Aliment Pharmacol Ther 2001; 15: 203±206.
Usefulness of Helicobacter pylori stool antigen test to monitor response to eradication treatment in children G. ODERDA, A. RA PA, D. MAR INELLO, B. RONCHI & A . ZAVALLONE Paediatric clinic, University of Piemonte Orientale, Novara, Italy Accepted for publication 20 September 2000
SUMMARY
Background: The monitoring of the results of eradication treatment is a crucial step for patients with Helicobacter pylori gastritis. A non-invasive test for H. pylori antigens in stools (HpSA) was recently validated for children. Aim: To evaluate the accuracy of HpSA in monitoring eradication treatment in children. Methods: In 60 children, H. pylori gastritis was diagnosed by endoscopy and the 13C-urea breath test. The children were treated and returned for a follow-up 13 C-urea breath test 6 weeks after the end of treatment. Children were considered cured when the 13C-urea breath test was negative. Stool were collected at
INTRODUCTION
In children treated for Helicobacter pylori infection, the response to treatment should be monitored with a reliable non-invasive test, as suggested by a Consensus Conference on the management of H. pylori infection.1 This is because: (i) symptom disappearance is not a reliable predictor of eradication; (ii) in children with gastric or duodenal ulcer there is a high recurrence rate if infection persists; (iii) endoscopy is not justi®ed when only H. pylori needs to be determined; and (iv) the physician gets feed back on the ef®cacy of the treatment regimen. To-date the most reliable test for this purpose seems to be the 13C-urea breath test but its cost is high and its availability is not diffuse in every Centre. Correspondence to: Dr G. Oderda, Clinica Paediatrica, UniversitaÁ del Piemonte Orientale, Via Solaroli 17, 28100 Novara, Italy. E-mail:
[email protected] Ó 2001 Blackwell Science Ltd
baseline, and at 2 and 6 weeks. Stool antigens were measured by HpSA. Results: According to 13C-urea breath test, 6 weeks after the end of treatment 49 children were cured and 11 were still H. pylori-positive. The sensitivity and speci®city of HpSA on stools collected 2 weeks after therapy were 100%. At 6 weeks speci®city was 93.9 and sensitivity 100%. Results by visual reading were concordant with the plate-reader in all but two cases at baseline. Conclusions: HpSA is accurate for monitoring treatment in children as early as 2 weeks after therapy, when information is most useful and unachievable with other tests. Results by visual reading are accurate, and this can make the test cheaper and more practical.
A new non-invasive commercially available ELISA test (HpSA, Meridian Diagnostics, Inc. Cincinnati, USA) measures H. pylori antigens in human stools, and is easy to perform and cheaper: in Italy the cost of one determination is approximately one half the average cost of one 13C-urea breath test. The test has already been evaluated against biopsy-based tests in children, and was found to be reliable,2, 3 but information about its accuracy after treatment is lacking in childhood, and results in adulthood are controversial.4±8 We evaluated the accuracy of HpSA to monitor the results of treatment in children with H. pylori gastritis, compared with the 13C-urea breath test. PATIENTS AND METHODS
In 60 children (26 males, median age 8.7 years, range 1±15 years) undergoing endoscopy for dyspeptic symptoms, H. pylori gastritis was diagnosed by histology 203
204
G. ODERDA et al.
(after Giemsa staining of biopsies from antrum and fundus), rapid urease test and/or culture (the baseline data of 17 of the children were reported in a previous study).2 Within a few days, before the start of treatment, a 13C-urea breath test was performed in all patients. After an overnight fast they were given 100 mL of orange juice (pH 3.6) and, after 10 min, a few millilitres of water with 50 mg of 13C-urea. Breath was collected before and 30 min after the tracer administration, and analysed by a mass spectrometer. In younger children (< 3 years), breath was collected with a dedicated mask. The 13C-urea breath test was repeated 6 weeks after the end of treatment. Both at baseline and after treatment the 13C-urea breath test was considered positive if the d over the baseline of 13CO2/12CO2 was > 5&. Baseline stools were collected by parents at home, frozen at ± 20 °C and taken to the laboratory on the same day as endoscopy. All children were treated with different schedules according to their physician prescription: treatments encompassed two antibiotics with or without a proton pump inhibitor and were given for 1 or 2 weeks. Two and 6 weeks after the end of treatment stools were collected and kept frozen at ± 20 °C. Children returned for a second 13C-urea breath test 6 weeks after the end of treatment. In a subset of 23 children, stools were also collected on the day after treatment. Stool specimens were blindly tested in one session, according to manufacturer instructions, except for the use of a disposable 10-lL loop to improve the reproducibility of stool sampling. Microwells were read with a plate-reader at 450 nm and 450/655 nm, and visually evaluated by the same technician: wells were identi®ed as positive if yellow or negative if white. Values are expressed as Optical Densities (OD) of the samples. The cut-off used was that previously established for children of the same age, of 0.182 OD at 450 nm, and 0.134 OD at 450/655 nm.2
Statistical analysis Con®dence Intervals (95% CI) were calculated by the exact method, Likelihood-Ratio by CATmaker (http:// cebm.jr2.ox.ac.uk/docs/2x2table.html). RESULTS
At baseline, both HpSA and the 13C-urea breath test were positive in 58 out of 60 children. Six weeks after treatment, the results of the 13C-urea breath test showed a d over the baseline of < 5& (median 0.87, range 0.6±3.4) in 49 (cured) children and > 5& (median 17.7, range 8.1±40) in 11 (H. pylori-positive) children. Stools of 52 children were tested for HpSA 2 weeks after treatment (45 cured-children and seven H. pyloripositive children). The results were always negative (range 0.038±0.131 OD at 450 nm, 0.005±0.090 OD at 450/655 nm) in cured-children and always positive (range 0.354±1.645 OD at 450 nm, 0.311±1.635 OD at 450/655 nm) in H. pylori-positive children. At 6 weeks, HpSA was negative (range 0.038±0.146 OD at 450 nm, 0.003±0.107 OD at 450/655 nm) in 46 out of 49 cured children and positive (range 0.265± 2.455 OD at 450 nm, 0.221±2.447 OD at 450/ 655 nm) in 11 H. pylori-positive children (see Table 1 and Figure 1). In three children with positive HpSA and negative 13C-urea breath test, both tests were repeated 1, 3 and 6 months later. This con®rmed eradication in one of them (6-week HpSA 0.243, subsequent HpSA 0.147, 13C-urea breath test: d 0.33), but not in a second child (6-week HpSA 0.892, subsequent HpSA 1.02, 13C-urea breath test: d 13.3). The results were still con¯icting in the third child, but in the opposite sense (6-week HpSA 0.377, 13C-urea breath test: d 0.82, subsequent HpSA 0.105, 13 C-urea breath test: d 12.6). The results of visual reading were always concordant with plate-reader results, except for two cases at
Table 1. Results of the HpSA test in 60 children with H. pylori gastritis, at baseline and 2 and 6 weeks after the end of eradication treatment
At baseline (n = 60) 2 weeks after the end of treatment (n = 52) 6 weeks after the end of treatment (n = 60)
Sensitivity (95% CI)
Speci®city (95% CI)
Likelihood ratio positive (95% CI)
Likelihood ratio negative (95% CI)
96.7 (88.5±99.6) 100 (59±100) 100 (71.5±100)
Ð 100 (92.1±100) 93.9 (83.1±98.7)
Ð Ð 16.33 (4.2±76.9)
Ð 0 (0±0.45) 0 (0±0.34)
Ó 2001 Blackwell Science Ltd, Aliment Pharmacol Ther 15, 203±206
H. PYLORI STOOL ANTIGENS IN CHILDREN
205
Figure 1. HpSA titres at baseline and at 2 and 6 weeks after the end of treatment.
baseline whose was OD above the cut-off, although the micro-wells looked white. In 23 children, stools were collected the day after treatment and HpSA was negative in 14 out of 18 cured children (speci®city 77.8%, 95% CI: 52.4±93.6) and positive in three out of ®ve H. pylori-positive children (sensitivity 60%, 95% CI: 15±94). DISCUSSION
There is a growing evidence that the new non-invasive test detecting H. pylori antigens in stools is reliable for diagnosing H. pylori infection and is useful in con®rming eradication after treatment, even if earlier studies gave unsatisfactory results.4±8 In our series of children from different age groups (infancy to adolescence), the results of HpSA for monitoring eradication were reasonably accurate, compared to the results of the 13C-urea breath test performed 6 weeks after treatment. We began to collect 1 stools as early as 2 weeks after treatment, when it is more likely to be still in contact with the patients and follow-up data are easier to obtain. Early after treatment patients are often still symptomatic and need to know whether eradication has been achieved. However, other tests, both invasive and non-invasive, are unreliable at this early stage, and so in symptomatic patients the decision about a second treatment has to be postponed. This is inconvenient both for patients and physicians. The high accuracy of the HpSA test performed at 2 weeks seems to overcome this problem. Ó 2001 Blackwell Science Ltd, Aliment Pharmacol Ther 15, 203±206
Additionally, we tried to test the stools collected as early as the day after treatment, but false-positive and falsenegative results were frequent, and although results are preliminary and the series too small, this ®nding suggests that the test is not reliable as early as this. According to the Consensus Conference on the management of H. pylori infection in children, we monitored treatment by the 13C-urea breath test, performing it 6 weeks after treatment because an early 13C-urea breath test can give false-negative results. We still obtained three con¯icting results, and suspect that two were false-negative, because when repeated a few weeks later the results were positive. When, after treatment, the results of a 13C-urea breath test or a biopsy-based test and HpSA are in con¯ict, we suggest they are repeated a few weeks later, to check for the possibility of false-negative results due to a low bacterial load early after treatment. The good correlation between the results of plate-reader and direct visual reading, may be helpful in making the test cheaper. The cost of the plate-reader can be saved and this may be crucial for epidemiological studies in the developing world, or its possible use as an in-of®ce test. In conclusion, the results of our study suggest that HpSA is an accurate, easy-to-perform and cheap test for monitoring the results of eradication treatment, even as early as 2 weeks after the end of treatment. ACKNOWLEDGEMENTS
Meridian provided free kits for HpSA determination on stool samples.
206
G. ODERDA et al.
REFERENCES 1 Drumm B, Koletzko S, Oderda G. & The European Paediatric Task Force on Helicobacter Pylori. Helicobacter pylori infection in children. Report of a consensus conference held in Budapest, September 1998. J Paediatr Gastroenterol Nutr 2000; 30: 207±13. 2 Oderda G, Rapa A, Ronchi B, et al. Detection of Helicobacter pylori in stool specimens by non-invasive antigen enzyme immunoassay in children: multicentre Italian study. Br Med J 2000; 320: 347±8. 3 Ni YH, Lin JT, Huang SF, Yang JC, Chang MH. Accurate diagnosis of Helicobacter pylori infection by stool antigen test and 6 other currently available tests in children. J Pediatr 2000; 136: 823±7. 4 Makristathis A, Pasching E, Schutze K, Wimmer M, Rotter ML, Hirschl AM. Detection of Helicobacter pylori in stool specimens by PCR and antigen enzyme immunoassay. J Clin Microbiol 1998; 36: 2772±4.
5 Trevisani L, Sartori S, Galvani F, et al. Evaluation of a new enzyme immunoassay for detecting Helicobacter pylori in feces: a prospective pilot study. Am J Gastroenterol 1999; 94: 1830±3. 6 Ishihara S, Kaji T, Kawamura A, et al. Diagnostic accuracy of a new non-invasive enzyme immunoassay for detecting Helicobacter pylori in stools after eradication therapy. Aliment Pharmacol Ther 2000; 14: 611±4. 7 Vaira D, Malfertheiner P, Megraud F, et al. Noninvasive antigen-based assay for assessing Helicobacter pylori eradication: a European multicenter study. Am J Gastroenterol 2000; 95: 925±9. 8 Braden B, Tauber G, Dietrich CF, Caspary WF, Lembcke B. Comparison of new faecal antigen test with C-13-urea breath test for detecting Helicobacter pylori infection and monitoring eradication treatment: prospective clinical evaluation. Br Med J 2000; 320: 148.
Ó 2001 Blackwell Science Ltd, Aliment Pharmacol Ther 15, 203±206
