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:

The expense of the 13C-urea breath test (UBT) to detect Helicobacter pylori infection is mainly due to the cost of 13C-urea and the analysis using isotope ratio mass spectrometry (IRMS).

Aim:

To test whether a UBT, using a lower dose of urea and lower-priced isotope-selective nondispersive infrared spectrometry (INIS), can preserve diagnostic efficacy in clinical practice.

Methods:

A total of 177 dyspeptic patients received endoscopy for H. pylori culture and histology. All of them received a UBT in which the duplicate baseline, 10 min, and 15 min breath samples after ingestion of 50 mg 13C-urea were collected to analyse the excess 13CO2/12CO2 ratio (ECR) by IRMS (ABCA, Europa Scientific, UK) and INIS (UBiT-IR200, Photal Otsuka Electronics, Japan), respectively.

Results:

Of the 177 patients, 84 were infected and 93 were uninfected with H. pylori. A close correlation of ECR was found between IRMS and INIS (r=0.9829 at 10 min; r=0.9918 at 15 min, P < 0.0001). Analysing the 15-min samples, UBT by both IRMS and INIS achieved the same sensitivity (96.4%) and specificity (98.9%).

Conclusions:

INIS is as effective as IRMS for UBT, and can use a lower dose of 13C-urea. This can provide an economic UBT, using the lower-priced INIS and a low dose of 13C-urea.


INTRODUCTION

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

With the development of the isotope-labelled 13C-urea breath test (UBT), a convenient and noninvasive test to detect Helicobacter pylori infection has become widely available in clinical practice. This noninvasive test can be used to screen for H. pylori infection prior to an endoscopic procedure, and it has been confirmed as being reliable in assessing whether H. pylori is eradicated after therapy.1[2][3][4]–5 In spite of its reliability, UBT remains a relatively expensive method of testing for H. pylori infection. The high cost of performing UBT is due to the expense of using a high dose of 13C-urea and the use of an expensive analyser which involves isotope ratio mass spectrometry (IRMS). In addition to the high costs of the analyser, skilled personnel must be trained to operate the analyser. The availability of UBT with IRMS as the analyser is therefore limited.

Recently, a less expensive isotope-selective nondispersive infrared spectrometer (INIS) has been developed and has been shown to correlate well with IRMS in terms of quantitative gas analysis.6[7]–8 Moreover, INIS is comparatively very convenient and easy to handle in clinics.6, 7 For these reasons, the INIS-analysed UBT is now showing promise of having broader applications in clinical practice.

Since lowering the dosage of test urea can greatly reduce the cost of UBT, we have investigated whether the diagnostic efficacy of a more economical UBT, one that makes use of a lower-dosed (50 mg) test urea, and less expensive analyser with INIS, could maintain the same high diagnostic efficacy in detecting H. pylori infection.

MATERIALS AND METHODS

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

Patients and study design

From February to December 1999, a total of 177 patients undergoing diagnostic upper endoscopy following dyspeptic symptoms were enrolled into this study. To avoid any impediment in the evaluation of H. pylori status, the patients were excluded if they had any of the following conditions: ingestion of bismuth salts, proton pump inhibitors, or antibiotic use within the previous 8 weeks, known allergy to penicillin, or previous gastrointestinal surgery. Patients known to have had a past history of anti-H. pylori therapy and gastric malignancy were also excluded. After obtaining consent from the patients, they all received gastric biopsies for histology and culture to evaluate the initial status of H. pylori infection during the endoscopy. Furthermore, after endoscopy, all patients completed the protocol for UBT. In this study, H. pylori infection was defined by a positive result from either culture or histology.

Endoscopy and gastric biopsy for histology and culture

Video endoscopes (Olympus XQ 200, Olympus Corp., Tokyo) and large-channel biopsy forceps (Olympus FB-25N, Olympus Corp., Tokyo) were used throughout the study to obtain a total of six strips for gastric biopsy. During each endoscopy, a pair of biopsies over the gastric cardia was first sampled under the retroversion of the endoscope, 3 cm within the E-C junction, for histological evaluation. Next, another two pairs of biopsies were obtained from the lower body (45–50 cm on the greater curve) and from over the antrum (2 cm within the pyloric ring), respectively.9 In each pair of biopsies, one was used for culture and the other for histology. Among patients with gastric ulceration, an additional four biopsy samples were taken in order to exclude the risk of malignancy. If a malignant ulcer was proven, the patient was excluded from the study. The three biopsy specimens for histology were fixed in 10% buffered formalin and embedded in paraffin and routinely stained with H&E. A positive histology result for H. pylori was defined as the presence of bacteria in any biopsy sample, but was rechecked by Warthin-Starry silver stain once the result of the culture was found negative.

H. pylori culture

The three pieces of gastric biopsy which were collected during endoscopy for culture were immediately transferred into brain/heart infusion (BHI, Difco Laboratory, Detroit, MI) in ice and inoculated on to CDC anaerobic blood agar (BBL, Microbiology Systems, Cockeysville, MD). The agar was incubated at 35 °C for 2 days in a microaerophilic gas mixture composed of 5% O2, 10% CO2 and 85% N2 (Difco Laboratories, Detroit, MI). H. pylori was identified on the basis of rapid urea hydrolysis, catalase, oxidase production and the API Campy kit (bioMerieux, Marcy-L’Etolle, France).10

13C-UBT assessed by INIS and IRMS

Within three days of the endoscopy, each patient received the same protocol for UBT, preceded by a 100-mL citric acid meal to prolong gastric emptying time after overnight fasting and mouth washing. The expired gas samplings were collected in duplicate at baseline, 10 min, and 15 min after the ingestion of 50 mg 13C-labelled urea (INER-Hp13C-tester, Institute of Nuclear Energy Research, Taiwan). In each patient, the duplicate gas samples were collected for an analysis of any excess 13CO2/12CO2 ratio (ECR) by IRMS (ABCA, Europa Scientific, UK) and INIS (UBiT-IR200, Photal Otsuka Electronics Co, Japan), respectively. The gas sample collected for IRMS was stored in a 20 mL glass test tube. In contrast, the exhaled gas for INIS was collected in a 200 mL gas storage bag. The longest gas collecting time after ingestion of the test urea was scheduled at 15 min, because one of our previous studies confirmed that the 15-min gas sampling achieved a high diagnostic yield when performing the UBT.11 In each UBT, the patient was scheduled to change their lying position every 3 min to ensure the coating of the urea agent. The cut-off point of ECR was selected as 3.5 to judge the diagnostic efficacy of UBT.11

Statistics

Correlation between INIS and IRMS results were tested descriptively by linear regression. The Pearson’s correlation coefficient was recorded and assessed using a two-tailed method. Agreement between the two methods was illustrated by plotting the difference between the measured ECR vs. their mean value.12

RESULTS

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

The mean age of the 177 patients (79 females and 98 males) was 45.9 years. The endoscopic findings were duodenal ulcer in 50, gastric ulcer in 28, and absence of ulcer in 99 cases. Based on the culture and histology results, of the 177 included patients, 93 were found to be H. pylori-negative and 84 were H. pylori-positive. Eighty-two of the 84 H. pylori-positive patients had positive results from histology, including 71 with positive culture results and 11 without. Only two patients had negative histology result but a positive culture result. The prevalence rate of H. pylori was 88% (44/50) in patients with duodenal ulcer, which were higher than in patients with gastric ulcer, who had 50% prevalence (14/28).

Quantitative correlation of ECR between UBT by INIS and by IRMS

Applying both the 10 min and 15 min gas samples to the UBT, the linear regression of ECR analysed by INIS and IRMS had an extremely high correlation coefficient (r=0.9829 at 10 min; r=0.9918 at 15 min, P < 0.0001). The difference in the data pairs are plotted vs. their average in Figures 1 and 2, which show ECRs of 10 min and 15 min breath samples for UBT, respectively. In Figure 1, only 4.5% (8/177) of the paired 10 min gas samples had a difference beyond two standard deviations (s.d.) of the mean value. Figure 2 showed that only 5.6% (10/177) of the paired 15 min gas samples had a difference beyond two s.d. of the mean value. This analysis revealed that nearly 95% of the differences between the ECR by INIS and IRMS were less than two s.d. of the mean value. Moreover, in neither Figures 1 or 2 was there a difference in tendency to increase with higher mean ECR.

image

Figure 1. . Agreement of two different methods when applied with 10 min breath samples. The difference in ECR of UBT analysed by isotope ratio mass spectrometry (IRMS) and isotope-selective nondispersive infrared spectrometer (INIS) is plotted vs. the average of these data pairs. The middle line represents the mean of difference, and the upper and lower lines represents ± two standard deviations (s.d.). ECR: excess 13CO2/12CO2 ratio.

Download figure to PowerPoint

image

Figure 2. . Agreement of two different methods when applied with 15 min breath samples. The difference in ECR of UBT analysed by isotope ratio mass spectrometry (IRMS) and isotope-selective nondispersive infrared spectrometer (INIS) is plotted vs. the average of these data pairs. The middle line represents the mean of the difference, and the upper and lower lines represents ± two standard deviations (s.d.). ECR: excess 13CO2/12CO2 ratio.

Download figure to PowerPoint

Qualitative diagnostic efficacy of UBT assessed by INIS and IRMS

In the description of breath samples taken at 15 min in Table 1, the sensitivity and specificity for the UBT by INIS were as high as those for the UBT by IRMS (sensitivity: 96.4%; specificity: 98.9%). Whereas, in those taken at 10 min, the specificity of UBT assessed by INIS was slightly lower than UBT assessed by IRMS (94.6% vs. 96.4%).

Table 1.  . Diagnostic efficacy of 13C-UBT by IRMS and INIS for H. pylori detection %(n) Thumbnail image of

DISCUSSION

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

As the 13C-UBT is now confirmed as being reliable in the detection of H. pylori infection in dyspeptic patients, a wider clinical application of this noninvasive tool will be important. However, the major cost of a 13C-UBT is due to the fairly expensive 13C-urea and the IRMS analyser. Moreover, skilled personnel must be trained to perform the analyses of UBT by IRMS. This high cost and complexity of analysis limits the use of 13C-UBT in clinical practice. However, we have found that by using a less expensive analyser and a lower dosage of 13C-urea, a high diagnostic efficacy of the test could be preserved and that it can be made more widely available.

The newly developed INIS, UBiT-IR200, is a comparatively inexpensive and compact device that does not require a dedicated experienced operator to provide the 13C-UBT analysis. The cost of INIS is around £15 000–20 000, about one-quarter to one-third of the cost of IRMS. Furthermore, the evaluation of gas samples for 13C-UBT by INIS can be completed within 30 min. With these advantages, INIS can be very helpful in overcoming the present limits on the wider clinical application of IRMS. By using a dose of 100 mg 13C-urea for UBT, Ohara et al. have demonstrated a high correlation of the ECR (r=0.996) between IRIS and IRMS.6 This data confirms that IRIS can provide a simple and accurate method for analysing the 100 mg-dosed 13C-UBT.

In an effort to further reduce the cost of the 13C-UBT, the 13C-urea dosage was reduced from 100 mg to 50 mg in this study. The quantitative results of 13C-UBT analysed by INIS and IRMS showed an excellent correlation (r=0.9829 at 10 min; r=0.9918 at 15 min, P < 0.0001). Moreover, using the Bland–Altman plot method, very few quantitative variations were found between these two methods ( Figures 1 and 2). These data indicate that the quantitative assessment of INIS and IRMS had similar results, even when using half the previously recommended 13C-UBT urea dosage (50 mg).

The sensitivity and specificity of the 13C-UBT analysed by INIS were as high as those of the 13C-UBT assessed by IRMS (sensitivity: 96.4%; specificity: 98.9%) in the analysis of the 15 min breath sampling ( Table 1). This finding confirmed that, in the detection of H. pylori infection, an economic 13C-UBT, which makes use of a lower-dosed test urea and a less expensive analyser such as INIS, can achieve the same high diagnostic efficacy as a UBT assessed with IRMS.

In our study, while in analysing the 10 min breath samples, the sensitivity of 13C-UBT assessed by INIS was the same as that of 13C-UBT assessed by IRMS, while the specificity of 13C-UBT assessed by INIS was slightly lower than 13C-UBT assessed by IRMS (94.6% vs. 96.4%). The exact reason why a slightly lower specificity of 13C-UBT using the 10 min samples for INIS was obtained is uncertain. We speculate that the bias was due to oral flora and a larger volume (200 mL) of the breath collecting bag for INIS, resulting in some false positive results. However, the specificity was improved if a prolonged breath sampling time of 15 min was applied for the INIS.

In summary, an economical 13C-UBT, one which makes use of INIS and a dosage of 13C-urea as low as 50 mg, can provide an accurate and cost-effective method of detecting H. pylori infection. This economical method has demonstrated the possibility of a wider application of the 13C-UBT in clinical practice.

Acknowledgements

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

This work was supported by a research grant from the Research Committee of Medical College of National Cheng Kung University, Taiwan (NCKU-M89006). The authors thank Mr Kajiwara Meguru for providing UBiT-IR200, Mr Lin Long-Tsun, and Tsai Ming-Pin of Taiwan Otuska Company for their technical assistance.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References
  • 1
    Logan RP. Urea breath tests in the management of Helicobacter pylori infection. Gut 1998; 43(Suppl. 1): S4750.
  • 2
    Atherton JC. Non-endoscopic tests in the diagnosis of Helicobacter pylori infection. Aliment Pharmacol Ther 1997; 11(Suppl. 1): 1120.
  • 3
    Van Der Hulst RW, Lamouliatte H, Megraud F, et al. Laser assisted ratio analyser 13C-urea breath testing, for the detection of H. pylori: a prospective diagnostic European multicentre study. Aliment Pharmacol Ther 1999; 13: 11717.
  • 4
    Cave DR, Zanten SV, Carter E, et al. A multicentre evaluation of the laser assisted ratio analyser (LARA): a novel device for measurement of 13CO2 in the 13C-urea breath test for the detection of Helicobacter pylori infection. Aliment Pharmacol Ther 1999; 13: 74752.
  • 5
    McColl KEL, El-Nujumi A, Murray L, et al. The Helicobacter pylori breath test: a surrogate marker for peptic ulcer disease in dyspeptic patients. Gut 1997; 40: 3026.
  • 6
    Ohara S, Kato M, Asaka M, Toyota T. The UBiT-100 13CO2 infrared analyzer: comparison between infrared spectrometric analysis and mass spectrometric analysis. Helicobacter 1998; 3: 4953.
  • 7
    Hildebrand P, Beglinger C. Nondispersive infrared spectrometry: a new method for the detection of Helicobacter pylori infection with 13C-urea breath test. Clin Infect Dis 1997; 25: 10035.
  • 8
    Braden B, Schafer F, Caspary WF, Lembcke B. Nondispersive isotope-selective infrared spectroscopy: a new analytical method for 13C-urea breath test. Scand J Gastroenterol 1996; 31: 4425.
  • 9
    Yang SB, Sheu BS, Su IJ, Chien CH, Lin XZ. Clinical application of gastric histology to monitor treatment of dual therapy in H. pylori eradication. Dig Dis Sci 1997; 42: 183540.
  • 10
    Sheu BS, Wu JJ, Yang HB, Huang AH, Lin XZ. One week proton pump inhibitor-based triple therapy is effective in eradicating residual H. pylori after failed dual therapy. J Formos Med Assoc 1998; 97: 26670.
  • 11
    Sheu BS, Lee SC, Yang HB, Lin XZ. Quantitative result of 13C-urea breath test at 15 minutes may correlate with the bacterial density of H. pylori in the stomach Hepato-Gastroenterology 1999; 46: 205762.
  • 12
    Bland JM, Altmann DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 30710.