Relation between Helicobacter pylori and atrophic gastritis in elderly: Estimation of serum (gastrin 17 and pepsinogen-I) as novel biomarkers

Available online 1 September 2012

Peer review under responsibility of Alexandria University Faculty of Medicine.

Abstract

Background

Serum levels of gastrin-17 (S-G-17) and pepsinogen I (S-PGI) are biomarkers of gastric antral and corpus mucosa, respectively. We determined whether these tests, together with the assay of Helicobacter pylori antibodies, are a nonendoscopic tool for the diagnosis of atrophic gastritis.

Objectives

To study the relation between H. pylori infection and atrophic gastritis in the elderly through the estimation of gastrin 17 and pepsinogen I as novel noninvasive diagnostic biomarkers.

Material and methods

The present study included sixty individuals suffering from gastritis there were divided into two groups. Group (I): Thirty patients having H. pylori infection aged above 65 years. Group (II): Thirty patients without H. pylori infection aged above 65 years. All individuals were subjected to diagnostic upper-gastrointestinal endoscopy and specific gastric biomarkers (serum gastrin-17, serum pepsinogen I/II).

Results

Serum pepsinogen I, II, ratio and gastrin 17 decrease with increasing grade of atrophy of the antrum respectively. Serum pepsinogen I, II, gastrin 17 were significantly lower in mild, moderate, and severe gastritis compared to superficial gastritis patients.

Conclusion

The diagnosis of atrophic gastritis obtained with the blood test panel of G.17, SPG I/II and H. pylori antibodies is in good agreement with the endoscopic and biopsy finding. The panel is a tool for nonendoscopic diagnosis and screening of atrophic gastritis.

Abbreviation:

• H. pylori
• Helicobacter pylori
• PGI
• Pepsinogen I
• PGII
• Pepsinogen II
• G
• Gastrin

Keywords:

• Helicobacter pylori
• Atrophic gastritis
• Gastrin-17
• Pepsinogen-I
• Elderly

1 Introduction

Aging is a basic natural and biological process that goes on continuously in every individual and is responsible for changes affecting the organs by the passage of time. It starts with the onset of life and is terminated by death.¹

H. pylori infection increases with age. Its prevalence was found to approach 75% in patients above 65 years of age.^2–4

A recent retrospective study has found that H. pylori infection is associated with the majority of cases of symptomatic gastritis in the elderly patients.⁵

Recent studies have demonstrated a clear association between gastric colonization with H. pylori, gastritis and peptic ulcer disease.^6,7 Other epidemiological studies have also shown an association between infections with H. pylori and gastric cancer, the evidence is so complete that the world health organization has recently categorized H. pylori as a category I carcinogen.^8–12

Gastric atrophy characterized by the loss of antrum glands is considered a precursor of gastric adenocarcinoma since H. pylori increases the risk of gastric adenocarcinoma. It was been classified as a class I oncogenic factor,¹³ despite the central role of atrophic gastritis as a major risk factor for gastric carcinoma. Its diagnosis and its grading and proper assessments are done poorly or not at all in clinical practice.¹⁴

Because H. pylori infection is a necessary causative factor in the majority of cases of gastric cancer, eradication of the infection will very likely reduce the incidence of gastric cancer.¹⁴

The assays of serum/plasma pepsinogen I, pepsinogen II and pepsinogen I/pepsinogen II ratio in addition to gastrin 17 are easy tools for assessment of the presence or absence of atrophic gastritis even in a clinical setting whereas endoscopy is not available.

1.1 The aim of this work

To study the relation between H. pylori infection and atrophic gastritis in the elderly through the estimation of serum (gastrin-17 and pepsinogen-I) as novel noninvasive diagnostic biomarkers.

2 Methods

The present study included 60 individuals suffering from gastritis divided into two groups.

Group (I): Thirty patients having H. pylori infection aged above 65 years.

Group (II): Thirty patients without H. pylori infection aged above 65 years.

All patients included in the study were recruited from Main University Hospital and Fever Hospital.

2.1 Exclusion criteria

Patients with any other disease causing gastric insult were excluded from the study as were patients with diabetes mellitus, autoimmune diseases, chronic systemic infection, liver disease, chronic renal diseases and neoplastic disorders and also patients receiving drugs known to affect the gastric mucosa.

After taking a written consent from all patients, they were subjected to:

2.2 Endoscopy and biopsy sampling

Diagnostic upper gastrointestinal endoscopy was performed in all patients, two biopsies were taken from antrum to detect H. pylori, one for rapid urease and the other one for histopathological examination.

The biopsy specimens were immediately placed in a fixative (10% buffered formalin). The formalin fixed specimens routinely processed and embedded in paraffin blocks were serially cut and mounted on glass slides. The slides were stained with the conventional H&E stain. The prepared section was carefully examined by the light microscope and scored semi-quantitatively according to the updated Sydney classification system.¹⁵

2.3 Blood samples

The basal blood samples for the measurement of PGI, PGII, and gastrin 17 were obtained with the subjects in recumbent position, after an overnight fast, after venipuncture, serum and plasma sample were centrifuged and therefore kept at room temperature for 2 h. Before analysis samples were frozen at −20 °C for the latter assay of gastrin and pepsinogen.

Diagnosis of atrophic gastritis by the blood test panel in delineation of patients with different topographic types of atrophic gastritis took place; with the blood test panel an algorithm and a computer program (Gastrosuft, Bioint Plc) were used.

Three criteria for the best discrimination of gastritis of different types were obtained from previous studies.^15,16

2.4 Laboratory tests

2.4.1 Gastropanel test

(a)	Estimation of H. pylori IgG by ELISA with references negative <30, positive >30.17
(b)	Estimation of pepsinogen I by ELISA with reference range of 30–105 μg/l.17
(c)	Estimation of gastrin 17 by ELISA with a reference range of 1–10 mmol.17
(d)	Estimation of serum pepsinogen II by ELISA with reference range of 4–9 ng/ml.17
(e)	Estimation of pepsinogen I/II ratio by ELISA with reference range of 3–20.

2.5 Statistical methods

The Data were collected and entered into the personal computer. Statistical analysis was done using Statistical Package for Social Sciences (SPSS/version 17) software.

Arithmetic mean, standard deviation, for categorized parameters, chi-square test was used while for numerical data t-test was used to compare two groups while for more than two groups ANOVA test was used, least significant differences (LSD) was basically a t-test used only when the F-value is significant to detect the presence of significance between each of the two groups. The level of significance was 0.05.

3 Results

The mean age was 69.4 ± 5.04 years in H. pylori +ve group and 70.1 ± 3.69 years in H. pylori −ve group. There was no significant difference between both groups (P is significant if <0.05) (Table 1).

Table 1 Demographic data in the studied groups.

	H. pylori	n	Min	Max	Mean	SD	t	P
Age (years)	Negative	30	62	81	69.4	5.04	0.585	0.561
	Positive	30	65	78	70.1	3.69
BMI (kg/m^2)	Negative	30	17	24	21.2	2.20	0.483	0.631
	Positive	30	17	24	20.9	2.07

^∗P is significant if <0.05.

The mean body mass index (BMI) was 21.2 ± 2.20 kg/m² in H. pylori +ve group and 20.9 ± 2.07 kg/m² in H. pylori −ve group. There was no significant difference between both groups. (P is significant if <0.05) (Table 1).

The mean serum pepsinogen I was significantly lower in H. pylori +ve group (23.9 ± 21.84) unit compared to H. pylori −ve group (117.4 ± 43.47). (P is significant if <0.05) (Table 2).

Table 2 Relation between degree of gastritis and the other parameters.

	Gastritis degree	n	Mean	SD	F	P	LSD
Pepsinogen 1	Superficial	27	122.4	41.75	56.052	0.000^*	Superficial vs others Mild vs moderate and severe
	Mild	9	66.0	22.36
	Moderate	11	19.6	4.34
	Severe	13	9.6	5.31
Pepsinogen 2	Superficial	27	52.3	27.31	10.399	0.000^*	Superficial and mild vs moderate and severe
	Mild	9	40.8	28.90
	Moderate	11	18.5	17.03
	Severe	13	13.3	14.23
Pepsinogen 1/2 ratio	Superficial	27	3.7	3.85	3.606	0.041^*	Superficial and mild vs moderate and severe
	Mild	9	3.2	3.23
	Moderate	11	1.8	1.01
	Severe	13	1.2	1.06
Gastrin	Superficial	27	31.4	25.29	3.169	0.031^*	Superficial and mild vs moderate and severe
	Mild	9	31.3	20.07
	Moderate	11	15.8	11.19
	Severe	13	15.3	5.22
H. pylori antibody	Superficial	27	16.6	4.09	37.575	0.000^*	Superficial vs others
	Mild	9	34.5	13.85
	Moderate	11	37.4	4.84
	Severe	13	41.4	10.60

* P is significant if <0.05.

The mean serum pepsinogen II was significantly lower in H. pylori +ve group (23 ± 23.31) unit compared to H. pylori −ve group (48.1 ± 28.94). (P is significant if <0.05) (Table 2).

The mean serum pepsinogen I/II ratio was significantly lower in H. pylori +ve group (1.4 ± 0.98) unit compared to H. pylori −ve group (4.1 ± 3.85). (P is significant if <0.05) (Table 2).

The mean serum gastrin was significantly lower in H. pylori +ve group (32.8 ± 25.92) compared to H. pylori −ve group (17.3 ± 8.24). (P is significant if <0.05) (Table 2).

The mean serum pepsinogen I was significantly lower in mild, moderate and severe gastritis patients compared to superficial gastritis patients and it was significantly lower in moderate and severe gastritis patients compared to mild gastritis patients. (P is significant if <0.05) (Table 3, Fig. 1).

Figure 1 Relation between degree of gastritis and pepsinogen I and II.

Table 3 Relation between H. pylori and degree of gastritis.

	H. pylori		X2	P
	Negative	Positive
Superficial	27
	90.0%
Mild	3	6
	10.0%	20.0%
Moderate		11	57.00	0.000^*
		36.7%
Severe		13
		43.3%
	30	30

* P is significant if <0.05.

The mean serum pepsinogen II was significantly lower in moderate and severe gastritis compared to superficial and mild gastritis patients. The mean serum pepsinogen I/II ratio was significantly lower in moderate and severe gastritis patients compared to superficial and mild gastritis patients. (P is significant if <0.05) (Table 3, Figs. 1 and 2).

Figure 2 Relation between degree of gastritis and pepsinogen I/II ratio.

The mean serum gastrin was significantly lower in moderate and severe gastritis patients compared to superficial and mild gastritis patients. (P is significant if <0.05) (Table 3, Fig. 3).

Figure 3 Relation between degree of gastritis and serum gastrin.

The mean H. pylori antibody was significantly higher in mild, moderate and severe gastritis patients compared to superficial gastritis patients (Fig. 4).

Figure 4 Relation between degree of gastritis and H. pylori antibody.

As regards serum pepsinogen I at cut off value <97, sensitivity was 100%, specificity 81.5%, positive predictive value 86.81, negative predictive value 100% and accuracy 91.7% in detecting atrophic gastritis (Table 4).

Table 4 Evaluation of different markers in prediction of atrophic gastritis.

	Cut off value	Sensitivity	Specificity	Positive predictive value	Negative predictive value	Accuracy
Serum pepsinogen 1	<97	100.0	81.5	86.8	100.0	91.7
Serum pepsinogen 2	<52	93.9	63.0	75.6	89.5	80.0
Serum pepsinogen 1/2 ratio	<1.82	66.7	66.7	71.0	62.1	66.7
Serum gastrin	<28	90.9	44.4	66.7	80.0	70.0

As regards serum pepsinogen II at cut off value <52 sensitivity was 93.9%, specificity 63% positive predictive value 75.6%, negative predictive value 89.5% and accuracy 80% in detecting atrophic gastritis (Table 4).

As regards serum pepsinogen I/II ratio at cut off value <1.82 sensitivity was 66.7% specificity 66.7%, positive predictive value 71.0% negative predictive value 62.1% and accuracy 66.7% in detecting atrophic gastritis (Table 4).

As regards serum gastrin at cut off value <28 sensitivity was 90.9%m specificity 44.4% positive predictive value 66.7%, negative predictive value 80%, accuracy 70% in detecting atrophic gastritis (Table 4).

4 Discussion

The present study indicates that, in association with H. pylori testing, the serum level of G-17 and PGI can be used as biomarkers of gastritis and atrophic gastritis in the antrum. In the present study S.G.17 decreased with increasing grade of antral atrophy among H. pylori infected patient. Correspondingly S-PGI decreased with increasing grade of antral atrophy in a way that has been demonstrated in many previous studies.^18–24

Studies by using all three blood tests (H. pylori antibody assay, SG17, and SPGI as panels were done and by using the gastro soft computer program to generate the diagnosis it was possible to delineate the patients with advanced atrophic gastritis with a reasonably high accuracy. The data from the present prospective study among gastritis patients were comparable with those from observational case, control study²⁵ and prospective Italian study,²⁶ these three studies suggested that the overall accuracy of the panel in the diagnosis of atrophic gastritis is approximately 80% when compared with the diagnosis from endoscopy and biopsies.

Serum PG known to be a marker for gastritis serum PG has been reported to be a marker for atrophic gastritis^27,28 and is used for the eradication of H. pylori^29–31 Moreover low PG concentration defines groups at high risk of gastric cancer and adenoma^32,33 in describing the relationship between gastritis and pepsinogen concentrations, Samloff et al.²⁷ used the term “serologic biopsy” and Miki et al.²⁸ reported that I:II ratio was correlated with the extent of atrophic gastritis.

Diagnosis of atrophic gastritis by biopsy specimen is often unreliable and the interobserver variation is large in particular in antral biopsy. This may be due to the fact that atrophic lesions are often patchy due to the fact that atrophic lesions are often patchy and a limited number of biopsy specimens may easily result in under or over diagnosis, the existing inflammation may also bias the microscopic appearance that leads to over estimation of atrophic gastritis due to artificial impression of loss of glands. It is conceivable that the blood test panel avoids these biases.³⁴

Therefore the blood test panel is not a direct test for cancer or peptic ulcer but it may reveal patients who are at risk for this disease and when endoscopy is mandatory.³⁴

5 Conclusion

From the study we concluded that H. pylori is common in the elderly and it is the most important cause of atrophic gastritis in the elderly. Atrophic gastritis in the elderly is considered as a major risk factor for gastric carcinoma. The diagnosis of atrophic gastritis obtained with the blood test panel of SG17 SPGI, ratio and H. pylori antibodies is in good agreement with the endoscopic and biopsy findings.

Our study recommends that the blood test panel is a noninvasive alternative in the initial examination of patient dyspepsia. The panel helps to identify easily and nonendoscopically patients with atrophic gastritis in the antrum.

Notes

Peer review under responsibility of Alexandria University Faculty of Medicine.

Available online 1 September 2012