Macroscopic Esophagitis Research Articles

Pathophysiology of gastro-oesophageal reflux disease: implications for diagnosis and management.

Gastro-oesophageal reflux disease (GERD) is a common gastrointestinal disorder in which retrograde flow of gastric content into the oesophagus causes uncomfortable symptoms and/or complications. It has a multifactorial and partially understood pathophysiology. GERD starts in the stomach, where the refluxate material is produced. Following the trajectory of reflux, the failure of the antireflux barrier, primarily the lower oesophageal sphincter and the crural diaphragm, enables the refluxate to reach the oesophageal lumen, triggering oesophageal or extra-oesophageal symptoms. Reflux clearance mechanisms such as primary and secondary peristalsis and the arrival of bicarbonate-rich saliva are critical to prevent mucosal damage. Alterations of the oesophageal mucosal integrity, such as macroscopic oesophagitis or microscopic changes, determine the perception of symptoms. The intensity of the symptoms is affected by peripheral and central neural and psychological mechanisms. In this Review, we describe an updated understanding of the complex and multifactorial pathophysiology of GERD. It is now recognized that different GERD phenotypes have different degrees of reflux, severity of mucosal integrity damage and type, and severity of symptoms. These variations are probably due to the occurrence of a predominant pathophysiological mechanism in each patient. We also describe the main pathophysiological mechanisms of GERD and their implications for personalized diagnosis and management.

Low-impedance Baseline Values Predict Severe Esophagitis.

The aim of the study was to determine whether esophageal baseline impedance (BI) values in children could be predictive of esophagitis. Multichannel intraluminal impedance (MII) tracings of children 3 to 17 years of age suspected of having gastroesophageal reflux and esophagitis, who had also undergone upper endoscopy with multiple esophageal biopsies, were reviewed. Patients with eosinophilic esophagitis were excluded. Esophagitis was assessed by macroscopic and microscopic parameters. Esophageal histology was reported by 2 blinded independent pathologists unaware of the MII results. Mean BI was automatically calculated in the different MII channels (ch) by the specific software without removing any episode of increased/decreased BI. BI results were plotted against macroscopic and histological scores for each channel. Tracings of 87 children, 53 boys, were evaluated. Mean age was 7.4 years: 45 had histologic esophagitis, 8 macroscopic. Histologic mild esophagitis (grade 1) was observed in 30, and 15 had moderate to severe esophagitis (grade 2-3). Ten had grade 3 esophagitis. Eight had macroscopic esophagitis as well. in channel 6 of the MII, all 10 patients with grade 3 esophagitis and the 8 with macroscopic esophagitis had a BI <900 Ω/s (positive predictive value 100% and negative predictive value 100%), whereas none of those having a biopsy score of 0 to 2 or no endoscopic evidence of esophagitis had a mean BI below 2000 Ω/s. The evaluation of the BI measured in channel 6 gave us 100% prediction of grade 3 and macroscopic esophagitis. BI on channel 6 may be useful to predict severe esophageal mucosa inflammation and could potentially be used for follow-up evaluation, rather than repeating an upper endoscopy. In addition, it would seem that grade 3 esophagitis even in the absence of macroscopic esophagitis affects the integrity of the esophageal epithelium.

Bioimpedance spectroscopy: a new tool to assess early esophageal changes linked to gastroesophageal reflux disease?

Bioimpedance spectroscopy can identify pathological changes related to precancerous lesions of the cervix uteri and esophagus. It therefore has the potential to detect early reflux-related changes in the esophageal mucosa, such as dilated intercellular spaces. The reliable detection of dilated intercellular spaces at the time of endoscopy would yield a significant diagnostic advantage for separating patients with functional heartburn from the large proportion of patients with gastroesophageal reflux symptoms but no macroscopic esophagitis or pathological acid exposure. The bioimpedance of the esophageal mucosa, measured with a small caliber probe, was evaluated in a series of preclinical experiments. First, sections of rabbit esophageal epithelium were mounted in Ussing chambers and exposed to solutions at pH 7.4 or pH 1.5 for 45 minutes. Impedance measurements were taken at varying probe pressures. Second, rabbit esophageal epithelia were perfused for 45 minutes in situ with pH 1.1 or control solutions and impedance measurements taken. Samples from both in vitro and in situ experiments were taken for morphological examination by light microscopy. Finally, esophageal bioimpedance was measured in awake dogs with permanent esophagocutaneous stoma. The in situ experiments demonstrated that morphological changes in the esophageal mucosa could be discerned by the use of bioimpedance spectroscopy. The variability in resistivity was species-independent but was affected by the pressure applied to the probe. The results suggest that evaluation of bioimpedance spectroscopy for use in a clinical setting is warranted. Small morphological differences in the esophageal mucosa may be detected by the use of bioimpedance spectroscopy.

Use of tobacco products and gastrointestinal morbidity: an endoscopic population-based study (the Kalixanda study)

The impact of snus (smokeless tobacco or snuff) on gastrointestinal symptoms and pathological findings is largely unknown. The authors aimed to investigate whether the exposure to different forms of tobacco influences upper gastrointestinal symptoms, histology and frequency of Helicobacter pylori infection. A random sample (n = 2,860) of the adult population of two northern Swedish municipalities Kalix and Haparanda (n = 21,610) was surveyed between December 1998 and June 2001 using a validated postal questionnaire assessing gastrointestinal symptoms (response rate 74.2%, n = 2,122) (The Kalixanda Study). A random sub-sample (n = 1,001) of the responders was invited to undergo an esophagogastroduodenoscopy (participation rate 73.3%) including biopsies, Helicobacter pylori culture and serology and symptom assessment and exploration of present and past use of tobacco products. No symptom groups were associated with snus use. Snus users had a significantly higher prevalence of macroscopic esophagitis univariately but snus use was not associated with esophagitis in multivariate analysis. Snus use was associated with basal cell hyperplasia (OR = 1.74, 95% CI: 1.02, 3.00) and with elongation of papillae (OR = 1.79, 95% CI: 1.05-3.05) of the squamous epithelium at the esophago-gastric junction. Current smoking cigarettes was associated with overall peptic ulcer disease (OR = 2.32, 95% CI: 1.04, 5.19) whereas snus use was not. There were no significant association between current Helicobacter pylori infection and different tobacco product user groups. Snus significantly alters the histology of the distal esophagus but does not impact on gastrointestinal symptoms or peptic ulcer disease.

Gastroesophageal Reflux in Children with Refractory Asthma

To investigate the association between clinical and macroscopic findings of GER and severity of refractory asthma (statistical study). A prospective study in a university-based practice of 75 children who were diagnosed with refractory asthma because they exhibited no satisfactory response for at least three months of treatment. Medical history, physical examination, spirometrical measurements and prick skin test were conducted on all patients. Endoscopic and macroscopic evaluations for esophagitis were performed on all patients regardless of the presence of GER symptoms. Endoscopy was done for 75 children with refractory asthma. GER was symptomatic in 65% of all patients with no statistical significance (p>0.05) and the most frequent symptom was abdominal pain (67%). The frequency of these symptoms was 50% in mild asthma, 58% in moderate asthma and 72% in severe asthma with no statistical significance (p>0.05). The frequency of macroscopic esophagitis was 71%, distributed in three asthmatic groups as in order of 75% in mild asthma, 58.3% in moderate asthma and 76.6% in severe asthma (p>0.05). Regarding the relationship between GER and nocturnal attacks or spasmodic cough, the frequency of the latest was 66.7%. The differences have no statistical significance regarding GER symptoms (p>0.05), but they are statistically significant regarding the reflux esophagitis (p<0.05). Also endoscopic reflux was as frequent as76.3% in non allergic patients, and this result is statistically significant (p<0.05). In spite of the fact that there was no relationship between the severity of asthma and the symptoms of GER, or its endoscopic findings, the frequency of GER in asthmatic children was higher than its frequency in other children.

Indicaciones de cirugía antirreflujo: ¿Buenos y malos candidatos para la cirugía? Revisión bibliográfica y experiencia personal

Antireflux laparoscopic surgery has excellent results in terms of improvement of symptoms, esophagitis, gastroesophageal sphincter competence and abnormal acid reflux. Indications for surgery are well established, however some of these are controversial. This is a review of the present indications for surgery in gastroesophageal reflux. The surgical indication should be the result of a complex clinical and laboratory work up. Patients with a clear cut surgical indication should be differentiated from those with doubtful indications, that require further analysis and those that are bad candidates for surgery. Young patients with macroscopic esophagitis, an incompetent sphincter, abnormal acid reflux test, that have a partial or negative response to treatment with proton pump inhibitors are those with the best surgical results. Bad candidates are patients with a psychiatric background, with atypical symptoms and those with a normal acid reflux test. In our experience with 935 patients, only 23% had a surgical indication.

Indications for antireflux surgery: a clinical experience and review of the literature

Antireflux laparoscopic surgery has excellent results in terms of improvement of symptoms, esophagitis, gastroesophageal sphincter competence and abnormal acid reflux. Indications for surgery are well established, however some of these are controversial. This is a review of the present indications for surgery in gastroesophageal reflux. The surgical indication should be the result of a complex clinical and laboratory work up. Patients with a clear cut surgical indication should be differentiated from those with doubtful indications, that require further analysis and those that are bad candidates for surgery. Young patients with macroscopic esophagitis, an incompetent sphincter, abnormal acid reflux test, that have a partial or negative response to treatment with proton pump inhibitors are those with the best surgical results. Bad candidates are patients with a psychiatric background, with atypical symptoms and those with a normal acid reflux test. In our experience with 935 patients, only 23% had a surgical indication.

The Yield of a Continuously Patent Gastroesophageal Junction During Upper Endoscopy as a Predictor of Esophagitis in Children

Over the last years we have noted an association between the endoscopic finding of a continuously patent gastroesophageal junction (GEJ) throughout the procedure and macroscopic or microscopic esophagitis. We could not find documentation for these endoscopic findings as a predictor of esophagitis in the literature. We aimed to find an association between these findings and microscopic and macroscopic esophagitis. During upper endoscopy, we routinely observe the GEJ for about 60 s and note the behavior of the GEJ and esophageal contractions. Patients with a persistently patent GEJ were recorded. A group of patients referred for upper endoscopy for reasons other than suspected reflux, whose esophagus was normal, and patients with reflux symptoms served as a control groups. We found 21 patients (3.0%) in whom a patent GEJ had been noted. No significant age differences were noted between study and control groups. Eighteen out of 21 patients (86%) in the study group had varying degrees of microscopic esophagitis ranging from mild to severe (ten with mild esophagitis, three with moderate esophagitis, and five with severe esophagitis). Interestingly, ten out of 18 (55%) study patients with esophagitis on biopsies had no evidence of additional esophageal abnormality. Although all control patients had a normally appearing esophagus on upper endoscopy, 8/26 (31%) had mild esophagitis on biopsies. Differences were statistically significant (p < 0.001). A continuously patent GEJ predicts quite accurately the presence of esophagitis in biopsies and may serve an additional endoscopic finding for the diagnosis of esophagitis especially non-erosive GER.

Effects of Ecabet Sodium on Acute Mixed Reflux Esophagitis in Rats

Objective. Ecabet sodium (ENa) is a drug that repairs epithelial cells in the digestive tract via selective adhesion to damaged tissue. While the principal cause of reflux esophagitis is gastroesophageal acid reflux, the involvement of duodenal juice has also been identified as an important factor. This study aimed to explore the effect of ENa in an acute mixed reflux esophagitis (AMRE) rat model. Material and methods. Eight-week-old male Wistar rats were used to prepare an AMRE model. There were four experimental groups: Group A (sham-operated rats), Group B (AMRE rats), Group C (AMRE rats dosed with ENa at 10 mg/kg), and Group D (AMRE rats dosed with ENa at 30 mg/kg). All rats were assessed for incidence of macroscopic esophageal lesions, esophagitis index, and pathological findings. Amylase activity, bile acid concentration in the digestive fluid retained in the esophagogastric lumen and ENa concentration in the esophageal mucosa were determined. Results. The incidence of esophagitis was 0% for Group A, 100% for Group B, 40% for Group C, and 20% for Group D. It was significantly lower for Groups C and D relative to Group B. The median esophagitis index was 0 for Group A, 58.2 for Group B, 0 for Group C, and 0 for Group D, and it was significantly lower for Groups C and D relative to Group B. The histological severity grade of esophagitis in Groups C and D was significantly less than that in Group B. Conclusion. Treatment with ENa inhibited the development of AMRE in rats.

The Pathology of Gastric Cardia

"Carditis" (inflammation of the gastric cardiac mucosa) may be associated with gastroesophageal reflux disease (GERD), whereas other studies argue that Helicobacter pylori could play a significant role in the chronic cardiac damage. We examined prospectively histologic features of gastric cardia, esophagitis, and H. pylori status in 204 consecutive subjects with GERD symptoms (57.3% male, 42.7% female mean age 49.2 y) undergoing upper gastrointestinal endoscopy with multiple biopsies in the distal esophagus, cardiac region, and stomach. These were assessed for esophagitis landmarks [Ismail Beigi grading (g0-3)], gastritis, and H. pylori infection (Sydney classification). The average symptom duration was 10.8 months. Endoscopy showed no erosive disease in 54.5% patients, grade "A" esophagitis in 37.6%, "B" in 8%, and "C" in 1 case. Histologic examination disclosed g0 in 8.3% patients, g1 in 78.4%, g2 in 12.8%, and g3 in 1; analysis of the cardia showed oxyntic mucosa in 27.9% patients and chronic cardiac mucosa inflammation in 72.1%. Carditis was significantly related to macroscopic esophagitis (P=0.044) and heartburn score (P=0.001). H. pylori cardiac infection was present in 27.4% cases (73.2% associated with cardiac mucosa). Gastric H. pylori infection was demonstrated in 35% patients. H. pylori in the cardiac region was associated with gastric H. pylori infection (P=0.001) and with paucity of GERD symptoms (P=0.05). A good correlation between carditis and GERD, concerning symptoms and macroscopic esophagitis was found in this study. H. pylori-related carditis is likely to be differently compared with the GERD-related type.

Maintenance Therapy for Erosive Esophagitis in Children After Healing by Omeprazole: Is It Advisable?

To evaluate the efficacy of acid-suppressive maintenance therapy for gastroesophageal reflux disease (GERD) in children, after the healing of reflux esophagitis. Forty-eight children (median age 105 months, range 32-170) with erosive reflux esophagitis were initially treated with omeprazole 1.4 mg/kg/day for 3 months. Patients in endoscopic remission were assigned in a randomized, blinded manner by means of a computer-generated list to three groups of 6-month maintenance treatment: group A (omeprazole at half the starting dose, once daily before breakfast), group B (ranitidine 10 mg/kg/day, divided in two doses), and group C (no treatment). Endoscopic, histological, and symptomatic scores were evaluated at: T0, enrollment; T1, assessment for remission at 3 months after enrollment (healing phase); T2, assessment for effective maintenance at 12 months after T0 (3 months after the completion of the maintenance phase). Relapse was defined as the recurrence of macroscopic esophageal lesions. After the completion of the maintenance phase, patients without macroscopic esophagitis relapse were followed up for GERD symptoms for a further period of 30 months. Of 48 initially treated patients, 46 (94%) healed and entered the maintenance study. For all patients, in comparison to T0, the histological, endoscopic, and symptomatic scores were significantly reduced both at T1 and T2 (P<0.0001, for each). No significant difference was found in these three scores, comparing group A, B, and C at T1 and T2. A relapse occurred in one patient only, who presented with macroscopic esophageal lesions at T2. Three months after the completion of the maintenance phase, 12 (26%) patients complained of symptoms sufficiently mild to discontinue GERD therapy, excluding the patient who showed macroscopic esophagitis relapse. Three of 44 (6.8%) patients reported very mild GERD symptoms within a period of 30 months after maintenance discontinuation. Our pediatric population showed a low rate of erosive esophagitis relapse and GERD symptom recurrence long term after healing with omeprazole, irrespective of the maintenance therapy.

Sandifer Syndrome Posturing: Relation to Abdominal Wall Contractions, Gastroesophageal Reflux, and Fundoplication

Sandifer syndrome designates abnormal posturing in patients with gastroesophageal reflux. To explore its mechanisms via examining relationships among Sandifer syndrome posturing, abdominal wall contractions, and reflux episodes, we studied an affected child in detail. The study utilized esophageal pHmetry, surface electromyography, and split-screen videography. The multichannel physiologic study demonstrated association of rectus abdominis contraction with onset of reflux episodes (P < 0.001) and association of reflux episodes with Sandifer syndrome posturing. This child's subsequent course confirmed his diagnosis and suggested mechanisms of the association of reflux and Sandifer syndrome. We conclude that abdominal wall contractions may induce reflux episodes. Sandifer syndrome may be due to gastroesophageal reflux even without hiatal hernia, macroscopic esophagitis, or reflux symptoms. Despite the absence of more typical reflux symptoms and failure to respond to very aggressive medical therapy, Sandifer syndrome may resolve after fundoplication.

The Impact of Antireflux Surgery on Oxidative Stress of Esophageal Mucosa Caused by Gastroesophageal Reflux Disease: 4-yr Follow-Up Study

Oxidative stress to esophageal mucosa plays a key role in the pathogenesis of gastroesophageal reflux disease (GERD), Barrett's esophagus, and adenocarcinoma. We investigated whether successful antireflux surgery eliminates oxidative stress. Oxidative stress of esophageal mucosa was measured in 20 GERD patients, before antireflux surgery and 6 and 48 months after it, and compared with normal controls' mucosa (N = 9). Preoperatively, 12 of the 20 had erosive esophagitis or Barrett's metaplasia. Postoperatively, healing of GERD was verified with endoscopy and 24-h pH monitoring. We measured oxidative stress by myeloperoxidase activity (MPA), superoxide dismutase activity, and glutathione content (GSH) in distal esophagus samples from endoscopy. No patient had reflux symptoms after surgery, and pH measurements had normalized. MPA in the distal esophagus decreased (p < 0.05) after successful antireflux surgery, but remained higher than that of controls both 6 months and 4 yr postoperatively (p < 0.05). At all time-points, MPA was higher in patients with preoperatively detected erosive reflux disease (ERD) as compared to non-erosive reflux disease (NERD) (p < 0.01, p < 0.05, and p < 0.05, respectively). GSH values decreased with time only in NERD. At all time-points, GSH levels in distal esophagus were lower than control levels. Antireflux surgery can heal macroscopic esophagitis but cannot fully reverse the oxidative stress (as reflected by MPA and GSH) upon the distal esophageal mucosa.

Oesophagogastrostomy without a drainage procedure in oesophageal carcinoma.

Ten patients with carcinoma of the lower third of the oexophagus who had oesophagogastrectomy followed by oesophagogastrostomy without a drainage procedure were investigated 6 months postoperatively. Postoperative symptomatic evaluation and modified Visick grading of the results were carried out by an independent observer. Objective assessment included fibre-optic endoscopy of the oesophagus, stomach and duodenum, gastric acid secretory studies and measurement of gastric emptying times following the ingestion of a barium food mixture. Eight of the 10 patients had an excellent or very good result and the remaining 2 were graded as satisfactory. No evidence of macroscopic oesophagitis or gastritis was detected and pyloric obstruction was never observed. Gastric emptying time was not prolonged. The results obtained support the concept that a drainage procedure is unnecessary after oesophagogastrostomy.

Heartburn in patients with gastro-oesophageal reflux disease in Germany and Sweden: a study on patients' burden of disease.

Gastro-oesophageal reflux disease (GORD) is widespread in the general population and presents in most cases with heartburn as the main symptom. The severity of symptoms is not necessarily related to erosive damage to the oesophagus due to acid reflux, but the frequency and severity of symptoms have an impact on the health-related QOL (HR-QOL) of the patient. To characterise patients with GORD who consult a physician because of heartburn with respect to medical background and burden of disease in Germany and Sweden. A total of 1011 patients who had been experiencing symptoms of GORD, including heartburn, for at least 1 year were recruited by physicians, who collected data on the patients' previous diagnosis and treatment. The patients themselves were interviewed by telephone about their heartburn. The EuroQOL 5-dimensional HR-QOL questionnaire (EQ-5D) and the Gastrointestinal Symptom Rating Scale (GSRS) were used to measure patient-reported outcomes. About half of the patients had been experiencing symptoms of GORD for >5 years. The majority (54%) perceived the severity of their heartburn as moderate. Thirty-eight percent of the patients had been examined by endoscopy during the last 12 months and 79% of these had macroscopic oesophagitis. Medical treatment for heartburn had been prescribed to 88% of the patients. The mean annual number of days with heartburn was assessed as 170 days per patient, with the majority of patients (64%) experiencing heartburn every week. HR-QOL was reduced in these patients as expressed by EQ-5D scores compared to scores for a general population. Patients with GORD had problems related to gastrointestinal symptoms as rated by GSRS scores. More severe heartburn was associated with a greater number of symptom days and reduced HR-QOL scores, whereas a relationship between the findings at endoscopy and the severity of symptoms or HR-QOL could not be found. This study demonstrates that heartburn confers a significant burden on patients with GORD, related to the frequency and severity of symptoms and reflected in the reduction in HR-QOL as perceived by the patients.

Study of intestinal metaplasia in Barrett’s esophagus

OBJECTIVE: The aim of the present study was to investigate the factors predisposing patients with Barrett’s esophagus (BE) to intestinal metaplasia (IM). METHODS: Forty‐seven BE patients were studied. By using endoscopic and histological methods, esophageal IM was diagnosed in 36 patients, who were compared with 11 patients without IM in regard to their age, the endoscopic appearance of the esophagus and esophageal motility. RESULTS: The patients’ age and endoscopic features of the esophagus, but not esophageal motor function and grade of macroscopic esophagitis, were risk factors for the development of IM. CONCLUSIONS: Patient age and endoscopic features of the BE mucosa are associated with IM and may be prognostic factors for BE.

Esophageal histology does not provide additional useful information over clinical assessment in identifying reflux patients presenting for esophagogastroduodenoscopy.

We prospectively evaluated the value of histology in identifying gastroesophageal reflux disease (GERD) in consecutive patients enrolled for upper endoscopy. GERD was defined as heartburn occurring at least weekly. Macroscopic esophagitis was graded and an esophageal biopsy was taken 2 cm above the gastroesophageal junction. Histological esophagitis was identified by: (1) basal cell hyperplasia >15%, (2) increased papillary length >66%, and (3) infiltration by leukocytes/eosinophils. The sensitivity, specificity, and positive and negative predictive value of histological esophagitis in patients with and without typical reflux symptoms, with and without endoscopic changes, or both were evaluated. Of 178 patients, reflux symptoms were present in 59% (N = 105) and esophageal erosions in 19% (N = 34); 75 patients had reflux symptoms but no erosions. While the specificity of histology was adequate (78%), it was insensitive (30%). The positive and negative predictive values were 67% and 44%, respectively. No single individual parameter was better than any other. Thus, histology appears to be of no additional value in identifying GERD.

Treatment of Cytomegalovirus Esophagitis in Patients With Acquired Immune Deficiency Syndrome: A Randomized Controlled Study of Foscarnet Versus Ganciclovir

Although several uncontrolled studies have shown that the response rate to ganciclovir and foscarnet for all forms of cytomegalovirus (CMV) infection in immunocompromised patients is almost similar, to date, no controlled clinical trial has been specifically designed to compare these two agents in the treatment of CMV esophagitis. The aim of this study was, therefore, to compare the efficacy and safety of these two drugs in the induction therapy of CMV esophagitis in patients with acquired immunodeficiency syndrome (AIDS). Thirty-nine of 211 (18%) consecutive AIDS patients undergoing endoscopy for esophageal symptoms had macroscopic esophagitis that proved to be sustained by CMV based on the documentation of typical intranuclear inclusions at histology; 23 were considered eligible for this study and were randomized to receive foscarnet 90 mg/kg b.i.d. or ganciclovir 5 mg/kg b.i.d. for 21 days. Twelve patients received foscarnet, whereas 11 were treated with ganciclovir. Clinical and laboratory evaluation was performed weekly, and endoscopy was repeated at the end of therapy. The two treatment groups were well balanced as to the following characteristics at entry: age, sex, absolute number of CD4 cells, duration of AIDS, Karnofsky score, frequency of concomitant Candida esophagitis (grade I or II), and severity of esophageal symptoms. Marked endoscopic improvement (complete disappearance of macroscopic lesions or significant reduction of the endoscopic score) was observed in eight of 11 (73%) of foscarnet and seven of 10 (70%) of ganciclovir-treated patients, and inclusion bodies disappeared from follow-up biopsies in 55% and 50% of patients, respectively. The symptomatic response was also similar for both treatments: 82% of patients who received foscarnet and 80% of those treated with ganciclovir had a complete or at least a good clinical response. Frequency of adverse events was comparable with both drugs: only one patient in each group suspended treatment because of severe side effects. Foscarnet and ganciclovir appear to be similarly effective and safe in the induction therapy of AIDS-related CMV esophagitis. Consequently, the choice of the anti-CMV agent should be tailored to the individual patient according to the different toxicity profiles of the two drugs.

Treatment of cytomegalovirus esophagitis in patients with acquired immune deficiency syndrome: A randomized controlled study of foscarnet Versus ganciclovir

Objective Although several uncontrolled studies have shown that the response rate to ganciclovir and foscarnet for all forms of cytomegalovirus (CMV) infection in immunocompromised patients is almost similar, to date, no controlled clinical trial has been specifically designed to compare these two agents in the treatment of CMV esophagitis. The aim of this study was, therefore, to compare the efficacy and safety of these two drugs in the induction therapy of CMV esophagitis in patients with acquired immunodeficiency syndrome (AIDS). Methods Thirty-nine of 211 (18%) consecutive AIDS patients undergoing endoscopy for esophageal symptoms had macroscopic esophagitis that proved to be sustained by CMV based on the documentation of typical intranuclear inclusions at histology; 23 were considered eligible for this study and were randomized to receive foscarnet 90 mg/kg b.i.d. or ganciclovir 5 mg/kg b.i.d. for 21 days. Twelve patients received foscarnet, whereas 11 were treated with ganciclovir. Clinical and laboratory evaluation was performed weekly, and endoscopy was repeated at the end of therapy. The two treatment groups were well balanced as to the following characteristics at entry: age, sex, absolute number of CD4 cells, duration of AIDS, Karnofsky score, frequency of concomitant Candida esophagitis (grade I or II), and severity of esophageal symptoms. Results: Marked endoscopic improvement (complete disappearance of macroscopic lesions or significant reduction of the endoscopic score) was observed in eight of 11 (73%) of foscarnet and seven of 10 (70%) of ganciclovir-treated patients, and inclusion bodies disappeared from follow-up biopsies in 55% and 50% of patients, respectively. The symptomatic response was also similar for both treatments: 82% of patients who received foscarnet and 80% of those treated with ganciclovir had a complete or at least a good clinical response. Frequency of adverse events was comparable with both drugs: only one patient in each group suspended treatment because of severe side effects. Conclusions Foscarnet and ganciclovir appear to be similarly effective and safe in the induction therapy of AIDS-related CMV esophagitis. Consequently, the choice of the anti-CMV agent should be tailored to the individual patient according to the different toxicity profiles of the two drugs.

Mucosal reactive oxygen species production in oesophagitis and Barrett's oesophagus.

Reactive oxygen species (ROS) produced by inflammatory cells can contribute to tissue destruction. ROS have been implicated in various gastrointestinal abnormalities, including the acid related peptic diseases. Although the development of oesophagitis and Barrett's columnar epithelium is associated with prolonged reflux of gastric acid, the exact mechanism by which tissue damage occurs is not known. To discover if ROS are involved in damage to the oesophageal mucosa, this study measured in vitro the mucosal ROS concentrations of biopsied mucosal samples taken from patients with reflux oesophagitis using luminol enhanced chemiluminescence (LECL). Mucosal biopsy specimens were taken from 83 patients: 19 with normal oesophageal mucosa (group I); 20 with macroscopic oesophagitis (group II); 20 with biopsy confirmed Barrett's epithelium without macroscopic oesophagitis (group III); and 24 with Barrett's epithelium with macroscopic oesophagitis (group IV). The mucosa from patients exhibited significantly higher LECL values than the mucosa from controls. But, there were no significant differences between groups II, III, and IV. Addition of the myeloperoxidase inhibitor, azide, or the hydrogen peroxide scavenger, catalase, to the tissue suspension caused a decrease in LECL values of 32% and 45%, respectively, suggesting that neutrophils--although important--are not the only source of mucosal LECL. These data are consistent with the proposal that ROS play an important part in the tissue injury associated with oesophagitis and Barrett's columnar epithelium.