Clinical Instrumentation (General Chemistry and Immunoassay

---

CLINICAL CHEMISTRY

Clinical Instrumentation (General Chemistry and Immunoassav " Analvzers) " Ronald G. Haas Marshfield CliniclSt. Joseph's Hospital Joint Venture Laboratory, 1000 North Oak Avenue, Marshfield, Wisconsin 54449

This review provides an updateon current instrumentation for routine clinical analysis during the period from January 1991 through November 1992. with a particular focus on automated instrumentation for clinical immunoassays. an area which has undergone signifirant rhange during the past two years. lnstrumentn fur near-patient testing in critical-care areas and in physician office lahoratories are also briefly reviewed. Since many of thesesystems werediscussed in the prior review (01).the reader is encouraged t o ciinsult this reference for continuity.

Ronald 0. Ham is a Slaff clinical chemist p F m w with tha Joint Venhne Laboratory of SI.

Josephktlospblandtha Marshfield Clinic In Marshfield. W I .

IMMUNOASSAY A N A L Y Z E R S Books a n d References. A number of monographs on immunoassays and immunoassay instrumentation are of interest (02, 03)including two on luminescence techniques and immunoassays ( 0 4 , 0 5 )and still another on fluorescence immunoassay (06).A monograph with information ongeneral immunoassay principles. available nonradioactive labels, and biosensor devires (07) is also availahle. A practical uide to immunoassay automation has appeared recently with details on many of the newer instrumenl;i. The Cliniral Ligand Assay Society also has a focus issue (091 reviewing automated immunoassay instruments. A review (0101 uf simple immunoassay techniques for near.patient testing is also of interest. The American Association for Clinical Chemistry Endocrinology Cuntinuing Education Series likewise provides on- oing information on instrumentation and immunoassay teckniques (011) in addition t o information on specific endocrine metaholites. Immunoassay Instrumentation. Asnoted,thepast two years have witnessed the announcement of a large number of completely automated immunoassay systems. Many of these systems have only recently been introduced and a number arestill undergoing evaluation at 'Vsitesor arestill in development. Characteristics of these automated immunoassay systems typically parallel those of other automated rout ine-chemistry systems. Included are such features as a large on-board test menu, theahility torun a battery of testsonacommonsample (random access capahility), continuous sample loading while a run is in progress, the ability to read bar-coded sample tuhes, direct sampling capability from primary collectiun containers. reagent bar-coding and tracking of reagent inventory, on-board reagent retrigeration, autodilution and autorepeat of out.of.range results, extended calibration stahility of several weeks, and the ability to communicate hidirectionally with host computers. A representativesampling of fully automated immunoassay systems and some of their operating characteristics is shown inTalile0-I. Allofthesesystems usesolid-phase,sandwichtype assays. In this mode, the sample incubates with an antibody attached to a solid phase such as microbeads, a coated tube,or glass-fiher matrix. Asecond antibody ,signal antibody) is then added as a liquid phase, which reacts at a secondsiteon theanalytemolecule to ivea bridgedsandwich complex. Alkaline phosphatase.labefed antihtidies are used as signal antibodies on several system3. On-line washing of the solid-phase matrix is incorporated to remove excess. unbound signal antibody. Measurement of the bound signal antibody is then accomplished by use of substrates whirh result in chemiluminesrent or fluorescent productn, some of which are detectable at extremely low levels. The dioxetane substrates. fa!r example, which are employed in the both the lmmulite (Diagnostic Productn Corp.) and Access (SanofiPasteur, systems produce unstable anionic products which deromposegivingcontinuouslightemissionandenablin the detection of 1000 molecules of alkaline phosphatase (812). Iletection limits for analvtes in the attomole zentomole Der lirerrange(l0 ~ I ~ ~ , a p p e a r f e a q i h l e ( 0inpr;nripleu&g 13i chemiluminesrence for decection

((hi

444R

ANALYTICAL CHEMISTRY, VOL. 65. NO. 12. JUNE 15. 1993

Not included in Table 0-1are a number of semiautomated systems which involve manual transfer of individual tuhes or microtiter plates. Examples of such systems include the Cyberfluor 615 immunoanalyzer (0141, which uses timeresolved fluorescence for detection; the KodaklAmbersham Amerlite system and Diagnostic Products Millenia. which usemicrowell plates and chemiluminescence and colorimetric detection, respectively; and the NicholsILondon Diagnostics LumaTag chemiluminescence analyzer, which employs a coated-tube solid phase. The Du Pont aca Plus also enables solid-phase immunoassays to be added t o t h e aca system by virtue of a stand-alone, off-line workstation which produces amicroparticle sandwichsuspension withinanaca pack. The bound, signal antibody in the sandwich complex is then detected colorimetrically by manual transfer of the test pack t o the aca instrument. Also of note are several additional fully automated systems, one from Ahbott Diagnostics for hepatitus marker and human immunodeficiency virw testing using chemiluminescence detection (015)and another prototype unit of British design, the Anagen System AN2000, which uses fluorescence detection and ma netic separation (016).AlsonotincludedinTableO-Iisthe8'allaclPharmacia AutoDelfia system. This fully automated system to he released tentatively in 1993 uses antibody-coated microwell plates and the same time-resolved fluorescence technology incorporated into earlier manual assays. Abbott Diagnostics has also announced the fully automated AxSYM system t o be released in 1994. Several of the systems in Table 0-1support multiple-assay formats enabling solid-phase single-antibody competitivetype assays and homo enous assays (not involving a solid phase) as well as duafantibody sandwich assays. These differing formats optimize sensitivity and speed of analysis to a particular analyte and facilitate the assay of both large and small molecules. The Abhott IMx utilizes fluorescence for ita solid-phase assays and fluorescence polarization for several homogenous assays. Likewise, the MilesITechnicon Immuno 1 utilizes colorimetry for ita solid-phase assays and turbidimetric measurements involvinglatexag lutination for homogenous assays. Several of the automatefsystems such as the Boehringer Mannheim ES300 (017)can also vary the incubation time as a means of optimizing sensitivity. Such automation expedites throughput of analytes not re uiring high sensitivity while accommodating the longer inc&ation times required for low-level analytes. Both the Stratus and IMx alsosupporttwo-step reactions. In this mode, incubation of the serum analyte and solid phase occurs initially followed by a wash step to remove endogenous serum proteins. Addition of the labeled antibody then follows in a second, independent incubation step. Thisapproacheliminates other binding roteins and reduces the possibility of falsely low results &e t o "hook" effects a t extremely high analyte (antigen) concentrations (018). Improved linearity is also frequently possible using a two-step incubation. Differing separation techniques are employed t o enable washing and removal of unreacted signalantibody. Magnetic separation using beads containing iron or chromium is a popular approach. Application of a magnetic field enables

CLINICAL CHEMISTRY

Table 0-1. Solid-Phase Automated Immunoassay Analyzers minutes instrument and max to fmt solid-phase/ detection method0 teetalh result separation method manufacturer ACS-180 Ciba-borning Acceea, Sanofi Diag. Pasteur Affiiity, BectonDickinson AIA-600, Tosoh-Medics

90-180

15

100

20-55

20-30

23-60

60

50

coated beads, magnetic

AIA-l200DX, Tosoh Medics Cobas Core, Roche Diagnostics ES 300, Boehriiger Mannhem Immulite, Diag. Prod. Corp. Immuno 1, Miles/ Technicon IMX, IMX Select, Abbott Diagnostics Luminomaster, sankyo co., JaDan Opus; PB Diag. Systems

120

62

fluor

100-150

35-120

coated beads, magnetic coated bead, magnetic

colorim

FS, RA, CL, PST, BC BT, RA, CL, ma

120

55-170

coated tube

colorim

pparticle, magnetic pparticle, magnetic coated cuvet

chemilum

attributea of system*

chemilum colorim fluor option fluor

BT, RA, CL, PST, BC BT, RA, CL, RR BT, RA, CL, ma

max avail analYtas/

cal frequency/ no. of calibratorad

13

1-2 wks, 2 pt

24

30 days, multipoint 2 wks, 6 pt

runc

no. of FDAapproved assays' 15-19 avail 1Q '93

man. loading 6/sample man. 30 days, 2-6 pt loading lO/nample 21 30 days, 2-6 pt

15-18

10+

30 days, 2-6 pt

1

BT, RA, ma, 12 PST/BC pending

2 wks, 5 pt,

18-20

BT, RA, CL,

12 5/sample

2 wks, 2 pt

8

16

30 days, 6 pt

1-5

1

2 wks, 6 pt

30

30 days, 6 pt

not avail in U.S.

BT, RA, CL, ma

1-Pffrun

11

15-18

120

45,75

coated bead, centrifugation

chemilum

120

7-32,72

particle, magnetic

32-48

17-43

pparticle, glass fiber

120

45

coated tube

colorim FS, RA, CL, PST, BC, RR and turbidimetric fluor BT, batch, and ma fluor polar chemilum FS, RA, CL, RR

20-80

6-27

multilaler, 'dry reag

fluor

BT, RA, CL, ma

Opus M + ~ u m , PB Diag. Systems Radi.us, Bio-Rad Diagnostics SR1, SeronoBaker Diagnostics Stratus IIntelect, Baxter Diag. System 7000, Biotrol Vidae, BioMerieux Vitek

50-190

6-27

multilayer, 'dry" reag

fluor

FS or BT, RA, CL

80-125

70-150

coated well

colorim

PST, BT, RA, ma

12

31-60

40-86

particles, magnetic

colorim

BT, RA, CL, ma

man. 4 wks loading

45-72

8-10

fluor

BT, batch, ma

1

2 wks, 6 pt

34

100

30-45

colorim

30-55

fluor

BT, RA, PST, BC BT.. RA.. CL. ma

20

36-45

glass-fiber, radial chromatop pbeads, magnetic coated,

man. loading

4 pt/lot, 2 ptlday 2 wks, 1pt

avail in France 11-15

Vita,Syva

38-76

fluor

BT, RA, RR

15

4-6 wks, 5 pt

5

78-144

%et-like CrOz pparticle, magnetic

ma

"Select": 3 20

2-8 wks, 6 pt man. loading 3/sample >20 2-8 wks, 6 pt 8lsample 2 wks,6 pt, 1Ptlrun

20-24 avail mid-'93 avail mid-'93 15

*

a Abbreviations: chemiluminescence,colorimetric, fluorescence,fluorescence polarization. BT (benchtop); FS (free-standing); RA (random access-performs multiple analytes per sample per run);CL (continuous loading of additional samples possible during a run); PST (allows we of primary-sample tube, Le., original blood collection tube); BC (sample bar coding); RR (on-board refrigerated reagent storage); ma (complete reagents not stored on-board necessitating manual addition of solid-phasehitized dose for each analyte at s t a r t of run). Maximum number of on-board analytes and maximum number of analytes available per sample. d Stability of calibration curve in weeks;number of calibrators required for a multipoint curve. Available assays as of 12/16/92, or date of instrument availability.

rapid isolation of the beads either for washin or for removal of the magnetic particles from the cuvet l i g b path during photometric measurements. The Diagnostic Products Immulite uses a uniquely designed, tapered tube containing a single l/r-in.-diameter, antibody-coated bead. Bead washing is accomplished quickly and efficiently on-line by s inning the tube about its vertical longitudinal axis, which Lads to rapid expulsion of wash li uids into an attached outer concentric reservoir (sump) $at ca tures the waste liquids (019). The tapered design is such t f a t several washings can be accomplished within a 4-8 interval (020). In selecting a system for routme use m the laboratory, several practical issues influence consideration by an enduser. Among these is the size of the test menu. Although developmentalcycles for new instruments can be accelerated, menu expansion and introduction of new analytes into the U.S. market is frequently slowed because of the len hy processes associated with ap roval by the U.S.Food and rug Administration. Accordingfy many of the analyzers (Table 0-1)have relatively small test menus. Two exceptions are

8

established batch anal zers such as the Abbott IMx and Baxter Stratus, which lave menus of 30 analytes or more. The combination of an IMx for high molecular weight analytes and an Abbott TDx using fluorescence polarization for low molecular wei ht analytes enables the assay of an extremely broad range of d r u g s and roteins. Large numbers of users likewise exist with these gatch systems providing the advantage of a homo enous, statistically stable peer group in federally mandate! proficiency testing surveys. Another practical issue in the working laboratory is the need for redundant instruments capable of providing backup when one system is down. This is particularly important for assays such as therapeutic drugs and for cardiac markers such aa CK-MB, which frequently require rapid turnaround. As laboratories move away from the batching of immunoassays to performin them on an as-received basis using this new generation o f analyzers, expectations of laboratorian8 and particularly physicians will increase for consistent (aswell as rapid) turnaround of results. On-site availability of two or more systems is thus a necessity. Analyzers which conserve ANALYTICAL CHEMISTRY, VOL. 65, NO. 12, JUNE 15, 1993

445R

CLINICAL CHEMISTRY

bench space are thus advantageous. In moderately sized laboratories, the availability of models having differing capacities and sharing common reagenta is likewise desirable. Thus, the PB Di ostics instrument series will accommodate high-volume a p x a t i o n s in the future on the Opus Magnum model while lower through ut can be achieved on the currently available Opus and Opus h s models for backup and/or for STAT testing. Redundancy is also built into the Opus Magnum, which has separate and independent sample rocessing and measuring channels within a common chassis. gimilarly, high throughput can be achieved on the freestanding Tosoh Medics AIA-1200 model while the benchtop AIASOO sup orta lower volume applications using identical reagents. T i e bioMerieux Vidas and MiniVidas models provide similar complementarity. The Vidas is also of interest in that ita throughput and capacit are scaleable by virtue of having stand-alone analytical a n i d a t a processin modules which are interfaced to form the working system. Thoughput can be increased severalfold by interfacingup to four analytical modules to the shared data-processing module. Interest in several of the analyzers listed in Table 0-1will also be stimulated by the availability or analytical characteristics of several noteworthy analytes or groups of analytes. These include a high-sensitivity prostate-specific antigen assay (021)as well as hepatitus markers on the IMx; bacterial antigens on the VIDAS; troponin-?', an interesting new cardiac-specific rotein marker for acute myocardial infarction, on the Boehnger-Mannheim ES300 (022);and rapid assays for myoglobin (also used as a cardiac marker) on the Stratus (023)and Opus (024)systems. Interest likewise remains high in developin automated assays for HIV antibodies and for ultralow revels of thyrotrophin. Because of the ability of these s stems to perform testa in a random manner, integration w i d other ra id-turnaround, routine chemistry analyzers is highly desirafle for efficiency of staffing. The current batch-testing mode employed for immunoassays resulta in a relatively long turnaround time (typically 1 day or more) for completion and reporting of results. Because of this slow turnaround, immediate access to laboratory specimens in the large centralized lab is usually not a high priority. Accordingly, immunoassay work areas are often located remotely from centralized specimen processing and routine chemistry areas. With many of the new analyzers, however, it is now feasible to simply transfer a bar-coded primary collection tube from the sampler of a highvolume, multichannel chemistry-paneling analyzer to the sampler of an adjacent immunoassay analyzer. Bringing the immunoassay assay work area into closer proximity to specimen processing and routine chemistry work areas now becomes a priority and may necessitate reexamination of traditional lab design. Personnel will likewise have to adjust. Diehard immunoassay technologists will now experience "future shock" in having to troubleshoot a closed system of an instrument lus as reagents in which previously discrete analytical steps pipeting, phase separation, and analytical measurement (traditionally of radioactivity) are no longer able to be examined in isolation but are now far removed from operator control and are dictated largely by software and hardware protocols within the automated system. Because of the user's inability to exercise manual control, instrument downtime and the need for redundancy (as noted earlier) become major concerns. From a patient pers ective, these systems will now enable the laboratory to compTete out-patient testing in a more rapid manner. In screening for prostatic cancer, testa such as prostate-specificantigen will in the future likelybe performed and reported while the physician is completing the patient's physical exam which typically includes a digital rectal exam. Having both lab and rectal exam resulta availableconcurrently will improve the sensitivity of detection (025)of this highrevalence cancer in aged males. Likewise oncology patients eing monitored for tumor reoccurrence following treatment may soon come to expect their tumor marker results being available while on-site during routine outpatient visits. One final consideration with these systems relates to cost, both the cost of operation for end-users as well as costa related to assay development for manufacturers. Given the plethora of systems, many of which have limited menus, end-users are reluctant to use capital equipment monies for multiple systems

sue!

E

448R

ANALYTICAL CHEMISTRY, VOL. 85, NO. 12, JUNE 15, 1993

from different manufacturers. Placement is thus contingent on bundling analyzer costa into the costa of reagents. Manufacturers on the other hand are faced with develo ment costa estimated at 50-80 million dollars (026)for introiucing a new immunoassay analyzer with a 12-test menu. Given the large number of analyzers available and the cost of development, it is unlikely that all of these systems will succeed in the marketplace. GENERAL CHEMISTRY ANALYZERS This portion highlights changes and improvementa in instruments for routine chemistry testing occurring during the intervening 20-month period since the last review (01). Although changes occurring in routine analyzers have been modest compared to immunoassay systems, incremental im rovements in analyzer desi are nevertheless evident. T a b e 0-11lists serum/plasma cemistry analyzers which are relatively new or have undergone significant im rovementa in the prior two years. Not included in the Talle 0-11are centrifugal analyzers, which were discussed in the prior review (01). Literature references relating to the evaluations of several of these systems have likewisebeen noted previously. Two analyzers evaluated recently include the Olympus AU510 (027) and Miles/Technicon Axon (028).One newly announced system is the Roche Diagnostics Integra, a random access, 700 test/h, disposable-cuvet, closed-reagent analyzer targeted for 1994 delivery and having capability for both fluorescence polarization and colorimetric assays. Interferences. Pigment interferences in photometric testing on routine chemistry analyzers from bilirubin, hemoglobin, and lipemia continue to be of concern. A second edition (029)of a user's guide is now availablewhich compares the influence of these interferences on different instruments. Also included in this source are interferences associated with nonhuman serum or plasma for veterinary testing. Of note is that many analytes show a much different interferant response in nonhuman specimens compared to human samples. Physician office analyzers are likewise evaluated regarding pigment interferences. These pigment interferences have also been examined (030)for five instrumenta performing homogenous immunoassays and one heterogenous immunoassay analyzer. In evaluatin bilirubin interference, unconjugated bilirubin has traditionefily been used for spiking studies. Water-soluble (conjugated) bilirubin forms have been shown to interfere to a greater extent however in assays such as creatinine. Because of the difficulty in obtaining con'uated bilirubin in urified form, the use of a synthetic soluble %ilirubin,ditaurofilirubin, has been sug ested as a surrogate for conjugated bilirubin in spiking stuiies (031). Laboratorians may thus need to assess the influence of both ditarurobilirubin and unconjugated bilirubin in the future in establishingmethod performance. The influenceof covalently bound &bilirubin has also been evaluated for creatinine (032). Measurement of serum indices (033)on Boehringer Mannheim/Hitachi analyzers by monitoring the concentrations of these pigments by direct spectrophotometry a t multiple wavelengths is likewise useful in providing an objective measure of interference. Another bothersome interference is ascorbic acid (vitamin C), which typically interferes in peroxide- eneratin reactions where it competes with chromogenic su%strates feading to falsely low results. High doses of vitamin C in atienta with malignancies, particularly intravenous doses (834),can lead to marked aberrations in chemistry- anel analytes such as uric acid, cholesterol, and triglycerig. Finally, interferences in ion-selective electrode (ISE) methods used in chemistry analyzers remain of interest. Because of the popularity of the Kodak and Hitachi analyzers, these ISE s stems have received considerable scrutiny, especially sohYlum (035-037) and carbon dioxide (038,039) on the Kodak, and sodium (040,041)and chloride (042)on the Hitachi. Design improvements b Kodak in the sodium ISE (043)and use of an enzymatic carton dioxide slide (044, 045)have mediated these problems. Citrate interference in the measurement of sodium in urine has also been noted (046') with the AVL Scientific/Beckman Lablyte analyzer. Data Handling, Method Validation, and Interpretation. The proceedings of a conference on validation of clinical laboratory methods and instruments (047)should be noted.

CLINICAL CHEMISTRY

Table 0-11. General Chemistry Instruments for Serum or Plasma Analyses, Grouped by Capacity. open/ reused/ closed dispos analyzer characteristics and instrument and alternative models in series reagenta cuveta t/h +ISEs manufacturer AU5223, Olympus

DAX-48, Miles/Technicon

747-100, Boehringer Mannheim/Hitachi

Chem 1, Miles/Technicon

Ektachem 750XRC, Eastman Kodak

Paramax 720ZX, Baxter Diagnostics

Axon, Miles/Technicon Dimension AR, Du Pont Nucleus, Nova Biomedical Reply, Olympus

Synchron CX7, Beckman Instrumenta

911, Boehringer Mannheim/Hitachi

Cobas Mira Plus, Roche Diagnostics Express 550 Plus, Ciba-Corning

Group A * Very High Volume Chemistry Analyzers >2oM) t/h R 165 X (1-32), + Na K fixed-sample throughput; cuveta dedicated to specific test; also models AU5231,330 X 24 t/h; AU5211,165 X 16 t/h; AU5221,330 X 16 t/h R 150 X (1-32), + Na K fixed-sample throughput; cuveta dedicated to specific test; also models DAX-72,300 X 24 t/h; DAX-24, 100 X t/h; DAX-36,300 X 32 t/h throughput varies with test mix; R 600-2400, + Na K C1 6-8 sample cycle; also higher throughput Model 747-200 (up to 6600 t/h); 33 2-reagent chemistries; 300 ISEsIh Group B:c High-Volume Chemistry Analyzers >500 t/h R 720, + Na K reagenta reconstituted on instrument; C continuous-flow 'capsule technology" using only 14 pL of reagent/t; 32-reagent cassette capacity 600-700, + Na K C1 COz 750 t/h with ISEs. Multilayered 'dry" slide C slide chemistries (40+) shared on Models 700,500,250, and office analyzer DT-60; 3-6 month cal. stability; no drain/plumbed water used D 600-700, + Na K C1 Li closed-tube sampling; 40+ tableted C chemistries reconstituted on instrument plus several open chemistries; 3-month cal. stability typically Group C:d Moderate-Volume Chemistry Analyzers >200 t/h R 480, + Na K bar-coded reagent/lot no. tracking; 0 'windows"-based software; 36 resident chemistries 40+ reagenta reconstituted on-board; D 300, + Na K C1 COz C 60-90 day cal. stability; Model ES, 180-200 t/h; no drain/plumbed water required; waled cuveta formed on-board varies, + Na K C1 COz Ca bench-top; 1-12 self-sampling chem C R modules; 1-2-min rapid STAT response; use of redundant modules enables throwhmt w test menu tradeoff 300+, + Na K C1 up to 3 ri&ent/t; 1 m c u v e t D 0 capacity; 400 t/h without ISEs; 5-6-min STATdwell time; ca. 40 resident testa with 30-day cal. stability; no reagent bar-coding includes user-modifiable photometric R 220, + Na K C1 COz o/c testa plus 8 additional rapid response (90e) closed-reagent testa used on CX3 model; bar-coded reagenta/lot no. tracking 360, + Na K C1 up to 3 reagentalt; 2 X 32 R 0 bar-coded reagents, reagenttlot no. tracking; higher throughput Model 717 (600 t/h) also available Group D:e Low-Volume Chemistry Analyzers >lo0 t/h D 132, Na K C1 300-cuvet capacity; on-board 0 reagent cooling; PST/BC; closed-tube sampling; on-board use of 14 3-reagent or 43 1-reagent chemistries D 180 capacity for 26 1-reagent room-temp 0 chemistries; bar-coded reagenta; cuveta loadable as needed; ISE testa require separate analyzer

Classified as open (0) or closed (C) reagent systems based on ability to use reagenta from multiple sources and employing either reuseable (R) or disposable (D) cuvets, t/h, photometric testa/hour excluding ISE (ion-selective electrode) testa; PST/BC, primary-tube sampling and sample bar-coding. All are free-standing models, support PST/BC, and require a drain and on-line deionized water for cuvet washing. All are free-standing models, support PST/BC; sample throughput ia test-mix dependent. All are freestanding except as noted; all support PST/BC. e Bench-top, disposable cuvet analyzers.

Instrumentation-related topics include assessing linearity, improving quality control, satisfying proficiency testing requirements, and establishing reference ranges. Similar

issues are also addressed in a Scandinavian consensus document on clinical test evaluation (048).An IBMcompatible,artificialintelligenceprogram (049)incorporating ANALYTICAL CHEMISTRY, VOL. 65, NO. 12, JUNE 15, 1993

4471

CLINICAL CHEMISTRY

some 4000 rules for interpreting and validatin biochemical data from multichannelchemistryanalyzers is a%o of interest. Ale0 noteworthy is a special focus monograph on laboratory and hospital information systems. This monograph covers a diversity of to ics including the application of bar coding for multichannefihemistry analyzers (050). Reference values have also been addressed for chemistry anels as well as other analytes for the 60-1OO-year age group (8511, a rapidly growing segment of our population. RAPID-TURNAROUND AND NEAR-PATIENT TESTING Critical-CareTesting. Whole blood analyzers continue to be of eat interest for critical-care testing. A number of whole bgod systems have been reviewed (0521, and the Yhydridlaboratory”has been proposed (053)in which central, satellite, bedside, and in-vivo monitoring strate ies are integrated using staff of different disciplinesand witg a focus on customized test menus and rapid turnaround time. Among whole blood analyzers for critical care use are the Nova Di ostics’ instruments, which provide sensors for either a wh%blood lactate (STATProfile 7) or glucose (STATProfie 5) (054) in addition to traditional blood gas and sodium/ potassium measurements. Measurement of ionized m sium has also been proposed in the future on either the % a! or AVL Scientific whole blood analyzers. The Radiometer ABL 520 blood gas anal zer is of interest in that in addition to traditional pH, pC& and p0z measurements, it also incorporates oximetry for oxy-, carboxy-, met-, and total hem0 lobin using multiwavelength spectrophotometry on 85 pL of%lood. Trends toward portability and miniaturization are also evident. The Mallinckrodt Sensor Systems Premier analyzer facilitates near-patient use by eliminating the usual clutter of compressed gas cylinders, tubing, and buffer bottles. Dis osable foil packs containing fluids pretonometered with caligrating gases are employed. The foil packs are stable for 7 days and incor orate a disposable assembly containing electrodesfor H, hood gases, sodium, potassium, hematocrit, and ionizedc8cium or chloride,which is replaced concurrently with the fluid ack (055,056). Another innovativeapproach is the i-STAd? portable clinical analyzer. This hand-held device, about the size of a telephone receiver,em l o p a sin leuse biosensor cartrid e to which 50 pL of whole bpood is adted. Calibrant, containet! in the cartridge, contacts the sensors just prior to the patient sample. Patient results for sodium, potassium, chloride,glucose, urea nitro en, and a hematocrit are available in about 90 s on the antfyzer’s liquid crystal display in both a graphic and numeric format (057). Results for u to 50 patients are stored and can then be uploaded to an IJM-compatible computer by infrared wireless transmission. Another interesting ortable analyzer is the StatPal (PPG Biomedical Systems Aiv., Sensors). This device, about the size of a hand purse, rovides blood gas/pH resulb, which are displayed on a sm& monitor and integral printer. Of final note is a blood ammonia checker (HiChem BAC 111) which may address the long-standing problems of measuring blood ammonia levels rapidly and without contamination when specimens are transported to a centralized laboratory (058). This bedside device uses fingerstick (arterial) whole blood and measures a reflectance signal produced on a disposable test strip (059). Improving Turnaround in the Central Laboratory. Although near-patient testing continues to show significant growth, a proaches to improving turnaround in large centralized la%scontinue to be explored. The use of pneumatic tubes dedicated to the transport of specimens remains an option for maintaining rapid turnaround in large institutions. A report (060)on a newer generation tube system which does not unduly jostle specimens is of interest. The use of plasma instead of serum for chemistry paneling is also beneficial (061) in reducin turnaround time. Unlike serum, plasma can be centrifqe! in a shorter time interval and does not require any clottmg delays. Total protein results on plasma chemistry panels remain problematic however because of the variable fibrino en content in plasma which may cause protein results to be figher by 1.0 g/dL or more compared to serum, particular1 in acutely ill patients. T w o recent reporta (062, 063) descrite an inexpensive,nonimmunologic,turbidimetric 448R

ANALYTICAL CHEMISTRY, VOL. 65, NO. 12, JUNE 15, 1993

fibrino en method for Hitachi analyzers which when incororatedi into a chemistry panel would enable on-line calcui t i o n of a serum protein result by subtracting the measured fibrinogen from the observed plasma protein value. Robotic handling of specimens (064) in the clinical lab have also been reviewed as a means of expediting sample processing. Integration of the blood collecting device into a desk-top instrument to enable on-line centrifugation and processing has been reported (065)using a prototype system. PhysicianOffice Laboratories. Advances in whole blood analyzersfor physician’soffice laboratorieshave also occurred. The Abaxis EPOC 2000 (066) utilizes centrifugal force on whole blood to obtain plasma, which is then metered into independent reagent wells enabling assay of 12 or more analytes per sample using a disposable rotor. The Biotrack (Ciba-CorningDiagnostics) multianalyte assa system (067) allows simultaneous measurement of hemogfobin, lucose, and cholesterol using a sin le-use plastic cartrifge and reflectance photometry. Simfar single-usecartrid ea are also available for single-analyte assay of dr 8 for t erapeutic monitoring (068). Several innovative x o l e blood assays should be noted on two large-repertoire, physician-office analyzers. These include the determinabon of glycated hem0 lobin for monitoring lucose control in diabetics on the A%bott Vision (069) an! whole blood measurement of HDL-cholesterol on the Boehringer Mannheim Reflotron (070) and Abbott Vision (071, 072). Unlike most HDLcholesterol assays, these systems do not require a preliminary off-line reci itation step. A dedicated office analyzer for hemoglo%in from Miles/Technicon has also been evaluated (073) along with a portable device for whole blood lucose measurements (074). The ability of a popular serum%asedoffice analyzer, the Kodak DT-60, to meet current regulatorystandards for proficiencytesting has been assessed (075) based on data from 400 laboratories. Another serumbased system, the Miles Clinistat analyzer, has also been evaluated for li id profilin in a multicenter study (076). Finally, the pubication (877) of a European community consensus document focusing on good practices in decentralized laboratories should be noted.

E

11,

LITERATURE CITED (01)HeeS R. G. A M I . 1991, 63, 246R-240R. (02) Rice, C. P., Newman, D. J., E L . Wnc~hssandRactlcBofImmuwssey: Stockton Press: New York, 1991. (03) Butler, J. E. Immcmochemlsby of solldshese Inwnunoeesel: CRC Proas: Boca Raton, FL, 1991. (04) Van Dyke, K. Lumbmscenm Immunoassay and Mobcdar Appuccldbns; CFiC Press: Boca Raton, FL, 1990. (05) Baeyens, W. R., De Keukeleire, D., Korkklb, K., E L . L t u n b m m w Techniques In Chem&al and Ebchmkal A n a W Marcel Dekker: New Yo*, 1990. (06) Hemmlla, 1. A. ApplVcebbne of Fluoreecence In I-ssam John Wlley 8. Sons Inc.: New York. 1991. (07) Nakamua, R. M.. Kasahara, Y., Recknk, (3. A,, E L . InwnurodwMJcel Assays and Bkwnsor Tedwkw for tJm 1990s; Amerlcen SwIety for Mlcrobldogy: Washington, DC, 1992. (08)Chen, D. W., Ed. I m m s s a y A m t b n , A RactlcelQulde: Academlc Press: Orlando, FL, 1992. (09) Ng, R. H., Ed. J. Ckh. I ~ ~ S W 1991, Y 74, 59-138. (010) Sutherlend. R. M.; Slmpson, B. Advances In Simple Immunoassaysfor Decentralized Testing. Adv. CM. Chem. 1990, 28, 93-106. (011) Kricka L. J.; Memandb, E. P.; Chrhtopoubs, T. K. €nhcdm&y InServlce Trehlngand ContInuIngEducatkn; AACC Press: Washlngtcm, DC, 1992; pp 3-26. (012) Schaap, A. P.;Akhavan, H.; Romano, L. J. Ckh. chem.1989,35, 18831864. (013) Krlcka, L. J. C/h.Chmn. 1991, 37. 1472-1481. (014) Papanastasbu, A Shankaran, P.; Khosravl, M. J. Ckh. Bbchm., 1992, 25, 255-281. (015) Khalll, 0. S.; et ai. CM. chem. 1991, 37, 1540-1547. (018) Clemente, J. A,; Forrest, G. C.; Jay, R. F.; Jeffery, M.; Kemp, P. M.; Kjddsen, N. J.; Rattle, S. J.; Smkh, A. CMn. them. 1992, 38, 1671-1677. (017) Camara, P. D.; Velletrl. K.; Krupskl, M.; Roener, M.; (.Lrmths, W. C. CMn. Bkchem. 1992, 25, 251-254. (018) H s k , B. L.; Fuller, K. A.; Brow, W. S.; Kmlnb, J. W.; Evdend, 8. J.; Scott, M. 8. CM. Chmn. 1992, 38, 437-438. (019) Babson, A. L.; Obon, D. R.; Pamkrl, T.; Ross. A. F.; Mer,D. M.; Mulqueen, P. J. CMn. chem.1991, 37, 1521-1522. (020) Babson, A. L. J. CMn. I - w y 1991, 14, 83-89. (021) Vessella, R. L.; Noteboom,J.; Lange. P. H. CM. Chmn. 1992,38,20442055. (022) Pentik, I.; Win, M.; Kerth, P.; Mkttlnen, M.; Rentanen, T. Ck.Chmn. 1990, 38, 965 (abstract). (023) Koskl, K.; Cobb, J.; Cheng, P. J.; Rodriguez, L.; Baw,R. CMn. Chmn. 1992, 38, 1101 (abstract).

CLINICAL CHEMISTRY (024) Foster, K.; Kundzlcz, H.; Harthus, H. P.; Saul, R.; Metzmann, E. Clln. Chm. 1992, 38, 1081 (abstract). (025) Catabna, W. J.; Smlth, D. S.; Ratllff, T. L.; Dodds, K. M.; Coplen, D. E.; Yuan, J. J.; Petros, J. A.; Andrlob, G. L. New €n@. J. W .1991, 324, 1156-1 161. (026) Tune, L. C h . Chem. News 1992, (Sept), 9. (027) Fam, C.; Vdesco, J.: Ramon, F. J. Autom. Chm. 1991, 13, 217-220. (028) Brmne. 8.; Prmclpe, L. C//n. chem.1992, 38, 20862100. User’soulde to (029) Wck, M. R.; Ryder, K. W.; Qllck, S. J. In-phs, I ~ h C M ~ b y I n s i W n e n t2ndd.;SclenceEnterprbs, s , Inc.: Indlanepob, IN, 1991. (030)R W , K. W.; Trundle, D. S.; Bode, M. A.; Cole, R. E.; Moorehead, W. R.; Qllck, M. R. CWn. Chem. 1991, 37, 1134-1135. (031)Franzlnl, C.; Morelll, A. M.; Cattouo, (3. CHn. chem. 1991, 37,236-238. (032) I W a , H.; Nakamura, H.; Yoshlda, M.; Wu, T. W. Clin. Chem. 1992. 38. 32 1-322. (033)Pearson, J. R.; Welb, R.; Welb, A. CUn. Chem. 1991, 37, 2014-2015. (034)Badrlck, T. C.; Campbell, B. C//n. chem.,1992, 38, 2160. (034hyckens, A.; SCMos, J.; Vandemplas, H.; Senesael, F.; Qoedhuys, W.; aoruS, F. K. C h . Chm. 1992, 38, 114-118. (036) StOW, J.; Frder, E. CWn. Chm., 1991, 37, 931 (abstract). (037) Bach, P. R. V h n Is a Sodlum not a Sodium. M & b : Lab. W ,1991, 0, 2. (038)Rlfal, N.; Hyde. J.; Jmedsohn, M; Qlasqow, A. M.; Soldln, S. J. Ann. Clln. 6bCbn. 1992, 20, 105-108. (039)Harrhn, S. P. Ann. CMn. Blochem., 1992, 20, 570-571. (040)Lary, N. 0.;Pembrok, A.; Duggan, P. F. CM. Chem. 1992,20,345-346. (041) Bouhnger, J., Paket, J. V.; En, P. CHn. Chm. 1992, 38, 166-187. (042) Faukner, A. M.; Peak, M. J. Ann. C//n. Blochem. 1991, 28, 107-108. (043) Dankl, D.; Bakoe, J.; Caspw, S.; Klssel, T.; Kozuch, S.; Mwawskl. J.; Roth, J. Toner, J.; Van Brunt, N. C//n. Chem. 1991, 37, 946 (abstract). (OU)Wd8mlth, B. M.; Palmer-Foust, A. C h . chem.1891,37,946 (abstract). (045)Daoud,E. W.;McClellan, A.; Scott,M. (3. CHn. Chem. 1992,38,168-169. (048) Hartby, S. J. CHn. Chem., 1992, 38, 788-787. (047) MuWb authors. Arch. Pethd. Lab. W .1992, 116, 701-799. (048) Multiple authors. Soand. J. C/h. Lab. Invest. 1992, 52 (Suppl. 208). 9-126. (049) Valdlguk, P. M.; Rogarl, E.; Phllllppe, H. CHn. Chem. 1992, 38, 83-87. (050) Wellefi, M; nlzer, L. L. CHn. Lab. W .1991, 11. 135-152.

(051)Tletz, N. W.; Shuey, D. F.; Wekstdn, D. R. C h . Chem. 1902,38, 11671185. (052) Kost, G. J. J. Int. Fed. a n . Chem. 1991, 3 (4), 160-172. (053) Kost, G. J. Arch. Palhd. Lab. Wed. 1982, 116, 1002-1003 (dltorlal). (054) Selem, M.; Chernow, B.; Burke, R.; Stacey, J.; Sbgoff, M.; Sood, S. J. Am. W . Aseoc. 1991, 266, 382-389. (055) Statland, B. E.; BrZyS, K. Cheet 1990, 07, 198S-203S (SUPPI). (056)Mlstlano, D. R.; Meyerhoff, M. E.; Cdllson, M. E. chsst 1990, 07,204s214s (Suppl). (057) Wilding, P.; Erlkson, K . A. CNn. Chem. 1990, 36, 1206 (abstract). (058)Conn, H. 0. Clkr. Chem. Acta 1992, 210, 1-3 (dltorlal). (059) Hulzenga, J. R.; Tangennann, A.; Qlps, C. H. C//n. chkn.A& 1992,210, 153-1 55. (080) Keshgeglan, A. A.; Bull, G. E. Am. J. C h . Pethd. 1992, 07, 535-540. (061) Nosanchuk, J. S.; Stull, R.; Keefner, R. Lab. M.1091, 22, 465-469. (062) Bakker, A. J.; Gorgele, J. P.; Draalsma, J.; Jongendljk, M.; Altene, L.; Hamersma, A.; Welland, A. CHn. Chm. 1992, 38, 2221-2223. (063) Pares, L.; Ceballero, L.; Pardo, C.;Serrano, J. C//n. Chm. News 1992, 18 (12). 20-21. (064) Ozawa, K.; Schlpelsky, P.; Pardue, H. L.; Place, J.; Truchaud, A. J. Int. Fed. CUn. Chem. 1992, 4, 174-181. (065) Columbus, R. L.; Palmer, H. J. C h . Chem. 1991, 37, 1548-1556. (066)Schembri, C. T.; Ostdch, V.; Llngane, P. J.; Burd, T. L.; Buhl, S. N. CHn. Chem. 1991, 38, 1665-1670. (067) Sterling, 6.;et al. C h . Chem., 1992, 38, 1858-1664. (068)Glbbons, I.;(krln,M.;Yasslnradeh,Z.;Petereon,P.;Be~~,D.;Dilbn, K.; Bud, T.: Hlllman, R.; Smoluk, G.; Cobb, M. C//n. Chm. 1989, 35, 18891873. (089) Kraft, C.; Olson, J.; Jackson, B.; Shavez, J. CHn. Chem. 1992, 39, 1034 (abstract). (070)Ng, R. H.; Sparks, K. M.; Statland, 8. E. CHn. Chem. 1991,37,435-437. (071) McGowan, M; ONell, J. C//n. Chem. 1990, 36, 967 (abstract). (072) Meloy, N.; Stein, E. CHn. Chem. 1990, 38, 963-964 (abstracts). (073) Ng, R. H.; Sparks, K. M.; Hlar, C. E. CHn. Chm. 1992, 38, 1647. (074) Ashworth, L.; Gibb, I.; A I M , K. G. CMn. Chem. 1992, 38, 1479-1482. (075)Mekw,S.;Ehnneyer,S.;Leesslg,R.; Woodlelf,G.;Lenphear,B.;Haswmer, D. Arch. Palhd. Leb. W .1992, 116, 524-530. (076) Hardy, R. W.; et al. W.Chem. 1992, 38, 1437-1439. (077) Dybkaer, R., Mattln, D. V., Rowan, R. M., E&. Sand. J. CHn. Lab. Invest. 1892, 52 (SUPPI.208), 7-168.

ANALYTICAL CHEMISTRY, VOL. 65, NO. 12, JUNE 15, 1993

4491