ABSTRACT
Introduction: Several tests are available for determining the presence of Helicobacter pylori (H. pylori) infection. These may be invasive or noninvasive. The carbon urea breath test (C-UBT) is generally considered to be a simple, non-invasive and accurate test for the detection of H. pylori infection both before and after treatment. Commercially available 13C-UBT kits are generally validated in their country of manufacture and the stated accuracy of their tests may not be applicable to our local population.
Aim: The aim of our study was to determine the accuracy of a commercial 13C-urea breath test kit, Hp-Plus (Utandningstester i Sverige AB, Sweden), in the Singapore population. 
Patients and methods: One hundred patients for oesophago-gastro-duodenoscopy (OGD) were recruited into this prospective study. Gastric biopsies were obtained for the biopsy urease test and histological examination. Blood samples were obtained for H. pylori serology. Breath samples were then obtained at baseline and after consumption of 100 mg of labelled 13C-urea. The presence of H. pylori infection was defined by a positive result on any two of the three tests (biopsy urease test, histology, serology) performed for the detection of H. pylori. Using this “gold standard”, the sensitivity, specificity, and positive and negative predictive values of the 13C-UBT were calculated.
Results: In the Singapore population, the 13C-UBT (Hp plus) has a sensitivity and specificity of 94.2% and 100% respectively for the detection of H. pylori infection. The positive predictive value and negative predictive value of the 13C-UBT is 100% and 88.6% respectively.
Conclusion: The 13C-UBT is a simple, safe, and accurate non-invasive test for the detection of H. pylori infection, making it a valuable tool in local clinical practice.
Keywords: Helicobacter pylori, urea breath test 
Singapore Med J 2002 Vol 43(8):408-411

INTRODUCTION
The aetiological role of Helicobacter pylori (H. pylori) infection in gastroduodenal ulcer disease and gastric malignancies is well recognised. Its role in other gastrointestinal conditions, such as gastrooesophageal reflux disease and non-ulcer dyspepsia, remains controversial.
Several tests have been developed to diagnose H. pylori infection. These range from invasive tests such as histological examination, culture, biopsy urease tests to non-invasive tests such as serology and urea breath tests. No single test is universally accepted as the “gold standard” for the diagnosis of H. pylori infection. Problems with invasive tests include the patchy distribution of H. pylori in the gastric mucosa, difficulty in culture, and human errors in interpretation. Non-invasive tests such as urea breath tests and serology are global rather than local and more accurately reflect H. pylori infection status. The urea breath test is generally regarded as the best non-invasive method for diagnosing H. pylori infection(1). Urea breath tests may utilise either 14C or 13C and there are commercial test kits available for both. The main issue surrounding the use of the 14C-urea breath test is the safety of 14C-urea. Being a radioactive isotope, concerns exist as to its safe handling, administration and disposal as well as the appropriateness of administration to children, pregnant women or women of childbearing age. Regulations governing the use of radioactive material also make its use restricted to selected sites. 13C is a stable, natural, non-radioactive isotope that does not require any special handling. Several 13C-urea breath test kits are commercially available. These 13C-urea breath test kits are generally validated in their country of manufacture and the stated accuracy of their tests may not be applicable to our local population. We studied the accuracy of a commercial 13C-urea breath test kit, Hp-Plus (Utandningstester i Sverige AB, Sweden), in the Singapore population.

PATIENTS AND METHODS
This was a prospective study in which 100 patients for oesophago-gastro-duodenoscopy (OGD) were invited to participate in the study. The study was approved by the Changi General Hospital ethics committee. The patients gave written informed consent. Demographic details, a history of smoking, alcohol consumption and non-steroidal anti-inflammatory drug (NSAID) usage were recorded. Exclusion criteria were recent (within one month) use of antibiotics, bismuth-containing compounds, proton-pump inhibitors and previous resective gastric surgery.
Patients were sedated with intravenous midazolam and endoscopy was performed. Endoscopic findings were recorded. Biopsy specimens were taken from the antrum and corpus of the stomach for histological examination as well as testing with a biopsy urease test kit.

Biopsy Urease Test
Two biopsy specimens, one each from the antrum and corpus, were placed into the yellow CLOtest (Delta West Ltd, Australia) gel. The specimens were examined at three hours and at 24 hours after being kept at the endoscopy room temperature of about 28ºC. Any colour change of the gel from yellow to shades of red was read as a positive CLOtest. The CLOtest has previously been validated in the Singapore population and found to have a sensitivity and specificity of 77% and 96% respectively(2).

Histology
Biopsy specimens were processed routinely, embedded in paraffin wax and stained with haematoxylin and eosin. The specimens were examined by histopathologists who were blinded to the results of the other tests (biopsy urease test, serology and 13C-UBT). H. pylori infection was described as present or absent.

Serology
An in-house enzyme-linked immunosorbent assay (ELISA) was used to determine the presence of H. pylori antibodies (IgG) quantitatively. This ELISA has previously been validated in the Singapore population (3). At a threshold for seropositivity at 45 ELISA units, our ELISA had a sensitivity and specificity of 94.2% and 92.3% respectively.

13C-Urea Breath Test (UBT)
A commercial 13C-UBT kit, Hp-Plus (Utandningstester i Sverige AB, Sweden) was used. The 13C-UBT was performed three hours after OGD as it was felt that the clinical effects of any drugs given during OGD would have been negligible by then. Gastrointestinal bleeding can lead to a false positive result and the manufacturer’s recommendation was to wait for at least one hour after gastric biopsy before performing the 13C-UBT. The patients were asked to drink a solution containing citric acid as a test meal. Two baseline expired breath samples were then collected. Patients were then asked to drink a second solution containing citric acid and labelled 13C-urea. They were then asked to lie on their left side for 30 minutes. At 30 minutes, two more samples of expired breath were collected.
The samples of expired breath were analysed using the Automated Breath 13Carbon Analyser (ABCA) (Europa Scientific Ltd, UK) which comprises an autosampler, a gas purification module and a mass spectrometer. The ratio of 13CO2 to 12CO2 in the expired breath samples was measured. Results were expressed as a Δδ13CO2 – value which was defined as the difference in 13CO2 in parts per thousand between the baseline pre-13Curea breath sample and the 30-minute breath sample. The test was considered positive for H. pylori infection when the Δδ13CO2 - value was greater or equal to 3.5 parts per thousand.
 

“Gold standard” definition for the presence of H. pylori infection
There is no single test accepted as the “gold standard” for the diagnosis of H. pylori infection. Instead, combinations of available tests, both invasive and non-invasive, are used as the “gold standard”(4). In our study, the presence of H. pylori infection was defined by a positive result in any two of the three tests (biopsy urease test, histology, serology) performed for the detection of H. pylori infection. Absence of H. pylori infection was defined as negative results in two or more of the tests. Using this “gold standard”, the sensitivity, specificity, positive predictive value and negative predictive value of the 13C-UBT were calculated.

RESULTS
A total of 100 patients (70 males and 30 females) with a mean age of 45 years (range 21 years to 75 years) were included in the study. Demographic details of the study population are listed in Table I. The most frequent endoscopic finding was that of duodenal ulcers. Table II lists the endoscopic diagnoses of the 100 patients.
The number of true positive cases for H. pylori infection was 69. The remainder 31 were considered true negative cases (see Table III). Of the true positives, 13C-UBT was positive in 65 cases (sensitivity = 94.2%) while of the true negative cases, 13C-UBT was negative in all 31 cases (specificity =100%). The positive predictive value and negative predictive value of the 13C-UBT were 100% and 88.6% respectively.
 

Table I. Demographic data of patient population.

Table II. Endoscopic diagnoses of patient population.

Table III. Results of tests for diagnosing H. pylori infection.

Legend: H = Histology, S = Serology, BUT = Biopsy Urease Test, 13C-UBT = 13C-Urea Breath Test

DISCUSSION
H. pylori infection is a chronic infection that has wide implications for public health, especially in countries such as Japan, Hong Kong and Singapore, where the high incidence of gastric carcinoma contributes significantly to the overall morbidity, mortality and healthcare expenditure of the society. The Asia-Pacific Consensus Conference on the management of Helicobacter pylori infection(7) recommended that in countries with a high incidence of gastric cancer, patients with uninvestigated dyspepsia above an age cut-off (depending on the national gastric cancer incidence) should be investigated by endoscopy. In Singapore, this age cut-off is 35 years(6). It was also recommended that young patients without alarm symptoms should be tested for the presence of H. pylori infection using a locally validated non-invasive test. Such a test should have a sensitivity and specificity of 90% or greater. If the patient is found to be positive for H. pylori infection, endoscopy should be performed. In those who tested negative for H. pylori infection, the probability of missing serious gastroduodenal disease is very low.
The commonly available non-invasive tests for the determination of H. pylori infection status are urea breath tests and serological tests. Direct comparisons between UBTs and serology have shown serology to be less accurate in the diagnosis of H. pylori infection(7-9). Serological methods are limited in their usefulness in assessing posttreatment H. pylori status as qualitative antibody assays remain positive for up to three years after bacterial eradication. In the follow-up of patients who have received H. pylori eradication therapy, serological evaluation is unreliable for up to six months post-treatment(7,10,11). In order to document cure of H. pylori infection serologically, an ELISAbased test of a six-month post-treatment sample and a stored pre-treatment sample would have to be analysed concurrently. This would be impractical. Several studies have demonstrated the efficacy of the 13C-urea breath test in the assessment of H. pylori eradication(12,13). The fact that the 13C-UBT can be used a month after completion of eradication therapy gives it an advantage over serological tests.
The use of labelled carbon urea breath test (C-UBT) was first described in 1987 by Graham DY et al (14). H. pylori produces a powerful urease. In the C-UBT, exhaled breath samples are obtained from the patient before and after the consumption of labelled urea. In the presence of H. pylori infection, hydrolysis of the labelled urea by the urease occurs,  liberating labelled carbon dioxide which is then measured. The C-UBT reflects the current bacterial status in patients with H. pylori infection. Studies done on Western populations have rated the sensitivity of the 13C-UBT at 90%-92% and its specificity at 96%-100% when compared against biopsy-based tests (4,7,15). Labelling of the urea can be done with either 13C or 14C. The main disadvantage of 14C is that it is a radioisotope and as with all radioactive material, there are strict regulations relating to the licensing, security, storage, recording and moving of stock. 14C has a half-life of 5.7 years and only 70% of the given dose is excreted, with the remaining 30% being incorporated into the patient’s tissue carbon pool (16). Safety considerations therefore do not permit its use in children or pregnant women and repeated testing should be avoided. 13C is a stable, natural, non-radioactive isotope found in about 1.11% of expired carbon dioxide. Special handling procedures are not required. The small volume of expired breath required for analysis also allows for a test-tube of expired air to be sent by post for commercial laboratory analysis. The non-radioactive nature of 13C allows for the use of the 13C-UBT in children as well as for repeat noninvasive assessments of H. pylori status. The main disadvantage of the 13C-UBT is the high cost of the gas chromatograph isotope ratio mass spectrometer (GC-IRMS) which is required to measure 13CO2 in expired breath.
In our study, the commercial 13C-UBT kit used  (Hp-Plus) had a sensitivity and a specificity of 94.2% and 100% respectively in the local Singapore population when validated against a combination of other tests (serology, biopsy urease test and histology). The positive predictive value was 100% while the negative predictive value was 88.6%. This met the standards recommended by the Asia Pacific Consensus on the management of Helicobacter pylori infection for a non-invasive test.

CONCLUSION
The sensitivity and specificity of the 13C-UBT for the diagnosis of H. pylori infection in the Singapore population is 94.2% and 100% respectively. It is a non-invasive test and the nonradioactive nature of 13C allows for its use in children and pregnant patients. The 13C-UBT is thus a simple, safe and accurate non-invasive test for the determination of H. pylori infection status, making it a valuable tool in local clinical practice.

REFERENCES
1.    Hunt RH, Thomson ABR. Canadian Helicobacter pylori Consensus Conference. Can J Gastroenterol 1998; 12:31-41.
2.    Ng TM, Fock KM, Ho J, Tan AL, Chia SC, Yap CK, Chew CN, Chee EN. Clotest (rapid urease test) in the diagnosis of Helicobacter pylori infection. Singapore Med J 1992 Dec; 33(6):568-9.
3.    Fock KM, Dhamodaran S, Ng TM, Khor CJL, Teo EK, Lee YM, Chua TS. Quantitative detection of Helicobacter pylori specific antibodies in a multi-ethnic asian population. Gastroenterology 2000; 118(4) Suppl (2):A504 (Abstract).
4.    Cutler AF, Havstad S, Ma CK, Blaser MJ, Perez-Perez GI, Schubert TT. Accuracy of invasive and noninvasive tests to diagnose Helicobacter pylori infection. Gastroenterology 1995; 109:136-41.
5.    Lam SK, Talley NJ. Report of the 1997 Asia Pacific Consensus Conference on the management of Helicobacter pylori infection. J Gastroenterol Hepatol 1998 Jan; 13(1):1-12.
6.     Management of Helicobacter pylori infection. Ministry of Health, Singapore, MOH Clinical Practice Guidelines 2/98; 10.
7.    Logan RPH, Polson RJ, Misiewicz, JJ, Rao G, Karim NQ, Newell D, Johnson P, Wadsworth J, Walker MM, Baron JH. Simplified single sample 13Carbon urea breath test for Helicobacter pylori: comparison with histology, culture, and ELISA serology. Gut 1991; 32:1461-4.
8.    Faigel DO, Childs M, Furth EE, Alavi A, Metz DC. New noninvasive tests for Helicobacter pylori gastritis: Comparison with tissue based gold standard. Digest Dis Sc 1996; 41:740-8.
9.    Lin SK, Lambert JR, Schembri M, Nicholson L, Finlay M, Wonz C, et al. A comparison of diagnostic tests to determine Helicobacter pylori infection. J Gastroenterol Hepatol 1992; 7:203-9.
10. Kosunen TU, Seppala K, Sarna S, Sipponen P. Diagnostic value of decreasing IgG, IgA and IgM antibody titres after eradication of Helicobacter pylori. Lancet 1992; 339:893-5.
11. Cutler A, Schubert A, Schubert T. Role of Helicobacter pylori serology in evaluating treatment success. Digest Dis Sc 1993; 38:2262-6.
12. Slomianski A, Schubert T, Cutler AF. 13C urea breath test to confirm eradication of Helicobacter pylori. Am J Gastroenterol 1995; 90(2):224-6.
13. Johnson PG, Duggan AE, Olson C. 13C-urea breath test — a reliable diagnostic technique for assessment of eradication. Gut 1996; 39 Suppl 3:A37.
14. Graham DY, Klein PD, Evans DJ, Evans DG, Alpert LC, Opekun AR, Boutton TW. Campylobacter pylori detected noninvasively by the 13C-urea breath test. Lancet 1987; i:1174-7.
15. Mock T, Yatscoff R, Foster R, Hyun JH, Chung IS, Shim CS, Yacyshyn B. Clinical validation of the Helikit: a 13C urea breath test used for the diagnosis of Helicobacter pylori infection. Clin Biochem 1999 Feb; 32(1):59-63.
16. Marshall BJ, Surveyor I. Carbon-14 urea breath test for the diagnosis of Campylobacter pylori associated gastritis. J Nucl Med 1988; 29:11-6.