Breath.tempurl.co.il

Aliment Pharmacol Ther 2004; 20: 117–122.
Masking of 13C urea breath test by proton pump inhibitors isdependent on type of medication: comparison between omeprazole,pantoprazole, lansoprazole and esomeprazole A . L E V I N E * , O . S H E V A H   , V . S H A B A T - S E H A Y E K   , H . A E E D   , M . B O A Z à , S . F . M O S S § , Y . N I V – ,Y . A V N I – & H . S H I R I N –*Pediatric Gastroenterology Unit, Departments of  Gastroenterology and àEpidemiology Unit, The E. Wolfson MedicalCenter, Tel-Aviv University, Tel Aviv, Israel; §Gastroenterology Division, Department of Medicine, Rhode Island Hospital/Brown University, Providence, RI, USA; –Department of Gastroenterology, Rabin Medical Center, Petach Tiqva, Tel-AvivUniversity, Tel Aviv, Israel Results: One hundred and seventy-nine patients, mean age 45.8 ± 16.8, completed the study. Treatment with Background: The need to withhold acid suppression omeprazole or pantoprazole prior to urea breath test therapy while awaiting urea breath test results is a (UBT) was associated with low false negative results, common clinical problem in symptomatic patients. It is while lansoprazole and esomeprazole caused clinically unclear at present if the dose or type of proton pump unacceptable high false negative rates (pantoprazole inhibitor or the type of test meal govern the apparent 2.2% vs. lansoprazole 16.6%, P ¼ 0.02, vs. esomepra- masking effect of proton pump inhibitors on the urea zole 13.6%, P ¼ 0.05; omeprazole 4.1% vs. lansopra- Aim: To prospectively evaluate Helicobacter pylori detec- Conclusions: Proton pump inhibitor-induced false neg- tion rates during treatment with four different proton ative results on high-dose citric acid based urea breath pump inhibitors, utilizing a high-dose citric acid-based test vary with the type of proton pump inhibitor used.
Selection of the appropriate test meal and proton pump Methods: Patients positive for Helicobacter pylori by urea inhibitor may allow symptomatic individuals to con- breath test were randomized to receive either omeprazole tinue their proton pump inhibitors prior to performing a 30 mg/day or esomeprazole 40 mg/day for 14 days. Arepeat breath test was performed on day 14 of treatment.
Inhibition of acid secretion from gastric parietal cells is achieved by blocking the H+, K+ adenosine triphos- The introduction of proton pump inhibitors (PPIs) phatase (ATPase) ion pump.1 PPIs are highly effective constitutes one of the most significant medical break- in the treatment and symptomatic relief of peptic ulcer, throughs in the treatment of acid related disorders.
gastro-oesophageal reflux disease, and as part ofcombination triple therapy for Helicobacter pylori erad-ication.2, 3 Few clinically significant differences have Correspondence to: Dr H. Shirin, Department of Gastroenterology, The been found in the efficacy or adverse events rate E. Wolfson Medical Center, Holon 58100, Israel.
E-mail: [email protected] between the three most prevalently prescribed and studied PPIs, lansoprazole (LAN), pantoprazole (PAN) prefer to start PPI treatment until the time of UBT.
and omeprazole (OME).1, 4 The recently introduced Based on previous observations, it may be possible to (S)-isomer of OME, esomeprazole (ESO), may provide start or continue short term PPI therapy if the more effective gastric acid control than standard doses appropriate PPI and test meal do not significantly alter of the other PPIs,5 but may not afford a better clinical UBT results. The aim of this study was to prospectively evaluate the false negative rates of four different PPIs The effect of PPIs on the results of the 13C UBT appears during UBT, using a high dose citric acid test meal.
to be due to a pH-dependent mechanism.8–10 Previouslypublished reports have described false negative rates of17–38% for 20 mg/day OME after 14 days. Similar results were reported with the use of 30 mg/day LAN (Table 1).2, 3, 11 The effect of PAN is more controversial.
Parente et al.12 demonsrated that the use of 40 mg/day A UBT (BreathID; Oridion, Jerusalem, Israel) was PAN for 14 days does not lead to false negative UBTs, performed after a 3-h fast in patients over 18 years of whereas Dulbecco et al., found significant false negative age with upper gastrointestinal symptoms. Those with a results using PAN.13 There is no data currently positive UBT were included in the study. Exclusion estimating the false negative UBT rate induced by ESO.
criteria included: (i) administration of antibiotics and/or Difference in the effect of PPIs on UBT may also depend bismuth preparations within 4 weeks before the date of on factors other than the choice of PPI. We and others entry to the study, (ii) administration of PPIs within have recently published that the use of a test drink 4 weeks before the date of entry to the study, (iii) containing high dose citric acid may significantly pregnant or breast-feeding women and (iv) previous decrease the false negative results associated with gastric or oesophageal surgery. Patients were random- 14 days treatment by OME or PAN, independent of ized to treatment with either OME 20 mg/day, PAN 40 mg/day, LAN 30 mg/day or ESO 40 mg/day, taken Based on previous studies, most centres currently at 08:00 hours, 30–60 min before breakfast. A repeat breath test was performed on therapy at day 14, 1–3 h UBT.17, 18 This requirement means that symptomatic after patients received their last PPI dose. Patients were patients have to defer therapy for a significant period of asked not to take antibiotics, bismuth compounds or to time in order to be tested. Ideally, for both clinical and alter the recommended dose of the PPI. Compliance was quality of life concerns, patients and physicians would checked by means of pill counts on day 14.
Patients with a negative UBT on day 14 underwent another UBT 2 weeks after PPI cessation, in order to Table 1. Summary of false negative urea breath test results clarify whether this was a false negative result, or induced by using standard doses of different proton pump alternatively, true eradication of the bacteria by PPI had inhibitors for 14 days. Comparison between omeprazole (OME), pantoprazole (PAN) and lansoprazole (LAN) Informed consent was obtained from each patient before enrollment in the study. The study protocol was approved by the Institutional Review Board of the E. Wolfson Medical Center. This study was not suppor- ted by a commercial company. The effect of PPIs on the detection of H. pylori was examined by continuous real time UBT, (BreathID; Oridion).19 All patients ingested a test drink provided by the manufacturer that included 75 mg 13C-urea (tablet form of 99% 13C-enriched urea) with 4.0 g citric acid granulated based powder dissolved a Using high-dose citric acid as a test meal.
in 200 mL water. The cut-off point or threshold for the Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 117–122 P R O T O N P U M P I N H I B I T O R S A N D U RE A B R E A T H T E S T BreathID test has been determined to be 5 delta over because of refusal to participate and 212 patients were enrolled in the study. None of the patients hadsignificant chronic medical problems. During the study33 patients were excluded because of PPI-induced side effects (mainly diarrhoea) or failure to return for Analysis of data was carried out using SPSS 9.0 follow-up testing after day 14 of PPI therapy. One statistical analysis software (SPSS Inc., Chicago, IL, hundred and seventy-nine patients (89 males and 90 USA, 1999). Distributions of continuous variables were females, mean age 45.8 ± 16.8, range 18–85 years) tested for normality using the Kolmogorov–Smirnov completed the study with a full set of test data.
test. The DOB distributions after 2 weeks of PPI Indications for H. pylori testing included, epigastric treatment significantly differed from normal, so non- pain (51), gastro-oesophageal reflux disease (50), parametric hypothesis testing was used. Spearman’s rho dyspepsia (28), peptic ulcer (14), vomiting (2), stool correlation coefficients were calculated to describe occult blood (1) and patient’s request (33). All patients associations between baseline and 2-week PPI (2w had complied fully with the medication schedule.
PPI) breath test values. Additionally, these associationswere tested within each treatment assignment individu- ally. UBT results were compared simultaneously acrossPPI treatment groups using the Kruskal–Wallis test and After 14 days of treatment 20 patients became negative.
followed with post hoc pairwise testing using the Mann– In a third breath test performed 2 weeks after cessation Whitney U. The Fisher exact test was used to compare of PPIs, all but four of these patients with negative tests the rate of false negative results by PPI treatment and to became positive, confirming that the initial result had determine whether false negative results differed by been a false negative result. Of these 16 patients, seven gender. The t-test for independent samples was used to received LAN (16.6%) and six received ESO (13.6%) examine age by false negative results at day 14. Logistic while only two patients on OME (4.1%) and one on regression analysis was used to determine whether the PAN (2.2%) demonstrated false negative results (overall baseline 13C excretion predicted false negative results.
P ¼ 0.04). Post hoc pairwise comparisons revealed that All tests were considered significant at P £ 0.05.
subjects treated with PAN had significant fewer falsenegative results than LAN-treated subjects (P ¼ 0.02)and significantly fewer false negatives than ESO-treated subjects (P ¼ 0.05). Additionally, subjects treated withOME had fewer false negatives than LAN-treated sub- jects (P ¼ 0.05). Significant differences in other pairwise Three hundred and eighty-six consecutive patients comparisons were not detected (Table 2). There were no were tested by BreathID and 271 (69.9%) were found unifying parameters that characterized patients with a to be H. pylori positive. Of these, 59 were excluded 14 after proton pump inhibitor treatment.
Comparison between omeprazole (OME), * OME vs. LAN P ¼ 0.05.
** PAN vs. LAN P ¼ 0.02, PAN vs. ESO P ¼ 0.05.
DOB, delta over baseline.
Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 117–122 Figure 1. Individual values of delta over baseline (DOB) 13CO2 excretion of 179patients before acid suppressive therapy and the corresponding values after 14 daysof treatment with omeprazole 20 mg/day, 30 mg/day and esomeprazole 40 mg/day.
Four patients (three on ESO and one on PAN) who PPIs that could provide symptom relief, because of the became negative after 14 days, remained negative risk of a false negative test. The results of the present 2 weeks later, suggesting either the first UBT was study show that PPI-associated UBT masking can be falsely positive or alternatively that ESO and PAN had kept to a minimum with judicious use of the appropriate truly eradicated H. pylori. In a logistic regression model, PPI. Both PAN and OME had very low false positive baseline UBT results did not predict the false negative rates (2–4%), whereas LAN and ESO had unacceptably results at day 14. The DOB at 2 weeks did not differ high false negative rates ranging from 13 to 16%.
significantly from baseline in the OME, PAN and LAN Variability between PPIs using the same dose of citric treatment groups, but it became significantly lower acid may be explained by two different mechanisms in the ESO group (23.8 ± 18.3 vs. 19.1 ± 17.5, P ¼ which influence detection of H. pylori. Inhibition of the 0.04). Figure 1 demonstrates the individual subject UBT bacterial urease activity secondary to alkalinization of values in each group at baseline and 14 days after PPI the gastric content is one of the possible explanations for the false negative UBT induced by PPIs.20–23 Bothurease activity and the transport of urea into thebacteria which is regulated by UreI-dependent specific H+-gated urea channels are pH dependent.24 Urease Symptomatic patients referred for 13C UBTs, prior to activity is low at neutral pH, but as the external pH H. pylori diagnosis, often have to refrain from taking decreases to between 6.5 and 5.5 there is a 10–20-fold Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 117–122 P R O T O N P U M P I N H I B I T O R S A N D U RE A B R E A T H T E S T increase in activity which remains high through citric acid to prevent LAN and ESO induced false approximately pH 2.5.20, 21 In this way, alkalinization negative UBTs may be explained by a combination of of gastric juice by PPIs may reduce both the entrance of marked gastric acid suppression and antimicrobial urea into H. pylori and the activity of its cytoplasmic activity of these compounds against H. pylori. Selection urease and consequently lead to false negative results.
of the appropriate test meal and PPI may obviate the Comperative studies demonstrated that PAN 40 mg/ need to withold therapy prior to performing UBTs.
day and LAN 30 mg/day were more effective ininhibiting acid secretion in healthy volunteers than OME 20 mg/day, but were equally as effective as OME40 mg/day.25, 26 probably reflecting dose differences.
1 Horn J. The proton pump inhibitors: similarities and differ- On the contrary, direct comparison between 30 mg ences. Clin Ther 2000; 22: 266–80.
2 Savarino V, Neri M, Vigneri S. PPI-triple therapy in the LAN and 40 mg PAN revealed that despite the different eradication of H. pylori infection. Gastroenterology 1999; doses, 30 mg/day LAN may produce a greater degree of acid inhibition than PAN 40 mg once daily.27 Anders- 3 Stedman CAM, Barclay ML. Review article: comparison of son et al. also demonstrated increased acid inhibitory pharmacokinetics, acid suppression and efficacy of proton effect of 20 mg ESO compared with 20 mg OME.28 pump inhibitors. Aliment Pharmacol Ther 2000; 14: 963–78.
An alternative explanation is that PPIs directly inhibit 4 Richardson P, Hawkey CJ, Stack WA. Proton pump inhibitors.
Pharmacology and rationale for use in gastrointestinal dis- the viability and growth of H. pylori.29 Formation of sulphides of benzimidazoles as demonstrated in culture 5 Scott LJ, Dunn CJ, Mallarkey G, Sharpe M. Esomeprazole: a media30 or direct inhibition of the bacterial urease review of its use in the management of acid related disorders.
activity may be the reasons for this selective antibac- terial effect31 although growth inhibition has been 6 Scholten T, Gatz G, Hole U. Once-daily pantoprazole 40 mg and esomeprazole 40 mg have equivalent overall efficacy in reported also in urease negative H. pylori derivatives.32 relieving GERD-related symptoms. Aliment Pharmacol Ther Among the three PPIs, OME, LAN and PAN that have been compared for antibacterial activity, LAN was 7 Chey W, Huang B, Jackson RL. Lansoprazole and esomepra- found to be the most potent and PAN the least active zole in symptomatic GERD. Clin Drug invest 2003; 23: 69–84.
compound in vitro.33, 34 Gatta et al. compared the 8 Laine L, Estrada R, Trujilo M, Kniggek K, Fennerty MB. Effect susceptibility of 52 H. pylori isolates to both OME and of proton-pump inhibitor therapy on diagnostic testing forHelicobacter pylori. Ann Intern Med 1998; 129: 547–50.
ESO. They found that 17 isolates were 2–64-fold more 9 Chey WD, Woods M, Scheiman JM, Nostrant TT, DelValle J.
susceptible to ESO compared with only two isolates that Lansoprasole and ranitidine affect the accuracy of the 14C-urea breath test by a pH-dependent mechanism. Am J The rationale for the use of citric acid as an optimal test drink in UBT is not new. This is based on two 10 Greig MA, Neithercut WD, Hossack M et al. Suicidal destruc- tion of H. pylori mediated by its urease activity. Gut 1990; 31: characteristics: decrease of gastric pH and delay of gastric emptying.15, 16, 36, 37 These characteristics may 11 Eaton KA, Brooks CL, Morgan DR et al. Essential role of urease optimize the test meal even during conditions which in pathogenesis of gastritis induced by Helicobacter pylori in normally may cause a false negative breath test, such as Gnotobiotic piglets. Infec Immun 1991; 59: 2470–5.
use of OME.14 Previous studies using standard test 12 Parente F, Sainaghi M, Sangaletti O, et al. Different effects of meals with the same PPIs, presented in Table 1, have short-term omeprazole, lansoprazole or pantoprazole on theaccuracy of the 13C-urea breath test. Aliment Pharmacol shown much higher and clinically unacceptable false negative rates varying from 17 to 38% for OME, 17 to 13 Dulbecco P, Gambaro C, Bilardi C, et al. Impact of long-term 20% for LAN and up to 10.7% on PAN. These ranitidine and pantoprazole on accuracy of [13C] urea breath differences may be related to the effect of high dose test. Dig Dis Sci 2003; 48: 315–21.
citric acid on test results, although we did not evaluate 14 Shirin H, Frenkel D, Shevah O, et al. Effect of proton pump inhibitors on the continuous real time 13C-urea breath test.
Am J Gastroenterol 2003; 98: 46–50.
We conclude that if a PPI needs to be administered 15 Chey WD, Chathadi KV, Montague J, Ahmed F, Murthy U.
when a UBT is performed, it is preferable to choose PAN Intragastric acidification reduces the occurrence of false-neg- or OME, even when using use a high dose of citric acid ative urea breath test results in patients taking a proton pump for the test. We surmise that the failure of high dose inhibitor. Am J Gastroenterol 2001; 96: 1028–32.
Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 117–122 16 Graham DY, Runke D, Anderson SY, Malaty HM, Klein PD.
30 Sjostrom JE, Kuhler T, Larsson H. Basis for the selective Citric acid as the test meal for the 13C-urea breath test. Am J antibacterial activity in vitro of proton pump inhibitors against Helicobacter spp. Antimicrob Agents Chemother 17 Savarino V, Bisso G, Pivari M et al. Effect of gastric acid sup- pression on 13C-urea breath test: comparison of ranitidine 31 Nagata K, Satoh H, Iwahi T, Shimoyama T, Tamura T. Potent with omeprazole. Aliment Pharmacol Ther 2000; 14: 291–7.
inhibitory action of the gastric proton pump inhibitor lan- 18 Savarino V, Tracci D, Dulbecco P et al. Negative effect of soprazole against urease activity of Helicobacter pylori: uniqe ranitidine on the results of urea breath test for the diagnosis of action selective for H. pylori cells. Antimicrob Agents Chem- Helicobacter pylori. Am J Gastroenterol 2001; 96: 348–52.
19 Shirin H, Kenet G, Shevah O et al. Evaluation of a novel 32 Nagata K, Takagi E, Tsuda M, et al. Inhibitory action of lan- continuous real time 13C urea breath analyzer for Helicobacter soprazole and its analogs against Helicobacter pylori: inhibition pylori. Alimentary Pharmacol Ther 2001; 15: 389–94.
of growth is not related to inhibition of urease. Antimicrob 20 Rektorschek M, Weeks D, Sachs G, Melchers K. Influence of pH on metabolism and urease activity of Helicobacter pylori.
33 Iwahi T, Satoh H, Nakao M, et al. Lansoprazole, a novel Gastroenterology 1998; 115: 628–41.
benzimidazole proton pump inhibitor, and its related com- 21 Scott DR, Weeks D, Hong C, Postins S, Melchers K, Sachs G.
pounds have selective activity against Helicobacter pylori.
The role of internal urease in acid resistance of Helicobacter Antimicrob Agents Chemother 1991; 35: 490–6.
pylori. Gastroenterology 1998; 114: 58–70.
34 Nakao M, Malfertheiner P. Growth inhibitory and bactericidal 22 Pantoflickova D, Scott DR, Sachs G, Dorta G, Blum AL. 13C urea activities of lansoprazole compared with those of omeprazole breath test (UBT) in the diagnosis of Helicobacter pylori: why and lansoprazole against Helicobacter pylori. Helicobacter does it work better with acid test meals? Gut 2003; 52: 933–7.
23 Meyer-Rosenberg K, Scott DR, Rex D, Melchers K, Sachs G.
35 Gatta L, Perna F, Figura N, et al. Antimicrobial activity of The effect of environmental pH on the proton motive force of esomeprazole versus omeprazole against Helicobacter pylori.
Helicobacter pylori. Gastroenterology 1996; 111: 886–900.
J Antimicrobial Chemother 2003; 51: 439–42.
24 Weeks DL, Eskandari S, Scott DR, Sachs G. A H+-Gated urea 36 Dominguez-Munoz JE, Leodolter A, Sauerbruch T, Malferthe- channel: the link between Helicobacter pylori urease and gas- iner P. A citric acid solution in an optimal test drink in the tric colonization. Science 2000; 287: 482–5.
13C-urea breath test for the diagnosis of Helicobacter pylori 25 Bruley des Varannes S, Levy P, Lartigue S, Dellatolas F, Lemaire M, Galmiche JP. Comparison of lansoprazole with 37 Shiotani A, Saeed A, Yamaoka Y, Osato MS, Klein PD, Gra- omeprazole on 24-hour intragastric pH, acid secretion and ham DY. Citric acid-enhanced Helicobacter pylori urease serum gastrin in healthy volunteers. Aliment Pharmacol Ther activity in vivo is unrelated to gastric emptying. Aliment 26 Florent C, Forestier S. 24 monitoring of intragastric acidity 38 Chey WD, Spybrook M, Carpenter S, Nostrant TT, Elta GH, between lansoprazole 30 mg and pantoprazole 40 mg. Eur J Scheinman JM. Prolonged effect of omeprazole on the Gastroenterol Hepatol 1997; 9: 195–200.
14C-urea breath tests. Am J Gastroenterol 1996; 91: 89–92.
27 Huang JQ, Goldwater DR, Thompson ABR, et al. Acid sup- 39 Connor SJ, Seow F, Ngu MC, Katekaris PH. The effect of dosing pression in healthy subjects following lansoprazole or pan- with omeprazole on the accuracy of the 13C-urea breath test toprazole. Aliment Pharmacol Ther 2002; 16: 425–33.
in Helicobacter pylori-infected subjects. Aliment Pharmacol 28 Andersson T, Rohss K, Bredberg E, Hassan-Alin M. Pharma- cokinetics and pharmacodynamics of esomeprazole, the 40 Adachi K, Fujishiro H, Mihara T, Komazawa Y, Kinoshita Y.
S-isomer of omeprazole. Aliment Pharmacol Ther 2001; 15: Influence of lansoprazole, famotidine, roxatidine and reb- amipide administration on the urea breath test for the diag- 29 Graham DY, Opekun AR, Hammoud F, et al. Studies regarding nosis of Helicobacter pylori infection. J Gastroenterol Hepatol the mechanism of false negative urea breath tests with proton pump inhibitors. Am J Gastroenterol 2003; 98: 1005–9.
Ó 2004 Blackwell Publishing Ltd, Aliment Pharmacol Ther 20, 117–122

Source: http://breath.tempurl.co.il/uploadimages/systemFiles/HP3.pdf