 min, and the length of the longest reflux event were all significantly reduced in the patients with achalasia (p < 0.03), but not in the scleroderma group (p > 0.20). These results suggest that a residual resting LES pressure greater than that demonstrated by patients with scleroderma (> 10 mm Hg) may be necessary for nasal CPAP to affect nocturnal reflux.

Esophageal reflux before and after isolated myotomy for achalasia.

Four patients with achalasia underwent 24-hour esophageal pH measurements as ambulatory patients before and after limited myotomy without fundoplication. Resting lower esophageal sphincter pressure was reduced from 24.3 +/- 1.3 mm Hg to 7.5 +/- 4.3 mm Hg. No significant differences (p greater than 0.05) were found before and after operation in the total 24-hour pH data distribution (pH 6.24 +/- 0.84 vs 5.75 +/- 1.03), the fraction of time below pH 4.0 (4.8% +/- 5.3% vs 8.0% +/- 6.9%), or the mean duration of reflux episodes greater than 5 minutes (22.8 +/- 18.8 minutes vs 23.0 +/- 10 minutes), all +/- SD. Effective relief of esophageal obstruction in achalasia is feasible by isolated limited myotomy without producing gastroesophageal reflux.

Reflux patterns in patients with achalasia without operation.

Five patients with achalasia who had not had an operation underwent esophageal manometry and 24-hour, ambulatory pH monitoring to determine the pattern and extent of esophageal reflux. One patient had reflux within normal limits. In 2 patients, reflux occurred 0.5% of the total time and no episodes of supine reflux were recorded. In the 2 remaining patients, reflux was measured 16.8% and 55.3% of the total time; however, in both patients, these results were influenced by lengthy bouts of supine reflux. These indices of reflux were not influenced by differences in resting lower esophageal sphincter tone, position, or length. Twenty-four-hour esophageal pH monitoring can be useful in the preoperative assessment of patients with achalasia, and the information obtained might influence the choice of operative procedure.

Retrosternal pain subsequent to sclerotherapy.

A controlled trial was conducted to determine the etiology of retrosternal pain following sclerotherapy. Sclerotherapy was performed on patients that had previously bled from varices. Esophageal motility studies were recorded from seven patients prior to sclerotherapy and immediately subsequent (within 10 min) to it. Motility studies were also recorded from seven normal subjects before and after undergoing gastroscopy to determine the effects of gastroscopy upon esophageal motility. Sclerotherapy appears to induce a transient pattern of diffuse esophageal spasm, the main features of which are significantly (p less than 0.05) prolonged wave duration, increased peak amplitude, and simultaneous contractions.

MRI in the diagnosis of Crohn's disease in two pregnant women.

Two pregnant women considered symptomatic for inflammatory bowel disease were referred for magnetic resonance imaging (MRI) for assistance in establishing a diagnosis. Breath-hold gradient echo and contrast-enhanced T1-weighted fat-suppressed spin echo sequences were used. The MR appearance of the terminal ileum was consistent with Crohn's disease in both patients. The diagnosis of Crohn's disease was subsequently confirmed histologically. MRI is a safe, noninvasive modality that may be useful in confirming a suspected diagnosis of inflammatory bowel disease, especially in patients who cannot be evaluated with standard techniques.

Magnetic resonance imaging evaluation of the local extent of colorectal mass lesions.

Magnetic resonance imaging (MRI) was conducted in 21 patients with known or suspected colorectal mass lesions. Imaging was carried out at 1.5 T using Fast Low Angle Shot (FLASH) and fat-suppressed (T1FS) gadolinium enhanced sequences. The lesions studied included adenocarcinoma (n = 16), lipoma (n = 2), villous adenoma with foci of adenocarcinoma (n = 1), malignant melanoma (n = 1), and rectal lymphoma (n = 1). Correlation was made with surgical findings in 14 patients, 10 of whom had tumors resected. Bowel wall involvement, tumor size, and extension demonstrated on MR images was correlated with histopathology findings. No significant differences were found (p > 0.05) among those parameters when compared to MR information from T1FS images. Lymph nodes were demonstrated on MR images but distinction between benign and malignant nodes could not be made.

Chronic pancreatitis: MR imaging features before and after administration of gadopentetate dimeglumine.

Magnetic resonance (MR) imaging was performed in patients with a history (> 1 year) of inflammatory pancreatic disease. Calcification was seen at recent computed tomographic examinations in 13 patients and was not seen in nine patients. On fat-suppressed spin-echo images, the signal-to-noise ratio of the pancreas was significantly lower (P < .001) in patients with pancreatic calcification (18.2 +/- 2.5 vs 38.1 +/- 6.1). On fast low-angle shot images, the percentage of contrast enhancement was also significantly lower (P < .001) in patients with calcification (26.1% +/- 5.8 vs 78.7% +/- 15.9). The results suggest that MR imaging may be useful in evaluating patients with a long history of pancreatic disease for the presence of irreversible disease.

Magnetic resonance imaging in inflammatory bowel disease.

Magnetic resonance (MR) images were acquired in 28 consecutive inflammatory bowel disease (IBD) patients undergoing medical treatment. The protocol employed included i.v. gadopentatate dimeglumine, pre- and post-contrast breath-hold Fast Low Angle Shot (FLASH), and fat-suppressed spin echo imaging. The percent contrast enhancement (% CE) of the fat-suppressed images was compared with severity of inflammation based on endoscopic and/or surgical findings. The %CE of the contrast-enhanced images was 169% +/- 63 in cases of severe inflammation (n = 16), 97% +/- 38 in moderate inflammation (n = 8), and 49% +/- 26 in mild inflammation (n = 4). Significant correlations were found between the clinicopathologic findings and (a) % CE, r = 0.74 (p = 0.0001); (b) length of affected bowel segment, r = 0.49 (p = 0.007); and (c) bowel wall thickness, r = 0.42 (p = 0.02). In a subsequent comparison, %CE was correlated with length of affected bowel and bowel wall thickness. The best correlation was bowel wall thickness, r = 0.53 (p < 0.004). Good correlation was found between MR findings and pathology/histology findings in the determination of bowel wall thickness, length of diseased bowel, and severity of inflammation in 10 patients who underwent bowel resection. The results of this study show that MR images demonstrate the extent and severity of inflammatory changes in the GI tract, which correlate with endoscopic and histological findings.

A new direct portosystemic collateral in canine portal hypertension.

Two different models of extrahepatic portal obstruction were prepared in dogs to examine the development of portosystemic collateral pathways. The collateral pathways that developed were of two main types, portosplenic communications with the azygos system and a direct communication between the portal vein complex and the intrathoracic part of the inferior vena cava (IVC). This direct communication between the portal venous system and the IVC was unknown to us prior to these studies. The channel we describe has a potential for directly inducing pulmonary embolism, a possibility not seriously considered so far.

Giant villous adenoma of second portion of duodenum: case report.

Villous adenomas of the duodenum are rare. In this report the authors describe the case of a 75-year-old woman who had an extensive adenoma involving the second portion of the duodenum and ampulla of Vater. The patient was successfully managed by a pancreaticoduodenectomy. The authors describe the histopathology of the lesion and suggest guidelines for surgical consideration.

Intestinal villus blood flow measured with carbon monoxide and microspheres.

The purpose of the present study was to compare the small intestinal blood flow that equilibrates with luminal CO (FLco) with simultaneous determinations of villus blood flow measured by a recent modification of the microsphere technique. These studies, carried out in rabbits, showed that FLco closely correlated (r = 0.83) with villus flow measured with microspheres over a three-fold range of flows, and the mean rate of flow to the villi by both techniques was about 0.08 ml/min X g of intestine. Thus, FLco appears to be a measure of villus blood flow. Based on previous studies of inert gas uptake from the rabbit small intestine, it appears that absorption of readily diffusible substances in the rabbit can be represented by a simple two-component model: a flow-limited component in which substances equilibrate with villus blood flow and are carried away without subsequent countercurrent exchange, and a diffusion-limited component which presumably represents uptake by the blood flow of the crypt region or submucosa.

Pseudocalculus of the common bile duct.

Common-bile-duct growths are rarely identified unless they cause chronic biliary obstruction. This case report describes a 71-year-old woman who had jaundice and epigastric pain. A cholecysto-colonic fistula was demonstrated by endoscopic retrograde cholangiopancreatography. The patient also had multiple filling defects in the common bile duct. The fistula was closed and stones were removed. A postoperative cholangiogram showed two calculi. One was removed with a basket through the T-tube tract, but the second, which did not appear completely free of the common-duct wall, could not be removed by the basket method. Subsequently at laparotomy this was found to be a benign pedunculated polyp, composed of collagenous and vascular tissue and with no surface epithelium. Surgeons should bear in mind the possibility of a common-bile-duct growth in cases of extrahepatic biliary obstruction.

Nasal continuous positive airway pressure. A new treatment for nocturnal gastroesophageal reflux?

Nasal continuous positive airway pressure (CPAP) reduces nocturnal gastroesophageal reflux (GER) in obstructive sleep apnea syndrome (OSAS) patients. The primary objectives of our investigation were to determine if CPAP could reduce reflux in non-OSAS patients and, if so, by what mechanism. Esophageal pH was monitored for 48 h in six nocturnal reflux patients. During the first 24 h, basal reflux data were collected; the second night, nasal CPAP was administered (pressure = 8 cm H2O). Esophageal manometry was obtained in six healthy adult volunteers both on and off nasal CPAP (pressure = 8 cm H2O) to ascertain CPAP's effects on esophageal pressure and peristalsis. The six reflux patients experienced less nocturnal GER while on CPAP. The mean percent time esophageal pH < 4 was reduced from 27.7 +/- 10.0 to 5.8 +/- 2.6 (p < 0.004); the mean reflux duration dropped from 2.1 +/- 0.6 to 0.9 +/- 0.5 min (p < 0.03); and the mean duration of longest reflux improved from 84.3 +/- 32.6 to 13.8 +/- 6.9 min (p < 0.01). The CPAP raised the mean resting midesophageal pressure by 4.4 cm H2O (p < 0.01) and the mean resting lower esophageal pressure (LES) by 13.2 cm H2O (p < 0.02) in the healthy volunteers. Nasal CPAP effectively reduced nocturnal GER in six patients with nocturnal reflux. The antireflux activity of CPAP is likely due to passive elevation of intraesophageal pressure and possibly to reflex LES constriction.

Bowel disease: prospective comparison of CT and 1.5-T pre- and postcontrast MR imaging with T1-weighted fat-suppressed and breath-hold FLASH sequences.

The potential of new high-field-strength magnetic resonance (MR) imaging sequences to evaluate bowel disease was investigated and compared with computed tomographic (CT) studies. Thirty-two patients were studied, 14 with known or suspected gastrointestinal tumors and 18 with known or suspected bowel inflammatory conditions. T1-weighted fat-suppressed spin-echo and breath-hold FLASH (fast low-angle shot) images were obtained before and after intravenous injection of 0.1 mmol/kg gadopentetate dimeglumine. Pathologic confirmation was obtained by biopsy (n = 18), surgical excision (n = 8), or endoscopy (n = 6). CT and MR images were analyzed separately in a prospective fashion and reviewed by consensus. Information from CT and MR images was comparable in cases of confirmed bowel neoplasia. CT scans had better spatial resolution, while fat-suppressed gadolinium-enhanced MR images had better contrast resolution. In the 18 cases of bowel inflammation, CT scans showed concentric wall thickening in 16, while MR images showed concentric wall thickening in 14 and increased contrast enhancement in 17. Contrast enhancement was better appreciated on fat-suppressed images than on FLASH images. The results suggest that MR imaging may play a role in the evaluation of bowel disease.

Pancreatic disease: prospective comparison of CT, ERCP, and 1.5-T MR imaging with dynamic gadolinium enhancement and fat suppression.

New magnetic resonance (MR) imaging techniques possess features desirable for imaging the pancreas. Computed tomography (CT) and endoscopic retrograde cholangiopancreatography (ERCP) were prospectively compared with breath-hold fast low-angle shot (FLASH) and fat-suppressed spin-echo techniques before and after enhancement with gadopentetate dimeglumine. Thirty-five patients underwent ERCP, CT, and/or MR imaging studies within a 1-month period. Correlation with surgical findings, histologic findings, or clinical and/or imaging follow-up was obtained in all cases. Quantitative measurements of pancreas, pancreas minus pancreatic lesion, and pancreas minus fat signal-to-noise ratios (SNRs) were performed on MR images. The highest measurements of pancreas minus pancreatic tumor SNR were on gadolinium-enhanced, fat-suppressed images (8.9 +/- 3.4). The 1-second postcontrast FLASH images most reliably showed enhancement of normal pancreatic tissue. Nonenhanced FLASH images depicted peripancreatic fluid and inflammatory changes most successfully. The findings from this study suggest MR imaging is effective for imaging inflammatory and neoplastic pancreatic disease and may be superior to CT.

Reflux in untreated achalasia patients.

We made a prospective assessment of acid exposure in the distal esophagus in 48 consecutive untreated patients with achalasia using 24-h ambulatory esophageal pH studies. The majority of patients (38/48) experienced reflux that was within reported values for normal controls (total time pH < 4.0, 1.8 +/- 1.9%). Approximately 20% (10/48), however, demonstrated abnormal acid exposure (total time pH < 4.0, 18.8 +/- 14.8%). The difference in reflux expressed by these two groups was not due to a significant difference in lower esophageal sphincter pressure (p > 0.05) or retained food. An in vitro model of lactobacillus fermentation supported the contention that true acid reflux accounted for changes in esophageal pH. Repeat pH studies were obtained in 23 patients following treatment: 15 underwent pneumatic dilatation and 8 underwent limited myotomy. Although no significant differences were found between pre- and posttreatment reflux, some patients undergoing either treatment were found to demonstrate increased acid exposure. In conclusion, we believe that patients with achalasia should be tested by pH study both before and after treatment. Most of the patients who demonstrated significant pretreatment reflux were asymptomatic, and both methods that were used to decrease resting sphincter pressure were shown to be able to increase distal acid exposure.

Comparison of magnetic resonance imaging and endoscopy in distinguishing the type and severity of inflammatory bowel disease.

Twenty consecutive patients with first-time presentation of suspected inflammatory bowel disease underwent both endoscopy with biopsy and magnetic resonance imaging (MRI) within a 3-day period; the relative abilities of endoscopy and MRI to distinguish ulcerative colitis (UC) from Crohn's disease (CD) and to determine the severity of the disease process were compared. In 18 of 20 patients, a diagnosis of UC or CD could be made on histological specimens. MRI correctly diagnosed 17 of these 18 patients using T1-weighted fat-suppressed spin echo and gadolinium enhancement. Endoscopy correctly diagnosed 15 patients. Overall, MRI was not significantly better (p > 0.05) than endoscopy in distinguishing UC from CD. MRI correctly graded the severity of inflammatory changes in 13 of 20 patients, and endoscopy did so in 11 of 20. MRI and endoscopy findings were within one grade of histology findings in seven patients each. No significant difference (p > 0.05) was found between MRI and endoscopy in the ability to estimate the severity of the disease (as determined from biopsies). Bowel wall thickness measured on MR images demonstrated good correlation with percentage of contrast enhancement: r = 0.61; p = 0.003. In sum, magnetic resonance imaging was shown to be comparable with endoscopy in differentiating UC from CD and in gauging the severity of disease. Transmural assessment, sagittal imaging, and the lack of invasiveness were attractive features of MRI.