Revised: 21 May 2013

Committees

Minutes of the 109th Medicines Adverse Reactions Committee Meeting - 27 March 2002

Held in Harbour Room One, 20th Floor, Grand Plimmer Tower, commenced at 9:30am.

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Descriptions of unpublished case reports have not been included to protect the privacy of those involved.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

The recommendations of the committee are in bold type face.

Minutes:

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 (Chair)
Dr M Tatley
Dr R Savage
Dr J Goldsmith
Professor P Ellis
Dr H Kingston
Dr F McClure
Dr M Rademaker
Dr N Rafter
Professor DCG Skegg
Dr M Harrison-Woolrych
Dr K Ronaldson (Secretary)

IN ATTENDANCE

All day
Ms S Von Afehlt (Editor, Prescriber Update)
Dr R Hill (Adverse Drug Reactions Unit, TGA)

Part of the day
Dr S Jessamine (Principal Technical Specialist, Medsafe)
Dr S Martindale (Team Leader, Business Support and Development, Medsafe)

1. MATTERS OF ADMINISTRATION

1.1 Welcome and apologies

The Chair welcomed Dr Hill from the Adverse Drug Reactions Unit of the Australian TGA to the meeting.

Apologies had been received from Dr Coulter.

1.2 Minutes of the 108^th meeting

The minutes of the 108^th meeting were signed by the Chair as a true and accurate record of the meeting.

The Secretary stated that Medsafe is wanting to have the minutes finalised more quickly than in the past. Medsafe would like to have the minutes finalised about 6 weeks after the meeting, and to have the Chair sign them off soon after that and prior to the next meeting. It will then be possible to publish the minutes on the web site more speedily, within 2-3 months of each meeting.

At present the minutes are finalised 2-3 weeks before the following meeting which means that it is not possible to seek sign off by the Minister's Delegate on the minutes and recommendations of the meeting until most of the recommendations have already been implemented. To speed up the process it will be necessary for members to observe the 2-3 week deadline given to them by the Secretary when she sends out the minutes for review. Members will also need to be sure to pick up any errors, or anything they want changed, at that review.

Members indicated they wished to be advised of any changes they suggest which are not included. The Secretary advised that she already got back to members about queries she had over proposed changes. She pointed out that changes proposed to minutes at meetings were always issues not previously raised rather than unresolved matters which had already been brought to the Secretary's attention.

Members agreed to comply with the requirements needed to achieve finalisation of the minutes within about 6 weeks of each meeting.

1.3 Dates of the next meetings

20 June, 11 September and 5 December were confirmed as the dates for the next meetings.

1.4 Conflict of interest

• Declaration of competing interests (guidelines and declaration form)
• Competing interests tables

Members with undeclared conflicts of interest submitted these to the Secretary.

1.5 Status statements for sections 5 and 9

• K Ronaldson. Status statements for sections 5 and 9. 1 Mar 2002

During the review of the minutes by members, a member had raised the inclusion in the minutes of a listing of documents which had not been discussed by the Committee. The Secretary had proposed that status statements be included in both the minutes and the agenda to describe the status of the material on the agenda for information or interest only. The statements for the minutes had been included in the minutes of the 108^th meeting with the agreement of the Chair. In addition, the Secretary had prepared a paper on the subject to present the matter to the members.

It was pointed out that Medsafe would insist that these documents be listed in the minutes because for its purposes Medsafe needs it to be clear that the material had been before the Committee even if it had not been discussed. In addition, the ADRAC minutes list documents not discussed but in the dossier.

Members accepted the status statements proposed by the Secretary, with one change to be made to the statement for the minutes to be included at the beginning of Section 5:

• From "it would be moved to section 3 or 4" to "it would have been moved to section 3 or 4".

Dr Tatley pointed out that although the Quarterly Report for CARM was under section 5, the status statement for that section did not cover it. It was not a "commentary, review article or preliminary material on an emerging issue." In addition, he considered it to be important that the Committee discuss the report, rather than simply have it before them for information. Dr Tatley asked for the permission of the Committee to move the Quarterly Report to Section 3 for this and subsequent meetings. Members gave their permission.

1.6 Name for CARM

DM Coulter. Global name change for the National Centre. Discussion paper prepared for MARC, 28 Feb 2002

Dr Coulter had prepared a paper proposing that the national centre for adverse reaction reporting in Dunedin be given the global name of "New Zealand Pharmacovigilance Research Group". Members did not like the proposed name, since "group" implies a loose conglomeration rather than the tighter aggregate suggested by "centre". In addition, using the word "research" implies that they should be producing research publications, but that is not a primary outcome. Members preferred the name "New Zealand Pharmacovigilance Centre", but recognised that the finalisation of the name was an issue for CARM and Medsafe to settle between them.

2. MATTERS ARISING

2.1 Report on the recommendations

• K Ronaldson. Secretary's report to the members of the MARC for the 109^th meeting, 11 Mar 2002
• Minutes and recommendations of the 108^th meeting of MARC, 11 Dec 2001. Report for the Minister's Delegate. 7 Mar 2002
• Recommendations of the 108^th meeting of the MARC

2.1.1 Prescriber Update (1.5, Dec 2001)

• Front cover/contents page Prescriber Update 23(1) Apr 2002

Members noted the new A4 format for Prescriber Update, and said they considered it an improvement. They indicated their approval for the innovations in Prescriber Update.

2.1.2 Correspondence with Pharmac (2.1.5, Dec 2001)

• T Maling. Nil response to letters to Pharmac. Letter to W McNee, Pharmac 12 Feb 2002
  .., Pharmac. Funding of EpiPen. Letter of 11 Dec 2002
  W McNee, Pharmac. Response to MARC. Letter of 8 Mar 2002

After considerable delay a reply had been received to the letters from the MARC to Pharmac concerning safety issues with bupropion, the funding of EpiPen (adrenaline for self-administration), venous thromboembolism (VTE) with Diane 35/Estelle 35, and cardiovascular effects with COX-2 inhibitors. The reply was in a letter from Mr McNee in response to a letter from the Chair of MARC expressing concern about the lack of response to earlier letters.

Members noted that the reply did not indicate that Pharmac intended to take any action as a result of the concerns expressed by the MARC. In particular, members were concerned at the apparent intention to do nothing about funding EpiPen for those with allergies to substances to which they become inadvertently exposed. It was agreed that the allergens in such cases were usually foods (e.g. peanuts) or bee stings. Hence, the adverse event would most often not be in response to a medicine. Although essentially the issue was outside of the remit of the Committee, it was agreed that it was nevertheless appropriate for the Committee to express its view. Members asked that a letter be sent on behalf of the Committee to the Minister of Health to indicate to her the need for EpiPen to be funded, and to advise her of the failure of Pharmac to act on repeated concerns on this subject expressed by MARC.

In addition, members mentioned the names of individuals who could be approached to provide advice on the need for the funding of EpiPen: ..

Members also noted the minute item on rofecoxib from the PTAC (see item 6.2), which indicated that the letter from the MARC had been passed to the Pharmac committee. However, it also indicated that PTAC had missed the force of the letter from MARC, that the evidence points to rofecoxib being thrombotic and that this is the explanation for the results of the VIGOR study (higher rate of myocardial infarction with rofecoxib than with naproxen).

2.1.3 Workshop on CIOMS IV (2.2, Dec 2001)

Dr Martindale advised that Medsafe had not accepted the recommendation of the Committee that someone from the FDA be invited to facilitate a workshop using the CIOMS IV methodology. Medsafe considered it preferable to arrange a joint workshop with ADRAC, given the proposal to establish a Trans-Tasman regulatory agency. This may occur at a time when there has been further progress towards the new agency.

2.1.4 Isotretinoin and reducing pregnancy risk (3.1.2.1, Dec 2001)

- FDA announces changes to the risk management program to prevent birth defects caused by Accutane. FDA Talk Paper 31 Oct 2001
- .., Roche. Reducing the risk of pregnancy with Roaccutane. Letter 21 Feb 2002
- NZ labelling for Roaccutane 10mg capsules
- NZ consent form for female patients taking Roaccutane
- Warning against pregnancy in Roaccutane consumer leaflet

Members noted the new restrictions to be adopted in the US to reduce the risk of pregnancy with isotretinoin, including 2 negative pregnancy tests before commencing isotretinoin and a further pregnancy test before each monthly dispensing of isotretinoin. The new restrictions are to take effect on 1 April. Members did not support introducing similar restrictions in New Zealand.

It was noted that while in the US any medical practitioner can prescribe isotretinoin, only specialists can prescribe funded isotretinoin in New Zealand. General practitioners can prescribe unfunded isotretinoin, and the generic is cheap enough, though still expensive, to make payment by the patient cost-effective. Members commented that the advantage of isotretinoin over Diane 35 or Estelle 35 is that isotretinoin provides long term remission, while recurrence is likely following discontinuation of Diane 35 or Estelle 35.

Members noted that Medsafe intends to ask the sponsor companies (Roche and Douglas) to send out a Dear Doctor letter concerning the need to prevent pregnancy in women taking isotretinoin.

In addition, members noted that the package insert for Roaccutane has a cross-reference to "Interactions" with no information to match the cross-reference. It appeared that the intention was to mention that progestogen-only pills do not provide adequate contraception. The Committee recommended pointing out to the company the omission of the information to match the cross-reference in the package insert for Roaccutane.

2.1.5 NSAIAs and post-operative haemorrhage (3.1.4.1, Dec 2001)

The previous meeting had reviewed 2 cases of serious post-operative haemorrhage associated with NSAIAs, and the Committee had recommended conducting a literature search on postoperative haemorrhage and NSAIAs. The literature identified provided no evidence that there is a clinically significant increase in the risk of post-operative haemorrhage with NSAIA use. Members observed that the study by Catella-Lawson et al (see item 4.6.4) indicated that platelet inhibition by NSAIAs is not irreversible or complete. Hence, there is unlikely to be an effect which persists long after each dose. Members agreed that there was no need to pursue this matter further at present.

2.1.6 Gabapentin and teratogenicity (3.1.5.1, Dec 2001)

- .. Gabapentin and multiple malformations. Comment 4 Feb 2002 Following a case of multiple malformations with intrauterine exposure to gabapentin, .. at the Anti-Epileptic Drugs Pregnancy Register, based at Massachusetts General Hospital had been contacted. He had checked the register and found that it did not have a signal for similar cases. However, as he pointed out the register is in its infancy and needs to record many more pregnancies exposed to gabapentin before any definitive statement could be made. Members noted both the helpfulness of the response and the potential value of the register in the future. The Committee recommended no additional action at this time.

2.1.7 Estelle 35/Diane 35

- K Ronaldson. The funding of Estelle-35. Letter to Pharmac, 29 Jan 2002
- W McNee. The funding of Estelle-35. Response from Pharmac, 13 Mar 2001
- .. Douglas Pharmaceuticals. Promotional activities for Estelle-35. Letter of 22 Feb 2002
- S Jessamine. Diane-35 indications. Letter to Schering, 27 Feb 2002
- Oral contraceptive use in NZ. Graph of pill use and abortion uptake. Graph prepared by Schering NZ Ltd

Members noted the letter sent to Pharmac making two specific suggestions to the listing of Estelle-35 in the Pharmaceutical Schedule in order to make it clear that it is not primarily an oral contraceptive (OC), but should only be used for contraception in women being treated for androgenic diseases. The letter had also questioned the fully funded status of Estelle-35 in view of the risk of VTE with this medicine. The letter had received a dismissive reply from Mr McNee.

Members asked that the issues raised in the letter to Pharmac concerning the manner of the listing of Estelle-35 in the Pharmaceutical Schedule be discussed with Pharmac by Medsafe at the next liaison meeting.

Members expressed some concern about the advertising for Estelle-35. Members considered that it could be understood to mean that Estelle-35 could be prescribed for the prevention of acne. The Secretary advised that the Compliance Team had reviewed the advertising and had found it to be in line with the legislation and guidelines. However, Medsafe had requested a change to the indications for Estelle-35 and Diane-35 to indicate that they are to be second line therapy for acne, and for contraception in women requiring treatment for androgenic diseases. The Secretary advised that the advertisement would not be consistent with the stricter wording for the indications. The new wording for the indications is in line with the indications approved for Diane-35 in other countries, specifically Australia and the United Kingdom.

The Committee was in agreement with the requested revision to the indications. The Committee recommended that the wording of the advertising for Estelle-35 be addressed once again when the change to the indications has been approved.

2.1.8 Anticonvulsants and bone mineral loss (4.4, Dec 2001)

• .. Dept Medicine, University of Auckland. Bone mineral loss with anticonvulsants. Letter of 15 Feb 2002

On the recommendation of the previous meeting, a letter had been written to .. asking for advice on the need for treatment of bone mineral loss with anticonvulsants. He had commented that loss of bone mineral density is a minor problem with anticonvulsants. However, anticonvulsants cause vitamin D deficiency which may become a problem with poor nutrition and little exposure to sunlight.

Members recommended that no further action be taken on anticonvulsants and bone mineral density loss or vitamin D deficiency at this stage.

2.1.9 Progestogen-only pills and venous thromboembolism

• .. Schering, NZ. POPs and use post-VTE. Letter, 20 Feb 2002
• .. Schering, NZ. Alternative wording for data sheets. 5 Mar 2002
• Progestogen-only oral contraceptives and VTE. Item 12.1, minutes of MARC, 5 Sep 2001

Professor Skegg had asked one of the members of the panel which drew up the WHO guidelines on contraceptives to review a draft Prescriber Update article on VTE with progestogen-only pills. The article included advice from the Committee formulated at the meeting in September 2001 and based on the WHO guidelines. The panel member had had no difficulty with the Committee's interpretation of the advice.

However, Schering AG had objected to the advice, particularly the suggestion that a progestogen-only pill could be considered an option for contraception in women who have experienced a previous VTE. Schering NZ had proposed an alternative wording, which stated that a progestogen-only pill was not contraindicated in women with a personal history of VTE, and need not be discontinued for major surgery, except in cases of prolonged immobilisation.

The Committee did not wish to change the advice on progestogen-only pills and VTE formulated at the September 2001 meeting.

The Committee considered the wording proposed by Schering NZ regarding use in women with a personal history of VTE was acceptable, but noted that the statement about use in surgery was not an accurate interpretation of the WHO Guidelines. The Committee recommended accepting the first proposed statement, but negotiating with Schering to revise the statement about surgery.

2.1.10 Nefazodone and hepatic toxicity

• Serzone (nefazodone) data sheet. Updated for hepatic reactions. 7 Mar 2002

The Committee noted the update to the Serzone data sheet regarding hepatic reactions.

3. ISSUES ARISING FROM General Reporting AND IMMP

3.1 Spontaneous reporting programme

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.

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.

3.1.1 Deaths

3.1.1.1 Atorvastatin and myocardial infarction, 49387

Comment: The WHO database holds 108 reports of myocardial infarction associated with atorvastatin, but no such cases are recorded in the CARM database. The Committee considered the use of atorvastatin to be coincident to the event in the present case. The causal association was designated "unlikely". No additional action was recommended.

3.1.1.2 Diazepam, fluoxetine and suicide, 48954

Comment: It was noted that fluoxetine has been associated with suicidal tendencies. The CARM database holds 2 reports of suicidal tendency and 8 of suicide attempt. No events of this nature have been reported with diazepam, but it may have contributed to disinhibition. The patient's schizophrenia may also have contributed to his suicide. The causal association for both diazepam and fluoxetine was designated "possible". The Committee recommended that no additional action be taken.

3.1.1.3 Diclofenac and vomiting, rigors, dyspnoea, faecal incontinence, skeletal pain, sepsis, 49093

Comment: Members observed that information from dissection of the muscles of the spine and back may have assisted in the difficult interpretation of this report. The CARM database holds 4 reports of necrotising fasciitis with diclofenac and the WHO database holds 49 reports of sepsis (out of a total of 20,815; IC-2SD 0.5). Each of the adverse events (vomiting, rigors, dyspnoea, faecal incontinence, skeletal pain and sepsis) was given the causal designation "possible". No further action was recommended.

3.1.1.4 Isotretinoin and suicide, 49497

Comment: Of a total of 68 reports with isotretinoin in the CARM database, 9 have been of depression, 3 of suicide attempt and 1 of suicide. Suicide is mentioned in the data sheets for both Roaccutane and Oratane. In the present case the association of the young man's suicide with use of isotretinoin was designated "possible". Since suicide is not mentioned in the Adverse Reactions section of the Oratane data sheet, members recommended that the company be asked to correct this omission.

3.1.1.5 Quinapril and pancreatitis, 49415

Comment: The CARM database holds 1 report of pancreatitis with quinapril. Reports of pancreatitis with ACE inhibitors were received when three of these medicines were monitored in the IMMP. The product data sheet for Accupril (quinapril) states that pancreatitis was reported in clinical trials. The present patient's pancreatitis was designated "possible" for causal relationship, and it was considered that the "medicine may be contributory" to the death. The Committee recommended no further action.

3.1.1.6 Ribavirin, interferon alpha, triazolam, diazepam, doxepin and suicide, insomnia, paranoid reaction, anxiety, 49052

Comment: Depression is a recognised adverse effect of interferon, and suicide attempt is listed in the data sheet for Roferon. The WHO database holds 14 reports each of suicide attempt with ribavirin and interferon alpha. Members considered that each of the medicines, including triazolam, doxepin and diazepam may have contributed, and designated the causal relationship for each adverse event "possible" for each medicine. No other action was recommended.

3.1.2 Hormones

3.1.2.1 Diane 35 and DVT, arterial embolism, amputation, 49383

Comment: The CARM database holds one report of DVT and arterial embolus with norethisterone/ethinyloestradiol 50mcg, but none with other combinations of oestrogen/progestogen or antiandrogen. The WHO database holds one case of renal artery stenosis in a male taking Diane-35 with additional cyproterone. The association between this patient's DVT and arterial embolus was designated "possible". No additional action was recommended.

3.1.2.2 Monofeme and DVT, pulmonary embolism, 49474

Comments: The CARM database holds 2 reports of DVT with Monofeme and 13 of DVT and 3 of PE with Microgynon (chemically identical). The association for the present case was designated "possible". Members were concerned at the lack of appropriate action following the positive scan in this case. Members asked CARM to write to the Royal Australasian College of Physicians and the College of Emergency Medicine citing this case and advising on appropriate management of DVT (discontinuation of the OC and treatment with anticoagulation therapy).

3.1.2.3 Oxytocin and therapeutic effect inadequate, product storage error, 49385

Comment: The Committee considered the appropriate designation for inade-quate therapeutic effect to be "probable". The Committee suggested that CARM write a letter to the New Zealand College of Midwives to advise of this case and highlight the need for appropriate storage of oxytocin ampoules.

3.1.3 Musculoskeletal

3.1.3.1 Diclofenac, quinapril, allopurinol and haematuria, proteinuria, 49108

Comment: Haematuria has previously been reported in New Zealand for diclofenac (5 cases) and allopurinol (1 case). In the WHO database there are 39 reports of albuminuria or proteinuria with diclofenac and 13 with allopurinol. Members observed that each of the medicines the patient was taking were associated with adverse renal effects. In the present case, details of renal function before commencement of diclofenac or after the development of the adverse reaction were not supplied in order to clarify the role. Both adverse events were given the designation "probable." No additional action was recommended.

3.1.4 Vaccines

3.1.4.1 DTaP, Hib/HepB, oral polio and meningitis, vomiting, fever, anorexia, 49478

Comment: Meningitis with hepatitis B vaccine is an event recorded on the WHO database (99 cases, 6 fatal; IC-2SD 1.02). No reports of this event have been recorded in association with the other vaccines, and the data sheets do not include meningitis as an adverse reaction. Members commented that the infant may have developed a coincidental adverse reaction. The reaction was designated "possible" for causality. The Committee recommended taking no further action.

3.1.4.2 Hepatitis A/Hepatitis B (Twinrix) and multiple sclerosis, 49005

Comment: A link between hepatitis B vaccine and multiple sclerosis was alleged about five years ago, but epidemiological studies (reviewed by MARC in June 2001, item 7.2.1 and March 2000, item 7.2.3) have not confirmed any association. This material is included in an article to be published in the April 2002 issue of Prescriber Update.

The WHO database holds 134 reports of multiple sclerosis-like syndrome with hepatitis B vaccine. The value for IC-2SD is 3.67, meaning the signal is significantly above background. However, it was noted that several years ago the possibility of this association was given considerable prominence. The publicity could have prompted reports. In addition, few other medicines would be associated with multiple sclerosis and then only rarely. These two effects would result in a significant value for IC-2SD. In the present case, the onset of symptoms was only 2 days after the vaccination, which would seem to be too early for a causal association. In addition, lesions would not appear on MRI scan until they had been developing for some time. Members considered a causality designation of "unlikely" to be appropriate for the present case. The Committee recommended that no additional action be taken.

3.1.4.3 Influenza vaccine and gastric ulcer, Guillain-Barré Syndrome, 48912

Comment: This is the sixth report of Guillain-Barré syndrome in association with influenza vaccine received by CARM. It appears that the event is associated in very rare cases - about 1 in a million doses of vaccine, according to an American study. In the present case the causal association was given the designation "possible for Guillain-Barré syndrome and "unlikely" for gastric ulcer. Members discussed conducting a study of the association between Guillain-Barré syndrome and influenza vaccine using New Zealand data, but no conclusion was reached.

3.2 IMMP

3.2.1 Selective COX-2 inhibitors

- DM Coulter. Early observations on the COX-2 cohorts. Report for MARC, 28 Feb 2002
- DM Coulter. Acute psychiatric events with COX-2 inhibitors. Report for MARC, 27 Feb 2002
- H-K Jiang, D-M Chang. Non-steroidal anti-inflammatory drugs with adverse psychiatric reactions: five case reports. Clin Rheumatology 18:339-45, 1999

The IMMP database now holds 2786 patients taking celecoxib and 1379 taking rofecoxib. Patients taking celecoxib are generally older than those taking rofecoxib (36% of those on celecoxib are ( 70 years compared with 25% for rofecoxib). Joint problems, mainly osteoarthritis, are the chief indications for celecoxib, but rofecoxib is most likely to be indicated for a pain syndrome.

13 reports of acute psychiatric reactions have been reported with the COX-2 inhibitors. Five of the reports were of confusion, 2 of depression and 3 of hallucination. Exacerbation of manic depressive psychosis was also reported in one case. In one case a 40-year-old woman developed a swollen face and dramatic change in personality, including irritability, crying for no reason, and irrational fears. Reducing the dose of rofecoxib to 12.5mg bid did not result in resolution of the symptoms, but she recovered following withdrawal. Depression, anxiety, confusion and hallucinations also feature for celecoxib and rofecoxib in the WHO database. Such events are known to be associated with the older NSAIAs.

The Committee recommended publishing a Prescriber Update article on acute psychiatric events with the COX-2 inhibitors.

3.2.2 Tolcapone

• Tolcapone. Monitoring summary. Interim report to 31 July 2001

Dr Tatley advised that 21 reports of liver-related adverse reactions out of a total cohort of 415 patients had been received. None of the cases were of serious liver problems. Most of them were of minor increases in liver function test parameters: 4 reports of abnormal liver function tests, 15 reports of alkaline phosphatase increased and 1 report of gamma GT increased. In addition, there was one report of cholelithiasis which was not thought to be associated with tolcapone. Some of the patients had liver cancer or gallstones.

Tolcapone had been taken off the market in Europe in 1998 because of reports of serious hepatic reactions. In New Zealand use had been restricted to prescription by specialists with liver function monitoring. In addition, monitoring in the IMMP had been commenced.

The Committee recommended taking no additional action at this time.

3.3 Intensive Vaccine Monitoring Programme

• M Tatley. Proposal for the intensive monitoring of the meningococcal B vaccine, 28 Jan 2002

Dr Tatley described the intended monitoring of the meningococcal B vaccine. Because of the meningococcal epidemic in New Zealand (12 deaths per month) caused by a strain not known elsewhere, the vaccine is to be introduced in South Auckland, where the epidemic is the most severe, without a phase III study. Although the vaccine is to be used in those up to the age of 20 years, the Intensive Vaccine Monitoring Programme is to cover the birth cohort to 6 months of age. The monitoring is to be carried out as for the IMMP with registration of each child vaccinated. In addition, it is intended to compare the results in the cohort who received meningococcal B vaccine with another cohort of infants who were not immunised with this vaccine.

The Committee asked that a letter be sent from the Committee to the Minister of Health to advise her that members whole-heartedly support the monitoring by CARM using the method proposed, and stating the view of the Committee that it is essential that the introduction of the meningo-coccal B vaccine be monitored in this way. The Committee asked that the need for a computer-based vaccination register be emphasised as well.

3.4 Quarterly report

- M Tatley. Quarterly Report, for quarter ending 31 Dec 2001.

Dr Tatley highlighted several things from the Quarterly report. Ten further reports of reduced therapeutic effect had been received with brand switches to the generic fluoxetine. Three reports of this nature had also been received with methylphenidate. Since the listing of Estelle-35 in the Pharmaceutical Schedule reports of intermenstrual bleeding and headache had been received. Some of these may relate to bioavailability.

For the December 2001 quarter, the number of reports for bupropion had fallen compared with the previous quarter, probably reflecting reduced uptake of the medicine. Of the frequently occurring reactions, there had been 3 reports of anxiety, 3 of aggressive reaction, 5 of insomnia and 3 of urticaria.

A total of 73 reports of injection site reaction with DTaP/HiB vaccines had been received. Most of these occurred following the fourth dose at 15 months. Typically the child has significant inflammation at the injection site with or without erythema, but experiences no pain or discomfort.

The medicine-related deaths in the quarterly report were those 6 included on the agenda (3.1.1), that is they covered only the quarter. Members asked that the Quarterly Reports cover the deaths for the entire 12 months preceding the date of closure of the report.

4. Pharmacovigilance issues

4.1 The safety of bupropion

• Zyban (bupropion) safety update, MCA, 10 Jan 2002

Glaxo SmithKline's submission in response to Medsafe's request:

• Executive summary
• Analysis of spontaneous reports associated with Zyban in New Zealand
• Discontinuation of therapy
• Sudden unexplained death
• Cardiovascular and cerebrovascular morbidity and mortality
• Neuropsychiatric effects
• Effects of timing of dose escalation on adverse events profile
• Comparison with nicotine replacement therapy/benefit-risk assessment for Zyban
• .. Zyban: Commentary on sponsor's response to Medsafe's concerns. Report for MARC 5 Mar 2002

Glaxo SmithKline had been asked to supply data relating to reports of sudden unexplained deaths, cardiovascular and cerebrovascular morbidity and mortality, and neuropsychiatric events with bupropion. The company had also been asked about the effects of timing of dose escalation on the adverse effects profile. Although section 36 of the Medicines Act 1981 had not been invoked, the company had agreed to comply with the terms of the section including complying with any restrictive action which may be the outcome of the consultative process.

.. had prepared a report on the material submitted by Glaxo SmithKline. He noted that the company's conclusions were as follows:

1. No new safety signals have been identified since the launch of bupropion as an antismoking aid.
2. In many of the deaths cardiovascular and other factors could have explained the event; a causal association with bupropion has not been established.
3. Morbidity due to withdrawal from smoking and/or nicotine is a significant confounding factor in determining a causal role for bupropion in smoking cessation.
4. Media attention and an effective pharmacovigilance system, particularly in the United Kingdom, contributed to a greater awareness and enhanced rate of reporting.
5. The reported rate of serious events such as death is lower in users of bupropion than in other smokers.
6. Approximately 2- to 4-fold more smokers need to be treated with nicotine replacement therapy (NRT) to achieve the same benefit as bupropion.

.. commented that while there is no evidence of a causal association between bupropion and the deaths, 75% of them are not sufficiently well-documented to be able to identify confounding factors or to exclude bupropion as a causative agent. The data supplied by the company highlighted the continuing uncertainty around the deaths, and the non-fatal cardiovascular and cerebrovascular events occurring with bupropion.

The Company had conducted a benefit-risk assessment. In one study, with bupropion treatment, smoking quit rates were 23% at 1 year, with NRT the rate was 12% and with placebo 8%. The Company claimed that the typical 1-year quit rate with bupropion was 30%. Using the real life quit rate, the number needed to treat with bupropion to prevent one morbid event was 49 and to prevent one fatal event was 114. The Company analysis concluded that:

"Examination of the benefits confirmed the effectiveness of Zyban in enabling patients to quit smoking and thus avoid the significant medical consequences of continued smoking. Examination of the risks did not reveal any new significant medical consequences of continued smoking."

Members questioned the quit rates found with NRT in the Company analysis. Members reported that the New Zealand experience with NRT and counselling was a 1-year quit rate of > 23%.

The United Kingdom had changed the timing for dosage escalation from day 4 to day 7. The company was unaware of the rationale for this change. It was thought to be a move to reduce adverse effects, particularly seizures, by producing a more gradual increase in plasma levels. However, the mean time to onset of seizures was 9 days under the original dosage schedule. Also 49% of these events occurred at steady state. There appears to be no reason to change the schedule for dose escalation.

It was noted that a prescription event monitoring study is being conducted for bupropion in the UK. Bupropion has also been recommended for consideration for the IMMP. Dr Hill advised that use of bupropion had dropped off in Australia. Some of the Australian adverse reaction reports described events which occurred in the first week of therapy when the users were still smoking. Smoking cessation could not then be a confounding factor. ADRAC has discussed the possibility of arrhythmias being a cause of death, and the company has been asked to address this. Members discussed whether the company should be asked to change the data sheet for Zyban to state that it should be used only as second-line therapy after other therapy including NRT has failed. The article in Prescriber Update (October 2001) which was prepared on Committee advice states as the recommendation of the Committee that bupropion should be used only as second-line therapy. At the time the company had refused to change its data sheet to reflect this advice. Members noted that the evidence of the adverse reaction reports is that bupropion is being prescribed responsibly in New Zealand. Hence the Committee decided to let its advice about use as second-line therapy stand, but agreed not to recommend that the company be asked to change the indication in the data sheet to reflect the Committee's advice.

4.2 Antipsychotics and sudden death

• W A. Ray, et al. Antipsychotics and the risk of sudden cardiac death. Arch Gen Psychiatry 58(12): 1161-1167, December 2001
• C A. Zarate, Jr, J Patel. Sudden cardiac death and antipsychotic drugs. Do we know enough? Arch Gen Psychiatry 58(12): 1168-71, December 2001
• H Cohen, et al. Association of autonomic dysfunction and clozapine. Heart rate variability and risk for sudden death in patients with schizophrenia on long-term psychotropic medication. Brit J Psychiatry 179:167-171, 2001
• K Ronaldson. Antipsychotics and sudden death. Report for MARC, 5 Mar 2002
• Death in association with antipsychotics. Item 4.2, minutes of MARC, 25 Nov 1999

Ray et al conducted a cohort study including a total of 26,749 person years of current moderate dose antipsychotics (>100 mg/day thioridazine equivalents) and 31,864 person-years of current low-dose use. In addition, 37,881 person-years of use of antipsychotics in the past year and 1,186,501 person-years of non-use were included as controls. The study was conducted prior to widespread use of atypical antipsychotics so that 21% used haloperidol, 20% thioridazine, 17% perphenazine and only 1% clozapine. Only 4% used multiple antipsychotic medication.

After standardisation for age and sex, current users of moderate dose antipsychotics had a lower baseline rate of cardiovascular disease as measured by use of medication and admissions than either of the control groups. Current users of moderate dose antipsychotics had a rate ratio of sudden death of 2.39 (95% CI, 1.77-3.22) compared with non-users. The rate ratio for users of low dose antipsychotics was significantly less than that for moderate dose users (p=0.003). The rate ratios for moderate dose users according to pre-existing cardiac disease were 1.60 (0.89-2.87) for no pre-existing disease, 3.18 (1.95-5.16) for mild disease, 2.12 (1.08-4.14) for moderate disease and 3.53 (1.66-7.51) for severe disease. A secondary analysis for the effect of smoking on sudden death found a non-significant 26% increase with smoking.

According to Zarate and Patel, this is the first pharmacoepidemiological study of sudden death with antipsychotics. The study has the following strengths:

• Case definition required documentation from medical records consistent with cardiac arrest (some potentially qualifying cases would be excluded);
• Pre-existing cardiovascular disease was controlled for (i.e. separate analyses) (there could, however, be underdiagnosis);
• Although it did not control for smoking it controlled for factors that would be associated with smoking's effect on the risk of cardiovascular disease;
• Deaths associated with evidence of substance abuse were excluded;
• Comparison with past users and low dose users suggests the measured effect is from the medication (although the moderate dose users may have had more severe mental illness and be subject to other factors increasing risk);
• 96% were on only a single medication.

The authors state that:

"greater attention to pre-treatment cardiac assessment and care to titrate dose to the lowest effective level seem warranted."

Zarate and Patel in their editorial question the wisdom of reducing the dose of antipsychotics merely to reduce the risk of sudden cardiac death since many patients require doses higher than the equivalent of 100 mg/day thioridazine to control their psychosis. Serious morbidity from the mental illness and even mortality is risked by lowering the dose.

Cohen et al conducted heart rate analysis on 21 psychiatric patients treated with clozapine 300-700 mg/day, 18 with haloperidol 5-10 mg/day and 17 with olanzapine 20 mg/day and 53 healthy controls. Patients taking clozapine had higher heart rate (mean 107; cf. 85 for haloperidol, 89 for olanzapine and 62 for the normal controls), lower heart rate variability, and lower high frequency and higher low frequency components than those in the comparative groups. Mean QT_c was higher in those on antipsychotics (503 with clozapine, 518 for haloperidol, 505 with olanzapine) than in the controls (370). The findings are consistent with clozapine decreasing parasympathetic activity, elevating sympathetic activity and prolonging QT-interval. These results have clinical implications, but their connection to an elevated risk of sudden cardiac death with clozapine is unclear.

A US epidemiological study of deaths with clozapine found a rate of 9 per 100,000 person-years for deaths from conduction disorders or sudden death for current users, 2 for recent users and 1 for past users (study cited in the report included on the agenda). Overall the rate of death was lower among current users than among past users. The reduction in death rate during current use was largely accounted for by a reduction in suicide rate compared with past use (relative risk 0.25; 95% CI 0.10-0.30). This study provides evidence of the value of antipsychotic treatment, despite the potential for life-threatening adverse reactions.

The CARM/IMMP database holds 20 reports of death (see 5.2.1), 10% of cases. Ten are either known to be or are possibly cardiac-related. These 10 cases include 4 cases of sudden death, 1 of sudden death associated with pancreatitis, 3 cases of myocardial infarction and 1 of cardiac arrhythmia associated with vomiting, diarrhoea and electrolyte abnormality. The additional case of cardiomyopathy is not relevant here.

Members noted that the Report of the Working Group of the Royal College of Psychiatrists' Psychopharmacology Sub-Group on antipsychotics and sudden death, reviewed at the meeting in November 1999, highlighted that paucity of data on the subject and the need to control serious psychiatric illness with medication in order for the patient to function in society and for the risks associated with disturbed behaviour to be reduced.

Concerning the study by Ray et al, members observed that most patients with psychoses would be on doses considerably higher than 100mg thioridazine. Hence, the study would have been more useful with a further breakdown of dosages at higher strengths to ascertain further the dose relationship. However, members questioned what could be done to reduce the risk to the patient requiring high dose treatment to control mental illness, if there is shown to be a strong dose relationship.

Members asked that any new material concerning sudden death with antipsychotics be brought to its attention.

4.3 Antiepileptics and sudden death

• MJ Hennessy, et al. Sudden withdrawal of carbamazepine increases cardiac sympathetic activity in sleep. Neurology 57:1650-1651, November 2001
• Lamotrigine and worsening epilepsy. Item 11.7, minutes MARC, 25 Nov 1999
• L Nilsson, et al. Risk factors for sudden unexpected death in epilepsy: a case-control study. Lancet 353:888-93, 13 Mar 1999
• K Ronaldson. Sudden unexpected death in epilepsy. Report for MARC, 11 Mar 2002

Hennessy et al studied the effects on cardiac sympathovagal balance during sleep of sudden lowering of carbamazepine dose in patients with medically intractable epilepsy. The authors' intention was to conduct a preliminary study to test the hypothesis that non-compliance may be a cause of sudden unexpected death in epilepsy (SUDEP).

ECG recording was conducted 24 hours before reduction in carbamazepine dose, and then again 4 days after the withdrawal or dose reduction at a time when carbamazepine levels would have fallen by at least 80% (half-life 16-24 hours after repeated administration). The authors examined the low frequency (LF; representing sympathetic and parasympathetic influence) and the high frequency (HF; representing parasympathetic influence) components of the heart rate variability spectrum and calculated LF/HF. Before withdrawal of carbamazepine mean LF/HF was 2.15, and after withdrawal it was 2.65 (19% increase; p=0.018).

The authors commented that an increase of 19% in LF/HF ratio appears to be of physiological importance (comparable changes have been observed with other medication), reflecting a change in sympathovagal balance. This change may be a cause of SUDEP. The risk may be reduced by gradual withdrawal of antiepileptic medication where possible.

Hennessy et al in part designed their study on the basis of the findings of a case-control study of risk factors for SUDEP by Nilsson et al. This study found that frequent changes in dose during the year (3-5) were a risk factor (relative risk 6.08; 95% CI 1.99-18.56, based on 12 cases out of 57). However, other factors were also important and may point to poor seizure control in general being the underlying cause.

Nilsson et al identified 57 cases of definite or probable SUDEP from a study population of 6880 people admitted to hospital at least once over a 10-year period with a discharge diagnosis of epilepsy. Cases had taken at least one of phenytoin, carbamazepine and sodium valproate for at least a year before death. Cases were matched to controls for year of birth, sex and assessment period (3 controls per case). The factor most strongly associated with an increased risk of SUDEP was seizure frequency > 50/year compared with 2 or less per year (RR 10.16; 2.94-35.18). Other factors were onset aged 0-15 years compared with onset aged > 45 years (7.72; 2.13-27.96) and 3 antiepileptic medications compared with one antiepileptic medicine (9.89; 3.20-30.6). None of the 3 medicines were individually associated with an increased risk of SUDEP. These authors concluded that

"The profile of the patient at risk for SUDEP that emerges from our study is a patient with early-onset epilepsy, poor seizure control or inaccurate seizure records, polytherapy with antiepileptic drugs, and frequent dose adjustments."

Members observed that the data for items 4.2 and 4.3 were looking at the same problem, sudden death, from different angles. The relationship between sudden death and QT-prolongation, reduced heart rate variability and disturbance of sympathetic/parasympathetic balance is unknown.

The Committee asked that any new material concerning sudden death with antiepileptics be brought to its attention.

4.4 Safety issues with oral contraceptives

4.4.1 Oral contraceptives and myocardial infarction

• B C. Tanis, et al. Oral contraceptives and the risk of myocardial infarction. New Engl J Med 345:25, 1787, 20 Dec 2001
• L Chasan-Taber, M Stampfer. Oral contraceptives and myocardial infarction - the search for the smoking gun. N Engl J Med 345:1841-2, 20 Dec 2001
• Oral contraceptives and myocardial infarction. Item 12.6.5, minutes MARC, 17 June 1999

Tanis et al conducted a population-based study in the Netherlands of 248 cases of women aged 18-49 years with a first myocardial infarction and 925 controls matched for age, calendar year of index event and area of residence. The odds ratio for myocardial infarction in women taking a second generation OC compared with non-users was 2.5 (1.5-4.1) and for women taking a third generation OC it was 1.3 (0.7-2.5). The presence of a prothrombotic mutation (factor V Leiden or G20210A) had minimal effect on the risk of myocardial infarction. The study found that smoking was a major risk factor for myocardial infarction, increasing the odds ratio to 13.6 (7.9-23.4) for users of OCs. Hypertension, hypercholesterolaemia and diabetes were also significant risk factors. The study included 59 cases taking second generation OCs and 20 cases taking third generation pills.

The results of this study differ from those of an earlier study by Dunn et al which found a higher risk of myocardial infarction with third generation compared with second generation OCs (odds ratio 1.78; 0.66-4.83). In this study there were 20 cases taking second and 20 cases taking third generation OCs. A member commented that the main message of this study was that smoking was an important risk factor. This study had been considered at the meeting in June 1999.

Members observed that the risk of VTE was more important for users of combined OCs, since most are aged < 30 years, when the risk of myocardial infarction is close to zero. For non-smokers aged > 30 years the risk of myocardial infarction is still very small.

The Committee recommended that no further action be taken on the new evidence about myocardial infarction and combined oral contraceptives.

4.4.2 Female hormones and microalbuminuria

• T B. M. Monster, et al. Oral contraceptive use and hormone replacement therapy are associated with microalbuminuria. Arch Intern Med 161:2000-2005, 10 September 2001.

In a case-control study, Monster et al found odds ratios for microalbuminuria, compared with non-users, (urinary albumin excretion of 30-300mg per 24 hours) of 1.90 (1.23-2.93) for users of OCs and 2.05 (1.12-3.77) for users of hormone replacement therapy (HRT), after adjustment for age, hypertension, diabetes, hyperlipidaemia and smoking. Microalbuminuria may be a predictor of cardiovascular risk, but the clinical relevance of the findings of this study are unknown.

Members recommended taking no further action following on from this study.

4.4.3 Funding for oral contraceptives

• P Moodie, Medical Director, Pharmac. Tendering of oral contraceptives. Letter, 25 Jan 2002

Dr Moodie had sent out a letter advising of Pharmac's intention to fully fund one of each chemical combination of the OCs, including the third generation OCs. Members considered that it would be better to fully fund the third generation OCs only on a special authority or in the presence of a "certified condition." Members preferred the "certified condition" option since it was less bureaucratic and required only endorsement of the script. The certified condition in this case would be difficulties of tolerance with a second generation OC.

Dr Jessamine reported that he had discussed the matter with Dr Moodie, Medical Director, Pharmac. Dr Moodie had advised that the Pharmac view was that there should be no barrier to low-income women without contraindications and with difficulties on a second generation pill having available to them a funded third generation alternative. Dr Jessamine had no objection to this reasoning. He reported that Dr Moodie had advised that Pharmac intends mentioning in the schedule entry a sentence to the effect that the third generation OCs are associated with a higher risk of VTE than the second generation OCs.

Members observed that having the third generation OCs unfunded or partially funded sent out a signal about safety, and also created a barrier to use. It was easier to transfer a woman from an unfunded medicine (third generation OC) to a fully funded medicine (second generation). However, members also observed that other second line medicines are funded.

The Committee recommended sending a letter to Pharmac:

• expressing the concern of the Committee at the intention to fully fund a third generation OC,
• suggesting that the third generation OCs be funded only where a woman without contraindications has difficulties with a second generation OC (to be indicated by endorsement of the script "certified condition"), and
• proposing that the listing of any third generation OCs in the Pharmaceutical Schedule mention the higher risk of VTE with these medicines, compared with the second generation OCs.

4.5 Safety issues with hormone replacement therapy

4.5.1 Hormone replacement therapy and dry-eye syndrome

• D A. Schaumberg, et al. Hormone replacement therapy and dry eye syndrome. JAMA 286:2114-2119, 7 November 2001

Using the Women's Health Study cohort (25,665 postmenopausal women), Schaumberg et al found that use of HRT with oestrogen alone was associated with an odds ratio of 1.69 (1.49-1.91) for dry eye syndrome and with oestrogen plus progestogen an odds ratio of 1.29 (1.13-1.48) compared with no HRT use. The risk of dry eye syndrome increased with duration of use (15% every 3 years). Dry eye syndrome was defined as either clinically diagnosed or characterised by severe symptoms (dryness or irritation) or both.

Members observed that the diagnosis was reported in a self-administered questionnaire.

The Committee noted that the study was of interest, but recommended that no additional action be taken at present.

4.5.2 Hormone replacement therapy and breast cancer

• C-L Chen, et al. Hormone replacement therapy in relation to breast cancer. JAMA 287(6): 734-741, 13 February 2002
• Hormone replacement therapy and breast cancer. Item 12.5.2, minutes MARC, 22 Mar 2000.

Chen et al conducted a nested case-control study of 705 post-menopausal women aged 50-74 years with primary invasive breast cancer and 692 randomly selected age-matched women. The incidence of breast cancer was increased by 60-85% in recent long-term users of HRT (oestrogen-only or combined). However, use for ( 57 months (odds ratio 3.07; 1.55-6.06) and current use (3.91; 2.05-7.44) of combined HRT were associated with a greater increased risk of lobular breast cancer. The authors concluded that the increased risk of breast cancer noted with HRT may largely be associated with an increased risk of lobular cancers, rather than the more common ductal cancers.

Members observed that the increasing risk of lobular cancer with duration of use suggested that a real effect was being observed. This study adds to the picture about HRT and breast cancer. It was agreed that there was no need to take further action at present.

4.5.3 Oral versus transdermal hormone replacement therapy and clotting factors

• GDO Lowe, et al. Different effects of oral and transdermal hormone replacement therapies on Factor IX, APC resistance, t-PA, PAI and C-reactive protein. A cross-sectional population survey. Thromb Haemost 86:550-556, 2001
  
• V Marque, et al. The effects of transdermal and oral estrogen/progesterone regimens on free and total protein S in postmenopausal women. Thromb Haemost 86: 713-714, 2001

Lowe et al found that oral HRT was associated with increased plasma levels of Factor IX, activated protein C resistance and C-reactive protein, but decreased levels of tissue plasminogen activator and plasminogen activator inhibitor activity. None of these effects were observed in women using transdermal HRT. Differences in levels of fibrinogen, factor VII and von Willebrand factor were not significant.

Marque et al found that oral but not transdermal estrogen significantly reduced total protein S. However, only free protein S has anticoagulant activity and there was no difference in this level between oral ad transdermal HRT.

Members noted that Lowe et al did not find all coagulation factors to be altered. The clinical significance of these findings needs to be investigated in further studies.

4.5.4 Hormone replacement therapy and quality of life benefit

• M A. Hlatky, et al. Quality-of-life and depressive symptoms in postmenopausal women after receiving hormone therapy. Results from the Heart and Estrogen/progestin Replacement Study (HERS) Trial. JAMA 287(5):591-7, 6 Feb 2002
• KM Rexrode, JE Manson. Postmenopausal hormone therapy and quality of life. JAMA 287:641-2, 6 Feb 2002

In a further publication of the results of the Heart and Estrogen/progestin Replacement Study (HERS) trial, Hlatky et al found that women with flushing at study entry who were randomised to HRT had improved mental health and fewer depressive symptoms compared with those given placebo. In contrast, women without flushing and assigned to HRT had greater declines in quality of life parameters (physical function and energy levels) than women randomised to placebo. Hence, HRT was beneficial only in those with menopausal symptoms.

Members were interested in these results from the point of view of the benefit-risk ratio for HRT, especially when it is used for a purpose other than the relief of menopausal symptoms. The Committee considered that no further action was necessary at present.

4.6 NSAIAs and cardiovascular effects

• Selective COX-2 inhibitors and cardiovascular events. Item 5.2, minutes MARC, 5 Sep 2001
• K Ronaldson. NSAIAs and cardiovascular effects. Report for MARC, 12 Mar 2002

4.6.1 Background

The VIGOR study found a significantly higher rate of myocardial infarction with rofecoxib than with naproxen. A debate has arisen as to whether the result was a random effect or was associated with a cardioprotective effect of naproxen or some effect of rofecoxib which increases the risk of cardiovascular disease (see MARC, 5 Sep 2001, item 5.2).

4.6.2 Cardiovascular effects of non-aspirin NSAIAs

• W A. Ray, et al. Non-steroidal anti-inflammatory drugs and risk of serious coronary heart disease: an observation cohort study. Lancet 359:1-2, 12 January 2002

Ray et al conducted a cohort-based analysis of rates of serious cardiovascular disease (acute myocardial infarction or death from myocardial infarction) in users of non-aspirin NSAIAs (NANSAIAs; 181,441 periods of new use) and matched controls (181,441 periods). The two groups were well-matched for cardiovascular disease risk, which was high in both groups: 67% of NANSAIA uses and 66% of controls had used cardiovascular medication in the past year.

The rates of serious cardiovascular disease were 12.84/1000 person-years for current NANSAIA use and 11.86/1000 person-years for the control cohort (rate ratio 1.05; 95% CI 0.97-1.14). The rate ratio for naproxen was 0.95 (0.82-1.09) and for ibuprofen it was 1.15 (1.02-1.28) - difference significant (p=0.03). The authors found no duration of use or dose association, and the rate ratio for high dose naproxen did not differ from that for ibuprofen. Reanalyses excluding members with previous stroke or myocardial infarction, those with baseline heart failure, and those who died from coronary heart disease did not significantly change the rate ratios.

The authors commented that they found no consistent evidence for the hypothesis that naproxen is cardioprotective, even at doses > 1000mg daily (thought to produce substantial and sustained antiplatelet effect). The size of the difference in rate ratio between naproxen and other NANSAIAs is too small to account for the results of the VIGOR trial. By the authors' estimate this group of patients had approximately 4-times the risk of serious cardiovascular disease of the patients in the VIGOR study. Ray et al did not have data on OTC aspirin use. Aspirin use would have introduced bias only if use had differed according to status of NANSAIA use. Exclusion of those with previous myocardial infarction or stroke, a group most likely to be using aspirin, did not significantly affect the results.

Ray et al concluded that naproxen and other NANSAIAs should not be used for cardioprotection without evidence of an effect from randomised controlled clinical trials.

4.6.3 Cardiovascular effects of rofecoxib

• M A. Konstam, et al. Cardiovascular thrombotic events in controlled clinical trials of rofecoxib. Circulation 104:2280-2288, 2001

Konstam et al (some authors were employees of Merck, others have been paid consultants although not for this work) have conducted a pooled analysis of cardiovascular thrombotic effects in clinical trials using rofecoxib. This study differs from that by Mukherjee et al in that it pooled individual patient data rather than comparing absolute rates in different studies. The endpoint in this analysis was cardiovascular, haemorrhagic and unknown death, non-fatal myocardial infarction and non-fatal stroke.

The relative risks for this endpoint in the various analyses were:

• for rofecoxib (n=6290) versus placebo (n=3482), 0.84 (0.51-1.38)
• for rofecoxib (4549) versus non-naproxen NSAIAs (2755), 0.79 (0.40-1.55)
• for rofecoxib (9083) versus naproxen (7870), 1.69 (1.07-2.69).

The authors concluded that their study had found no evidence that rofecoxib was associated with an increased risk of cardiovascular events compared with placebo or non-naproxen NSAIAs. The data were suggestive of a cardioprotective effect for naproxen.

4.6.4 Effect of NSAIAs on the cardioprotective effects of aspirin

• F Catella-Lawson, et al. Cyclooxygenase inhibitors and the antiplatelet effects of aspirin. New Engl J Med 345(25): 1809-1817, 20 December 2001

Catella-Lawson et al conducted a series of pharmacodynamic studies to ascertain the effects of NSAIAs and paracetamol on inhibition of platelet aggregation by aspirin. Measurements were taken on day 6 after 6 days on the same regimen.

The effect of aspirin on serum thromboxane B₂ and platelet aggregation was attenuated by ibuprofen administered 2 hours before. Ibuprofen administered in the opposite order did not alter the effects of aspirin. Paracetamol and rofecoxib in either order had little effect on aspirin-mediated inhibition of serum thromboxane B₂ or on inhibition of platelet aggregation.

The inhibitory effects of aspirin were not altered by diclofenac administered 2 hours later, but the effects of aspirin on serum thromboxane B₂ were attenuated 33% at 24 hours by ibuprofen (despite the order of administration), recovering to 8% at 2 hours after administration of aspirin. The inhibition of platelet aggregation was attenuated by about 90% at 24 hours by ibuprofen.

Catella-Lawson et al concluded that these results show that ibuprofen inhibits the cardioprotective effects of aspirin. Indomethocin may have a similar effect. However, diclofenac and rofecoxib do not attenuate the effects of aspirin. Paracetamol is a weak, reversible non-specific COX inhibitor.

4.6.5 Effect of aspirin on mortality

• JGF Cleland. No reduction in cardiovascular risk with NSAIDs-including aspirin? Lancet 359:9301 1-3, 12 January 2002

Cleland in his editorial suggested that aspirin may not in fact reduce the risk of cardiovascular death. He claimed that aspirin as prophylaxis has increased the risk of sudden death in every secondary prevention study. The evidence of the benefit of aspirin is based on meta-analyses, which may be subject to publication bias, because of an absence of small published trials not finding a protective effect. He said aspirin may be effective in the first 6 weeks following acute myocardial infarction, but he doubted its long term value, especially in view of its adverse effects and if it is relied on without proper use of other agents of proven effectiveness.

Members asked that published (web site or print) responses to this editorial by Cleland be brought to the next meeting.

4.6.6 Other studies

The Secretary reported that two further re-analyses of earlier studies had been published on thrombotic events with celecoxib and rofecoxib. The second study was from Merck Research Laboratories. The authors concluded that neither COX-2 inhibitor is associated with a higher risk of cardiovascular thrombotic events compared with other NSAIAs.

4.6.7 Conclusions and recommendation

The Committee recognised that the association of NSAIAs with cardiovascular events is an area of intense interest at present. Members considered it would be premature to publish any comment, and recommended that the article on adverse reactions of COX-2 inhibitors do no more than acknowledge that the cardiovascular effects of COX-2 inhibitors are currently an area of study and debate. It was noted that cardiovascular events with COX-2 inhibitors are listed with the adverse reactions of current concern.

4.7 Selective COX-2 inhibitors and renal disease

- S Z. Zhao, et al. A comparison of renal-related adverse drug reactions between rofecoxib and celecoxib, based on the World Health Organisation /Uppsala Monitoring Centre Safety Database. Clinical Therapeutics 23(9):1478-1491, 2001.

This article was included on the agenda because it illustrated use of the methodology used by the WHO to test the statistical significance of an adverse reaction signal, that is the degree to which it appears above the background noise. Using the number of reports of that adverse reaction in the WHO database, the number of adverse reactions recorded for the medicine in question and the number of occurrences of the combination of that adverse reaction and that medicine, the "information component" (IC) is calculated. A figure >1 means the signal appears above the background. The IC is further tested for significance by subtracting 2 standard deviations. If the result is still >1 the signal is statistically significant. The method is one aspect of an approach to investigating the possibility of an association between an adverse event and a medicine. A limitation of the method is illustrated in section 3.1.4.2.

The analysis presented in the study by Zhao et al found that rofecoxib may be associated with greater renal toxicity than celecoxib. These renal effects, if they are confirmed, may contribute to adverse cardiovascular and cerebrovascular outcomes.

One limitation of the study is that all of the reports included in the study would have been reported before the end of 1999, and hence do not cover a very long duration of use of either agent.

Members noted the results of the study and the fact that use would be made of this methodology at future meetings.

The Committee asked CARM staff to prepare for the next meeting a presentation on IC values, to assist members to understand their significance.

4.8 Sibutramine

- S Olsson, WHO. Sibutramine. Email 8 Mar 2002
- Abbott Laboratories. Summary of two Italian cases of death with sibutramine
- Details of sibutramine problems emerge. Scrip No.2728, p.22, 13 Mar 2002

Advice had been received on 11 March 2002 that sibutramine (Reductil) had been taken off the market in Italy following 2 deaths in people taking this medicine. Further information has been scanty. A Scrip article published in March advised that 50 reports of adverse reactions associated with sibutramine had been received in Italy. 15% of these reports were of serious or life-threatening events. The fatal events seemed unlikely to be caused by sibutramine since one occurred in a patient exposed for only 3 days and other causative factors were present in the second patient including gastrointestinal bleeding and haematological abnormalities.

A company report identified a total of 36 deaths worldwide, giving a death rate of 2.86 reports per 100,000 patient-years. Among these fatal cases, there was considerable heterogeneity for cause of death, most cases had alternative etiologies and complicating conditions and in other cases insufficient information was available to identify cause of death.

Dr Rafter commented that cardiovascular events including tachycardia and QT-prolongation, were identified as a matter for concern in a report prepared for the MAAC.

Sibutramine has been monitored in the IMMP since June 2001. At the date of the meeting there were about 1000 patients in the IMMP cohort, with data entry to be completed for the final 4 months of 2001. The only significant adverse reaction reported so far is a case of acute renal failure in a 56-year-old woman who had been taking sibutramine for 25 days. No information was available on prior cardiovascular or renal disease.

The Committee asked that further information on the safety of sibutramine including the MAAC reports, an update from the IMMP monitoring and any further information that becomes available internationally be provided at the next meeting.

4.9 Oral contraceptives and cervical cancer

- V Moreno et al for the International Agency for Research on Cancer Multicentre Cervical Cancer Study Group. Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 359:1085-92, 30 Mar 2002
- N Muñoz et al for the International Agency for Research on Cancer Multicentre Cervical Cancer Study Group. Role of parity and human papillomavirus in cervical cancer: the IARC multicentre case-control study. Lancet 359:1093-101, 30 Mar 2002
- DCG Skegg. Oral contraceptives, parity, and cervical cancer. Lancet 359:1080-81, 30 Mar 2002

On the date of the meeting the Lancet had published on its web site 2 studies on cervical cancer due to be published in the issue of 30 March 2002. The studies which were tabled at the meeting had attracted media attention in the United Kingdom. Medsafe and the Committee considered it to be prudent to prepare a media release in response to these studies to be used should the issue arise in the New Zealand media.

Both studies had been conducted by the International Agency for Research on Cancer Multicentre Cervical Cancer Study Group. The studies pooled data from 8 case-control studies of patients with invasive cervical carcinoma and from 2 studies with carcinoma in situ. The studies included 1676 patients with squamous cell carcinoma and 255 controls who were positive for human papillomavirus (HPV) DNA. The data were from Thailand, Philippines, Morocco, Brazil, Peru, Paraguay, Colombia and Spain.

In the first study, the relative risk of developing cervical cancer in those who had used oral contraceptives (OCs) at some time was 1.42 (95% CI 0.99-2.04). Use for < 5 years was not associated with increased risk. Women who had used OCs for ≥ 5 years had a relative risk of cervical cancer of 3.42 (2.13-5.48). Except where use had persisted for ≥ 5 years, women who had ceased use > 5 years previously had a relative risk for cervical cancer at levels close to those seen in women who had never used OCs. Adjustment for parity and sexual behaviour did not change the risk estimates by much.

The second study looked at the effect of parity on the risk of cervical cancer. The relative risk of cervical cancer with ≥ 7 full term pregnancies was 3.8 (2.7-5.5) compared with nulliparous women.

In his commentary on the studies, Professor Skegg highlighted the finding of the studies that multiparous women and those taking OCs were not more likely to have HPV infection, suggesting that these factors act not by affecting the risk of infection but by promoting progression of infection to invasive cancer. Professor Skegg observed that there may have been some use of injectable contra-ceptives in the studies which was not distinguished from OC use. The studies checked for HPV infection only once, but infection can come and go. He noted that the persistence of any effect after discontinuation of OCs needs to be studied further. He concluded that the results have their greatest significance for the developing world where cervical cancer is common and there is limited access to screening of reliable quality.

Members noted that HPV infection is essential to the development of cervical cancer. Most women are infected at some time but the infection may not persist, depending perhaps on the presence of cofactors. Smoking also appears to be a significant cofactor. The importance of the first of these studies is in that it shows an association between OC and cervical cancer that is not confounded by HPV and sexual behaviour.

Members considered that the study does not give reason to provide different advice regarding cervical screening to those who have taken or are taking an OC compared with those who are not. It was agreed that women who have an abnormal smear should not be advised to stop taking the OC. The OC has protective effects on the risk of cancers of the uterus and ovaries, as well has having other beneficial effects including prevention of unplanned pregnancy and reducing the likelihood of painful and heavy periods.

It was noted that the National Cervical Screening Programme had been successful in reducing the incidence of cervical cancer and the rate of death from cervical cancer. The increase in risk of cervical cancer associated with the OC demonstrated by the study would be reduced by having a regular (every 3 years) smear test. Women most likely to develop cervical cancer are those at risk who are not regularly screened.

The Committee recommended preparing a media statement based on these studies, Professor Skegg's editorial and the comments of the Committee to be released in response to media enquiries.

A postal consultation was held following the meeting. As a result of this consultation the Committee recommended publishing a Prescriber Update article (in the November 2002 issue) on OCs and cervical cancer based on these studies and Professor Skegg's editorial, and reinforcing advice about the importance of cervical screening. Members considered that a low-key approach (publication of no more than a Prescriber Update article) without attaching any urgency was appropriate. The study results reinforce the importance of existing advice about cervical smear testing. There was no need to modify or add to the advice that had already been given.

5. ISSUES FOR INFORMATION ONLY - usually full text supplied

This material was not discussed by the committee. It includes updates on issues already known to the committee, commentaries, review articles and preliminary information on emerging issues. Members were asked to read this material, with the option of requesting that it be discussed, in which case it would have been moved to section 3 or 4, as appropriate.

5.1 Spontaneous reporting programme

5.1.1 Antacids - Omeprazole and Stevens Johnson Syndrome, 48895

5.1.2 Dermatologicals - Isotretinoin and abortion, 49212

5.1.3 Hormones

• Kliogest and DVT, 49066
• Oxytocin and therapeutic effect inadequate, product storage error, 49385

5.1.4 Immunosuppressives

• Thalidomide and peripheral neuropathy, 49472

5.1.5 Vaccines

• DTaP, Hib/HepB and convulsions grand mal, encephalopathy, character change, 49354

5.2 IMMP

5.2.1 Clozapine

• Deaths with clozapine in CARM/IMMP database

5.2.2 COX-2 inhibitors

• Rofecoxib and psoriasis worse, 49777

5.3 Adverse reaction monitoring

• E Heeley, et al. Prescription-event monitoring and reporting of adverse drug reactions. Lancet 358:1872-1873, 1 December 2001.

5.4 Cerivastatin and fatal rhabdomyolysis

• J A. Staffa, et al. Cerivastatin and reports of fatal rhabdomyolysis. New Engl J Med 346(7): 539-540, 14 February 2002.

5.5 Effect of multiple vaccination on the immune system

• P A. Offit, et al. Addressing parents' concerns: do multiple vaccines overwhelm or weaken the infants immune system? Pediatrics 109(1): 124-129, January 2002.

Members asked that a copy of this article be sent to Dr N Turner, Immunisation Advisory Centre.

5.6 Cardiovascular effects of sildenafil during exercise

• A M. Arruda-Olson, et al. cardiovascular effects of sildenafil during exercise in men with known or probable coronary artery disease. A randomized crossover trial. JAMA 287(6):719-725, 13 February 2002
• T H. Marwick, MD, PHD. Safe sex for men with coronary artery disease. exercise, sildenafil, and the risk of cardiac events. JAMA 287(6): 766-767, 13 February 2002.

5.7 Hormone replacement therapy

• MG. Shlipak, et al. Hormone therapy and in-hospital survival after myocardial infarction in postmenopausal women. Circulation 104:2300-2304, 2001.
• JC Stevenson, DCG Skegg. Hormone therapy and heart disease. Lancet correspondence 359:795-6, 2 Mar 2002

5.8 Antipsychotic medication and venous thromboembolism

• R Thomassen, J P. Vandenbroucke and F R. Rosendaal. Antipsychotic medication and venous thrombosis. Brit J Psychiatry 179:63-66, 2001.

5.9 Aspirin and gastrointestinal haemorrhage

• L A. Garcia Rodriguez, et al. Association between aspirin and upper gastrointestinal complications: Systematic review of epidermiologic studies. Br J Clin Pharmacol 52:563-571, 2001.

5.10 Pseudoephedrine and mania

• H Wilson, D Woods. Pseudoephedrine causing mania-like symptoms. NZ Med J 115:86, 22 Feb 2002

5.11 Medication error

• L Eaton. Adverse reactions to drugs increase. BMJ 324:8, 5 January 2002
• UK ADR deaths increase five-fold. Scrip, p.3, 2/4 Jan 2002
• UK Audit Commission. A spoonful of sugar: Medicines management in NHS hospitals. http//www.audit-commission.gov.uk/publications/spoonfulsugar.shtml
• R Voelker. Hospital Collaborative Creates Tools to Help Reduce Medication Errors. JAMA 286(24):3067-3069, 26 December 2001.

5. 12 Global drug safety

• R Balkrishnan, CD Furberg. Developing an optimal approach to global drug safety. J Int Med 250:271-9, 2001

6. Other NZ activities

6.1 Prescriber Update

• April 2002 issue of Prescriber Update - final text
  
• Schedule of planned Prescriber Update articles
• M Tatley. Pulmonary reactions with nitrofurantoin. Draft Prescriber Update article.

6.2 PHARMAC

• Minutes of the Pharmacology and Therapeutics Advisory Committee meeting 15 Nov 2001

6.3 Drug information at Christchurch hospital

• S Gardiner, et al. Fifteen years of drug information in Christchurch Hospital. NZ Med J 114:393-395, 14 September 2001.

6.4 Publications

• DM Coulter. Privacy issues and the monitoring of sumatriptan in the New Zealand Intensive Medicines Monitoring Programme. Pharmacoepidemiology and Drug Safety 2001, Vol 10.
  DCG Skegg. Evaluating the safety of medicines with particular reference to contraception. Statistics in Medicine 20:3557-69, 2001

6.5 CME points for IMMP questionnaires

• DM Coulter. CME points for reporting by general practitioners. Report for MARC, 1 March 2002

7. Other countries

7.1 Australia

• Aust Adv Drug Reactions Bulletin 21:1-4, Feb 2002
• Adverse Drug Reactions Advisory Committee, minutes 2 Nov 2001 meeting
• .. TGA. Leflunomide and pancytopenia. Letter of Feb 2002

7.2 Canada

• Canadian Adv Drug React Newsletter 12:1-6, Jan 2002

8. general reporting

• Case reports considered since Nov 1997, by medicine class
• Case reports considered since Nov 1997, by recommendation
• Case reports for alternative therapies
• Vaccine reports considered since Nov 1997
• Definitions of terms

9. ISSUES FOR INTEREST ONLY - abstract or reference only

This material was not discussed by the committee. It includes updates on issues already known to the committee and preliminary information on emerging issues. This material differs from that in section 5 in being of more peripheral relevance or importance. The material was provided for the interest of the members. Reading it was optional. Members were able to request that the material be discussed at the present or subsequent meeting. Items discussed were moved to section 4.

9.1 Abstract only

• L Letrilliart, et al. Postdischarge adverse drug reactions in primary care originating from hospital care in France. A nationwide prospective study. Drug Safety 24(10): 781-792, 2001.
• C Lacoste-Roussillion, et al. Incidence of serious adverse drug reactions in general practice: A prospective study. Clin Pharmacol Ther 69:458-462, 2001.
• P A. Thurmann. Methods and systems to detect adverse drug reactions in hospitals. Drug Safety 24(13): 961-968, 2001.
• J Hippisley-Cox, et al. Primary Care. Antidepressants as risk factor for ischaemic heart disease: case-control study in primary care. BMJ 323:666-669, 22 September 2001.
• The American Lung Association Asthma Clinical Research Centers. The safety of inactivated influenza vaccine in adults and children with asthma. New Engl J of Med 345:1529-1536, 22 November 2001
• N Eid, et al. Decreases morning serum cortisol levels in children with asthma treated with inhaled fluticasone propionate. Pediatrics 109:217-221, Feb 2002
• E. Ekman, E. Vinge. Zyprexa (olanzapine) dilated cardiomyopathy. Medical Products Agency 12: 76-77, December 2001. Reactions abstract

9.2 Reference only

• P S. Wang, et al. Zolpidem use and hip fractures in older people. J Am Geriatr Soc 49:1685-1690, 2001
• S P. Johnsen, et al. Hospitalisation for upper gastrointestinal bleeding associated with use of oral anticoagulants. Throm Haemost 86:563-568, 2001
• N Shakoor, et al. Drug-induced systemic lupus erythematosus associated with etanercept therapy. Lancet 359: 579-580, 16 February 2002.
• French Perinatal Cohort Study Group. Risk of early febrile seizure with perinatal exposure to nucleoside analogues. Lancet 359(9306): 583-584, 16 February 2002.

The meeting ended at 5pm.