De fire abstractene er publisert i Drug Safety:

• Vorren S, Småbrekke L, Nordmo E. Phu Chee – a herbal product containing dexamethasone. Drug Saf 2007; 30(10): 929.
• Røed J, Schjøtt J. Hepatitis Caused by a Natural Product? Drug Saf 2007; 30(10): 944.
• Frost SK, Schjøtt J. Drug Use in Pregnancy: Are There Differences Between Common Sources of Information? Drug Saf 2007; 30(10): 954.
• Narum S, Solhaug V, Myhr K, Kringen MK, Johansen PW, Brors O. Anticoagulant-related bleeding episodes: an analysis of reports from the Norwegian pharmacovigilance database. Drug Saf 2007; 30(10): 978.

Phu Chee – a herbal product containing dexamethasone
S. Vorren, L. Småbrekke, E Nordmo.

During 2004 the Regional Pharmacovigilance Centre of North Norway received 11 reports on adverse drug reactions after use of a herbal product named Phu Chee. The product was sold over the internet, by mail order and local salesmen, and was claimed to relive symptoms from rheumatoid arthritis and arthrosis. According to the package Phu Chee only contained herbal extracts. Rheumatic patients found the product to be effective against some symptoms of these diseases, and Phu Chee soon became highly popular in a local area in North Norway. Some patients who had used Phu Chee as long as two years, developed moon face, increased body hair, weight gain, and lowered glucose tolerance. Frequent infections and increased tendency for bruises and wounds was also reported. Patients who had stopped using Phu Chee abruptly, developed symptoms consistent with low cortisol, like low blood pressure, dizziness, nausea, vomiting, and stomach ace. The symptoms disappeared after treatment with prednisolone 10-15 mg.

Analyses of Phu Chee by The Norwegian Medicines Agency (NoMA) found each tablet to contain 0,4-0,5 mg dexametasone and an antihistamine. The patients had used 3-9 tablets a day, resulting in a daily dosage of 1,2 to 4,5 mg dexamethasone. NoMA also found dexametasone in two other herbal products named Lin Chee and Active Rheuma Plus. NoMA banned the sale of these products, and decided to press charges.

These cases illustrate serious adverse drug reactions from a herbal product after illegal addition of dexamethasone and an antihistamine. This emphasises that there are reasons for scepticism when herbal product gives rapid relief of symptoms from serious diseases.

Hepatitis caused by a natural product?
J. Røed, J. Schjøtt.

Background: Natural products are often marketed as safe. Both national and international reports show that natural products can cause serious adverse reactions. The reactions can be due to the claimed constituents, pesticides, residues from the production processes or added drugs. Green tea has in some cases been associated with hepatotoxicity and is a constituent of several natural products marketed in Norway.

Method: Presentation of a case of hepatitis and jaundice associated with the natural product Lotus-f3. Search for reports or questions associating natural products with hepatitis or jaundice. Searches were made in the national adverse drug reaction database and the database of queries to the drug information centres in Norway.

Case report: A 56-years old woman got raised hepatic test results and developed jaundice 3 weeks after starting treatment with the natural product Lotus-f3, in doses recommended by the manufacture. The woman had rheumatic arthritis treated with etanercept for over a year. She was hospitalized with test results consistent with hepatitis and jaundice. Viral or immunologic explanation were ruled out. Both etanercept and the natural product Lotus-f3 were withdrawn and 5 weeks later the liver values were near normalized without any other treatment. The patient was later restarted on etanercept without return of liver affection.

Results: Lotus-f3 is marketed as a weight reducing product and the claimed constituents are: Lotus arabicus, Citrus aurentium, Camellia sinensis, Fraxinus excelsior, Betula pendula and chrome.
Only green tea (Camellia sinensis) is former associated with liver damage. Hepatotoxicity is primarily linked to green tea extract products in pill form and most cases of toxicity have had a hepatocellular-cholestatic presentation
In the Norwegian adverse drug reaction database there were no former reports on the natural product Lotus-f3, but a similar case was recently reported. Other natural products containing green tea are suspect drugs in 3 out of 6 reports of hepatitis or jaundice since 2003.
The regional drug information centres in Norway had registered six questions concerning the possible association between natural products and hepatitis or jaundice since 1995. In to of these questions the suspected natural products contained green tea, among other constituents.

Conclusion: Both the national pharmacovigilance database and the database containing queries to the regional drug information centres are sources of information about possible associations between natural products and serious adverse reactions.

Drug use in pregnancy – are there differences between common sources of information?
S. Frost, J. Schjøtt.

Background: Safety regarding use in pregnancy is not established for many drugs[1]. Inadequate or inconsistent information from different sources can lead to confusion and thereby counteract the intention of the drug information[2]. It is therefore important to measure clinically important differences between drug information sources.

Objective: To compare two easy accessible Norwegian sources providing advice on drug use in pregnancy, the product monographs in Felleskatalogen (FK) and the regional Drug Information Centres (DIC`s). 
Methods: Advice on drug use in pregnancy provided by DIC`s in 2003 and 2005 were compared to advice in the product monographs for the respective drugs in FK. Any advice were placed in one of four defined categories:
1. The drug can be used by pregnant women
2. The drug should only be used by pregnant women after a risk-benefit assessment
3. The drug should not be used by pregnant women
4. No information on drug use in pregnancy.

Results: 443 drug advice were compared. Seven out of ten of the most frequently enquired drugs were drugs acting on the nervous system. For 208 comparisons (47%), advice on drug use in pregnancy differed between DIC`s and FK. In the case where a DIC gave the advice “can be used” (Category 1), FK gave the same advice in only 23% of the cases, and advised against the use (Category 3) in 21% of the cases. However, when FK gave the advice “can be used” (Category 1), DIC`s agreed in 94% of the cases, and provided no advice against use (Category 3) in any of these cases.

Conclusions: We have demonstrated considerable differences between two Norwegian sources providing advice on the use of drugs in pregnancy. Based on the knowledge that health care providers choose sources randomly, our results are troublesome. The implications are therapeutic consequences for the pregnant woman and her foetus. The present results suggest a need for regulatory actions to coordinate and quality-assess sources of information on clinical important issues, such as the safe use of drugs in pregnancy. We believe that the problem with heterogeneous drug information on this subject is not confined to Norway, and that our results could be of international interest.

References:
1. Lagoy CT, Joshi N, Cragan JD, Rasmussen SA. Medication use during pregnancy and lactation: an urgent call for public health action. J Womens Health (Larchmt). 2005 Mar;14(2):104-9.
2. Bjerrum L, Foged A. Patient information leaflets–helpful guidance or a source of confusion?
Pharmacoepidemiol Drug Saf 2003 Jan-feb; 12 (1): 55-9.

Anticoagulant-related bleeding episodes: an analysis of reports from the Norwegian pharmacovigilance database.
S. Narum, V. Solhaug, K. Myhr, MK Kringen, PW Johansen, O. Brors.

Introduction: Oral anticoagulants are effective in stroke prevention in patients with atrial fibrillation [1], but are also the drugs most frequently implicated in bleeding events in Norway [2]. Several studies have shown that the initial phase of therapy conveys the highest risk of haemorrhage [3,4]. This was also seen in a previous Norwegian study based on an analysis of reports from the Norwegian pharmacovigilance database [4].

Aim of the study: The aim of our study was to describe the pattern of spontaneously reported adverse drug reactions to vitamin K antagonists by analysing data from the national spontaneous reporting system in Norway.

Methods: We performed a search in the Norwegian pharmacovigilance database and identified all spontaneous reports on anticoagulant-related bleedings received during a 3 year period from 2003 to 2006.

Results: There were 289 case reports of haemorrhages associated with vitamin K antagonists (46 % of the total number of reports on bleeding events). The mean age was 75.9 years and 40% were women. Of the reported anticoagulant-related bleedings; 60% were cerebral, 24% gastrointestinal and 16% localized to other organ systems. Forty-eight percent of the bleedings were fatal. In 22% of the cases warfarin was the only medicine used, 48% used 2-5 medicines and 30% used more than 5 concomitant medicines at the time of bleeding. Most of the reported haemorrhages (57%) occurred after more than one year of warfarin use. Ten percent of the bleedings occurred during the first month of therapy. Among the early bleedings, 59% occurred during the first five days.

Conclusion: In our study the majority of bleedings (57%) related to warfarin occurred after more than a year on therapy. This is in marked contrast to a recent Norwegian study, in which most of the bleedings were reported to occur during the first month. Our findings indicate a change in the reported bleeding events in Norway. The difference may be explained by altered use of warfarin and comedications or altered reporting by physicians. Our study also shows that in most of the reported haemorrhages the patients used one or more concomitant medicines.