Helicobacter Pylori Infections: Culture from Stomach Biopsy, Rapid Urease Test (Cutest®), and Histologic Examination of Gastric Biopsy

Home , Diagnostic microbiology, Rapid urease test

Available online at www.annclinlabsci.org 148 Annals of Clinical & Laboratory Science, vol. 45, no. 2, 2015 An Efficiency Comparison between Three Invasive Methods for the Diagnosis of Helicobacter pylori Infections: Culture from Stomach Biopsy, Rapid Urease Test (CUTest®), and Histologic Examination of Gastric Biopsy

Avi Peretz1, Avi On 2, Anna Koifman1, Diana Brodsky1, Natlya Isakovich1, Tatyana Glyatman1, and Maya Paritsky3

1Clinical Microbiology Laboratory, 2Pediatric Gastrointestinal Unit, and 3Gastrointestinal Unit, Baruch Padeh Medical Center, Poria, affiliated to the Faculty of Medicine, Bar Ilan University, Galille, Israel

Abstract. Background. Helicobacter pylori is one of the most prevalent pathogenic bacteria in the world, and humans are its principal reservoir. There are several available methods to diagnose H. pylori infection. Disagreement exists as to the best and most efficient method for diagnosis. Methods. In this paper, we report the results of a comparison between three invasive methods for H. pylori diagnosis among 193 patients: culture, biopsy for histologic examination, and rapid urease test (CUTest®). Results. We found that all three methods have a high sensitivity and specificity for the diagnosis of infections caused by H. pylori. However, the culture method, which is not used routinely, also showed high sensitivity, probably due to biopsies’ seeding within 30 minutes, using warm culture media, non-selective media, and longer incubation. Conclusions. Although not a routine test, culture from biopsy can be meaningful in identification of antibiotic-resistant strains of H. pylori and should therefore be considered a useful diagnostic tool.

Keywords: Helicobacter pylor, Culture, Urease test, Gastric biopsy.

Introduction

Helicobacter pylori is one of the most prevalent Recently, a close association was found between H. pathogenic bacteria in the world. Humans are the pylori infection and the development of gastric principal reservoir of this bacterium. Epidemiologic cancer and gastric mucosa-associated lymphoid- studies indicate a much higher incidence in under- tissue (MALT) lymphoma. This heightens the im- developed countries (70-90%) than in industrial- portance of H. pylori diagnosis [5]. ized countries (25-50%)[1]. However, the exact route of transmission remains unknown. The most There are several available invasive methods to diag- likely mode of transmission is fecal-oral route in nose H. pylori infection, including biopsy of the under-developed countries and gastro-oral in in- gastric epithelium for histologic examination, mo- dustrialized countries [2,3]. H. pylori has a low oxy- lecular tests, and culture. The use of culture occurs gen tolerance and harbors a urease enzyme that en- mostly after the failure of antibiotic treatment. In ables the hydrolysis of urea from body fluids to recent years, there has been an increased use of a ammonia and CO2, thus alkalizing its environ- rapid urease test that enables the physician to get a ment. These two characteristics enable the bacteri- bedside result and thus make medical decisions rap- um to adapt to the highly acidic environment of the idly. It is important to emphasize that despite the gastric epithelium and to induce an immune relatively wide variety of methods, there is dis- response that results in chronic inflammation and agreement in regard to the best and most efficient peptic ulcers [4,5]. method for diagnosing H. pylori infection. The preference for a specific diagnostic method is based Address correspondence to Dr. Avi Peretz, PhD.; Clinical Microbiology Laboratory, Baruch Padeh Medical Center, Poria, affiliated to the on the equipment available in the medical institu- Faculty of Medicine, Bar Ilan University, Galille, Hanna Senesh 818/2 tion and in the specific laboratory that performs the Tiberias, Israel; phone: 972 4 6652322; fax: 972 4 6652531; e mail: [email protected] test. The sensitivities of the different methods, as

were incubated in microaerophilic conditions Table 1. A comparison between culture, histology and rapid (CampyGenTM, Oxoid, UK) for 14 days at a tempera- urease test as diagnostic methods for the diagnosis of H. pylori ture of 35ºC. Cultures with bacterial growth were tested from biopsy with oxidase, catalase, and urease activity tests, all of which were positive for H. pylori. Gram staining was per- Method Number Number formed in order to look for characteristic morphology, of positives of negatives which includes Gram-negative bacteria with a gull wing-shape. Culture 104 (53.8%) 89 (46.2%) Histology 96 (49.7%) 97 (50.3%) Histology. Biopsies were obtained for histologic exami- Rapid urease test 98 (50.7%) 95 (49.3%) nation in a collection tool which contained 4% formal- dehyde. In the laboratory, the specimens were subjected to tissue processing procedures including embedding determined in many studies, are 80-87% for culture and cutting before staining with hematoxylin and eosin from biopsy, 95-99% for PCR, 85-90% for histology, (H&E staining). All preparation procedures were per- and 85-95% for different immunological tests formed using automatic instruments (Leica Tp 1050, (Immunochromatographic EIA) [6,7]. Leica Eg 1160, Leica Autostainer XI, Germany).

In this study, we report the results of a comparison that All samples were examined by two expert pathologists in we performed between three methods for H. pylori diag- order to ensure high expertise in the histologic examina- nosis that are available in our laboratory. We found that tion. A positive biopsy sample was defined as a sample in culture from biopsy has a higher sensitivity than the which there were bacteria with the characteristic mor- other methods tested. Although not a routine test, cul- phology of H. pylori (gull wing-shape). ture from biopsy can prove meaningful in identification Rapid urease test. Rapid urease test was performed us- of antibiotic-resistant strains. ing a commercial kit (CUTest®, Temmler Pharma GmbH & Co. KG, Marburg, Germany). The kit supplies Material and Methods Eppendorf tubes that contain 1.5 mL urea solution with the pH indicator phenol red (49 mg urea, 0.015 mg phe- Patients' characteristics. A total of 193 patients were enrolled nol red). The biopsy samples were kept in these tubes up in the study. Biopsy samples were taken from all patients dur- to 24 hours at 37ºC. A color change indicates a pH ing gastroscopy. Among the 193 participants, 21 patients change in the tube and the hydrolysis of urea to ammo- (10.8%) had ulcer disease and 172 patients (89.2%) had non- nia, thus evaluating the presence of urease. ulcer disease (such as gastritis, atrophic gastritis, dyspepsia without relevant lesions in stomach and/or duodenum); 98 Statistical Methods. Sensitivity and specificity analysis were adults (50.7%), with an average age of 48.5 years, and 95 and 95% confidence intervals (CIs) were computed us- were children (49.3%), with an average age of 11.7 years. We ing a 2x2 table for outcomes: collected 2 biopsies from each patient for each diagnostic ➢ culture vs. histology method (histology, culture and urease test) – one from the an- ➢ culture vs. rapid urease test trum and the other from posterior corpus areas. None of the ➢ histology vs. rapid urease test patients enrolled in this study received antibiotic treatment or Test sensitivity (the conditional probability that the test PPI before the test. Pregnant women and patients with coagu- is positive if the condition is positive) was calculated ac- lation disorders were not included. This study was ovappr ed cording to the following formula: by the hospital’s Ethics Committee. Sensitivity = (True Positive) / (True Positive + False Negative) X 100 Culture. Biopsies were in the laboratory within (at most) half Test specificity (the conditional probability that the test an hour of sampling, in sterile Eppendorf tubes that contained is normal if the condition is normal (negative) was calcu- a physiologic solution. In the laboratory, the biopsies were lated using the following formula: pounded and seeded on two growth media that were pre-heat- Specificity = (True Negative) / (True Negative + False ed in 35ºC blood agar (BD Diagnostics, Sparks, MD) and Positive) X 100 helicobacter-selective agar, which contains the antibiotics co- The data was analyzed using the SAS® version 9.1 (SAS listin and polymyxin (Hy-Laboratories, Israel). The cultures Institute, Cary North Carolina). 150 Annals of Clinical & Laboratory Science, vol. 45, no. 2, 2015

Table 2. Number of positive samples for: all 3 tests, 2 out of 3 tests, 1 of 3 tests and number of negative samples for all 3 tests. positive negative 3 out of 3 tests 2 out of 3 tests 1 out of 3 tests 0 out of 3 tests

Number of samples 84 20 6 83

Results present study. The first factor is the time between biopsy sampling and seeding in the laboratory. In Of 193 biopsies, 104 cultures (53.8%) were posi- our study, we insisted on seeding as soon as possible tive for H. pylori growth. On the other hand, only after sampling (up to half an hour later), and seed- 96 biopsies (49.7%) were found to be positive in ing was done on pre-heated agar plates. The second the histologic examination, and 98 biopsies factor is the incubation duration. Many protocols (50.7%) were positive to urease activity in the rapid for growing H. pylori note that the incubation du- urease test (Table 1). Among the positive samples, ration has to range between four and seven days 84 samples were positive in all three tests, 20 sam- [11,12]. In contrast, in our study, only 64 cultures ples were positive in two of the three tests, and only were found positive in this range of time. Moreover, 6 samples were positive in one of the three tests in a considerable number of cultures, H. pylori (Table 2). Sensitivity and specificity for the meth- growth was detected only on the eleventh day. ods culture vs. histology were 93.75 (87.03-97.1) and 85.57 (77.2-91.2), respectively. For culture vs. In regard to the growth media, it should be men- rapid urease test the sensitivity was 93.88 (87.3- tioned that the growth delay of certain strains in 97.2), and the specificity was 87.37 (79.2-92.6). selective growth media that contain different anti- For histology vs. rapid urease test, the sensitivity biotics is a known phenomenon. It is accepted that was 91.84 (84.7-95.8), and the specificity was at least 5% of H. pylori strains will not grow on 93.68 (86.9-97.1). In regard to culture, it should be such media. Therefore, we should seed the culture noted that H. pylori growth was found in all the on a non-selective medium such as blood agar, in cultures that were taken from the antrum and the addition to the selective medium, in order to raise posterior corpus that seeded on blood agar. the test’s sensitivity. In this study, we showed that However, only 98 of the 104 positive cultures were non-selective media was better than selective me- also positive in the helicobacter-selective agar. Of dia. However, in some cases, selective media pre- the other 6 cultures, four were taken from the an- vent contamination from normal flora that can trum and two from the posterior corpus grow on non-selective media and create difficulties in isolating H. pylori colonies [13,14]. Discussion The success of H. pylori cultures is very important, Overall, there was very strong concordance between as growing this bacterium will enable hospitals to the three methods. The culture method is not in- routinely perform antibiotic sensitivity tests. Since cluded in routine testing, as many physicians prefer most H. pylori strains appear to be resistant to the to use histological examination or rapid urease tests. antibiotics that are used as primary treatment, per- Typically, culture is performed only in cases of fail- forming cultures from biopsy should be considered ure of an empirical antibiotic treatment, in order to as soon as the biopsy is taken for the purpose of detect resistant strains [8]. rapid urease test and histologic examination, espe- cially because this does neither endangers the pa- In many studies, the success rates of culture are tient nor requires a special invasive procedure. much lower than in our study [9,10]. It is impor- With respect to cases in which the histologic ex- tant to note that culture's success possibly depends amination gives a negative result, perhaps such ex- on two factors that can increase the sensitivity of amination depends on the bacteria concentration culture, and we included these factors in the in the sampling region. The higher the bacteria Invasive methods for H. pylori diagnosis 151 concentration, the more easily the bacterium can 5. Blaser MJ, Atherton JC. Helicobacter pylori persistence: biology and disease. J Clin Invest 2004;113:321-333. be identified. This is also true for the rapid urease 6. Holmes KP, Fang JC, Jackson BR. Cost-effectiveness of six test, which depends on the bacteria concentration strategies for Helicobacter pylori diagnosis and management in in the sampling region as well as on the use of uninvestigated dyspepsia assuming a high resource intensity practice pattern. BMC Health Serv Res 2010;10:334-440. Bismuth-containing compounds, proton-pump in- 7. McNulty CAM, Lehours P, Me´graud F. Diagnosis of hibitors, and antibiotics in close proximity to the Helicobacter pylori Infection. Helicobacter 2011;16:10-18. test. 8. De Francesco V, Giorgio F, Hassan C Gianpiero M, Vannella L, Panella C, Ierardi E, Zullo A. Worldwide H. pylori Antibiotic Resistance: a Systematic Review.J Gastrointestin Liver Dis In conclusion, all three methods – histologic ex- 2010;19:409-414. 9. Monteiro L, De Mascarel A, Sarrasqueta AM, Bergey B, amination, rapid urease test, and culture – have a Barberis C, Talby P, Roux D, Shouler L, Goldfain D, high sensitivity and specificity for the diagnosis of Lamouliatte H, Mégraudgraud F. Diagnosis of Helicobacter py- infections in the duodenum or stomach that are lori Infection: Noninvasive Methods Compared to Invasive Methods and Evaluation of two New Tests. Am J Gastroenterol caused by H. pylori. However, performing cultures, 2001;96:353–358. insisting on seeding as close as possible to the sam- 10. Lim SK, Lambert JR, Schembri M, Nicholson L, Finlay M, pling time, and allowing prolonged incubation du- Wong C, Coulepis A. A comparison of diagnostic tests to de- termine Helicobacter pylori infection. J Gastroenterol Hepatol ration of the growth media should be emphasized 1992;7:203-209. to identify antibiotic-resistant strains. 11. Forbes BA, Sahm DF, Weissfeld AS. Bailey and Scott's Diagnostic microbiology. 12th edition. Missouri: Mosby. 2007: 416-23. References 12. Isenberg HD. Clinical microbiology procedures handbook. 2nd edition. Volume 1. Washington, D.C: ASM press. 2004; Chapter 3.8.4. 1. Kenneth EL McColl. Clinical practice: Helicobacter pylori 13. Miendje Deyi VY, Van den Borre C, Fontaine V. Comparative Infection. N Engl J Med 2010;362:1597-1604. evaluation of 3 selective media for primary isolation of 2. Magalhaes Queiroz DM, Luzza F. Epidemiologyof Helicobacter Helicobacter pylori from gastric biopsies under routine condi- pylori infection. Helicobacter 2006;11:11-15. tions. Diagno Micro infec disea 2010; 68: 474-476. 3. Delport W, van der Merwe SW. The transmission of 14. Hartzen SH, Andersen LP, Bermmelgaard A, H Colding, M Helicobacter pylori: the effects of analysis method and study Arpi,J Kristiansen, T Justesen,F Espersen,N Frimodt-Møller, population on inference. Best Pract Res Cl Gl O Bonnevie. Antimicrobial susceptibility testing of 230 2007;21:215-236. Helicobacter pylori strains: importance of medium, inoculum, 4. Gerrits MA, Van Vliet HM, Kuipers EJ, Kusters JG. and incubation time. Antimicro agents chemoth 1997; Helicobacter pylori and antimicrobial resistance: molecular 41:2634-2639. mechanisms and clinical implications. Lancet infectious dis- eases 2006;6:699-709.