The basics of biochemistry instrumentation (Proceedings)

Introduction

This session will discuss basic methodologies used by selected common biochemistry analyzers and quality assurance issues which may result in better understanding of the advantages and disadvantages of various instrument types and ultimately assist with instrument selection. Additional quality issues are discussed in the "QA Tips" and "Lab Errors" sessions.

Methods

Common methods include photometry and electrochemistry. The origins of photometry go back to the mid 1600's when Sir Isaac Newton observed that a prism separates light into colors and that light retains its color whether it is reflected, scattered or transmitted. This is the main principle behind the majority of clinical biochemistry methods. Systems relying on the principles of light may use reagents that are liquid, reconstituted liquid, or dry. Liquid chemistry systems are considered the most traditional and are what most published reference intervals and interpretive guidelines are based upon. The spectrophotometry of liquid chemistry analyzers is similar in principle to the hemoglobin concentration measurement as part of a CBC analysis. These systems rely on a chemical reaction that occurs between the sample and one or more liquid reagents in a controlled environment. In this case, the sample is typically serum or plasma. The reaction creates a color change in the mixture and light of a specific wavelength is passed through the reaction mixture. The light that is not absorbed by the reaction mixture is transmitted to a photometer and is measured. Since other conditions, as the characteristics of the reaction/measuring chamber, remain constant, the intensity of that light is proportional to the concentration. Assays may be one of two general types: endpoint and rate. Endpoint assays require a photometer that measures absorbance at the beginning and end of the reaction. The concentration of the analyte is proportional to the magnitude of change. Because sample quality issues as lipemia, icterus or hemolysis (known as serum quality indices) can alter the light stream, systems typically incorporate a blanking system to minimize the interference. The degree to which these interferences affect results varies with the method, instrument and level of interferent. The user should be aware of if and when these interferences might occur in their chemistry system.

Rate assays involve the creation of a product or depletion of a substrate. How fast the reaction occurs over a specified time is the slope (rate) of the reaction and is proportional to the activity of the analyte, usually an enzyme. The substrate, reaction temperature and pH can affect the reaction rate and thus reference intervals may vary significantly between instruments types and methodologies.

Liquid systems tend to offer the largest test selection, have the highest amount of flexibility for customized panels, and cost-effectiveness is usually proportional to workload. In other words, the more tests performed, the lower the cost per test. A higher level of technical expertise is necessary to maintain the analyzer and the complex reagent system. Control materials must be run and evaluated regularly to insure integrity of the process, periodic calibration may be required and reagents may need manual reconstitution. Liquid systems are the most common form of instrument in large referral laboratories.

Efforts to simplify the instrumentation for users have resulted in a number of variations of this theme including removing the on-site calibration process. Reconstituted liquid systems are one such variation (VetScan® , Abaxis, Union City, California) where reagent handling and calibration requirements are simplified. Heparinized whole blood or serum is added to a rotor containing freeze-dried reagent pellets that are converted to liquid after the addition of a rotor-contained diluent. Built-in quality control materials indicate the integrity of rotor handling during shipment and storage as well as integrity of the lamp and other operational issues. Calibrations for new lot numbers are performed by the manufacturer and changes are automated via matching the barcode on each rotor to information installed with a magnetic strip on a card provided to the customer. Results are suppressed if sample quality issues interfere significantly (>10%) with the accuracy of result. Test flexibility is accomplished with selection from several predefined rotors providing test menus designed for several species and common organ assessments. A similar unit (Analyst® , Hemagen Diagnostics, Columbia, Maryland) requires dilution of the serum or plasma with the automatic diluter provided. Due to the panel arrangement, cost per test is intermediate and remains relatively constant regardless of the number of rotors used.

Dry chemistry systems use reflectance photometry rather than absorption. In this case, the reagents are present on a test strip in a dried form. A manually applied or automatically dispensed sample reconstitutes the reagents and a reaction occurs with subsequent color development. A photometer detects the color change as it is reflected off the surface. The test strips may have one or several reagent systems on one strip allowing for testing a single analyte or a mini-panel of predetermined tests (SpotChem® , Heska Corporation, Loveland, Colorado; RefloVet Plus® , scil Animal Care Company, Grayslake, Illinois). Those using dry slides (Dri-Chem® , Heska Corportion), VetTest® , Idexx Laboratories, Portland, Maine; Vitros DT60 II, Ortho Clinical Diagnostics, Raritan, New Jersey) accommodate a single test, or by loading several desired test slides into the testing unit, a customized panel of any number of tests. Some of these also offer slides into predetermined panel packets. The test menus may be more limited than larger liquid analyzers, typically offering common chemistries but not electrolytes (Heska Dri-chem excluded). Single tests are more economical than a rotor or cartridge system and reagent handling tends to be simple. Interference from lipemia is minimized since reflected light is less affected than transmitted light with dry slides but hemolysis and icterus can still interfere.

Electrochemistry methods are based on the principles of measuring the electrical potential difference (potentiometry), or level of current generated with an applied voltage (amperometry) or following a chemical reaction (coulometry). The sample, electrodes, a measuring device, membranes, and electrolyte solutions are common features of these systems that are arranged to provide electrical contact between electrodes. In potentiometry, the membrane contains chemistry to exclude the ion of interest, thus is an "ion-selective electrode", and other ions equilibrate across it. This exclusion creates a potential difference across the membrane that is proportional to the concentration of the ion being measured. The potential difference is measured, and the concentration is calculated based on the Nernst equation for ions as electrolytes, pH, ionized calcium and blood gas parameters. The arrangement has miniaturized into cartridges and slides that simplifies reagent handling and covers common critical care analytes.

Conductometry involves the measurement of resistance to electrical conduction between two electrodes. Since erythrocyte membranes contain lipid, there is a proportionate decrease in the current passing between the electrodes. This phenomenon allows for the measurement of hematocrit in whole blood samples and this application is common in critical care instruments. It is also the basis for the Coulter principle used in counting and sizing blood cells by impedance-based hematology analyzers.

Examples of analyzers using potentiometry, amperometry and/or conductometry include the i-STAT (Abaxis), Dri-Chem for electrolytes (Heska Corporation), VetLyte (Idexx Laboratories), IRMA TRUpoint (ITC, Edison, New Jersey) and EasyLyte Plus (Hemagen Diagnostics). The VetStat (Idexx Laboratories) uses optodes (optical electrodes) and fluorescence (methodology not discussed). Glucometers utilize a variety of electrochemistry methods linked to one of several possible enzymatic reactions that generates ions to measure glucose. The AlphaTrak (Abbott Laboratories), GlucoPet and GlucoVet (ADM) are marketed specifically for animals and there are multiple units marketed for humans. Several articles and abundant web-sites discuss the validity and/or necessity of these units.

Instrument selection

These systems generally provide for the use of fresh samples and improved turn-around-time and patient management. Instrument selection is often based on the needs of the user as glucose monitoring, acid-base and electrolyte monitoring, rechecks of specific organ systems and/or health and preoperative screens. Chemistry instruments range in complexity from those dedicated to measuring a single analyte as glucose, those capable of performing predetermined panels, and to those that have the flexibility to provide both. Some systems offer the unique ability to accommodate ambulatory testing or allow the use of whole blood thus by-passing the steps and time necessary to harvest serum. Sample volume requirements may also contribute to the instrument selection process.

Quality assurance

While generally fairly reliable, these systems are not without issues related to stability, sensitivity and selectivity. Because of the logarithmic relationship in electrochemical methods between the electrical potential and the activity of the electrolyte measured, small errors in the measured potential are magnified in the results and may affect precision and accuracy. Electronic noise, selectivity of the membrane, and other anions can interfere. Most systems provide a means for regularly monitoring the electrical and data management functions of the unit. These "surrogate" controls and electronic checks satisfy most human certification requirements. Unfortunately, this does not verify function of the chemistry within each cartridge or rotor. The typical daily QC check recommended for hematology analyzers is not practical for these systems. Instead, it is currently recommended that QC material is tested once on each shipment and periodically thereafter whenever questionable results are obtained. This will likely concurrently test each lot number as well, which is also recommended. Recommendations may change once finalized by the ASVCP Quality Assurance Committee (http://www.asvcp.org/pubs/qas/index.cfm); check the website for updates.

Sample collection and handling along with the patient's condition and therapies are important pre-analytical links in the chain. Potential interferences are numerous and are reviewed in more depth in the quality assurance sessions. Since the effects are method dependent and improvements are ongoing, it is best to contact your distributor regarding specific concerns. Issues related to a delay in testing are minimized by using in-house systems allowing the use of whole blood. Even so, most errors in testing are related to pre-analytical issues. If samples must be collected from an IV line, it is recommended that at least 3 times the dead space volume is discarded when sampling from catheters. Dilution with protein-free therapeutic or saline flush fluids may falsely lower the hematocrit and potassium and falsely elevate the sodium and chloride. Air bubbles in the sample can falsely increase pO2 while venous contamination of arterial samples can falsely decrease it. Clots may form if the sample is insufficiently mixed immediately after collection and can affect the accuracy of any of the analytes in the current run and possibly future samples. Contamination of the chemistry sample with EDTA will create a specific pattern of erroneous results including apparent hypocalcemia and hypomagnesemia through chelation, low alkaline phosphatase activity through chelation of ions used in the reaction, and hyperkalemia directly from potassium in the anticoagulant.

Reference intervals are based on samples collected from fasted animals with no hemolysis, lipemia or icterus. Hemolysis can be created by a traumatic venipuncture, delayed separation of serum/plasma from cells, or severe lipemia; and can occur in vivo from various hemolytic disorders. Hemolysis can interfere by modifying the color reaction and from leakage of erythrocyte contents. This leakage, if significant enough, can dilute the sodium and calcium and the color may interfere with multiple tests including albumin and bilirubin. Lactate may be falsely increased from stored transfused red cells. Elevated potassium concentration may be observed in hemolyzed samples from Akitas, Shar Pei dogs, Shibas, and other East Asian breeds; and increased AST, ALT and LDH activities may occur because of higher levels within erythrocytes compared to plasma. CK may be increased if the system does not utilize adenlyate kinase inhibitors to counteract the effect of this erythrocyte enzyme.

Lipemia can be found in samples collected from insufficiently fasted healthy animals, as well as patients with diabetes mellitus, hypothyroidism, hypercorticism, pancreatitis, and primary hyperlipidemia disorders. It creates turbidity in the sample and may increase results from liquid-based methods including AST, bilirubin, creatinine, lipase, and phosphorus. Dry-slide and reagent strips tend to minimize the effects of such interferences and electrochemical methods that do not involve diluting the sample are not affected. High volume analyzers dilute the sample prior to electrochemical analysis. Lipemia effectively takes up space during the dilution process with the ultimate effect of falsely lowering the sodium concentration.

Icterus can affect some bilirubin and creatinine determinations. Ascorbic acid and NSAIDs can interfere with some glucose measurements, and osmolality changes and abnormally high protein levels can interfere with the hematocrit measurement by conductometry.

The level and type of interferent and the extent of that interference on any particular test will vary between instruments and their methods. Note that therapies, including antibiotics can also interfere. Sample quality should be documented and complimented by a good working knowledge of these issues as they relate to the instrument in use. Human-based studies have found that the analytical link, that primarily influenced by the instrument, only contributes 7-13% of the total analytical error. Studies of this nature would be beneficial to the veterinary field; however this data suggests that the instrument really is the strongest link.

References and suggested reading

Vap LM, Weiser MG; Field chemistry analysis; Vet Clinics North America, Food Animal, 23;2007.

Beckman, A., et al., History of spectrophotometry at Beckman Instruments, Inc. Analytical Chemistry, 1977. 49 (3): p. 280-300.

Cohen TA, et al;Evaluation of six portable blood glucose meters for measuring blood glucose concentration in dogs, JAVMA 235:3; Aug 1, 2009

Contributors, W., Isaac Newton. 2007, Wikipedia, the Free Encyclopedia.

Contributors, Merck Veterinary Manual, in Diagnostic Procedures for the Private Practice Laboratory. 2006, Merck & Co, Inc.; Merial Ltd.

Stockham, S.L. and M.A. Scott, Enzymes, in Fundamentals of Veterinary Clinical Pathology. 2002, Iowa State Press. p. 433-459.

Laessig, R.H. and S.S. Ehrmeyer, Quality management and administration of point-of-care testing programs, in Principles and practice of point-of-care testing, G.J. Kost, Editor. 2002, Lippincott, Williams and Wilkins: Philadelphia.

Burkhard, M.J. and D.J. Meyer, Causes and Effects of Interference with Clinical Laboratory Measurements and Examinations, in Kirk's Current Veterinary Therapy, J.D. Bonagura, Editor. 1995, W.B. Saunders: Philadelphia. p. 14-20.

Wennecke, G. and G. Juel, Avoiding preanalytical errors in blood gas testing. 2005, Radiometer Medial ApS: Denmark.

Battison, A., Apparent pseudohyperkalemia in a Chinese Shar Pei dog. Veterinary Clinical Pathology, 2007. 36(1): p. 89-93.

Tang, Z., R.F. Louie, and G.J. Kost, Principles and Performance of Point-of-Care Testing Instruments, in Principles & Practice of Point-of-Care Testing, G.J. Kost, Editor. 2002, Lippincott William & Wilkins: Philadelphia. p. 67-92.

Hopfer, S.M., et al., Effect of Protein on Hemoglobin and Hematocrit Assays with a Conductivity-Based Point-of-Care Testing Device: Comparison with Optical Methods. Ann Clin Lab Sci, 2004. 34(1): p. 75-82.