Prototype micropump for insulin administration based on electrochemical bubble formation.

As part of the development of a percutaneous artificial pancreas islet, an insulin injection micropump was fabricated using micromachine techniques. The major components of the device were a thin-film, two-electrode system incorporated in a closed compartment, a silicone rubber diaphragm to separate an electrolyte solution from an insulin solution, a reservoir for insulin and a microneedle attached to the outlet. Hydrogen bubbles were formed on a platinum working electrode when current was applied. This caused the diaphragm to deform and to exert pressure on the insulin solution in the reservoir on the other side of the diaphragm. The injection of insulin was smooth and the injection rate could be controlled by the electrode potential or current. When the insulin solution was injected into streptozotocin-induced diabetic rats, a decrease in plasma glucose level (PGL) was observed which was dependent on the dose of insulin. No substantial difference was observed compared to manual injection.

Determination of the activities of glutamic oxaloacetic transaminase and glutamic pyruvic transaminase in a microfluidic system.

A microfluidic system for the analysis of the activities of glutamic-oxaloacetic transaminase (GOT) and glutamic-pyruvic transaminase (GPT) was fabricated. The device consists of a glass chip with a micro-electrochemical L-glutamate sensor and a polydimethylsiloxane (PDMS) sheet with a Y-shaped micro-flow channel. A sample solution and a substrate solution for the enzymes were introduced from two injection ports at the end of the flow channel. When the flows were stopped, substrates in a solution mixed immediately with either of the enzymes by diffusion in a mixing channel. L-glutamate produced by the enzymatic reaction of GOT or GPT in the flow channel was detected by using the L-glutamate sensor. A distinct current increase was observed immediately after mixing, and the initial slope of the response curve varied in proportion to the activity of GOT or GPT. The relation between the slope of the response curve and the enzyme activity was linear between 7 and 228 U l-1 for GOT and 9 and 250 U l-1 for GPT. The quality of the response curve was improved with an increase in the channel height. The measurement based on the rate analysis in the micro-flow channel facilitated the reduction of the influence of interferents. The influence of the viscosity of the sample solution was also checked for the analysis of real samples. The determination of the enzyme activities was also conducted in a system with micropumps fabricated for a sample injection. Two solutions could be mixed in the mixing channel, and the activity of the enzymes could be measured as in the experiments using microsyringe pumps.