Corrosion is a prime concern for active implantable devices. In this paper we review the principles underlying the concepts of hermetic packages and encapsulation, used to protect implanted electronics, some of which remain widely overlooked. We discuss how technological advances have created a need to update the way we evaluate the suitability of both protection methods. We demonstrate how lifetime predictability is lost for very small hermetic packages and introduce a single parameter to compare different packages, with an equation to calculate the minimum sensitivity required from a test method to guarantee a given lifetime. In the second part of this paper, we review the literature on the corrosion of encapsulated integrated circuits (ICs) and, following a new analysis of published data, we propose an equation for the pre-corrosion lifetime of implanted ICs, and discuss the influence of the temperature, relative humidity, encapsulation and field-strength. As any new protection will be tested under accelerated conditions, we demonstrate the sensitivity of acceleration factors to some inaccurately known parameters. These results are relevant for any application of electronics working in a moist environment. Our comparison of encapsulation and hermetic packages suggests that both concepts may be suitable for future implants.

Hermeticity is crucial for the long-term implantation of electronic packages. Pushed by advances in micromachining, package volumes are decreasing and current leak detection methods are no longer sensitive enough. This article reviews the limits of the most common methods and exposes their inadequateness for medical electronic applications where the device's life is 50 years or longer.

This paper presents an integrated stimulator that can be embedded in implantable electrode books for interfacing with nerve roots at the cauda equina. The Active Book overcomes the limitation of conventional nerve root stimulators which can only support a small number of stimulating electrodes due to cable count restriction through the dura. Instead, a distributed stimulation system with many tripole electrodes can be configured using several Active Books which are addressed sequentially. The stimulator was fabricated in a 0.6-μm high-voltage CMOS process and occupies a silicon area of 4.2 × 6.5 mm(2). The circuit was designed to deliver up to 8 mA stimulus current to tripole electrodes from an 18 V power supply. Input pad count is limited to five (two power and three control lines) hence requiring a specific procedure for downloading stimulation commands to the chip and extracting information from it. Supported commands include adjusting the amplitude of stimulus current, varying the current ratio at the two anodes in each channel, and measuring relative humidity inside the chip package. In addition to stimulation mode, the chip supports quiescent mode, dissipating less than 100 nA current from the power supply. The performance of the stimulator chip was verified with bench tests including measurements using tripoles in saline.

No abstract

Gastrointestinal stimulator implants have recently shown positive results in treating obesity. However, the implantation currently requires an invasive surgical procedure. Endoscopy could be used to place the gastric stimulator in the stomach, hence avoiding the riskier surgery. The implant then needs to go through the oesophagus and be located inside the stomach, which imposes new design constraints, such as miniaturization and protecting the electronic circuit against the highly acidic environment of the stomach. We propose to protect the implant by encapsulation with silicone rubber. This paper lists the advantages of this method compared to the more usual approach of a hermetic enclosure and then presents a method to evaluate the underwater adhesive stability of six adhesive/substrate couples, using repeated lap-shear tests and an elevated temperature to accelerate the ageing process. The results for different adhesive/substrate couples tested, presented on probability plots, show that FR4 and alumina substrates with MED4-4220 silicone rubber are suitable for a first implantable prototype. We then compare these with the predicted lifetimes of bonds between historical standard silicone rubber DC3140 and different substrates and describe the encapsulation of our gastrostimulator.

No abstract

Background Minimally invasive surgical (MIS) techniques are considered the gold standard of surgical interventions, but they have a high environmental cost. With global temperatures rising and unmet surgical needs persisting, this review investigates the carbon and material footprint of MIS and summarizes strategies to make MIS greener. Methods The MEDLINE, Embase, and Web of Science databases were interrogated between 1974 and July 2021. The search strategy encompassed surgical setting, waste, carbon footprint, environmental sustainability, and MIS. Two investigators independently performed abstract/full-text reviews. An analysis of disability-adjusted life years (DALYs) averted per ton of carbon dioxide equivalents (CO2e) or waste produced was generated. Results From the 2456 abstracts identified, 16 studies were selected reporting on 5203 MIS procedures. Greenhouse gas (GHG) emissions ranged from 6 kg to 814 kg CO2e per case. Carbon footprint hotspots included production of disposables and anaesthetics. The material footprint of MIS ranged from 0.25 kg to 14.3 kg per case. Waste-reduction strategies included repackaging disposables, limiting open and unused instruments, and educational interventions. Robotic procedures result in 43.5 per cent higher GHG emissions, 24 per cent higher waste production, fewer DALYs averted per ton of CO2, and less waste than laparoscopic alternatives. Conclusion The increased environmental impact of robotic surgery may not sufficiently offset the clinical benefit. Utilizing alternative surgical approaches, reusable equipment, repackaging, surgeon preference cards, and increasing staff awareness on open and unused equipment and desflurane avoidance can reduce GHG emissions and waste.

Objective. Finite element modelling has been widely used to understand the effect of stimulation on the nerve fibres. Yet the literature on analysis of spontaneous nerve activity is much scarcer. In this study, we introduce a method based on a finite element model, to analyse spontaneous nerve activity with a typical bipolar electrode recording setup, enabling the identification of spontaneously active fibres. We applied our method to the vagus nerve, which plays a key role in refractory epilepsy. Approach. We developed a 3D model including dynamic action potential propagation, based on the vagus nerve geometry. The impact of key recording parametersinter-electrode distance and temperatureand uncontrolled parametersfibre size and position in the nerveon the ability to discriminate active fibres were quantified. A specific algorithm was implemented to detect and classify action potentials from recordings, and tested on six rat in-vivo vagus nerve recordings. Main results. Fibre diameters can be discriminated if they are below 3 µm and 7 µm, respectively for inter-electrode distances of 2 mm and 4 mm. The impact of the position of the fibre inside the nerve on fibre diameter discrimination is limited. The range of active fibres identified by modelling in the vagus nerve of rats is in agreement with ranges found at histology. Significance. The nerve fibre diameter, directly proportional to the action potential propagation velocity, is related to a specific physiological function. Estimating the source fibre diameter is thus essential to interpret neural recordings. Among many possible applications, the present method was developed in the context of a project to improve vagus nerve stimulation therapy for epilepsy.