Circulating acylated ghrelin level decreases in accordance with the extent of atrophic gastritis
- First Online:
- Cite this article as:
- Kawashima, J., Ohno, S., Sakurada, T. et al. J Gastroenterol (2009) 44: 1046. doi:10.1007/s00535-009-0120-0
We attempted to clarify the significance of atrophic change of gastric mucosa for reduction of plasma ghrelin concentration irrespective of Helicobacter pylori (Hp) infection.
Plasma acylated (acyl-)ghrelin concentration in 220 subjects, both with and without atrophic gastritis, was measured with an enzyme immunoassay kit. The extent of atrophic change of gastric mucosa was assessed and graded endoscopically. Hp infection was determined by assay of the anti-Hp antibody with an ELISA assay kit.
Plasma acyl-ghrelin concentration was significantly lower in the Hp positive than the Hp negative group, and Hp eradication significantly increased plasma acyl-ghrelin concentrations in Hp positive subjects. Plasma acyl-ghrelin was significantly lower in subjects with severely atrophic gastritis than in those with mild or moderate atrophic gastritis, irrespective of Hp infection or age group (<60 years old or ≧60 years old). In male subjects with a normal body mass index, plasma acyl-ghrelin concentrations in subjects with severely atrophic gastritis were significantly lower than those in subjects with mildly or moderately atrophic gastritis, suggesting that the results of the study are independent of emaciation or obesity. Logistic regression analysis showed that gastric atrophy is the key factor that modulates plasma acyl-ghrelin levels.
The results suggest that plasma acyl-ghrelin concentration decreases in accordance with the extent of atrophic change in gastric mucosa irrespective of Hp infection, indicating that the low plasma acyl-ghrelin level of subjects with Hp infection is mainly caused by the progress of atrophic changes in gastric mucosa.
KeywordsGhrelin Helicobacter pylori Eradication Atrophic gastritis
Ghrelin, a novel peptide recently identified in the stomach of rodents and humans[1, 2], is a potent stimulator of growth hormone release and food intake [1, 3, 4, 5, 6, 7, 8, 9] and also stimulates the mobility and acid secretion of the stomach [10, 11]. We have previously shown that ghrelin stimulates the secretion of gastric acid through an increase in the production of histamine . Ghrelin is as potent as gastrin in the stimulation of gastric acid secretion in rats. Since the level of circulating ghrelin is maximal during fasting and decreases after feeding , it is possible that ghrelin levels might be related to acid secretion and gastric motility during fasting. Because ghrelin has such important effects on stomach function, a decrease in ghrelin secretion could induce dysfunction of the stomach and thereby contribute to hypochlorhydria in patients with atrophic gastritis or to the disturbed mobility of patients with functional gastrointestinal dyspepsia with symptoms such as abdominal distension, easy satiety, and appetite loss. Therefore, in order to understand the pathophysiology of these diseases, the dynamics of circulating ghrelin levels in these patients need to be clarified. Recent reports showed that Helicobacter pylori (Hp) infection and atrophic gastritis decrease the level of circulating ghrelin and its production in gastric mucosa [14, 15, 16, 17, 18]. Thus, plasma ghrelin concentrations have been shown to increase after eradication of an Hp infection . Additionally, atrophic gastritis was shown to impair the production of gastric ghrelin leading to a decrease in plasma ghrelin concentration . As the stomach is the main source of circulating ghrelin , these reports suggest that the level of circulating ghrelin is possibly dependent on the secretory function of the stomach [14, 17]. To date, however, the regulation of gastric ghrelin secretion and the mechanism underlying the dysfunction in its secretion are poorly understood. As mentioned above, in previous studies inflammation and damage to the gastric mucosa are believed to affect the secretion of ghrelin. Either functional or histological damage of gastric mucosa, or both, might induce impairment of the secretion of ghrelin. Thus, the production of inflammatory cytokines and the loss of ghrelin-producing cells may be responsible for the decrease in ghrelin production and secretion. In this study, we attempted to clarify the effect of Hp infection and atrophic change of gastric mucosa on plasma ghrelin concentrations. We further aimed to clarify which of these two factors is decisive. In this study, we evaluated the level of active plasma ghrelin by assay of acylated (acyl-)ghrelin, because only acyl-ghrelin can stimulate feeding, acid secretion and motility of the stomach [20, 21, 22], which are important for the maintenance of normal food intake.
Subjects and acyl-ghrelin assay
Characteristics of subjects
Endoscopic diagnosis of atrophic gastritis
Diagnosis of Hp infection and eradication treatment
Hp infection was determined by assay of the presence of an anti-Hp antibody with a commercially available assay kit (Eiken Chemical Co, Ltd., Tokyo, Japan). Of the 163 Hp positive subjects, 130 had a peptic ulcer. Fifty-four subjects with a peptic-ulcer accepted treatment to eradicate Hp and were treated with a triple therapy consisting of a daily regimen of 30 mg lansoplazole bid, 750 mg amoxicillin bid and 400 mg clarithromycin bid for seven days. After six weeks or more (average duration of 22.9 ± 2.7 weeks) diagnosis of Hp eradication was determined by a 13C-urea breath test, and plasma samples were collected to evaluate the effect of Hp eradication on plasma acyl-ghrelin levels.
Correlation between endoscopic assessment and histological assessment of atrophic gastritis
To confirm the endoscopic diagnosis of the location of the atrophic border, a histological examination was performed on biopsy specimens of some subjects (n = mild atrophy, 8; moderate atrophy, 7; severe atrophy, 11) that were taken from the following four different sites in the stomach: #1: at the middle of the lesser curvature of the antrum, #2: at the angle, #3: at the center of the lesser curvature between the angle and the cardia and #4: near the cardia. Histological evaluation of gastric atrophy was done according to the Updated Sydney system . To compare histological atrophy, scores were assigned to each biopsy specimen according to the grade of mucosal atrophy (no atrophy: 0, mild: 1, moderate: 2, severe: 3).
The study protocol was approved by the ethics committees of Saitama Medical Center, Saitama Medical University. Written informed consent was obtained from all subjects prior to enrollment.
Data values were expressed as means ± SE. Analysis of the statistical significance of observed differences was performed using the Wilcoxon signed-rank test. Comparison of data between more than three groups was performed using one-way analysis of variance followed by Scheffe’s F test. To compare the deviation of the atrophic state, Mann–Whitney’s U test was performed. To clarify the key factor that determines plasma acyl-ghrelin levels, the relationship between acyl-ghrelin and the variables that appeared to affect plasma acyl-ghrelin levels was assessed by multivariate analysis (logistic regression analysis). The variables examined in logistic regression analysis were: age, body mass index (BMI), gender, Hp infection and gastric atrophy. All statistical analyses were performed with STATVIEW for Windows version 5.0 (SAS Institute, Cary, NC, USA). P values of less than 0.05 were considered significant.
Correlation between endoscopic and histological assessment of atrophic gastritis
Histological atrophic scores for each grade of endoscopic atrophic gastritis at both biopsy sites
1.13 ± 0.23
0.5 ± 0.27
0.25 ± 0.25
1.71 ± 0.30
1.71 ± 0.30**
1.57 ± 0.33**
0.57 ± 0.42
2.1 ± 0.1
1.90 ± 0.16***
2.18 ± 0.17***
2.0 ± 0.1**, +++
Relationship between Hp infection and plasma acyl-ghrelin levels
Effect of Hp eradication on plasma acyl-ghrelin levels
Relationship between the severity of atrophic gastritis and plasma acyl-ghrelin levels
The number of each grade of atrophic gastritis in Hp positive and negative groups
Relationship between the severity of atrophic gastritis and plasma ghrelin levels in subjects with different ages
Relationship between the severity of atrophic gastritis and plasma ghrelin levels in male subjects with a normal BMI
Multivariate analysis of factors that affect plasma acyl-ghrelin levels
Multivariate analysis (logistic regression analysis) of variables associated with plasma acyl-ghrelin level in all subjects
In recent years, a relationship between plasma ghrelin levels and stomach diseases such as Hp infection and atrophic gastritis has been reported [14, 15, 16, 17, 18]. However, the significance of atrophic gastritis independent of Hp infection for determination of plasma ghrelin levels remains unclear. In this study, we undertook to clarify a role of atrophic change of gastric mucosa in the determination of plasma ghrelin levels. First, we studied the relationship between Hp infection and plasma ghrelin levels and elucidated that plasma levels of acyl-ghrelin in Hp-positive subjects were significantly lower than those in Hp-negative subjects. We also confirmed that eradication of Hp led to a significant increase in plasma acyl-ghrelin levels. Since it has been elucidated that subjects with Hp infection have active gastritis and those without Hp infection have no active gastritis, the results indicate that gastritis may induce a decrease in plasma acyl-ghrelin levels.
Based on the fact that the Hp-positive group in this study had significantly more subjects with severely atrophic gastritis than the Hp-negative group, we hypothesized that there might be a strong relationship between the severity of atrophic gastritis and plasma ghrelin levels. In this study, the extent of atrophy of the gastric mucosa was determined by endoscopic examination and the severity of atrophic change was determined according to the classification of Kimura–Takemoto . A positive correlation between endoscopic data regarding atrophic gastritis and histological data concerning glandular atrophy has previously been confirmed by examination of biopsies taken from multiple gastric sites . Using similar methods, our data confirm these findings. These results suggest that endoscopic findings are a good indicator of histological change due to mucosal atrophy.
In this study, we have shown that the severity of atrophic gastritis strongly decreases the plasma acyl-ghrelin level irrespective of Hp infection. Previous studies have demonstrated negative correlations between age [26, 27] or BMI  and plasma ghrelin levels. The present study shows that the decreased plasma acyl-ghrelin level in subjects with severe atrophic gastritis is also independent of age and BMI. Our data show that atrophic gastritis decreases plasma ghrelin levels to the same extent whether the subjects are under or over 60 years old, and occurs when the subjects are all male with a normal BMI. Of course, age and BMI are factors that could possibly affect plasma ghrelin levels. In particular, plasma ghrelin concentration is related to body weight, being low in obese individuals, high in lean individuals and patients with anorexia nervosa [28, 30, 31, 32], and returning to normal levels upon weight gain and recovery from disease [30, 33]. However, our results indicate that atrophic change in gastric mucosa exerts a strong influence on plasma ghrelin levels independent of Hp infection, age, or body weight. Logistic regression analysis in this study also showed that atrophy of gastric mucosa is the only significant factor that affects plasma acyl-ghrelin levels. This result suggests that atrophy of gastric mucosa is a key factor modulating the decrease in plasma acyl-ghrelin levels.
As shown in the present study, Hp infection has previously been shown to decrease plasma ghrelin levels [14, 15, 16, 17]. However, a study by Gokcel indicated that there was no relationship between plasma ghrelin levels and Hp infection . The discrepancy between the data of these previous reports has not been explained. However, when evaluating the effect of Hp infection of plasma ghrelin levels it is clearly necessary to consider the severity of atrophic gastritis as a contributing factor. Since the ghrelin-producing cells are located in the corpus of the stomach , inflammation and atrophy of the fundic mucosa could potentially impair the function of ghrelin-producing cells. A study by Takeguchi showed that the number of ghrelin-positive cells in the stomach was lower in Hp-infected patients than in healthy controls . The study by Osawa et al. , also indicated that Hp infection significantly lowers the expression of ghrelin mRNA and the number of ghrelin-producing cells as well as the plasma ghrelin levels . These results suggested that Hp infection impairs the production of ghrelin in gastric mucosa, resulting in a decrease in plasma ghrelin levels.
The correlation between ghrelin levels and the histological severity and topographic extent of Hp-associated chronic gastritis has been assessed . The study revealed that there is a stepwise decrease in ghrelin levels that descends from the normal level to the antrum-predominant pattern and to corpus-predominant gastritis, that plasma ghrelin levels correlated significantly with pepsinogen I levels and with the pepsinogen I/II ratio, which are used as the markers of gastric mucosal atrophy . Furthermore a significant decrease in both pepsinogen I levels and the pepsinogen I/II ratio, that occurred together with the decrease in plasma ghrelin levels in Hp-positive subjects, has also been reported in another study . Those data suggested that the extent of atrophic change in gastric mucosa is a cause of decreased plasma ghrelin levels and this fact was also shown in a study by another group [18, 35]. The report by Suzuki also indicated that the serum level of pepsinogen I and the pepsinogen I/II ratio correlated with plasma levels of total and active ghrelin, respectively. Furthermore a relationship between total or active ghrelin levels and gastric atrophy was studied based on endoscopic examination. It was shown that the plasma levels of total and active ghrelin were significantly lower in patients with endoscopically diagnosed open-type atrophy (severe atrophy) than in those with closed type atrophy (mild or moderate atrophy) . However, the number of subjects recruited for the endoscopic study was relatively (69), and the low level of ghrelin in patients with atrophic gastritis was seen only in the Hp-eradicated group and not the Hp-positive group. Based on these results, the authors proposed that plasma ghrelin levels might be a potentially useful non-invasive marker for chronic gastritis. While our results are almost the same as the results of that study, the present study was based on data from a higher number of subjects (220 subjects). Furthermore, in our study the relationship between plasma ghrelin levels and gastric mucosal atrophy was observed not only in Hp-negative subjects but also in Hp positive subjects, indicating that gastric mucosal atrophy is a strong and independent factor for determination of plasma ghrelin levels.
It is well known that patients with atrophic gastritis have decreased secretion of gastric acid . Although the effect of ghrelin on gastric acid secretion has not been determined in humans, it has been elucidated in in vivo experiments with rats [10, 11, 12] making it very likely that ghrelin has the same effect in humans. Impaired production of ghrelin is a possible mechanism leading to gastric hyposecretion  and dysmobility  in subjects with atrophic gastritis caused by Hp infection.
In this study, we have shown that the plasma acyl-ghrelin level is significantly lower in Hp positive subjects than in Hp negative subjects and also in subjects with severely atrophic gastritis diagnosed by endoscopic examination. Endoscopic examination is the most popular tool for evaluating the severity of atrophic gastritis and is a good indicator of gastric atrophy that possibly induces the loss of ghrelin-producing cells. Therefore, it may be possible to predict plasma ghrelin levels based on endoscopic findings. We believe that the strong relationship between plasma ghrelin levels and the severity of atrophic gastritis that we observed might provide a mechanism to explain symptoms, such as abdominal distension and anorexia, that are seen in patients with severely atrophic gastritis . These symptoms are therefore possibly related to insufficiency of secretion of gastric ghrelin due to chronic gastritis.
In conclusion, our results suggest that plasma acyl-ghrelin concentration decreases according to the extent of atrophic change of gastric mucosa irrespective of Hp infection, indicating that the low level of plasma ghrelin in subjects with Hp infection is mainly caused by the progress of atrophic change of gastric mucosa.