The Discrepancy Between Histology And PCR Methods for the Detection of Helycobacter Pylori In Patients With Dyspepsia Without Proper Preparation Before Endoscopy
* Maruni Wiwin Diarti, ** Haris Widita, * Soewignjo Soemohardjo, ** Weny Astuti, *** Troef Sumarno, ***** Yunan Jiwintarum,* Zainul Mutaqin, and **** Retno Handayani
* Biomedical Research Unit Mataram General Hospital
** Department of Internal Medicine Mataram General Hospital
*** Department of Pathology Medical Faculty Airlangga University
**** Department of Biochemistry Medical Faculty Airlangga University
***** Department of Laboratory Technology, Institute of Health Technology West Nusa Tenggara Mataram
Background : Detection of H. pylori in gastric tissue by tests based on urease enzymatic activity needs that the patients stop antibiotics or acid suppressor drugs two weeks before endoscopy to avoid false negative result. The objectives of this study is to compare the result of the detection of H. pylori in gastric biopsy by histology and Ure C PCR in patients with dyspepsia underwent upper GI endoscopy without preparations other than 6 hours fasting before endoscopy.
Material: 156 paraffin blocks of gastric endoscopic biopsy sample taken from antrum and corpus of patients with dyspepsia underwent endoscopy in the endoscopy Unit Biomedika Hospital Mataram.
Methods: All of the biopsy samples were stained with Hematoxylin and Eosin for tissue diagnosis and the Giemsa stain for the detection of H. pylori. PCR Ure C was done on all blocks. PCR for Cag a was done on all PCR Ure C positive samples.
Results: From 156 paraffin blocks, only 17 blocks (10.9%) were positive for H. pylori by histology. PCR was 100% positive from all 17 samples with positive histology. From 156 blocks PCR for Ure C was positive in 73% (45.9%). The PCR method has increased the positivity rates of H. pylori more than four times compared with histology. This study showed that the rate of cag a was 63.0%.
Conclusion : This study showed is that PCR ureC was superior to histology in patients without stopping acid supressor drug and antibiotic 2 weeks before endoscopy. This results can be explained by the change of spiral form into coccoid form that is difficult to detect using Giemsa stain.
Keywords : Helicobacter pylori, histology, ureC, Cag a, PCR.
One of the important methods of Helicobacter pylori detection in endoscopic gastric biopsy sample is the histological method. The most popular stain used is Giemsa which is very simple, easy, and cheap. For a long time, histological method was known to be sensitive and specific for the detection of H pylori in gastric tissue (1,2) but the histologic detection is difficult if the bacterial density of the specimen is low, and it becomes very difficult if the bacteria is in a cocoid form. The expertise of the pathologist is a very crucial factor and histological method is also considered to be relatively subjective. In recent years PCR method become popular for Helicobacter pylori detection. PCR using the fragment of ureC (glmM) gene as primers was proven to be the most specific and sensitive compared with PCR using the other primers such as ure A and ure B. (3,4,5) The superiority of PCR method over histology in the detection of Helicobacter pylori has been reported by Cesar et al (7).
Although most endoscopist realize the importance of proper preparation before endoscopy for the detection of Helicobacter pylori such as stopping acid suppressor drugs and antibiotics at least two weeks before endoscopy, if the test based on urease activity such as CLO will be used, to avoid false negative result. In daily practice it is very difficult to stop acid suppressor drug or antibiotic at least two weeks before endoscopy, because the patient cannot wait for two weeks and they want prompt endoscopy and prompt relief of the symptoms. This situation might also influenced the result of the diagnosis of H pylori by histology.
The objectives of this study is to compare the result between the detection of Helicobacter pylori in gastric biopsy sample using histological methods and using ureC (glmM) PCR in patients with dyspepsia underwent upper GI endoscopy without special preparation other than six hours fasting before the procedure.
Materials and Methods
In the period from January to December 2007 there were 177 upper GI endoscopies in patients with dyspepsia in endoscopy unit Biomedika Hospital Mataram. Endoscopic gastric biopsy was taken from antrum and corpus of the stomach from all of the patients. There was no special preparations of the patients before endoscopy except six hours fasting before the procedure. The patients were not told to stop any medications including acid suppression drugs nor antibiotics. From 177 paraffin blocks only 156 can be recovered and sufficient for this study.
All of the biopsy samples were stained with HE for tissue diagnosis and Giemsa stain for Helicobacter pylori detection(1,2). PCR UreC was done by method described by Lu et al(5). PCR of Cag A was done using primer and method described by David et al(6). Described histologic examination was done by an experienced pathologist. From originally 177 paraffin blocks, 23 ( 12.99% ) blocks showed positive Helicobacter pylori and 154 were negative by histology. The material of this study was 156 blocks remained. The 156 blocks included 17 blocks that was positive for Helicobacter pylori by histology. So 139 blocks were negative for Helicobacter pylori by histology. PCR ureC (glmM) were done on blocks with Helicobacter pylori positive and negative. PCR for cag a was done on all PCR ure C positive samples.
The sample consisted of paraffin blocks of 100 male and 56 female patients aged 15 years to 88 years. All of the patients had dyspepsia complaint for more than 6 months. Table 1 showed endoscopic diagnosis of the cases and the result of H. pylori diagnosis by PCR and Histology.Picture 2 showed the electrophoresis bands of H. pylori positive by PCR ureC (glmM).
Table 1 shrouded that most frequent endoscopic diagnosis was antral gastritis, followed why gastric ulcer, duodenal ulcer, malignancy, chronic gastritis, acute gastritis and normal endoscopy. From 27 gastritis case none of the block was positive by histology (0%). From 12 duodena ulcer only 1 (8,33%) was positive by histology. From 98 antral gastritis case only 13 (13,2%), while from 12 gastric malignancy 2 was positive 16,67%. Histology was negative in patients with acute gastritis 0%, of 2 patients with chronic gastritis only one was positive by histology ( 50% ).
From 17 blocks with Helicobacter pylori positive by histology all were Helicobacter pylori positive by PCR (100%). From 139 blocks negative for Helicobacter pylori by histology 56 were positive by PCR (40,7 %). From 17 blocks with Helicobacter pylori positive by histology and PCR, 14 were positive for cag a ( 82,38 % ), while from 56 samples with positive PCR but negative histology, 32 were positive for cag a (57,14 %). So in total from 156 patients with dyspepsia, 73 patients were positive for Helicobacter pylori by histology or PCR (45,9%).
From table 1 we can see that using the result of histology only minority of patient with gastric ulcer and duodenal ulcer shows positive Hp, while using PCR the positive rate is much more higher (around 50%)
Table 2 shows H.pylori by histology combined with PCR, that showed much more higher result compared with table 1, for example from this table gastric ulcer showed 51,58% positive for H.pylori and 50% of duodenal ulcer was positive for H.pylori.
Figure 1. Showed H. pylori positive by PCR ure C (glmM)
- 1. G316A Marker;
- 2. Negative control ATCC 53726 H. pylori 294 bp;
- 3. Negative control (aquadest);
- 4. Sample B88 à Positive for H. pylori (294 bp);
- 5. Sample B99 à Positive for H. pylori (294 bp);
- 6. SampleB104àPositivefor H. pylori (294 bp);
- 7. Sample B112 à Positive for H. pylori (294 bp);
- 8. Sample B118 à Positive for H. pylori (294 bp)
Table 3. Patients with dyspepsia underwent endoscopy whose parafin blocks available for the study
Table 4. Patients with dyspepsia underwent endoscopy whose parafin blocks available for the study With H. pylori positive by histology and positive by PCR result.
Using Chi Square analysis with Fisher’s test with confidence level of 95% it was found that the H. pylori positive rate of male is significantly higher compared with female patient (p < 0.05).
From table 4 it is indicated that the rate of cag a in blocks with PCR ure C positive was 63.01%. The difference of cag a positivity between the group with H. pylori positive by histology and the group with H. pylori negative by histology but positive by PCR are analyzed using T test paired samples with 95% confidence interval and p = 0.05. According to the test the difference of cag a positivity between the two groups is markedly significant (p = 0.000).
This study shows that H. pylori in the biopsy specimens with positive by histology has a higher rate of cag a positive compared to the specimens positive for H. pylori by PCR, but negative by histology. The reasons for this phenomenon are not known, but it might be due to the higher H. pylori DNA contends in the specimens with histology positive.
The result of this study in which 45.9% of patients with dyspepsia was positive for H. pylori by PCR ure C is almost similar to the results of the study by Soemohardjo et al. in 2007,8 that reported the frequency of H. pylori detected by PCR ureC in 41.13% of the patients with dyspepsia in Mataram general hospital. Arinton in his dissertation in 2008 reported that only 2% out of 104 patients with dyspepsia showed positive H. pylori infection by PCR ure C but the infection rate is 36.5% by PCR ure C.9
The positive rate of H. pylori infection done by a PCR method as high as 45.9% in contrast to only 10.9% by histology is really astonishing and there must be a reason for this large discrepancy.
Several studies showed that histologic diagnosis of H.pylori infection is accurate with high specivity and sensitivity. But it should be remembered that the histological detection of H. pylori depend on the appearance of spiral bacteria commonly seen near the area of gastric inflammation.1,2 The spiral bacteria can be seen in abundance in cases of H. pylori infection in the patients who didn’t take antibiotics or acid suppressant for a long time.10,11 It is recommended the patient should not take antibiotic and acid suppressant drugs at least for two weeks before endoscopy to ovoid false negative result. But it becomes difficult because it was known that not too many patients with dyspepsia who can stop acids suppressant drugs for two weeks. The histologic diagnosis for H. pylori infection is difficult if the bacterial density the sample is low and the spiral bacteria can be seen only after a long search and even it can not be detected. The procedure become very difficult if due to environmentals stress the bacteria change it appearance from spiral form in to curved form or cocoid form. In that case the immunohistochemistry is needed. But it was known that the staining is costly and complicated. Reports showed that in patients with H. pylori infections who took PPI for a long time the diagnostics based on the presence of urease enzymatic activity such as UBT or CLO test will show false negative results. The urease enzymatic activity reappears several days after PPI was stoped.10 the reason of this phenomenon is still not clear. Jekti at all reported that spiral form of H. pylori under condition several stresses such as prolonged incubation without new nutrient supplementation, aerobiosis, and with the addition of amoxicillin, showed clearly that the spiral form change to coccoid form with the stress. In all stages of coccoid development, the urease enzymatic activity was negative.12 The change of H. pylori in to coccoid form may explain the negative result of H.pylori infection by histology and UBT. Renz et al in 2,000.11 Reported 35 patients with chronic gastritis undergoing endoscopy. H. pylori infection was determined by CLO test, culture, antibody, and PCR for ure C, and histology. Fifteen patients were positive for CLO test and culture. All the patients showed numerous helical form of H. pylori. Nine subjects were CLO negative, culture negative, and positive PCR. Histology of these subjects showed cocoid forms, or mixed cocoid form with scant helical form, or mixed cocoid form with helical form in the equal proportion.
This study shows that in the diagnosis of H. pylori infection PCR ure C is superior to histological diagnostic especially in patients without preparation before endoscopy. In the future PCR may be used routinely in the detection of H. pylori detection in endoscopic biopsy sample because it can routinely detect all form of H. pylori including coccoid form.
In patients without proper preparation prior to endoscopy. The accurate of histology is less due to high false negative result.
To conclude, this study showed discrepancy between the results of H. pylori detection by histology and PCR methods in patients with dyspepsia without proper preparation before endoscopy. Only minority of patients show positive result using histology (10.9 ) while using PCR the positive rate is much higher (45.9). It is recommended that histological method cannot be used when proper preparation before endoscopy is not done and it should be replaced by PCR. The reason of this finding is that majority of patients with dyspepsia might be still taking acid suppressor drug or even antibiotics before endoscopy. This situation may caused the change of spiral form into coccoid form that decrease the possibility of histological detection of H. pylori, while PCR can detect the genome of both spiral and coccoid form of H. pylori.
1. Rotimi O, Cairns A, Gray S, Moayyedi P, Dixon MF. Histological identification of Helicobacter pylori: comparison of staining methods. J Clin Pathol 2000;53:756-59.
2. Logan RPH, Walker MM. Epidemiology and diagnosis of Helicobacter pylori infection (Clinical Review). BMJ 2000;323:920-22.
3. De Reuse H, Labigne A, Mengin-Lecreulx D. The Helicobacter pylori ureC gene codes for a phosphoglucosamine mutase. J Bacteriol 1997;179:11:3488-93.
4. Bickley J, Owen RJ, Fraser AG, Pounder E. Evaluation of the polymerase chain reaction for detection for detecting the urease C gene of Helicobacter pylori in gastric biopsy samples and dental plague. J Med Microbiol 1993;39:338-44.
5. Lu JJ, Perng CL, Shyu RY, Chen CH, Lou Q, Chong SKF, Lee CH. Comparison of five PCR methods for detection of Helicobacter pylori DNA in gastric tissue. J Clin Microbiol 1999;37:3:772-4.
6. David E, Canavez FC, Carvalho CM, Miteldorf CATZ, Lopes LHC, Leite KMR. Helicobacter pylori and cag a gene detected by polymerase chain reaction in gastric biopsi: correlation with histological findings, proliferation, and apoptosis. Sao Paulo Med J 2005;123(3):113-8.
7. Cesar ACB, Cury PM, Payao SLM, Liberatore PR, Silva AE. Comparison of histological and molecular diagnosis of Helicobacter pylori in benign lessions and gastric adenocarcinoma. Brazilian J Microbiol 2005;36:12-6.
8. Soemohardjo S, Palgunadi I, Gunawan S, Muttaqin Z, Widita H, Achwan W. The detection of Helicobacter pylori in gastric mucosal biopsy specimen used PCR ureC in dyspepsia patient. Indones J Gastroenterol Hepatol Dig Endos 2007;8(2):40-3.
9. Arinton IG, Level of serum gastrin and ratio pepsinogen I to pepsinogen II as a biomarker of chronic gastritis due to Helicobacter pylori (Dissertation). PhD Program in Medicine, Diponegoro University Semarang 2008.
10. Gatta L, Vakil N, Ricci C, Osborn JF, Tampieri A, Perna F, Miglioli M, Vaira D. Effect of proton pump inhibitors and antacid therapy on 13C urea breath tests and stool test for Helicobacter pylori infection. Am J Gastroenterol 2004;99:5:823.
11. Ren Z, Pang G, Batey R, Routiey D, Russeli A, Musicka M, Dunkley M, Beagley K, Ciancy R. Non – urease producing Helicobacter pylori in chronic gastritis. Aust N Z J Med 2000;30(5):578-84.
12. Jekti D, Soemohardjo S, Muttaqin Z, The Absence of Urease Enzymatic Activity of Helicobacter pylori coccoid form. Indones J Gastroenterol Hepatol Dig Endos 2008;9(2):35-41.
( Presented in Congress of Indonesian Society of Gastroenterology, Hematology, and Endoscopy. Denpasar August 13-16, 2009 ).