| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Oak Ridge Poster Sessions |
AVL Medical Instruments, Hans List Platz 1, A-8020 Graz, Austria
a author for correspondence: fax 43-316-787-635, e-mail Bernhard.Schaffar{at}avl.com
Fast point-of-care testing of metabolite concentrations has become more common in critical care analysis (1)(2)(3). We developed an integrated biosensor cassette for the determination of glucose, lactate, and urea in microsamples of undiluted whole blood or plasma. The biosensor cassette requires no maintenance and can be used for 1 week or 500 samples after insertion into an AVL OMNI® blood gas and electrolyte instrument. The linear analytical ranges of these sensors include the physiological ranges and most of the pathological ranges, e.g., glucose up to 40 mmol/L, lactate up to 20 mmol/L, and urea up to 40 mmol/L.
This integrated planar solid state sensor (38 x 20 mm) is
produced using thick-film technology, allowing high reproducibility at
very competitive manufacturing costs. The sensor cassette consists of
12 electrode spots (each 1.2 mm in diameter); 9 spots are used for
glucose and lactate, and 3 are used for urea. The total fill volume of
a cassette is 44 µL, including 6 µL for an external
maintenance-free reference electrode for the potentiometric urea sensor
system. The integrated biosensor is thermostated in the instrument at
25 °C to avoid interference by varying sample temperatures. Results
from a whole blood or plasma sample can be obtained within 90 s
after sample application. A schematic of the planar sensor is shown in
Fig. 1
.
|
The amperometric three-electrode glucose sensor, including working, counter, and reference electrodes, is based on the glucose oxidase method using hydrogen peroxide detection with electrochemical recycling of the oxygen, thus making the sensor essentially free from interference from the oxygen content of the sample. The lactate sensor is constructed identically but with lactate oxidase immobilized rather than glucose oxidase. To avoid interference problems for both sensors, an amperometric interference compensation electrode is used, which compensates for the electrochemical interference of substances such acetaminophen (paracetamol), uric acid, and ascorbic acid. This sensor uses an inactive immobilized protein rather than glucose oxidase. All three amperometric sensors are calibrated with acetaminophen to achieve perfect correction of interference. Acetaminophen is a good model substance for other exogenous and endogenous interfering substances. [For further details, see Ref. (4).] The potentiometric sensor for urea determinations is based on an ammonium ion-selective electrode (ISE) based on nonactin with immobilized urease to convert urea to bicarbonate and ammonium ions. To avoid interference from native ammonia in the sample, an additional ammonium ISE is used as well as a potassium ISE based on the ionophore valinomycin to compensate for interference from potassium ions at the ammonium sensors. Because the detection principle is via ammonium ISE and the ammonium/ammonia equilibrium is pH dependent, the system will give falsely low results at pH >8.0 and falsely high results at pH <6.8. (e.g., a sample with 9.0 mmol/L urea will typically read 8.4 mmol/L at pH 8.0 and 9.8 mmol/L at pH 6.8). However, these numbers further depend on the buffer capacity of the sample (5). The enzymes of this integrated biosensor are immobilized via a novel immobilization procedure based on acrylate prepolymers (6).
The sensors are calibrated automatically in the reference interval every hour and in the high pathological range as well as for interferents every 8 h. A recalibration with the flushing solution after each measurement compensates for potential baseline drift. Changes in the liquid junction potential of the potentiometric reference electrode are avoided by renewal of the bridge electrolyte solution after each measurement. Because these sensors read directly in undiluted media, the results are normalized to plasma results, regardless of the sample analyzed, e.g., plasma, heparinized whole blood, or aqueous samples. This is important for direct comparison with the other, mainly photometric methods currently used in clinical chemistry. Basically, similar corrections are made, such as those made to convert directly read ISE results to values obtained with flame photometry (7)(8)(9). In addition, whole blood samples are corrected for hematocrit, determined by conductivity measurements within the sample path.
When working with these new direct-reading metabolite sensors, it is important to note that metabolite determinations in heparinized whole blood must be performed soon after the sample is collected because glycolysis in the red blood cells decreases glucose and increases lactate concentrations, respectively. The use of glycolysis blockers such as fluoride or oxalate/EDTA is not possible because they either have detrimental effects on the biosensor itself [e.g., fluoride is known to inhibit urease (10)] or interfere in the simultaneous determination of additional analytes in our system (e.g., sodium or potassium as counter ions of fluoride or EDTA).
Some of the data obtained with the highly miniaturized,
integrated biosensor in the AVL OMNI during clinical studies are
summarized in Table 1
; these data show good agreement with existing clinical
methods. Additional analytes, such as ions and creatinine, will
be integrated.
|
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
