Revised: 21 May 2013

Committees

Minutes of the 105th Medicines Adverse Reactions Committee Meeting - 14 March 2001

Preface:

The material listed as being considered on an issue is not intended to be exhaustive.

Descriptions of unpublished case reports have not been included to protect the privacy of those involved.

Names of individuals have been deleted where the contribution is not in the public domain and will not shortly be so. For example the names of those to be approached to write an article are deleted, but not usually the names of those who have contributed a draft article. Names are not usually deleted when a contribution has been made in an official capacity.

In addition, some data have been deleted for reasons of commercial sensitivity and in order not to hinder the supply of information from the same source in future.

The recommendations of the committee are in bold type face.

MINUTES OF THE 105th MEETING OF THE
MEDICINES ADVERSE REACTIONS COMMITTEE
HELD 14 MARCH 2001 IN MEDSAFE MEETING ROOMS, 18th FLOOR, GRAND PLIMMER TOWER COMMENCING AT 9.30AM

Note relevant to these minutes: All the recommendations of this meeting were accepted by the Delegate of the Minister of Health.

PRESENT

Associate Professor TJB Maling (Chairperson)
Dr M Tatley (Medical Assessor)
Dr R Savage (Medical Assessor)
Dr DM Coulter
Dr J Goldsmith
Dr H Kingston
Dr S Jessamine
Dr F McClure
Dr M Rademaker
Professor DCG Skegg
Dr N Rafter (Medical Advisor, Medsafe)
Dr K Ronaldson (Secretary)

IN ATTENDANCE

Dr J McEwen (Head, Adverse Drug Reactions Section, Therapeutic Goods Administration, Australia)
Ms S Von Afehlt (afternoon only)
Dr GR Boyd (item 9.1 Medical error only)

1. WELCOME AND APOLOGIES

Dr McEwen was welcomed to the meeting.

Apologies were received from Professor Ellis.

2. MINUTES OF THE 104^th MEETING

- Minutes of the 104^th meeting

One correction was made to the minutes ("affect" was changed to "effect" in line 6 of the first paragraph of item 12.2). The minutes were then signed by the Chairperson as a true and accurate record of the meeting.

3. DATES OF THE NEXT MEETINGS

14 June (Thu), 5 September (Wed) and 6 December (Thu) were confirmed as the dates of the next meetings.

4. MATTERS OF ADMINISTRATION

4.1 Conflict of interest

- Declaration of competing interests MARC members - guidelines
- Form for declaration of competing interests - MARC members
- Tables of members' declarations

The Secretary advised that she had made an addition to the second paragraph of the form for declaration of competing interests:

"In addition, please supply dates of other transactions (e.g. receipt of a grant, initiation of a study, etc)."

The revised process for declaration of competing interest and the response form were ratified by the Committee.

4.2 Trans-Tasman harmonisation

Dr Jessamine reported on progress towards a proposed single regulatory agency for Australia and New Zealand. He suggested holding a joint meeting between the MARC and its Australian counterpart, ADRAC. Members agreed such a meeting may be rather cumbersome. No undertaking was made, but it was understood that Medsafe would investigate the possibility of setting up such a meeting.

4.3 Minutes on the Medsafe web site

Members noted that the minutes of the September 2000 meeting had now been published on the Medsafe web site. Members asked whether there was any protocol or guidelines for preparation of the minutes in this way. The Secretary advised that there were no guidelines as such but that the preface to the minutes explained what had been deleted prior to publication. There was also discussion as to whether the Committee should review the version for publication prior to its release. Members agreed that the version of the minutes for publication on the web site should be checked by the Chairperson prior to publication.

5. MATTERS ARISING

5.1 Report on the recommendations

- K Ronaldson. Secretary's report 26 Feb 2001
- Report to the Minister's delegate on the minutes and recommendations of the 104^th meeting of MARC, 21 February 2001

5.1.1 Tracking recommendations

- Recommendations tables

The Secretary had set up a series of tables by which to track progress on committee recommendations with different tables for different types of recommendations, e.g. one for data sheets, and another for articles. Members found the tables helpful.

The table of outstanding articles was noted in particular. It was agreed that some of these should be dropped from the list. Members asked the Chairperson and the Secretary to decide together on what action should be taken on some of the older outstanding articles.

5.1.2 Cross-communication between MARC and PTAC

Members agreed that the only way there could be suitable and effective communication between the MARC and the Pharmacology and Therapeutics Advisory Committee (PTAC) was for a current member of the MARC to become a member of the PTAC. The need the MARC saw was for someone with an understanding of adverse reactions and monitoring processes in New Zealand and a knowledge of emerging safety issues to bring that perspective to the PTAC, and bring PTAC issues back to MARC for discussion. It was necessary for the person to be a full member of PTAC, so that he or she would be reading the material presented to PTAC, to pick up on issues independent of direct discussion. No member volunteered for membership of the PTAC. Members agreed that the Chairperson and Dr Jessamine should discuss nomination of a member and proceed accordingly.

5.1.3 Inhaled/intranasal corticosteroids and growth retardation

- N Rafter. Inhaled/intranasal corticosteroids - data sheet information regarding growth retardation. Feb 2001

The Committee had discussed growth retardation with inhaled/intranasal corticosteroids at the previous meeting, but had made no recommendation, other than that there should continue to be a watching brief on the issue. At that time the first two studies providing reliable evidence that growth retardation occurs had been considered by the Committee.

Dr Rafter had conducted a review of data sheets of inhaled or intranasal corticosteroids to ascertain whether they included advice that systemic effects could occur, growth retardation had been reported and monitoring for growth retardation should be conducted. Members were impressed with the quality of the data sheets of products manufactured by GlaxoSmithKline and asked that the company be advised of this with the next communication about the data sheets. However, it was noted that growth retardation and systemic effects were not mentioned in the data sheet for Flixonase (intranasal fluticasone), but they were mentioned for Flixotide (inhaled fluticasone), both of which are manufactured by GlaxoSmithKline.

Members recommended:

• publishing a Prescriber Update article on growth retardation with inhaled and intranasal corticosteroids; and
• asking sponsors of inhaled and intranasal corticosteroids to include warnings in the data sheets about systemic effects and possible growth retardation.

The Committee recommended not advising any monitoring procedure for growth retardation.

The Prescriber Update article would be formulated as the advice of the Committee. The advice would be finalised at the next meeting or by postal consultation.

5.1.4 Use of medicines in lactation

- S Gardiner, E Begg. Drug safety in lactation. Draft Prescriber Update article

The Committee considered the article prepared by Gardiner and Begg to be very useful but members wished to see other medicines covered in the table:

• cotrimoxazole and trimethoprim
• other steroids, in addition to oral prednisone, and including inhaled and intranasal
• oral contraceptives and Depo-Provera
• newer anticonvulsants

In addition, it was recommended that a sentence be included in the text about transfer of active substance from topical agents into breast milk.

5.1.5 NSAIAs and intestinal disorders - R Savage. Nonsteroidal anti-inflammatory agents cause various disorders in the large and small intestine. Draft Prescriber Update article

Members noted the draft article on NSAIAs and intestinal disorders. The article had portions highlighted which were intended to be deleted in the short versions. However, members were advised that Medsafe had decided that in general there would be only one version of each article and that it would be a succinct presentation of the issue. Only in exceptional circumstances would a longer version be published on the web site.

With respect to the present article members were advised that it had undergone extensive editing since it had been put in the agenda papers. Hence the Committee recommended no changes.

5.1.6 Updates to the data sheet for nefazodone

- Correspondence with Bristol-Myers Squibb with data on withdrawal effects
- Serzone data sheet, updated with changes marked

The updates to the Serzone (nefazodone) data sheet in respect of hepatic reactions, and withdrawal and anticholinergic effects were noted.

5.1.7 Hyperprolactinaemia with antipsychotics

- N Rafter. Hyperprolactinaemia with antipsychotics. Draft Prescriber Update article

Members noted the article had been peer reviewed and was ready to be published.

5.1.8 Consumer representative on the Committee

The Secretary advised the Committee that she had been advised that the Ministry of Consumer Affairs and the Ministry of Women's Affairs are able to supply lists of possible candidates for consumer positions if they are approached with person-specifications and an indication of the role they are expected to fill. Members suggested that it may be necessary to advertise the position of consumer representative on the Committee in order to be seen to be proceeding appropriately.

5.1.9 Tetracycline and benign intracranial hypertension

- H Kingston. Tetracycline and benign intracranial hypertension - a headache rare but real. Published web site Jan 2001

Members noted that the article had now been published.

5.1.10 Sildenafil and angina

- Information for Viagra data sheet
- N Rafter. Viagra reminder - Ask! Don't be shy! Draft Prescriber Update article

Members had no additional comments to make on the information for the Viagra (sildenafil) data sheet or the draft article.

5.1.11 Case reports for alternative therapies

- Table of case reports for alternative therapies

Members agreed with the manner in which the alternative therapies had been broken down into further categories with terms such as bee product, dietary supplement and herbal remedy.

5.1.12 Risk factors project for venous thromboembolism with oral contraceptives

The previous meeting had recommended asking ... whether she would be interested in updating her report on risk factors for venous thromboembolism (VTE) with oral contraceptives. She had agreed to do so, but the update would not be completed before 17 April 2001. The new version would include information on the relative risk of VTE with second and third generation oral contraceptives, economy class syndrome and new information on congenital and acquired haemophilia.

The interested professional bodies had been advised that an update was being prepared. The New Zealand Medical Association had advised that it had been in agreement with the first report, but would like to see the update before assenting to being involved in its dissemination.

5.2 Perioperative use of OCs and HRT

- K Ronaldson. Perioperative use of OCs and HRT. 22 Feb 2001
- Drugs in the peri-operative period. 3. Hormonal contraceptives and hormone replacement therapy. Drug and Therapeutics Bulletin 37: 78-80, Oct 1999
- D Grady, et al, Postmenopausal hormone therapy increases risk for venous thromboembolic disease. The Heart and Estrogen/progestogen Replacement Study. Ann Int Med 132:689-96, 2 May 2000
- Extract from draft Summary Guideline for The Appropriate Prescribing of Hormone Replacement Therapy. Intended to update Hormone Replacement Therapy: A Consensus Development Conference Report. Nov 1993
- . Stopping combined oral contraceptives/HRT prior to major orthopaedic hip and knee surgery to minimise the risk of DVT. Letter 17 Dec 2000

The previous meeting had recommended publishing the article from Drug and Therapeutics Bulletin on hormonal contraceptives and hormone replacement therapy (HRT) in the perioperative period in Prescriber Update. However, Medsafe had considered it did not fit with the new image of Prescriber Update. An attempt had then been made to redraft guidelines, but the task and the issues around it were found to be more complex than was initially thought.

The article in Drug and Therapeutics Bulletin recommends stopping use of combined oral contraceptives from 4 weeks before elective surgery, but it gives no advice about when they can safely be recommenced. Advice about to be promulgated by the New Zealand Guidelines Group on the appropriate prescribing of HRT recommends stopping HRT 4 weeks prior to elective surgery and restarting it 90 days after surgery.

This 90-day post-operative period is based on the results of the HERS study (Grady et al) which was a randomised (1:1) placebo-controlled study of HRT in menopausal women with coronary heart disease. The study found an increased relative risk of VTE in the 90-day post-operative period of 4.9 (2.4-9.8) on the basis of 944 women undergoing 1454 surgical events. The study had not conducted a breakdown of risk for different post-operative periods between 1 and 90 days.

Medsafe had asked the Committee for advice on how to proceed in the face of the paucity of data on which to base best practice, and the complexity of the issue.

The Committee recommended:

• summarising the guidelines from the Guidelines Group concerning use of HRT in the perioperative period in a box in Prescriber Update, and including mention of the risk of VTE with oral contraceptives;
• communicating with the Royal Australasian College of Surgeons, Australian and New Zealand College of Anaesthetists and the Royal New Zealand College of General Practitioners concerning this issue; and
• sending copies of the 2 articles and the draft guidelines to ... as references for her report.
  

5.3 Non-sedating antihistamines and sedation warning

- Sedation with non-sedating antihistamines. Item 12.7 minutes of Dec 2000 meeting of MARC
- RD Mann, et al. Sedation with "non-sedating" antihistamines: four prescription-event monitoring studies in general practice. BMJ 320:1184-7, 29 Apr 2000
- JG Ramaekers, A Vermeeren. All antihistamines cross blood-brain barrier. BMJ 321:572, 2 Sep 2000
- JF O'Hanlon, JG Ramaekers. Antihistamine effects on actual driving performance in a standard test: a summary of Dutch experience, 1989-94. Allergy 50:234-42, Mar 1995
- Labelling for Telfast, Claratyne, Zyrtec

This subject had been on the agenda of the December meeting, but had not been discussed. A postal consultation had indicated that it was important that the material be discussed at a meeting. Hence it had been brought to this meeting. The material was presented in the minutes of the December meeting.

An addition to the material on the agenda at the December meeting was the study by O'Hanlon and Ramaekers on the effect of different antihistamines on driving performance. The second generation antihistamines were found to be less impairing than the earlier antihistamines, but all were found to be impairing at 1-2 times the recommended dose. The authors of this study had concluded:

Finally, warnings about antihistamines' possible adverse effects on driving and other potentially dangerous activities should not be waived for the second-generation drugs. It is unlikely that the majority of patients taking recommended doses of any of these drugs will experience untoward reactions affecting their driving safety, but if any fraction will, all patients should receive an appropriate warning.

Using prescription event monitoring, Mann et al had studied 4 "nonsedating" antihistamines and of these found cetirizine and acrivastine to be associated with a greater frequency (8.5 and 7.9 events per 1000, respectively) of reports of sedative effects than fexofenadine and loratadine (3.1 and 2.6 per 1000, respectively). However, the differences in sedative events did not translate into a higher rate of accident or injury.

At the meeting in March 1995, the MARC had considered this issue and had recommended that the following statement be included in the labelling for "nonsedating" antihistamines:

Although the medicine is unlikely to affect your ability to drive or operate machinery, a few people may be impaired and care should be taken.

This statement is in line with Regulation 22 which requires a warning about driving and operating machinery on the labelling of all antihistamines. Members considered it was appropriate to continue to require such a statement.

The use of the statement "non-sedating" or "non-drowsy" on the packages for the second generation antihistamines was discussed. The use of such statements had been adopted to contrast with the earlier sedating antihistamines. Members agreed to allow continued use of these statements.

The Committee recommended that the statement formulated and recommended for use at the meeting in March 1995, or words of similar meaning, be used in the labelling (carton or package insert) for all of the second generation antihistamines. The Committee agreed that the second-generation products should be allowed to continue claiming to be non-sedating, non-drowsy or similar. Members recommended having no differentiation in labelling between the different second-generation antihistamines despite evidence that some may be a little more sedating than others.

6. General Reporting

- Case reports considered since Nov 1997, by medicine class
- Case reports considered since Nov 1997, by recommendation
- Glossary of terms

All spontaneous reports presented to the MARC meeting have been assessed by CARM and replies have been sent to the reporters. The purpose of these responses is to assist the practitioner to discharge his/her responsibility to patients. These individual replies include as appropriate:

• comment about causality;
• information about similar suspected adverse reactions reported with the same or related medicines;
• prescribing advice;
• advice related to the care of the patient, including information that may assist the practitioner to make a risk/benefit assessment for future treatment; and
• any specific action being taken by the Centre, including entry of the reaction into the National Health Index against the patient's name, presenting the reaction to the MARC, including it in an article to be published, etc.

6.1 Deaths

Note: These deaths were discussed under the therapeutic group of the medicine involved. Each of the deaths has an associated report number which is also a hyperlink to the comment on that case. To access the comment click on the hyperlink.

Diethylpropion hydrochloride / medroxyprogesterone and convulsions, hypoglycaemia, pancreatic neoplasm, 46043
Klean-Prep and sudden death, 46125
Metformin and acidosis lactic, renal failure chronic aggravated, sepsis, neutropenia, disseminated intravascular coagulation, 46309
Tetracycline / diuretic and renal failure NOS, melaena, haematemesis, 46387
Celecoxib and melaena, disseminated intravascular coagulation, circulatory shock, infection, 46014
Celecoxib / metoprolol and pancreatitis, 46196
Amiodarone and interstitial pneumonitis, 46148
Azathioprine and myelodysplastic syndrome, septicaemia, 46044
Clozapine and myocardial infarction, 46073
Bambuterol and headache, dizziness, palpitations, myocardial infarction, 46383
Nefazodone and hypoglycaemic coma, 45875

Note: In the comment associated with each report the case has been given a causality designation using terms and definitions developed by the WHO. The precise definitions are available on the web site of the WHO Collaborating Centre http:www.who-umc.org These designations, certain, probable, possible, unlikely, unclassified and unclassifiable, refer to the degree of certainty about the relationship between the medicine and the adverse event. The terms should not be understood literally. So "certain" means the appropriate elements are present to match the international definition. It does not mean there is absolute certainty that the medicine caused the adverse event.

Explanations of the terms used by CARM and MARC can be accessed by hyperlink at each causality designation, certain, probable, possible, unlikely, unclassified and unclassifiable.

6.2 Alimentary

6.2.1 Diethylpropion hydrochloride / medroxyprogesterone and convulsions, hypoglycaemia, pancreatic neoplasm (death), 46043

Comment: Seizure is listed as an adverse reaction of diethylpropion in the Tenuate dospan data sheet, and it is recorded that it may reduce insulin requirements. Cases of both convulsions and hypoglycaemia are recorded in the WHO database for diethylpropion (9,1) and medroxyprogesterone (97,10). It was likely that the patient's pancreatic hyperplasia had caused the hypoglycaemia. Because there was sufficient cause without invoking the medicine, for each event the causal association was designated "unlikely" and the death was recorded as unrelated to the medicine. The Committee recommended no further action.

6.2.2 Klean-Prep and sudden death (death), 46125

Comment: Dr Jessamine reported that residue of the solution taken by the patient had shown the presence of some anomalies, but product taken off the shelf had been analysed at the request of Medsafe and the results were within the specifications of the product. Members commented that without further facts, it was impossible to know what the causative factors were in this case. Taking insufficient liquid, for example, can have serious consequences with a product like this. The patient instructions with the pack were considered by Medsafe to be adequate. A previous report associated with this product in the CARM database was for hypokalaemia and bradycardia in a child. The association was designated "possible" in the present case. No additional action was recommended.

6.2.3 Metformin and lactic acidosis, chronic renal failure aggravated, sepsis, neutropenia, disseminated intravascular coagulation (death), 46309

Comment: The patient's creatinine in 1991 was above the maximum level stated in the data sheets for Metomin (0.16mmol/L) and Apo-Metformin (0.136mmol/L). Above these concentrations metformin is contraindicated. CARM had received 12 previous reports of lactic acidosis with metformin. Although other factors may have contributed, the causal association for lactic acidosis and worsened chronic renal failure were designated "possible". However, it was considered that the other events were unlikely to be associated. Because the role of the patient's lactic acidosis in the fatal outcome was "uncertain", the cause of death was designated "medicine may be contributory." It was noted that Prescriber Update articles were published on lactic acidosis with metformin in March 1994 and April 1998. No action was recommended at this time.

6.3 Antiepileptics

6.3.1 Lamotrigine and toxic epidermal necrolysis, 46294

Comment: The data sheet for Lamictal (lamotrigine) and the entry in New Ethicals Catalogue specify a starting dose of 25mg on alternate days for the first 2 weeks in patients taking sodium valproate to reduce the risk of cutaneous reactions. In this case it appeared that the recommendations for initiating therapy had not been followed. However, members observed that, because toxic epidermal necrolysis is an immune-mediated reaction, commencing lamotrigine at a reduced dose would not have lowered the risk. The CARM database holds 12 previous reports of cutaneous reactions with lamotrigine, including one case of toxic epidermal necrolysis and one of Stevens Johnson syndrome. The causal relationship in the present case was designated "possible". No further action was recommended.

6.4 Antiinfectives

6.4.1 Minocycline and systemic lupus erythematosus, hepatic function abnormal, arthritis, 46298

Comment: It was noted that arthritis and abnormal hepatic function are part of the syndrome of SLE. The CARM database holds no previous reports of SLE with minocycline, but arthralgia or arthritis (9 reports - 4 with positive antinuclear antibodies) and abnormal hepatic function (12) have been reported. The WHO database holds 158 reports of SLE with minocycline, and this adverse reaction was widely reported in the literature,^ and is recorded in the data sheet for minomycin. Members commented that SLE usually occurs after about 50 weeks of therapy and resolves more quickly than in this case. However, the antibody serology is consistent with minocycline-induced disease. The causal association was designated "possible" for each of the events. The Committee recommended no additional action.

6.4.2 Terbinafine and acute generalised exanthematous pustulosis, fever, hepatic enzymes increased, 46353

Comment: The CARM database holds 32 previous reports of skin reactions including 3 cases of erythema multiforme and 5 of maculopapular rash with terbinafine. 36 cases of pustular rashes have been recorded in the WHO database. The present case with its combination of features appears to be a hypersensitivity reaction. Because of the temporal association with the initiation of the medicine and the rarity of the type of rash, this adverse reaction was designated "probable" for association. No further action was recommended.

6.4.3 Tetracycline / diuretic and renal failure NOS, melaena, haematemesis (death), 46387

Comment: Members observed that tetracycline can cause renal impairment and this may well have been exacerbated by the diuretic in this patient. The CARM database holds 5 reports of renal failure with tetracycline. The causal association was designated "possible" for each of the events. No additional action was recommended by the Committee.

6.4.4 Nitrofurantoin and pulmonary fibrosis, 46443

Comment: The CARM has received 19 previous reports of various types of pulmonary infiltration with nitrofurantoin. The association was considered to be "probable" in the present case. Members noted that the meeting in September 2000 had recommended publishing an article about nitrofurantoin and interstitial pneumonitis. The article was in progress. No further action was recommended.

6.5 Analgesics

6.5.1 Celecoxib and melaena, disseminated intravascular coagulation, circulatory shock, infection (death), 46014

Comment: Celecoxib is not thought to be associated with gastrointestinal ulceration to the same degree as the non-selective NSAIAs. Members observed, however, that very few patients in clinical studies have been aged > 80 years. CARM has received 7 reports of haematemesis or melaena out of a total of 70 reports with celecoxib. It appears that celecoxib is being preferentially prescribed to patients at risk of these events. In this case the adverse events, except infection were rated "possible".

6.5.2 Celecoxib / metoprolol and pancreatitis (death), 46196

Comment: There are no reports of pancreatitis in the CARM database with either celecoxib or metoprolol, but one such case is recorded in the WHO database for celecoxib. Dr McEwen advised that one such case has recently been reported in Australia. Pancreatitis is listed in the Celebrex (celecoxib) data sheet as a serious adverse reaction which has occurred rarely (0.1%) in patients taking Celebrex. No claim about causal association is made. It was noted that pancreatitis may prove to be a rare adverse reaction of celecoxib. The causality designation in the present case was "possible".

6.5.3 Comment

A previous meeting had recommended publishing an article on the adverse effects of celecoxib and rofecoxib. The article was in progress. In addition both celecoxib and rofecoxib are now being monitored in the IMMP. No additional recommendations were made.

6.6 Cardiovascular

6.6.1 Amiodarone and interstitial pneumonitis (death), 46148

Comment: The CARM database holds 34 previous reports of respiratory adverse reactions including 3 of interstitial pneumonitis, 7 of pneumonitis and 7 of pulmonary fibrosis. The causality designation of the present case was rated "possible". The data sheet for Cordarone X advises conducting regular chest X-ray in patients taking amiodarone, and the data sheet for the generic, Aratac recommends checking lung function and conducting chest X-ray should dyspnoea develop without signs of heart failure. In the present case lung function had not been monitored and the identification of the cause of early symptoms may have been confounded by the fact that the patient was elderly and a heavy smoker. It was noted that it is intended to include mention of other medicines that may cause interstitial pneumonitis in the Prescriber Update article on this adverse reaction with nitrofurantoin. Members recommended that amiodarone and methotrexate be included in the article as other causes of interstitial pneumonitis.

6.7 Hormones

6.7.1 Trisequens and deep vein thrombosis, 46063

Comment: This is the first case of venous thrombosis with Trisequens but these kinds of events have been reported with other hormone replacement therapies. The association was designated "possible" in the present case. Members questioned how long the woman had been thought to have asthma, and wondered whether symptoms thought to be associated with asthma may have been caused by pulmonary emboli. The Committee suggested that CARM ask the reporter about the date of onset of asthma.

6.7.2 Depo-Provera and bronchospasm, angioedema, fever, hypertension, tachycardia, paraesthesia, tremor, 46137

Comment: The data sheet for Depo-Provera records that cases of anaphylaxis and anaphylactoid reactions have been reported with Depo-Provera. The association for each of the events occurring in this case, except hypertension, was designated "probable" for Depo-Provera. The hypertension was a peculiar feature and may have related to the woman being 24 hours post-partum, but because of the temporal association it may have had a causal relationship to salbutamol (rated "possible" for salbutamol). No additional action was recommended.

6.7.3 Femodene and pulmonary embolism, 46155

Comment: CARM has received 10 previous reports of pulmonary embolism with Femodene. In the present case the causal association was designated "possible". Many alerts have been sent out by Medsafe on the risk of VTE with oral contraceptives (most recent June 2000). Other than the risk factors project (see 5.1.12) the Committee recommended no additional action.

6.7.4 Postinor-2 and rash, oedema, bronchospasm, 46201

Comment: Similar hypersensitivity effects are not listed in the data sheet for Postinor-2, and similar events have not been reported to CARM with either this product or with progestogen-only contraceptives with levonorgestrel as the active ingredient. The causality of each of the events was designated "probable".

The Committee suggested that CARM:

• write to the reporter to advise that oral contraceptives containing levonorgestrel should be avoided by this woman in the future; and
• ensure that the appropriate medical danger alert was entered against her name in the national patient database.

6.7.5 Diane 35 and pulmonary embolism, 46343

Comment: CARM has received 12 previous reports of VTE with Diane-35 including 8 reports of pulmonary embolism. The causal relationship was designated "possible" in this case. VTE with Diane-35 is listed as an adverse reaction of current concern. No further action was recommended.

6.8 Immunosuppressives

6.8.1 Azathioprine and myelodysplastic syndrome, septicaemia (death), 46044

Comment: The WHO database holds 8 reports of myeloproliferative disorder and 28 of sepsis with azathioprine. The causal association for this case was designated "possible". The postmortem report and further information on the timing of therapy dates and reaction onset had been requested. The Committee recommended no additional action.

6.9 Psychiatric

6.9.1 Bupropion and myocardial infarction, 46381

Comment: Cardiovascular effects with bupropion for smoking cessation have been reported elsewhere (see 12.7), but a causal relationship has not been established. There are 38 cases of myocardial infarction in the WHO database. Those taking bupropion for smoking cessation are likely to have a greater pre-treatment risk of myocardial infarction than other members of the population of a similar age. Because it is not known whether a causal association may exist between bupropion and myocardial infarction, the causal association in this case was designated "possible". The Committee recommended no additional action to the watching brief placed on adverse reactions to this medicine (see 12.7).

6.9.2 Clozapine and myocardial infarction (death), 46073

Comment: The WHO database holds 269 reports of myocardial infarction and 250 of cardiac arrest with clozapine, although neither is known to be causally associated with the medicine. Because of the scarcity of information, the present case was designated "unclassified". No further action was recommended.

6.9.3 Fluoxetine / St Johns wort and violent thoughts, judgement impaired, 46334

Comment: No similar events have been reported previously in New Zealand with St Johns wort, but 7 cases of aggressive reaction and 5 of impaired judgement have been reported with fluoxetine. Impairment of thought processes is listed as an adverse reaction in the data sheet for Prozac. Because of the close temporal relationship, the present case was assigned the designation, probable. It is also "possible" that there may have been an interaction between St Johns wort and fluoxetine, although no symptoms of serotonin syndrome were reported. The Committee recommended no further action.

6.10 Respiratory

6.10.1 Bambuterol and headache, dizziness, palpitations, myocardial infarction (death), 46383

Comment: The WHO database holds 21 reports of headache, 6 of dizziness and 18 of palpitations with bambuterol. These events were designated "certain," for causal relationship, because of the positive rechallenge. The myocardial infarction was considered "unlikely" to be associated. No further action was recommended.

6.11 Alternative medicines

6.11.1 Hayfever & Allergy Formula / cilazapril and dizziness, palpitations, dyspnoea, malaise, drug interaction, 46045

Comment: It was considered that an interaction was a possibility but no information (e.g. independent data consistent with an interaction occurring) was available to the Committee to permit such a designation to be made with any confidence. The causal association was rated "possible". No additional action was recommended.

6.12 CARM Quarterly report

- CARM quarterly report for Oct to Dec 2000

7. Vaccines

7.1 New Zealand adverse reaction reports

- Vaccine adverse reactions since Nov 97

No new cases of adverse reactions to vaccines were presented.

7.2 Articles from the literature

- K Ronaldson. Vaccine articles. Report for MARC 1 Mar 2001

7.2.1 MMR vaccine and autism

- JA Kaye, M Del Mar Melaro-Montes, H Jick. Measles, mumps and rubella vaccine and the incidence of autism recorded by general practitioners: a time trend analysis. BMJ 322:460-3, 24 Feb 2001
- Medsafe editorial team. MMR vaccine not associated with bowel disease & autism. Prescriber Update No.19, p37-9, Feb 2000

In January 1998 Wakefield et al² published a case series claiming an association between MMR vaccine and inflammatory bowel disease and autism. Evidence has subsequently accumulated³ which is not consistent with there being a causal association between MMR vaccine and autism. The evidence has been summarised in 2 Prescriber Update articles (April 1998 and February 2000).

Hershel Jick's group have conducted another study of the association between MMR vaccine and autism. They have used the UK General Practice Research Database (GPRD) and examined time trends for MMR vaccination and diagnosis of autism. The GPRD has been found by Jick et al (unpublished data) to be almost complete for the recording of vaccinations.

The study identified 305 children aged ≤ 12 years with a diagnosis of autism first recorded between 1988 and 1999. During this period the annual incidence of diagnosed autism increased 7-fold from 0.3 per 10,000 person-years in 1988 to 2.1 per 10,000 person-years in 1999.

The further analysis was restricted to 114 boys born 1988-93 with a diagnosis of autism at age 2-5 years. The 4-year prevalence of diagnosed autism rose from 8 per 10,000 (95% CI 4-14) for the 1988 birth cohort to 29 (20-43) per 10,000 for boys born in 1993. For children registered on the GPRD the uptake of MMR vaccine was virtually constant (about 97%) for each successive year. Among the 110 boys with autism born in 1988-93 for whom MMR vaccination could be assessed, the age distribution at first vaccination was almost identical to that of the general population and the uptake of vaccination was 99% (109), close to the population uptake.

Hence, no time correlation was found between the incidence of diagnosis of autism and the prevalence of MMR vaccination. The exposure then is not an explanation for the rise in the incidence of diagnosis of autism.

The Committee recommended not publishing a further Prescriber Update article about MMR vaccine and autism.

The Committee asked that material about MMR vaccine and idiopathic thrombocytopenic purpura be included on the agenda of the June meeting.

7.2.2 Hepatitis B vaccine and arthritic reactions

- MR Geier, DA Geier. Arthritic reactions following hepatitis B vaccination: an analysis of the Vaccine Events Reporting System (VAERS) data from 1990 through 1997. Clinical and Experimental Rheumatology 18:789-90, Nov-Dec 2000

Geier and Geier used the US Vaccine Adverse Events Reporting System (VAERS) to investigate the occurrence of arthritic reactions following hepatitis B vaccination. They found more of each of arthralgia, arthritis and arthrosis being reported with hepatitis B vaccine than with any other vaccine (total for hepatitis B vaccine 1,941, compared with 784 for MMR or rubella vaccines, where this is a recognised adverse reaction of rubella vaccine). Incidence data were not available. The female/male ratio was 3.5, most of the arthritic reactions occurred in persons in their thirties and the mean time to onset was within 2 days. No data were presented on severity or duration of symptoms.

Arthralgia, myalgia (rare) and arthritis (very rare) are listed as adverse reactions in the Engerix-B data sheet. The Committee did not consider that any further action was necessary.

7.2.3 Influence of vaccination on mortality in developing countries

- F Shann. Non-specific effects of vaccines in developing countries. BMJ 321:1423-24, Dec 2000

Shann has commented on the results of a study⁴ of mortality following vaccination in Guinea-Bissau. Both BCG and measles vaccines were found to halve child mortality but DTP and polio vaccines seemed to increase mortality (mortality ratio 1.84 95% CI 1.10-3.10). Most of the unvaccinated group eventually were vaccinated, so that the difference in mortality is an effect of delaying vaccination. Further the vaccinated and unvaccinated groups were both in the same area where vaccination was the norm, and there was a degree of herd immunity. Shann postulated that the reduction in mortality with BCG and measles vaccination was greater than the number of deaths from tuberculosis or measles and conversely vaccination was resulting in a decrease in deaths from diphtheria, pertussis, tetanus and polio, but these gains were offset by increased mortality from other causes.

No action was considered necessary by the Committee.

8. IMMP

8.1 Rofecoxib

8.1.1 Rofecoxib and paroxysmal atrial fibrillation, 46461

- R Pierson. US regulators likely to scrutinize Vioxx adverse events. Reuters medical news. 5 May 2000

Comment: The WHO database holds 129 reports of cardiac failure with rofecoxib and 99 with celecoxib. These represent 1.4% and 0.5% of reports, respectively, for these medicines. A study (reported in Pierson) had found a significantly higher rate of cardiovascular events with rofecoxib than with naproxen.

The Committee recommended asking the sponsor company to comment on the cardiovascular events with rofecoxib (Vioxx).

8.2 Celecoxib

8.2.1 Celecoxib and tachycardia, interaction with amitriptyline, 46655

Comment: 67 reports of tachycardia with celecoxib are recorded in the WHO database. In addition, four reports of interaction between celecoxib and amitriptyline have been received. Tachycardia and palpitation were each reported in two cases. Celecoxib inhibits CYP 2D6 which metabolises amitriptyline. The causal association was rated "possible".

8.2.2 Celecoxib and cerebral events, 46395

Comment: Coma (16), encephalopathy (8), stupor (15) and ataxia (27) have been recorded in the WHO database with celecoxib, and CARM has 3 previous reports of similar events. The Committee rated the causal association "possible" in the present case.

8.3

Nefazodone

8.3.1 Nefazodone and hypoglycaemic coma (death), 45875

- JK Warnock, F Briggs. Nefazodone-induced hypoglycemia in a diabetic patient with major depression. Am J Psychiatry 154:288-9, Feb 1997 (Micromedex abstract)

Comment: Members observed that, because of the patient's underlying poorly controlled diabetes this event may not be medicine-related. Alternatively, it may be an interaction between nefazodone and glipizide. Nefazodone by itself has the potential to lower blood sugar, and there are 6 reports of this in the WHO database, and a case published by Warnock and Briggs.

8.4 Olanzapine

8.4.1 Summary of the monitoring of olanzapine

- Olanzapine monitoring summary

The monitoring of olanzapine currently covers a period of 1304 patient-years with a total of 2456 patients. To date the relative risk for incidents for women versus men is 2.9 which appears to be an anomaly, possibly related to low patient numbers. However, it is also possible that some of these "incidents" are unrecognised adverse reactions or reflect some unrecognised confounder.

8.4.2 Olanzapine and weight gain

- T Wetterling. Bodyweight gain with atypical antipsychotics. A comparative review. Drug Safety 24(1):59-73, 2001

The most common adverse reaction identified in the IMMP monitoring has been weight gain (10 patients; 1.9%). One patient experienced a weight gain of 15kg. Wetterling conducted a review of studies of atypical antipsychotics to compare rates of weight gain. He found olanzapine to be associated with a weight gain of 2.3 kg/month, clozapine with 1.7 kg/month and risperidone with 1.0 kg/month. He found bodyweight gain to be particularly marked in the first 12 weeks of therapy.

8.4.3 Olanzapine and neuroleptic malignant syndrome, 44011

8.4.4 Olanzapine and Monofeme and deep vein thrombosis, 46332

Comment: Because of the possible association of VTE with clozapine, and hence with olanzapine, and the known association with oral contraceptives, this adverse event was listed as an interaction. The WHO database holds 162 reports of pulmonary embolism with clozapine and 19 with olanzapine. Deep thrombophlebitis has been reported 52 and 14 times, respectively.

8.4.5 Olanzapine and benign intracranial hypertension, 46364

Comment: The WHO holds 2 reports of BIH with olanzapine, and 2 of papilloedema. There are a total of 9 reports in the WHO database of BIH with the atypical antipsychotics. In the absence of further data on whether a causal association exists for this adverse reaction, the causality of this case was considered to be "possible".

8.4.6 Recommendation

The Committee suggested that psychiatrists and mental health teams be advised of these early results of the IMMP monitoring of olanzapine.

8.5 Mirena

8.5.1 Mirena and benign intracranial hypertension, 45192

- JB Alder, et al. Levonorgestrel implants and intracranial hypertension. NEJM 332:1720-1, 22 Jun 1995
- AJ Sunku, et al. Benign intracranial hypertension associated with levonorgestrel implants. Program and Abstracts American Neurological Soc p.299, 19 Oct xxxx

Comment: The WHO database holds 85 reports of intracranial hypertension and 45 of papilloedema with levonorgestrel. There is no reference to intracranial hypertension in the data sheet. Alder et al and Sunku et al have each reported 2 cases of BIH with Norplant, an implanted device (not intrauterine) which also releases levonorgestrel. It appears that there may be a causal association between Mirena and BIH. The designation of this case was considered to be "possible".

The Committee recommended asking the sponsor company to comment on this possible association.

9. Other NZ activities

9.1 Medical error

- LT Kohn, JM Corrigan, MS Donaldson (Ed). To err is human. Building a safer health system. Institute of Medicine, National Academy Press, Washington, DC. 2000:
- Those involved in preparation
- Preface
- Foreword
- Contents
- Executive summary
- Chapter 7, Setting performance standards and expectations for patient safety.
- Chapter 8, Creating safety systems in health care organizations
- Appendix C, Literature summary
US Government response:
- Doing what counts for patient safety: federal actions to reduce medical errors and their impact. Report of the Quality Interagency Coordination Task Force (QuIC), To The President, Feb 2000:
- Members of the Task Force and acknowledgements
- Contents
- Executive summary and actions
- Safe use of drugs and devices
- Implementing safety systems in health care organizations
- Using decision-support systems and information technologies
- Using standardized procedures, checklists, and the results of human factors research
- D A Peitro, et al. Detecting and reporting medical errors: why the dilemma? BMJ 320:794-796, Mar 2000
- P Barach, and S D Small. Reporting and preventing medical mishaps: lessons from non-medical near miss reporting systems. BMJ 320:759-763, Mar 2000
- S N Weingart, et al. Epidemiology of medical error. BMJ 320:774-777, Mar 2000

The New Zealand study on medical error by Dr Peter Davis had been reported in the media the previous day. The study results were similar to those found in American and Australian studies. 16.6% of hospital patients suffered a treatment-related adverse event, of which half were preventable. Of all the adverse events identified in the study, 15% were medication-related.

Dr Boyd advised that the Sentinel Events Project being initiated by the Ministry of Health is focussed on reducing the frequency of adverse events particularly in hospital settings. The intention is to draw up guidelines which define reportable events. It will then be possible to conduct root cause analyses of these events, and modify procedures and systems as a result so that the likelihood of reoccurrence is reduced. Dr Boyd advised that the District Health Boards were taking initiatives to put safety controls in place to reduce the occurrence of adverse events.

The MARC could contribute to discussions on reducing medicine-related error. The material on the impact of warnings about the use of cisapride in the US (item 10.3) indicates how little effect dear doctor letters and data sheet changes may have. The MARC needs to review the means of communication it uses in order to reduce medicine-related error occurring following the issuing of safety alerts.

Medication error could be reduced by putting alerts in software used by prescribers to avoid the use of medicines in situations in which they are contraindicated. Data-linkage (e.g. to ensure all conditions and medication of a patient are in the system) would be essential to make such a system fully functional and effective.

Standards New Zealand has a committee set up to reduce medical error in New Zealand. Members considered that the CARM should be represented on the committee. It was agreed that CARM should not initiate, but should be involved in determining and setting up procedures and systems to reduce medication error.

The Committee asked that Dr Boyd take to the Sentinel Events Project the comments of MARC regarding the importance of CARM involvement in any discussion of means to reduce medication error.

9.2 ACC medical misadventure levy

- ACC report, 10 Oct 2000

The ACC report was prepared for the board of ACC. It discussed the proposal to initiate a levy on healthcare providers for medical misadventure (medical error and mishap). At present no decision had been reached regarding the levy, but the necessary legislation was in place. The MARC had been concerned at news of the levy because it was thought that it may cause a climate of greater defensiveness and reduce willingness to report adverse reactions to medicines.

The report was partly inspired by a desire to reduce the incidence of medical misadventure claims being lodged with the ACC, but it was realised that because the account was not yet mature the rate would continue to increase and the very nature of medical progress meant the claims would increase with time because increasingly risky procedures were being attempted.

An issue the report discussed was whether all healthcare providers should be levied at the same rate, or whether those working in more at-risk areas should be levied according to the level of risk (risk rating). It was thought that initiating a differential levy for medical practitioners may reduce willingness to practise in those areas associated with greater risk, resulting in a reduced availability of those services. However, it was considered likely that the cost of premiums would be passed on to employers and/or consumers of the services. Hospitals would then need to find an additional $21,000-$26,000 per specialist per year.

Another possibility was to levy individual providers according to the cost of medical misadventure claims levied against that individual (experience rating). Such an approach could make it financially difficult for those with high cost claims to continue to practise. On philosophical grounds and in a different context, the Government has opposed experience rating of employers because it penalises with no evidence of fault. The Government prefers to reward for improving safety rather than penalising after accidents. To date only around 15% of medical misadventure claims have been designated "medical error" by ACC.

The document also discussed using financial incentives to encourage employers of healthcare providers to put in place systems to increase the level of safe practices.

It was noted that introducing a levy would increase the cost of delivering healthcare. This cost would be passed on inevitably to the taxpayer. It would also increase the cost of medical misadventure claims to ACC.

Members foresaw many difficulties with the proposed levy, but there was a potential problem that was central to the activities of the Committee. It was anticipated that the levy may create a defensive attitude among medical practitioners. They would then be reluctant to report adverse reactions, some of which may have been caused by medication error, and may highlight a wide-spread ignorance that needs to be addressed, and some of which may contribute to knowledge about the adverse reactions profile of the medicine. Introduction of the levy has the potential to seriously undermine the work of both CARM and the Committee.

The Committee recommended preparing a comment around the concern that introducing such a levy may inhibit reporting of adverse reactions to CARM, because of the fear of blame being attributed.

At least to some degree the paper had been prepared in response to a comment by the Health and Disability Commissioner, Ron Paterson, suggesting that ACC should initiate the levy. The Committee asked ... to discuss with Mr Paterson the difficulties the levy would cause.

9.3 CARM News

- CARM news report

10. Other countries

10.1 Australia

- Aust Adv Drug Reactions Bulletin 19(4), Dec 2000
- Aust Adv Drug Reactions Bulletin 20(1), Feb 2001

10.2 Canada

- Canadian Adv Drug React Newsletter 11(1), Jan 2001

10.3 US

10.3.1 Impact of cisapride safety alert 1998

- W Smalley, et al. Contraindicated use of cisapride. Impact of food and drug administration regulatory action. JAMA 284:3036-3039, Dec 2000
- R Woosley. Drug labelling revisions - guaranteed to fail? JAMA 284:3047-9, Dec 2000

Smalley et al evaluated the impact of the FDA in June 1998 requiring a black box warning in the labelling of cisapride, issuing a media release and having the manufacturer send out a Dear Health Care Professional advising of the risk of QT-prolongation and that cisapride is contraindicated with certain medicines and in patients with certain conditions predisposing to QT-prolongation. The authors checked records in 3 health districts for contraindicated use during the 1-year periods immediately before and immediately after the communications.

Contraindicated medicine use was said to have occurred if dispensed prescriptions indicated at least one day of concomitant use. Use was considered contraindicated for medical condition if a contraindicated condition was recorded on a medical file in the 180 days preceding dispensing of a cisapride prescription. The authors found almost no change in contraindicated use in each of the 3 health districts after (24%, 28% and 58%, respectively) the communications compared with before (26%, 30% and 60%, respectively). Even for new users of cisapride only minor reductions in contraindicated use were found.

The editorial by Woosley blamed the lack of impact of the FDA's efforts in part on the wide variety of means and venues for dispensing medicines in the US, which reduces the ability to put controls to ensure compliance with safety requirements on the system. Woosley called for electronic means to increase compliance with safety warnings.

Members noted the low impact of these safety alerts in the US. Information on the impact of similar activity in New Zealand is not available.

11. Publications

12. Current topics

12.1 Thioridazine and QT-prolongation

- Risk management and the cardiotoxicity of thioridazines. Report for Medsafe, 27 Feb 2001
- M Tatley. Thioridazine: review of the CARM database. Feb 2001
- Prof A Breckenridge. Thioridazine: restricted indications and new warnings on cardiotoxicity. UK dear Health Professional letter. 11 Dec 2000
- Advice to Fellows concerning thioridazine, Royal Australian and New Zealand College of Psychiatrists. 4 Dec 2000
- JG Reilly, et al. QTc-interval abnormalities and psychotropic drug therapy in psychiatric patients. Lancet 355:104852, 25 Mar 2000
- K Hartigan-Go, et al. Concentration-related pharmacodynamic effects of thioridazine and its metabolites in humans. Clin Pharmacol & Therapeutics 60:543-53, 1996
- B Warner. Potential for cardiac arrhythmias and events resulting from prolonged QTc under Melleril treatment. Novartis Pharma AG. 3 Nov 2000
- Current NZ data sheet for Melleril
- Pharmac data for usage of thioridazine in NZ
- G Seifert Jones, Medical Officer (Psychiatry), Wairarapa Health. Thioridazine restrictions. Letter 19 Dec 2000

12.1.1 Background

In 2000, because of evidence of QT-prolongation, the indications for thioridazine were restricted in the US, UK and Australia. In Australia it is now approved only for the management of persons with schizophrenia who have failed to respond adequately to treatment with appropriate courses of at least 2 other antipsychotic agents. The other countries have restricted its use in a similar way and imposed contraindications and ECG monitoring requirements.

In New Zealand thioridazine is currently indicated for a wide range of conditions including schizophrenia, anxiety, depression, sedation and behavioural disorders in mentally retarded individuals.

The sponsors (Novartis and Pacific) had been sent notices under s.36 of the Medicines Act advising that the Director-General of Health had concerns about the safety of thioridazine. The Act allows the sponsors 60 days after the date of the notice to respond. The 60 days were to expire on 19 March 2001. After that date the Delegate of the Director-General, if he still has concerns, may refer the matter to the appropriate committee. Since the meeting had preceded the expiry date, the Secretary advised the Committee that, provided the Delegate of the Director-General wishes to so refer the matter, it will need to be discussed by teleconference to finalise it. Novartis had submitted a considerable amount of material prior to the date of the notice and the company had indicated that it had no additional material that it wished to submit. The material submitted by Novartis had been supplied to ..., psychiatrists who were engaged by Medsafe to prepare an expert report. Pacific had submitted no material either before or after the date of the notice.

12.1.2 The data

The study by Reilly et al reviewed the QTc-interval of 495 psychiatric patients compared with those of 101 healthy reference individuals. Risk factors for prolonged QTc-interval (>456ms) were found to be age > 65 years, use of droperidol, thioridazine (odds ratio 4.9; 95% CI 2.3-10.6) or tricyclic antidepressants. No patients were taking > 600mg thioridazine. Of the 4 taking 600mg, 2 had lengthening of the QTc-interval. The data were consistent with a dose-relationship.

The pharmacodynamic study in 9 healthy male volunteers by Hartigan-Go et al found QTc-prolongation at thioridazine doses of 10mg and 50mg (mean increase for 50mg 22ms, and 9ms for 10mg). Despite its limitations (e.g. small subject numbers), this study provides evidence of a dose-related effect on QTc-interval.

The WHO database holds 27 reports of QT-prolongation and 45 of other arrhythmia among a total of 7,128 reactions for thioridazine. Only 2 cases of QT-prolongation were reported for droperidol. CARM has received 2 reports of QT-prolongation and 5 other reports of cardiac events. A total of 2 cases (arrhythmia and sudden death) were fatal. From 1971 to 16 May 2000, 147 reports of heart rate and rhythm disorders with thioridazine were reported to the manufacturer. Six reports (4 fatal) of QTc-prolongation were received by the company from 16 May to 3 November 2000.

12.1.3 Expert report

The report by ... weighed the benefits and risks of thioridazine therapy and made recommendations about the future use of thioridazine in New Zealand. The authors had obtained data from Pharmac on the use of thioridazine. These data indicated that the use of thioridazine is declining, most prescribing is by general practitioners, and it is the most widely prescribed classical antipsychotic, but was in 2000 surpassed in use by an atypical antipsychotic. 1.7% of prescriptions were for dosages in excess of 400mg daily and 10% for dosages of ≥ 200mg daily.

Despite the evidence that the risk of harm from QT-prolongation with thioridazine is very low, ... recommended that the use of thioridazine be restricted (presented in brief):

• Patients should be initiated on thioridazine only as third line therapy and with specialist endorsement;
• Thioridazine should be prescribed only after baseline testing of ECG, serum potassium and CYP2D6 genotype (for slow metabolisers);
• Patients already established on thioridazine should be assessed for risk factors for QT-prolongation and switched to another medication if risk factors were identified, but continued if not;
• ECG monitoring should be initiated and be conducted at baseline and after a dosage increase of >25%;
• Doses above 200 mg/day should be avoided in all patients and a maximum of 50 mg/day should be observed in slow metabolisers.

... did not recommend a narrowing of the indications. They considered that such action would be an "over-reaction", but thought that the fact that other well-recognised regulatory authorities had restricted the indications may result in pressure being exerted for New Zealand to do likewise.

They did not recommend that specialist endorsement be required for continuation of thioridazine therapy. They commented that thioridazine appears to be well tolerated in existing patients and that only those considered to be at high risk of QT-prolongation should be switched to other therapy.

12.1.4 Committee discussion

Members agreed that a causal association between thioridazine and QTc-prolongation has been demonstrated. However, since the evidence was that harm occurred very rarely as a result of QT-prolongation with thioridazine some considered that it was not appropriate to take any more than minimal restrictive action. They questioned whether, if the use of thioridazine were to be severely reduced, the antipsychotics used in its place would be safer. It was considered that if no action were taken, the use of thioridazine in New Zealand would drop by natural attrition. Some members wanted the indications to be unchanged, and none asserted a view that they should be restricted. Dr Tatley, in his report on the CARM data, also proposed that the indications not be restricted. However, members were interested in reducing the prescribing of high doses of thioridazine, and in introducing appropriate contraindications and precautions. Members considered that the maximum daily dose should be 200mg and administration should commence at the lower end of the range and be gradually titrated upwards until an effective dose or 200mg daily is reached. The Committee believed that introducing ECG monitoring requirements, as proposed by the Royal Australian and New Zealand College of Psychiatrists and supported by..., was impractical, but members did support requiring baseline ECG checks. In addition, it was considered that checking patients for slow metabolism of CYP 2D6 was also impractical.

12.1.5 Recommendations

The Committee recommended:

• asking the company to update the data sheet with the dosage restrictions, new contraindications and precautions;
• publishing a Prescriber Update article to notify of the changes, and give advice about switching patients should this be deemed appropriate;
• asking the sponsor to send out a dear doctor letter advising of the changes.
• The Committee asked that the wording to be requested for the data sheet be passed by members and discussed at a teleconference before it is finalised. The formulation of the article and dear doctor letter would depend on the wording required for the data sheet. The teleconference would serve the need of the legislation should the Delegate of the Director-General refer the matter of the safety of thioridazine to the Committee.

12.2 Isotretinoin - safety and usage

- , Dermatologist, UK. Isotretinoin and general practitioner prescribing. Email of 17 Feb 2001
- . Roaccutane. Comment on behalf of the RNZCGP. 27 Feb 2001
- , Professor of Dermatology, UK. Isotretinoin and general practitioner prescribing. 23 Feb 2001
- M Bigby, RS Stern. Adverse reactions to isotretinoin. A report from the Adverse Drug Reaction Reporting System. J Am Acad Dermatol 18:543-52, Mar 1988
- M David, E Hodak, NJ Lowe. Adverse effects of retinoids. Med Toxicol Adverse Drug Exp 3:273-88, Jul-Aug 1988
- S S Jick, et al. Isotretinoin use and risk of depression, psychotic symptoms, suicide, and attempted suicide. Arch Dermatol 136:1231-1236, Oct 2000
- D K Wysowski, et al. Depression and suicide in patients treated with isotretinoin. NEJM 344, 8 Feb 2001
- . Isotretinoin use and risk of depression, psychotic symptoms, suicide attempts and suicides. Report for Medsafe, Sep 1999
- M Rademaker. Acute depression and isotretinoin. Prescriber Update No.17, p.17-19, Dec 1998
- New US restrictions likely on Roche's Accutane. Scrip 2577:20, Sept 2000
- Labelling in US, Jan 2001
- NZ Pregnancy data for Roaccutane users, Roche, Feb 2001
- NZ labelling for Roaccutane
- NZ labelling for Oratane
- NZ Package insert for Oratane

A member who declared that since he was a dermatologist he had a potential conflict of interest for this topic was permitted to participate fully.

The 2 dermatologists from the United Kingdom (...) were firmly of the view that isotretinoin should be available only on a specialist prescription because of safety issues associated with isotretinoin. Both mentioned that in their departments dermatology registrars receive considerable special training in the treatment of severe acne and in the use of isotretinoin before they are permitted to prescribe it. Both consider isotretinoin to be the supreme medication for acne ("most effective drug for the treatment of all but the mildest acne"). mentioned that skill is involved in selecting the optimum dose, and incorrect use can lead to a flare up of acne and significant scarring. He has had patients affected with these problems referred to him from dermatologists with experience in the use of isotretinoin.

Isotretinoin is associated with cutaneous side effects which affect almost all users. Patients need to be counselled about these effects and how they should be managed. Teratogenicity (50% spontaneous abortions; 50% term babies have major congenital anomalies), depression, arthralgia, myalgia, fatigue were other adverse effects mentioned by the dermatologists.

Both associated specialist-only prescribing in the UK with the low rate of pregnancies occurring while the woman is taking isotretinoin, or in the 4-week period after stopping. Conversely they considered the high rate in the US to be associated with more liberal prescribing rights (2.8 per 1000 women).

In contrast an opinion from the Royal New Zealand College of General Practitioners asserted that general practitioners are better placed than dermatologists to deal with some of the issues around the prescribing of isotretinoin. Acne is commonly encountered in general practice and currently general practitioners refer patients to dermatologists for isotretinoin. In other words they already have the experience to know when it is appropriate to prescribe it. But most importantly the general practitioner has a knowledge of the patient, is able to appreciate the potential for psychiatric adverse effects and is well experienced in giving contraceptive advice and prescribing contraceptives. If general practitioners were permitted to prescribe isotretinoin, they should do so according to agreed guidelines. The College would support development of a detailed patient information sheet with patient consent form.

In New Zealand, according to the records of Roche there have been 23 pregnancies in women taking isotretinoin from 1989 to 2000 inclusive and Douglas is aware of 1 pregnancy in a woman taking its product.

In New Zealand up until now and currently, funding has been only on the prescription of a specialist dermatologist. However, the price of the generic isotretinoin is low (about $100 for a month's supply), such that it is now affordable to obtain it on prescription by a general practitioner and for the patient to pay for the product. At present it seems that the proportion of general practitioner prescribing is about 5%.

... commented that isotretinoin is the only medicine available for which almost all users experience adverse effects, in particular mucocutaneous reactions. In addition to these effects, the adverse effects of isotretinoin cover a broad range, including ocular toxicity, teratogenicity, central nervous system effects including depression, liver toxicity, muscular effects and hyperlipidaemia. In New Zealand, dermatology registrars are not permitted to prescribe it until they have 2-3 years experience with the treatment of acne. There is a risk of using isotretinoin before the severity of the disease warrants it, and hence unnecessarily subjecting patients to adverse reactions. In addition, patients with less severe acne will be less motivated and less willing to accept and or take appropriate precautions in the face of the side effects. Those to whom isotretinoin is prescribed need competent counselling concerning the adverse effects.

The Jick study used the GPRD and Saskatchewan databases and found no evidence of an association between isotretinoin and depression, psychotic symptoms, suicide or attempted suicide. An expert report prepared for Medsafe September 1999 on this study and other data concerning the association between depression and isotretinoin presented the view that a very small number of patients may develop clinical symptoms of depression and related problems with isotretinoin.

There was some discussion of whether it would be acceptable to make isotretinoin available on prescription of a general practitioner, if there were certain controls in place, such as obtaining a particular qualification set up for the purpose. There was no support from any member for liberalisation of isotretinoin prescribing except under controlled conditions. The Committee asked that Pharmac be advised of this view of the MARC.

Members were concerned that the patient information leaflets for Roaccutane and Oratane did not provide enough information and advice about the risk of depression and suicide with isotretinoin. The Committee recommended that the respective sponsors be asked to address this deficiency.

12.3 Propofol, excessive dose for prolonged sedation and cardiac failure

- O L Cremer, et al. Long-term propofol infusion and cardiac failure in adult head-injured patients. Lancet 357:117-118, Jan 2001
- Diprivan NZ data sheet extract

Cremer, et al conducted a study of head injury patients requiring long term sedation with propofol following admission to a neurosurgical ICU between 1996 and 1999. 67 patients were eligible for analysis. In this group they identified 7 fatal cases of propofol-infusion syndrome. The syndrome did not occur in patients given propofol at < 5 mg/kg/h. The odds ratio for occurrence increased at 1.93 (95% CI 1.12-3.32) for every mg/kg/h increase in propofol dose above 5 mg/kg/h. All cases had received propofol at > 5 mg/kg/h for > 58 hours.

The authors recommended discouraging infusion of propofol at rates higher than 5 mg/kg/h.

The current data sheet for Diprivan (propofol) advises that 0.3-4.0 mg/kg/h should achieve satisfactory sedation, but there is no recommendation about a maximum rate of administration. The Secretary advised that she had received a verbal opinion from , pain specialist. He had reviewed the data and considered that it would be wise to set a maximum of 4.0 mg/kg/h on the use of propofol in sedation in intensive care settings.

The Committee recommended asking the sponsor to put in the data sheet a warning to observe a maximum dosage of 4.0 mg/kg/h in long term (> 24 hours) sedation.

12.4 GI Haemorhage and low dose, long term aspirin

- S Derry, and Y Kong Loke. Risk of gastrointestinal haemorrhage with long term use of aspirin: meta-analysis. BMJ 321:1183-1187, Nov 2000
- M R Tramèr. Aspirin, like all other drugs, is a poison. BMJ 321:1170-1171, Nov 2000
- H T Sorensen, et al. Risk of upper gastrointestinal bleeding associated with use of low-dose aspirin. Am J Gastroenterlogy 95:2218-2224, Aug-Sept 2000

Derry and Loke conducted a meta-analysis of 24 randomised placebo-controlled trials involving long term treatment with aspirin. A total of 65,987 patients were included. The daily dose of aspirin was 50-1500mg with a mean duration of 28 months. Patients were excluded if they had a history of peptic ulcer or gastrointestinal haemorrhage.

The pooled odds ratio for gastrointestinal haemorrhage with aspirin was 1.68 (1.51-1.88) (2.47% of aspirin users developed gastrointestinal haemorrhage). A meta-regression to test for a linear association between dose and risk of gastrointestinal haemorrhage was conducted. The estimated reduction in risk per 100mg reduction in dose was 1.5% (p= 0.3). Similarly controlled release formulations showed no clear benefit (odds ratio cf. placebo 1.93; 1.15-3.23).

The study found the number needed for harm to be 248 per year. In the primary prevention of myocardial infarction the number needed to treat was 555 per year and for patients with hypertension it was 794. However, myocardial infarction is more likely to be fatal than gastrointestinal haemorrhage.

Sorensen et al conducted a population-based study in the Danish County of North Jutland looking at the risk of gastrointestinal bleeding with low-dose aspirin. The study population included 27,694 users of low dose (100mg or 150mg) aspirin for a total of 42,054 person-years of observation. The standardised incidence ratio (SRI; ratio of observed rate versus expected rate based on rate for entire population) for gastrointestinal haemorrhage was 2.6 (2.2-2.9) for users of low dose aspirin alone, 5.6 (4.4-7.0) for aspirin plus other NSAIAs and 4.7 (2.9-7.2) for aspirin plus other medicines. There was no difference in risk between the 2 strengths of aspirin tablet nor between coated and uncoated tablets. The increase in risk of gastrointestinal bleeding when low dose aspirin was taken with other NSAIAs seemed to be at the same level as that observed for NSAIA treatment alone.

These authors concluded that further work needed to be done to ascertain the risk versus benefit of prophylactic low dose aspirin.

The Committee commented that these studies were not able to answer the question whether aspirin at 75 mg daily may be associated with a much lower risk of gastrointestinal haemorrhage and be as effective as 100mg to 150mg daily. The Committee recommended no action.

12.5 Folic acid antagonists and birth defects

- S Hernández-Díaz, et al. Folic acid antagonists during pregnancy and the risk of birth defects. NEJM 343:1608-1614, Nov 2000

Hernández-Díaz et al conducted a study including 6932 case infants with congenital defects known to be reduced in frequency with multivitamin supplementation (cardiovascular and urinary defects and oral clefts) and 8387 infants with other types of congenital anomalies. Mothers were interviewed by nurses blinded to the purpose of the study for use of medication during the pregnancy. The relative risk of all 3 types of defects (for cardiovascular defects 2.2 [95% CI 1.4-3.5]; for oral clefts 2.5 [1.5-4.2] and for urinary tract defects 2.5 [1.2-5.0]) was found to be elevated by maternal exposure to antiepileptic medication (phenobarbitone, phenytoin, primidone and carbamazepine). Similar elevations in relative risk were found for dihydrofolate reductase inhibitors (trimethoprim, triamterene and sulphasalazine) for cardiovascular defects (3.4 [1.8-6.4]) and oral clefts (2.6 [1.1-6.1]), but case numbers were too low (2 cases) to estimate the relative risk for urinary tract defects.

For the dihydrofolate reductase inhibitors the elevated risk was found to be associated with exposure during the second and third months of pregnancy. The relative risk of cardiovascular defects with dihydrofolate reductase inhibitors was reduced from 7.7 (2.8-21.7) to 1.5 (0.6-3.8) by administration of folic acid during the critical period. The number of cases with oral cleft was too low to estimate any reduction in relative risk with administration of folic acid. No reduction in risk of congenital defects was found if folic acid was used during pregnancy in those women taking antiepileptic medication.

The authors observed that the results were consistent with the hypothesis that folic acid helps prevent not only neural tube defects but also cardiovascular defects caused by dihydrofolate reductase inhibitors. The finding that carbamazepine may cause urinary tract defects is new. It appears that folic acid in vitamin supplements taken by pregnant women may not protect against congenital cardiovascular and urinary tract defects or oral clefts caused by antiepileptic medication. The teratogenic effects of phenytoin and phenobarbitone may not be occurring through depletion of folic acid antagonism but through a direct toxic effect.

The Committee noted the material but recommended no action.

12.6 Items of interest

The items in this section (12.6) were not discussed by the Committee.

12.6.1 Necrotising fasciitis and NSAIAs

- N Forbes, et al. Necrotizing fasciitis and non steroidal anti-inflammatory drugs: a case series and review of the literature. NZMJ 114:3-5, Jan 2001

12.6.2 Immunosuppressive therapy and non-Hodgkin's lymphoma

- R J Farrell, et al. Increased incidence of non-Hodgkin's lymphoma in inflammatory bowel disease patients on immunosuppressive therapy but overall risk is low. Gut 47:514-519, Feb 2000

12.6.3 Pregnancy-outcomes in diethylstilbestrol-exposed offspring

- R H Kaufman, et al. Continued follow-up of pregnancy outcomes in diethylstil-bestrol-exposed offspring. Obstetrics & Gynecology 96:483-489, Oct 2000

12.6.3 OC Use in the UK

- OC scripts still below 1995 levels in UK. Scrip No.2590, 8 Nov 2000

12.6.4 Just for fun

- Drug side effects: a comedian's version. NZMA insert in NZMJ

12.7 Safety of bupropion (Zyban)

- Zyban (bupropion hydrochloride) - safety update, Medicines Control Agency, 1 Dec 2000
- Anti smoking drug deaths. The Mail (London) 18 Feb 2001
- Safety profile of Zyban tablets. GlaxoSmithKline Medical Services Dept. Undated
- Bupropion hydrochloride, Type A report from CARM, 20 Feb 2001

The 18 February 2001 issue of The Mail (London) reported 18 deaths in the United Kingdom in patients taking bupropion for smoking cessation. Four of these had died of a heart attack. A spokesperson for the Medicines Control Agency was quoted as saying that the contribution of Zyban to these deaths was unknown. She said, many people attempt to stop smoking because of concerns about an underlying disease that may cause death.

Included in the dossier was a printout of adverse reactions to bupropion reported to CARM. 36 reports had been received as at 20 February 2001. As at 13 March a further 12 reports had been received. There had been no deaths but one case of myocardial infarction (see 6.9.1) had been reported. Zyban had been given approval in New Zealand in May 2000.

Bupropion is contraindicated in patients with a current or past seizure disorder. It may interact with other psychotropic agents. Some of those taking it, since they are long-term smokers, will be at risk of myocardial infarction.

The Committee asked that any new information about the safety of bupropion, particularly data about the relationship between bupropion and myocardial infarction, be brought to its attention.

The meeting closed at 5.05pm.