Infusion devices postmarketing survey: analysis of the reports made in 1998 to the French agency for medical devices.

Objectives: To analyze retrospectively the data base of the French national vigilance reporting system concerning infusion devices during the year 1998. Methods: Each report has been reviewed and classified, except implantable catheter ports. Results: The study included 309 reports. Among them, 28% concerned infusion pumps and 72% single use devices, i.e. catheters, infusion lines, taps, connectors, etc. We observed six deaths and 23 severe incidents. Only 25% of devices could be formally investigated by manufacturers. The origins of incidents were very different between infusion pumps and single use devices. The first showed simple failures, software errors, maintenance omissions or errors, and problems related to obsolescence of the device. The second revealed manufacturer quality insurance failures, design and utilization errors, and problems related to the technique itself rather than device. Discussion: Our findings are similar to the few data available in international literature,. They address the need to obtain an adequacy between the performances of devices which vary with generations of devices and the clinical risks: newest and safest devices should be preferred for the infusion of the most critical drugs. (C) 2000 Editions scientifiques et medicates Elsevier SAS.