Invagination of a part of the stomach into the esophagus has been described in autopsies since the beginning of the 19th century. Case reports on this subject have described the condition as "herniation of gastric mucosa into the esophagus" (2), "mucosal prolapse at the esophagogastric junction" (3) and "retrograde prolapse of gastric mucosa into the esophagus" (4).Young et al, identifying the condition as inflamed gastric mucosa below the esophagogastric junction, reported it to be a cause of hematemesis (6). Thomas and Khatak reported a case of hematemesis after alcohol ingestion, who had previously undergone multiple endoscopies where a similar lesion was identified as gastritis and they pointed out that the lesion may sometimes be confused with gastritis (7). A case of recurrent hematemesis due to mucosal prolapse in association with hiatus hernia has been reported by Laforret (8).
In 1984 Shephard, proposing the term "prolapse gastropathy syndrome", reported 22 patients who presented with epigastric pain alone or with hematemesis associated with a previous history of recurrent early morning retching or postprandial retching (5). He defined the characteristic endoscopic finding in these patients as a knuckle of inflamed and sometimes bleeding gastric mucosa which repeatedly prolapses into the esophagus lumen during retching. In 1999, Byfield reported seven cases with similar findings (9).
Endoscopically there is a localized patch of hyperemic, congested, bright red mucosa varying in size from 1 to 6 cm in diameter. The lesion is found in the proximal stomach several centimeters distal to the gastroesophageal junction, frequently on the lesser curve. Active oozing of blood may be seen along with coffeee ground material in the stomach (4).
It has been suggested that the force of the repeated retching causes gastric mucosal trauma. The prolapsing gastric mucosa becomes edematous and inflamed, and a localized lesion in the knuckle is caused by a pressure effect on the mucosa and mucosal capilleries while it is being jammed into the esophagus lumen. Gastric irritability caused by this inflammation and from the retching itself probably perpetuates the syndrome.
The lesions detected have commonly been located in the greater curve (46%), lesser curve (43%), anterior wall (7%) and sometimes posterior wall (5,6). In the present study the location of the lesions was the greater curve in 10 patients (50 %), lesser curve in seven patients (35 %) and anterior wall in three patients (15 %).
A history of NSAID intake and alcohol ingestion has been reported at a rate of 46% in patients with PG but in this study, it was only 15%. Other causes such as gastroenteritis, hyperemesis gravidarum, uremia, malignancy, and duodenal ulcer have been reported (9). In the present study, it was found that chemotherapeutic medication and intestinal obstruction can also cause PG.
The incidence of PG has been reported as between 2% and 5.5% (5). Shephard found PG in 22 (2.4 %) of 914 upper gastrointestinal endoscopies, but it was found in only 20 (0.4 %) of 5054 endoscopic examinations in the present study.
Prolapse gastropathy has been reported in almost 2% of adults who undergo upper endoscopy for the evaluation of hematemesis and Bishop described it as a cause of hematemesis in seven of 27 pediatric patients (25.9%). We detected six cases (0.6%) among 914 patients on whom endoscopy was performed for evaluation hematemesis, but in none of these cases was hematemesis related to PG, because other lesions were identified as the cause. Our findings correlate with the results of large series evaluating 3294 and 2225 cases for upper gastrointestinal bleeding, which failed to mention PG as a cause of bleeding (14, 15).
Prolapse gastropathy is detected mostly in patients with retching and vomiting and when this group is considered, the incidence was 25.9%, with none of the patients having upper gastrointestinal bleeding.
Evaluation of other studies suggests that PG is mostly found in male patients at a rate of 87% (5, 9-11). Our results were similar, with a male predominace of 90%.
The need for blood transfusion has been reported in only five patients with PG (6-8). None of the 14 patients in our retching and vomiting group required blood transfusion, while two out of six patients in the hematemesis group required blood transfusion, but both of these patients had duodenal ulcers, which led to the need for transfusion.
In conclusion although PG has been reported as a cause of hematemesis, in this study ıt was found that it was not a cause of hematemesis when patients with hematemesis accompanied by retching and vomiting were evaluated. The present authors suggest that when PG is detected in a patient with hematemesis, there should be evaluation for other causes.
){kind=link}
){kind=link}
){kind=link}
){kind=link}
){kind=link}
){kind=link}