SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. References

Background:

Several modifications of the 13C-urea breath test have been published. For reasons of cost efficiency and practicability, the urea dose and measurement duration should be reduced while still maintaining excellent diagnostic accuracy.

Aims:

To establish a validated simple protocol for the urea breath test with 50 mg 13C-urea and to compare this protocol with the conventional urea breath test with 100 mg 13C-urea.

Methods:

Conventional urea breath test with 100 mg 13C-urea was performed on 152 dyspeptic patients. Full-cream cow's milk was used as the test meal. Breath tests were repeated using 50 mg 13C-urea and the breath samples were collected at baseline and at 10 (protocol t10), 15 (protocol t15) and 30 min (protocol t30). Helicobacter pylori status was assessed by rapid urease test, histology and conventional urea breath test with 100 mg 13C-urea.

Results:

With protocol t15, the best combination of sensitivity (99.1%), specificity (97.3%) and accuracy (98.7%) was obtained with a cut-off of 2.5‰. There was an extremely high correlation coefficient between the three protocols and conventional 13C-urea breath test (all P < 0.001).

Conclusions:

A urea breath test with 50 mg 13C-urea using a simple test meal and a 15-min sampling interval with a low cut-off seems to be cost-effective and convenient. In a well-standardized laboratory, this modification is not associated with any loss of diagnostic accuracy.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. References

The 13C-urea breath test (13C-UBT), first described by Graham et al. in 1987,1 has been proposed as the most important non-invasive method for the detection of Helicobacter pylori infection.2 For the test, urea marked with the stable carbon isotope 13C is applied orally. If H. pylori's usually potent enzyme urease is present in the stomach during the test, urea will be hydrolysed to form ammonia and labelled carbon dioxide. The appearance of labelled carbon dioxide in the breath indicates that infection is present. Although 14C can be used for the same purpose,3, 413C has the advantage of being non-radioactive and can safely be used in children and women of child-bearing age.

Several papers describing the methodology of 13C-UBT have been published,5–20 but they differ with regard to the dose of 13C-urea, timing of the sample collection, type of test meal used and the equipment required to analyse the breath samples. However, only the dose of 13C-urea and the measuring equipment are directly related to the costs of the test.

For reasons of cost efficiency and practicability, the urea dose and measurement duration should be reduced while still maintaining excellent diagnostic accuracy. Therefore, the primary aim of this study was to establish a validated simple protocol for 13C-UBT with a low dose of 50 mg 13C-urea, a simple test meal and a shortened measurement time. The secondary aim was to compare the protocol with conventional UBT using 100 mg 13C-urea.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. References

Patient recruitment

During a 2-year period, 152 consecutive out-patients presenting for an oesophagogastroduodenoscopy because of dyspeptic symptoms were enrolled in the study.

To avoid interference in the evaluation of H. pylori status, the following patients were excluded: those who had taken antibiotics, bismuth-containing compounds or proton pump inhibitors in the 4 weeks preceding the study, and those with previous partial or complete gastrectomy. All of the procedures were performed with informed consent.

Endoscopy and biopsy

All of the endoscopic examinations were performed and interpreted by the same group of experienced endoscopists at Cathay General Hospital and Medical Centre, Taipei. For endoscopy and biopsy, Olympus instruments were used (GIF XQ200 or XQ230 endoscopes and biopsy forceps type FB-25K). At endoscopy, three antral biopsy specimens were obtained within 2–3 cm from the pylorus. The rapid urease test (CLO test; Delta West Pty, Ltd, Bently, Australia) was brought to room temperature before use, after which the first specimen was implanted in the agar gel. The CLO test was monitored for colour change up to 24 h, and was considered to be positive with the appearance of an appropriate colour change (yellow to magenta). The remaining two biopsy specimens were stained with haematoxylin and eosin and modified Giemsa's stain if necessary. The histological assessment was carried out by an experienced pathologist, who was unaware of the identities of the patients and the clinical diagnosis of H. pylori status.

13C-UBT protocols

13C-UBT was carried out on fasting patients after they had undergone endoscopy, and was performed under the supervision of a study nurse in all patients. The conventional 13C-urea used was 99% 100 mg 13C-urea (U100) produced by the Institute of Nuclear Energy Research in Taiwan. A very simple test meal of 200 mL full-cream cow's milk was given to delay gastric emptying. Baseline samples were collected by blowing through a disposable plastic straw into a 20-mL vacutainer 5 min after consuming the test meal, until condensation appeared on the vacutainer wall. Thereafter, the straw was removed, and the vacutainer was immediately resealed as described previously.6 Ten minutes after consuming the test meal, the patient drank the urea solution that had been prepared by dissolving U100 in 50 mL sterile water. Immediately after taking 13C-urea, patients gargled with water three times in order to avoid oral urease activity, after which they rested, lying on their sides, changing sides every 3 min for a total of three times. Fifteen minutes after ingestion of 13C-urea, a second breath sample was collected in the same way. All the samples were sent to the Institute of Nuclear Energy Research, where a continuous flow isotope ratio mass spectrometer (Europa Scientific, Crewe, UK) was used for analysis. Excess δ13CO215U100) greater than 5‰ between the two samples was considered to be a positive result. The U100 of conventional 13C-UBT has been proven to possess good sensitivity (97%) and specificity (95%) in Taiwan.14

On the second or third day after endoscopy, breath tests were repeated with a low dose of 50 mg 13C-urea (U50) after an overnight fast. The breath samples were collected at baseline and at 10 (protocol t10), 15 (protocol t15) and 30 min (protocol t30) after the ingestion of 13C-urea in the same way. The excess δ13CO2 values of the three protocols were expressed as δ10U50, δ15U50 and δ30U50, respectively.

Criteria of H. pylori infection

The patients were regarded as H. pylori-positive if two or more of the following tests were positive: CLO test, histology or U100 of conventional 13C-UBT; other patients were considered to be H. pylori-negative.

Statistical analysis

The mean ± s.d. and 95% confidence interval (CI) of the excess δ13CO2 values in H. pylori-positive and H. pylori-negative patients were calculated separately for the three protocols and conventional 13C-UBT. The cut-off values of the three protocols were calculated separately according to the receiver operating characteristic curves. The test quality criteria, including sensitivity, specificity, positive predictive value, negative predictive value and accuracy, were also calculated in the three protocols. The most suitable cut-off value of excess δ13CO2 in each protocol was determined by graphically plotting the false-alarm probability (1 – specificity) and discovery probability (sensitivity). Linear regression analysis was performed and the correlation coefficient r (Pearson's r) was calculated from the excess δ13CO2 of the three protocols and conventional 13C-UBT. The linear regression and correlation statistics were performed with Winks software (version 4.5, TexaSoft, Cedar Hill, TX, USA).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. References

A total of 152 dyspeptic patients were enrolled in the study. The mean age was 48.6 ± 14.4 years (range, 16–80 years). There were 86 men (mean age, 48.1 ± 15.0 years; range, 16–79 years) and 66 women (mean age, 49.2 ± 13.7 years; range, 23–80 years). On the basis of the criteria of H. pylori infection, 115 patients were H. pylori-positive and 37 patients were H. pylori-negative. The prevalence of H. pylori infection was 76% in dyspeptic patients. Among the H. pylori-negative patients, there were two patients with a single positive test. One was positive for CLO test alone and another was positive for conventional 13C-UBT with borderline δ15U100. These two patients were both negative for histology, including haematoxylin and eosin and modified Giemsa's stain, which was rechecked by the pathologist.

The endoscopic findings of the study patients were duodenal ulcer (43%), gastric ulcer (17%) and absence of active ulcer (40%). The corresponding percentages for H. pylori-positive patients were 94%, 73% and 57%, respectively.

The excess δ13CO2 values expressed as the mean ± s.d. (95% CI) for δ10U50, δ15U50, δ30U50 and δ15U100 were 17.3 ± 12.7 (14.9–19.6), 18.9 ± 12.3 (16.6–21.2), 18.5 ± 10.4 (16.6–20.4) and 24.9 ± 18.6 (21.5–28.4), respectively, in H. pylori-positive patients, and 1.3 ± 1.2 (0.9–1.7), 0.9 ± 1.1 (0.5–1.3), 0.9 ± 1.1 (0.5–1.2) and 1.1 ± 0.9 (0.8–1.4), respectively, in H. pylori-negative patients.

According to the receiver operating characteristic curves, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy of the three protocols at various cut-off values are given in Table 1. With protocol t15, the best combination of sensitivity (99.1%), specificity (97.3%) and accuracy (98.7%) was obtained with cut-offs of 2.5‰ and 3.0‰. Using protocol t10 or protocol t30, the best test quality criteria were obtained with the same cut-off of 2.5‰.

Table 1.  . Sensitivity (Sen), specificity (Spe), positive predictive value (PPV), negative predictive value (NPV) and accuracy (Acc) of 13C-urea breath test (13C-UBT) with varying cut-off values, performed with three protocols, in 152 dyspeptic patients Thumbnail image of

Regression analysis between the excess δ13CO2 values of the three protocols and conventional 13C-UBT is shown in Figure 1. Comparing δ10U50, δ15U50 and δ30U50 with δ15U100, there is an extremely high correlation coefficient, with values of Pearson's r of 0.8563, 0.8384 and 0.7694, respectively (all P < 0.001).

image

Figure 1. . Scattergrams showing the correlation between the excess δ13CO2 of the three protocols and conventional 13C-urea breath test (13C-UBT): (a) regression analysis between δ10U50 of protocol t10 and δ15U100 of 13C-UBT; (b) regression analysis between δ15U50 of protocol t15 and δ15U100 of 13C-UBT; (c) regression analysis between δ30U50 of protocol t30 and δ15U100 of 13C-UBT.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. References

Since the original report of 13C-UBT by Graham et al.1 who used 350 mg of 13C-urea and breath samples collected every 10 min for 3 h, several modifications have been described with regard to the dose of 13C-urea, breath sampling intervals, cut-off values and type of test meal used. These repeated modifications have been used to optimize the simplicity and minimize the costs of the test, as well as to make it applicable on a large scale.

The 13C-UBT as described in a standardized European protocol by Logan et al. in 1991,5 using 100 mg 13C-urea, is an accurate method for the detection of H. pylori infection; however, 75 mg 13C-urea has been shown to be equally effective.6 Klein et al. validated the protocol with 125 mg 13C-urea, which is currently employed in the USA.7

In order to render the test less expensive and more convenient, the use of lower doses (38–50 mg) of 13C-urea in gelatine capsules or in citric acid-containing tablets seems to be promising, as it shortens breath sampling intervals to 10–20 min.8–10 Enclosing the isotope in a gelatine capsule or in a rapid-releasing tablet shields the urea from exposure to the feeble ureases of oral bacteria that can produce a small early rise in exhaled CO2 tracer. In this way, it is possible to shorten the measurement time to 10 min. Although the solid dosage form of 13C-urea has been marketed, it is still not available world-wide, especially in underdeveloped or developing countries, due to its extra cost. Moreover, the short dissolving duration of capsules and tablets (10–15 min) might be difficult in patients with hypochlorhydria during H. pylori infection. Rigorous prospective studies are required to clarify this important issue. With our procedure, oral urease activity was reduced by gargling with water three times instead of using a solid dosage form. This method seems to be effective in avoiding oral urease activity and maintains excellent diagnostic accuracy, even at 10-min breath sampling intervals.

The need to provide a test meal in 13C-UBT has been demonstrated in several studies, as a meal increases the contact time between the tracer and H. pylori urease inside the stomach. Different nutrient meals have been proposed, including fatty test meals, pure orange juice or even full-cream milk.15, 16 Recently, citric acid, which acts by lowering duodenal pH, and in turn reduces antral motility and relaxes the gastric fundus, has been shown to be an optimal test meal.17, 18 However, citric acid solutions are bitter and unpleasant. In our study, 200 mL full-cream cow's milk, which is high in lipids, was used as a simple test meal, with satisfactory results being reported by Johnston et al.15 Furthermore, cow's milk is preferred to citric acid because of the better acceptance in patients with abdominal pain.

In accordance with the standardized European protocol, our patients laid on their sides, changing sides every 3 min for a total of three times; this provided maximal contact between the tracer and the H. pylori urease to obtain good hydrolysis of 13C-urea.

With regard to the breath collection times, there is general agreement on the use of two breath samples regardless of how much isotope is chosen (75–125 mg): one collected at baseline and another collected 30 min after ingestion of the tracer.5, 6, 11, 13 However, as shown in Table 1, we chose a lower cut-off value of 2.5‰ in the three protocols, compared to 5.0‰ as the cut-off for U100 of conventional 13C-UBT, which appears to be accurate for all three protocols. Thus we obtained a validated simple protocol with reduced urea dose (50 mg) and shortened measurement duration (15 min) for the detection of H. pylori infection, with a maximum sensitivity of 99.1%, specificity of 97.3% and accuracy of 98.7%. Furthermore, the shortened sampling interval may also be of diagnostic value, especially in patients with gastric motility disorder, as well as being convenient for the operator on a large scale.

For economic reasons, only half the amount of 13C-urea was used compared with conventional 13C-UBT. A very simple test meal of 200 mL full-cream cow's milk was used instead of a solid dosage form of 13C-urea. These modifications are definitely less expensive and more practical, with a cost saving of approximately 36% compared to U100 of conventional 13C-UBT.

In conclusion, 13C-UBT for the diagnosis of H. pylori infection can be performed using a validated simple methodology with a reduction in urea dose, shortened measurement time and a simple test meal. In a well-standardized laboratory, this modification is as accurate as the conventional 13C-UBT for the diagnosis of H. pylori infection.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. References

This study was supported by a research grant from the Research Committee of Cathay General Hospital, Taiwan.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. References
  • 1
    Graham DY, Klein PD, Evans DJ, et al.Campylobacter pylori detected noninvasively by the 13C-urea breath test. Lancet 1987; 1: 11747.
  • 2
    Atherton JC, Spiller RC. The urea breath test for Helicobacter pylori. Gut 1994; 35: 7235.
  • 3
    Rauws EAJ, Royen EAV, Langenberg W, Woensel JV, Vrij AA, Tytgat GN. 14C-urea breath test in Campylobacter pylori gastritis. Gut 1989; 30: 798803.
  • 4
    Marshall BJ, Plankey MW, Hoffman SR, et al.A 20-minute breath test for Helicobacter pylori. Am J Gastroenterol 1991; 86: 43845.
  • 5
    Logan RP, Dill S, Bauer FE, et al.The European 13C-urea breath test for the detection of Helicobacter pylori. Eur J Gastroenterol Hepatol 1991; 3: 91521.
  • 6
    Eggers RH, Kulp A, Tegeler R, et al.A methodological analysis of the 13C-urea breath test for detection of Helicobacter pylori infections: high sensitivity and specificity within 30 min using 75 mg of 13C-urea. Eur J Gastroenterol Hepatol 1990; 2: 43744.
  • 7
    Klein PD, Malaty HM, Martin RF, et al. Noninvasive detection of Helicobacter pylori infection in clinical practice: the 13C urea breath test. Am J Gastroenterol 1996; 91: 6904.
  • 8
    Bielanski W, Konturek SJ. New approach to 13C-urea breath test: capsule-based modification with low-dose of 13C-urea in the diagnosis of Helicobacter pylori infection. J Physiol Pharmacol 1996; 47: 54553.
  • 9
    Vaira D, Ricci C, Tampieri A, et al.A new 10 minute, 50 mg urea tablet-based 13C UBT for diagnosis and monitoring Helicobacter pylori (HP) infection. Gut 2001; 49(Suppl. 11): A102A102(Abstract).
  • 10
    Hamlet A, Stage L, Lönroth H, et al. A novel tablet-based 13C urea breath test for Helicobacter pylori with enhanced performance during acid suppression therapy. Scand J Gastroenterol 1999; 34: 36774.
  • 11
    Klein PD, Graham DY. Minimum analysis requirements for the detection of Helicobacter pylori infection by the 13C-urea breath test. Am J Gastroenterol 1993; 88: 18659.
  • 12
    Thijs WJ, Thijs JC, Kleibeuker JH, et al. Evaluation of clinical and home performance of the 13C-urea breath test for the detection of Helicobacter pylori. Eur J Gastroenterol Hepatol 1995; 7: 6037.
  • 13
    Ellenrieder V, Glasbrenner B, Stoffels C, et al.Qualitative and semi-quantitative value of a modified 13C-urea breath test for identification of Helicobacter pylori infection. Eur J Gastroenterol Hepatol 1997; 9: 10859.
  • 14
    Wang WM, Lee SC, Ding HJ, et al.Quantification of Helicobacter pylori infection: simple and rapid 13C-urea breath test in Taiwan. J Gastroenterol 1998; 33: 3305.
  • 15
    Johnston BJ, Levi S, Johnson P. Effect of test meal on 13C-urea breath test results. Gastroenterology 1996; 110(Suppl. 4): A144A144(Abstract).
  • 16
    Bazzoli F, Cecchini L, Corvaglia L, et al.Validation of the 13C-urea breath test for the diagnosis of Helicobacter pylori infection in children: a multicenter study. Am J Gastroenterol 2000; 95: 64650.
    Direct Link:
  • 17
    Domínguez-Muñoz JE, Leodolter A, Sauerbruch T, et al. A citric acid solution is an optimal test drink in the 13C-urea breath test for the diagnosis of Helicobacter pylori infection. Gut 1997; 40: 45962.
  • 18
    Graham DY, Runke D, Anderson SY, et al. Citric acid as the test meal for the 13C-urea breath test. Am J Gastroenterol 1999; 94: 12147.
    Direct Link:
  • 19
    Braden B, Gelbmann C, Dietrich CF, et al. Qualitative and quantitative clinical evaluation of the laser-associated ratio analyser for detection of Helicobacter pylori infection by 13C-urea breath tests. Eur J Gastroenterol Hepatol 2001; 13: 80710.
  • 20
    Sheu BS, Lee SC, Yang HB, et al.Lower-dose 13C-urea breath test to detect Helicobacter pylori infection — comparison between infrared spectrometer and mass spectrometry analysis. Aliment Pharmacol Ther 2000; 14: 135963.