The integrated blood-collection system as a vehicle into complete clinical laboratory automation.

A rationale is offered and methodology illustrated for integrating the fundamental steps involved in the collection and processing of blood for laboratory evaluation. The approach taken in the development of these concepts and components greatly extends the possibilities of laboratory systems integration without upsetting established modalities. A prototype design of the integrated blood-collection system integrates blood collection, cellular separations, sample transfer to stable storage without chemical mediators, and sample presentation for chemical analysis (e.g., precision metering) while preserving patient identification. A sophisticated, multi-chambered blood-collection container is the site of all blood sample processing and transfer steps. This device is supported by a compact, robotic centrifuge of unique design and a transfer mechanism to facilitate sample delivery for analysis within a diagnostic instrument. The confluence of these individual components into a single integrated system provides the means to completely automate the processing of blood samples, after sample collection, eliminating all manual transfer steps and any external exposure of blood interfaces outside the diagnostic instrument. Configurational derivatives of the Integrated Blood-Collection System offer choice of skin or venipuncture procedure, rapid plasma extraction for micro- or macro-collected volumes, and sample delivery by either aspiration or direct metering of discrete 10-microL samples from the collection container. The skin-puncture configuration provides the opportunity within a single device to collect and process up to 500 microL of sample by capillarity from a skin prick.

"Architextured" fluid management of biological liquids.

Novel "architextured" liquid-spreading and capillary flow systems are described as an adjunct to diagnostic procedures that range from the fully automated to techniques requiring no instrumentation. These systems were developed because of the observation that local features of macroscopic surfaces--roughness and chemical heterogeneity--influence both the spreading rate and equilibrium shape of liquid drops on solid surfaces. By use of organized patterns of surface structure complemented by surface chemical modification, this uncontrolled and undirected capillary flow is recast into flow behavior that is remarkably reproducible. Flow fronts are engineered to have prescribed shapes, and flow rates that are insensitive to the rheological properties of the biological liquids are obtained. To demonstrate the utility of fluid-management principles, we describe the implementation of architextured capillary configurations in microvolume liquid dispensers and potentiometric analytical systems.

Multilayer film elements for clinical analysis: general concepts.

Dry, thin films containing all necessary reagents for clinical analysis by colorimetry have been designed. Reagents in a matrix of hydrophilic polymer are coated on top of a transparent plastic base. A white isotropically porous polymer spreading layer, 80% void volume, is coated over the reagent layer(s). In the analysis, a drop (typically 10 microliter) of undiluted serum or other fluid is touched to the spreading layer. The fluid spreads rapidly and uniformly through the pore structure, filling a void volume corresponding to the drop volume. Water and low-molecular-weight components diffuse from the spreading layer into the reagent layer(s), initiating the reaction sequence. The spreading layer acts also as a white optical diffuser for reflection densitometry. Optical reflection density is linearized through use of the function developed by Williams and Clapper [J. Opt. Soc. Am. 43, 595 (1953)] to convert reflection to transmission density. A wide variety of chemical assays are compatible with this format. As an example, for the glucose film we found coefficients of variation of 1.5% in predicting glucose concentrations in control sera during 20 days. Results for glucose concentrations in several hundred patients' sera by the present method were very cose to those obtained with the Center for Disease Control's hexokinase reference method.