Article
Serious Blood Dyscrasias and Drug-Attributable Fraction: Estimates from the Berlin Case-Control Surveillance Study - FAKOS
Search Medline for
Authors
-
E. Bronder
- Institut für Klinische Pharmakologie, Charité Campus Mitte, Universitätsmedizin Berlin
-
F. Andersohn
- Institut für Klinische Pharmakologie, Charité Campus Mitte, Universitätsmedizin Berlin
-
A. Klimpel
- Institut für Klinische Pharmakologie, Charité Campus Mitte, Universitätsmedizin Berlin
-
E. Garbe
- Institut für Klinische Pharmakologie, Charité Campus Mitte, Universitätsmedizin Berlin
| Published: | September 30, 2004 |
|---|
Outline
Text
Background
Drugs are an important cause of serious rare blood dyscrasias. The drug-attributable fraction has been estimated for acute agranulocytosis to range from 70% to 90% and for aplastic anaemia from 5% to 86%. For other blood dyscrasias, data on the proportion of drug-induced cases are not available.
Objective
To estimate the drug-attributable fraction of serious rare blood dyscrasias using data from the ongoing Berlin case-control surveillance study of serious rare blood dyscrasias (FAKOS), which is supported by a grant from the Federal Institute for Drugs and Medical Devices, Germany.
Material and Methods
Adult patients with first occurrence of serious rare blood dyscrasias are identified in 52 hospitals and 14 haematological practices in Berlin, Germany. Patients with acute agranulocytosis (AGR), aplastic anaemia (AA), immune haemolytic anaemia (IHA), immune thrombocytopenia (ITP) or thrombotic thrombocytopenic purpura / haemolytic uraemic syndrome (TTP/HUS) were included. Causality between drug intake and each blood dyscrasia is assessed by applying the criteria of the World Health Organization causality assessment method. Causality assessment includes the categories „certain", „probable", „possible" and „unlikely".
Results
The analysis included 171 cases (October 2000 to November 2003): 31 x IHA, 30 x AGR, 19 x AA, 81 x ITP, and 10 x TTP/HUS. The drug-attributable fraction based on „possible" is estimated for IHA n=8 (26%; CI 12%-45%), AGR n=29 (97%; CI 83%-100%), AA n=4 (21%; CI 6%-46%), ITP n=20 (25%; CI 16%-36%) and TTP/HUS n=2 (20%; CI 3%-56%).
Conclusion
These data, to our knowledge, provide the first systematic estimates of the drug-attributable fraction for IHA, ITP and TTP/HUS. Since our causality assessment is based on „possible" instead of „probable", the drug-attributable fraction may have been overestimated. However, the category „probable" may lead to an underestimate of the drug-attributable fraction for the following reasons: (1) an improvement/recovery after drug withdrawal (positive „dechallenge") cannot be expected in the case of AA; (2) a positive dechallenge may be masked, since other treatment is often started at the same time so that the effect of drug withdrawal cannot be evaluated. Overall, our analysis suggests that a substantial fraction of blood cytopenias may be attributable to drug therapy. It is therefore important to ascertain a detailed medication history in case of a blood dyscrasia to avoid unintended re-exposure to a causative agent.
Conflict of interest: The Berlin Case-Control Surveillance Study - FAKOS is supported by a grant from the Federal Institute for Drugs and Medical Devices (BfArM), Germany.
