Editor,—The one minute unbuffered rapid urease test,1 previously described in your journal,2was adopted for use at the Royal Melbourne Hospital endoscopy day ward because of its affordability, ease of use, and rapidity. Over time, we had noticed a number of cases where the test had been negative at the one minute mark but later became positive. As we were unsure of whether these “late” positive results represented true or false positives, we decided to run a short study to assess the accuracy of the urease test compared with the “gold standard” of histology.

To this end we read and recorded the urease test at one and 10 minutes and compared the results with histological demonstration ofHelicobacter pylori on a single antral biopsy. This was carried out on 90 unselected patients undergoing upper gastrointestinal endoscopy for varied indications. Forty one patients were found to have H pylori on histology. The urease test was positive in 20 of these 41 when read at one minute compared with 34 at 10 minutes. There were two false positive results at the one minute mark and four at the 10 minute mark. The performance of the urease test at one and 10 minutes is compared in table1.

We have demonstrated a significant disparity from published data2 in the sensitivity of the ultra rapid urease test in our ward. Previous reports have shown a difference between the test results at one minute compared with 15 minutes but this was attributed to the lower initial temperature of the test solution as it was kept refrigerated until just prior to use.3 In our ward the test solution is made up in batches and stored at 4°C in the refrigerator but the test tubes are put out at the beginning of the day and thus start off at room temperature. There is evidence to suggest that storage at 4°C for a number of days has no deleterious effects on the performance of the rapid urease test3 but this factor may explain the poor performance of the one minute test in our hands.

These factors aside, it is important to point out that we have concluded that the rapid urease test is quite accurate, with sensitivity and specificity comparable with published values4-6 for other urease tests, if the reading time is modified to 10 minutes. There are other instances3 ,4 of variability of urease test performance depending on the time interval at which it is read. It may be that, prior to use, these tests need to be validated as conditions may vary from the prescribed ones under which the test was designed.

At 10 minutes the unbuffered urease test still provides results quicker than most rapid urease tests and in fact allows us to inform patients and organise further management for them prior to discharge from the endoscopy suite. Given the overall performance of the test, we are quite happy to plan the treatment of H pylorion the basis of its results. Histology can be reserved for those cases where urease testing is equivocal or other signs, such as mucosal abnormalities, are being sought.