Prescription of proton pump inhibitors

British Medical Journal, Feb 20, 1999

Ulcer healing drugs affect gastroscopy results

EDITOR--Bashford et al report a 10-fold rise in prescriptions for proton pump inhibitors between 1991 and 1995, mainly because of increased unlicensed use for non-ulcer dyspepsia.[1] Diagnosis of non-ulcer dyspepsia will be increased by the frequent use of ulcer healing drugs before referral for gastroscopy because these drugs reduce diagnostic yield.[2]

We audited 270 dyspeptic patients referred for gastroscopy to determine the ideal time to stop ulcer healing drugs before investigation. In all, 154 (57%) patients had taken ulcer healing drugs in the 3 months before gastroscopy; 77 of these received proton pump inhibitors. When we compared patients treated with ulcer healing drugs with those who had not taken these drugs in the 3 months before gastroscopy, treated subjects had a significantly reduced occurrence of mucosal inflammation at gastroscopy--for example, oesophagitis and peptic ulcer disease. The odds ratio of not having mucosal inflammation when ulcer healing drugs were taken within 2 weeks of gastroscopy was 3.1 (95% confidence interval 1.3 to 7.1, P [is less than] 0.01). The odds ratio when these drugs were taken 2 to 4 weeks before gastroscopy was 2.0 (1.0 to 3.9, P [is less than] 0.04). Use of ulcer healing drugs more than 4 weeks before the test did not affect diagnostic yield. Patients taking ulcer healing drugs within 1 month of endoscopy were also less likely to test positive for Helicobacter pylori.

We performed a telephone survey of 31 endoscopy units in July 1997. Fourteen (45%) gave no advice about discontinuing ulcer healing drugs before gastroscopy, and only one recommended stopping a month before the procedure.

In summary, ulcer healing drugs used up to 1 month before gastroscopy reduce both diagnostic yield and detection of H pylori.[3] Many endoscopy units do not advise patients to discontinue these drugs. This must exaggerate diagnosis of non-ulcer dyspepsia and will result in repeat prescriptions for patients who might have a curable problem such as duodenal ulcer.

Ian Ross Consultant gastroenterologist ross@newark.cnht.cnhc-tr.trent.nhs.uk

June Edgehill Sister in charge, endoscopy unit Newark Hospital, Newark, Nottinghamshire NG24 4DE

[1] Bashford JNR, Norwood J, Chapman SR. Why are patients prescribed proton pump inhibitors? Retrospective analysis of link between morbidity and prescribing in the General Practice Research Database. BMJ 1998;317:452-6. (15 August.)

[2] Mitchell RMS, Tham TCK, Collins JSA, Watson RGP. Differences in diagnostic yield of upper gastrointestinal endoscopy in dyspeptic patients on proton pump inhibitors and H2 receptor antagonists. Gut (suppl 1)1997;40:A60.

[3] Dickey W, Kenny BD, McConnell JB. Effect of proton pump inhibitors on the detection of Helicobacter pylori in gastric biopsies. Aliment Pharmacol Ther 1996; 10:289-93.

Indications may be more specific than suggested by GP records

EDITOR--Bashford et al report the reasons that general practitioners prescribe proton pump inhibitors, using data derived from the General Practice Research Database.[1] They note that one of the limitations of their study is that there may be imprecision in how diagnoses are recorded and that the diagnosis might change from that which was initially entered. We report data from one centre participating in a postmarketing surveillance study of omeprazole which allow quantification of this potential source of error.

During 1993-4, 42 Oxfordshire practices agreed to participate in the study. In all, 892 patients who were current or recent users of omeprazole gave consent for their medical records to be reviewed.[2] From these records we determined the reason for prescription of omeprazole and the results of any endoscopies and barium meal investigations; 705 (79%) patients had undergone at least one of these investigations. The diagnoses were coded according to ICD-10 (international classification of diseases, 10th revision). The table shows the reasons for prescribing omeprazole and diagnoses taking into account the results of investigations.

Based on the recorded reason for omeprazole, only 464 (52%) patients were being prescribed the drug for licensed indications--a similar proportion to that observed by Bashford et al. This figure rose to 554 (62%) if results of investigations were taken into account. The principal difference is that 88 (26%) patients with the nonspecific diagnosis of indigestion were subsequently given a more precise diagnosis. However, the commonest reasons for prescribing omeprazole remain reflux oesophagitis and indigestion, regardless of whether results of investigations are taken into account. Therefore, it is unlikely that imprecision of recorded diagnoseis will have been responsible for significant bias in Bashford et al's study.

Reasons for prescription of omeprazole recorded by general practitioner and diagnosis after upper gastrointestinal imaging (barium studies or endoscopy), 1993-4. Values are numbers (percentages) of patients