EDITOR,-The datasheet for Inderal (propranolol) states specifically that the drug is contraindicated in patients with asthma or a history of bronchospasm. Despite this, and despite the well known risks of non-selective I blockers in patients with asthma, over the past few years Zeneca has received a number of reports of cases in which an asthmatic patient died as a result of being prescribed propranolol.Since propranolol was first marketed in 1965 the estimated exposure to it has been about 56 million patient years. The Medicines Control Agency has 51 reports of bronchospasm in its database of reports of adverse reactions to propranolol; 13 of the cases are recorded as having been fatal. Of more interest, however, are six reports in which it is stated that the patient had a history of asthma, bronchospasm, or wheeze; five of these cases were fatal. To help prevent further occurrences of this sort in asthmatic patients who might be prescribed propranolol erroneously, Zeneca has decided to highlight the warnings concerning asthma in the patient information leaflet for Inderal and related products. Pack labels will also carry warnings. Doctors should prescribe original packs so that a last line of defence against incorrect prescribing is not breached.Current We looked for a possible regional effect in a cohort of 7400 babies from West Sussex, who were measured between birth and 35 weeks (courtesy of Dr Ann Wallace). As in Newcastle, there was no sex difference in the standard deviation score for weight at birth, but thereafter the boys' weight centiles tended to exceed the girls' (by a mean score of 0.31, compared with 0.41 in the authors' study). Thus regional differences alone are unlikely to explain the finding, and a bias in the growth reference must exist. We believe that this bias arose during the fitting process, when several datasets were merged after adjustment for regional imbalances and secular trends. This process distorted the relation between the sexes, particularly in infancy.We have now eliminated the bias by modifying the fitting process.3 We have also added data on a nationally representative sample of 1.5-4.5 year old children to the reference dataset.4 Compared with the original reference, the revised median weight for girls is reduced by a standard deviation score of up to 0.2 (180 g) at 9 months, while that for boys is unchanged. The line indicating a standard deviation score of -2.0 for girls (close to the third centile used by Wright and colleagues) is reduced by up to 0.3 (200 g), while that for boys is reduced by 0.07 (40 g), a net difference of 0.24. This accounts for about three fifths of the sex discrepancy in Newcastle and rather more in West Sussex. We are confident that the remainder is a genuine regional difference.Wright and colleagues point out that the same imbalance between the sexes occurred with the 1966 British standards. The same is true of the 1980 Dutch standards,5 which were based on a large sample (n = 8301). This emphasises the variable nature of the sex difference in w...

No abstract

This paper provides an insight into the methods used by the international headquarters of Zeneca Pharmaceuticals for the production of a core prescribing text. Information is provided on thesources of information used for screening for adverse drug events, including the worldwide computer adverse drug experience (ADE) database safety experience collated using reported events (SECURE).The method of database interrogation by means of a formalized process-the safety and efficacy review meeting (SERM)-will be presented, together with how proposed changes are justified and conveyed to the business. The remit of SERM is to "provide coordinated and professional product reviews, provide confidence to the business, and control the frequency of prescribing information change."

SUMMARY A total of 4676 patients and 1759 patients were treated with lisinopril and nifedipine respectively in a post‐marketing surveillance study conducted in general practice in the UK. Patients were followed up for 12 months. Most of the lisinopril patients had hypertension, but a small number (180) had heart failure. Most of the nifedipine patients had uncomplicated hypertension, but some (22.57%) had other cardiovascular disease with or without hypertension. Lisinopril and nifedipine were equally effective in reducing blood pressure. During the study, 1.5% of hypertensive patients assigned to lisinopril died compared with 1.8% of patients assigned to nifedipine, and 15.1% of lisinopril patients compared with 19.7% of patients in the nifedipine group withdrew because of adverse events. Cough, malaise and fatigue, nausea and vomiting were more frequent causes of withdrawal from lisinopril than nifedipine. Conversely, headaches, pallor and flushing, oedema and palpitations caused more frequent withdrawals from nifedipine. Anaemia was more often encountered on nifedipine treatment than on lisinopril. In hypertensive patients, the frequency of first‐dose hypotension was similar on both treatments. Serious events occurred in 0.8% and 0.5% of patients given lisinopril and nifedipine respectively. Lisinopril was well tolerated by heart failure patients: 16 patients (8.88%) died and an incidence of 4.44% of serious adverse events was reported, a pattern to be anticipated in such patients; dizziness, giddiness, dyspnoea, cough, nausea and vomiting were the most frequent causes of withdrawal; the incidence of first‐dose hypotension was low (2.22%). Overall, this survey found no adverse events not already recognised in controlled clinical trials, and shows lisinopril to be an effective and well‐tolerated treatment for hypertension and heart failure.