Encapsulation Materials for Implantable FES Systems – a Case Study. LPJ Kenney 1 , H Hermens 1 , D Francis 2 , G Bultstra 3 , J Holsheimer 3 and AJ Verloop 3 1 - Roessingh Res. & Dev. Bv, Enschede, The Netherlands; 2 -Finetech Medical Ltd, Welwyn Garden City, Herts. UK; 3 - University of Twente, Enschede, The Netherlands. Email: l.kenney@rrd.nl Abstract - This paper describes problems encountered with the encapsulation of an implantable receiver in epoxy resin and describes the solution adopted. The implant concerned is a receiver unit for a new peroneal nerve stimulator. The original receiver consisted of two ‘D’-shaped receiver coils and associated electronic components, mounted on two ceramic substrates. The components were encapsulated in a layer of epoxy resin, coated with silicon rubber. Evidence was found questioning the suitability of the epoxy encapsulation. The strongest evidence came from inspection of a receiver that had been explanted from an animal 3 ½ years earlier and left in water. When the receiver was examined at the end of this period a crack was noticed in the epoxy and corrosion was found on a receiver coil. Other samples also showed similar problems with cracks. Based on the evidence described above, it was decided to change to an alternative material, silicon rubber as the encapsulant. There is extensive experimental and theoretical evidence to show that silicon rubber adheres well to components both during manufacture and during sterilisation. The reliability of this bond between encapsulant and components has been shown to be a major factor determining the success of encapsulations. From a more practical perspective, there is also considerable clinical evidence demonstrating the effectiveness of this material as an encapsulant and this eases the regulatory problems. Keywords: encapsulants, epoxy, silicon rubber, implant. 1. Introduction and background This paper describes a case study of problems encountered with the encapsulation of an implantable FES device. The implant is part of a novel two channel device for the treatment of dropped foot [1]. The receiver consists of two galvanically separated channels with passive components placed inside each of two D-shaped coils. Two leads leave the receiver, each of which connects the receiver to a pair of miniature sub-epineural bipolar electrodes. One electrode pair is to be sited on the superficial peroneal nerve and the other is to be sited on the deep peroneal nerve. The original design work began in the early 1990s and a low shrinkage epoxy resin was chosen as the main encapsulant. Overlaying the epoxy was a thin layer of silicon rubber. The original design is shown in figure 1. 2. Implant failure In the early 1990s, animal testing of the new implant began. A number of implants were tested in goats and rabbits and satisfactory function was observed [2]. At the end of the animal tests, the implants were removed and histological examination of the nerve tissue was carried out. The removal entailed cutting of the lead wires between the body of the receiver and the electrodes. Out of interest, two of the explanted receivers were placed in a beaker of water and left. In 1996, the trials were about to commence, following approval from the relevant Ethics Committees. The remaining problem was finding an authority willing to insure the trials. This proved to be extremely difficult and, due to a shortage of manpower, the project came to a halt. In December 1998 concerted work on the project began again with the appointment of a full time researcher. It was assumed at the time of appointment that the device was more or less ready for use and hence that trials could begin within a few months. However, this proved to be a rather over-optimistic assumption. A number of outstanding practical problems quickly became apparent and a major problem with the implant Figure 1: Original epoxy-encapsulated implant