Printer Friendly

A rational way to select a chemistry analyzer.

A rational way to select a chemistry analyzer

Selecting a new chemistry analyzer is a big decision for any laboratory. At stake is a large investment both in dollars and in the personnel time that will be spent working with the instrument over its lifetime.

The choice calls for serious study, but that is not an easy thing to carry out, given the variety of analyzers on the market and the claims made for each. In this article, I will outline a three-step approach to the process of selecting a chemistry analyzer, pointing out criteria that are especially useful to look at in each stage. By following these steps, you should be able to locate a system that performs well and meets the needs of your laboratory:

Define what you want in an analyzer. Certain standard factors need to be considered, and it is helpful to set these down early. You can determine your needs and instrument-selection criteria by following the outline in Figure I.

The first group of criteria concern the hospital and how the instrument will be used in your laboratory. Hospital bed size is one predictor of workload, but perhaps more important are the types of services your hospital provides. Specialized units should be noted and reviewed in relation to the kinds of demand they place upon the lab.

For example, active critical care services such as large ICUs and trauma, cardiac surgery, and other special services create heavy demand for Stat testing. If your hospital has a large pediatric service, capacity for handling small specimen volumes may be important to look at.

What types of testing will the new analyzer be used for? Is your workload heavy with profiles or large batches of tests, or do you run most specimens one at a time? Do you expect to use the analyzer just during the day or on all three shifts? Do you need a system with the capacity to run electrolytes, or do you have other instruments for that purpose? Will the analyzer be required for special tests such as quantitation of proteins or therapeutic drug monitoring? Note all needs on your list of selection criteria.

Consider the instruments already in the lab. Note their strong and weak points and how a new system will complement them in getting the daily work out. If you are replacing an existing system, note why and what improvements you desire.

Also define your price range. If this is a more or less fixed amount, a constraint imposed by your hospital administration, you may be able to justify a larger outlay on the basis of an expanded range of capacities. At this point, you should be able to decide whether you are interested in a $50,000 analyzer that does 100 tests an hour or a $180,000 instrument with a capacity of 600 to 800 tests per hour.

Gather data on available systems. You are probably already familiar with a number of instruments through reading, word of mouth, and other contacts. To widen your possibilities, touch base with colleagues at other hospitals in town or other labs, and find out which instruments they have had good experience with. Check CAP proficiency survey and quality control program data summaries to see how many laboratories are using a given instrument and how consistent their results are.

Advertisements in journals and mailings, exhibits at meetings and conventions, and calls by salesmen are all perfectly valid means of learning what is on the market. You must, however, thoroughly investigate the instruments that are so promoted to determine if they live up to their promises.

Once you have narrowed down the field to systems you want to study, compile some basic information about each on a specifications sheet (Figure II). Most of this material, some of it quite technical, can be obtained from brochures and salespeople. I'll explain what to look for and why it is pertinent.

The array of chemistry analyzers and the variety of mechanical, optical, and other arrangements used in them appear almost endless, but most systems fall into three categories: random access, batch, and profile. Random access analyzers use a single wheel or other series of cuvettes, either reusable or disposable, into which specimens and reagents for various tests are pipetted in the order the tests are entered into the instrument. Batch instruments run multiple specimens simultaneously for a single test, while profile instruments run multiple tests in separate analytic channels. No one approach is best; different modes of operation may work better than others in different situations.

The most important aspect of an instrument is its throughput, or the number of tests performed in a given time. Manufacturers' tests-per-hour numbers have to be analyzed carefully to understand how they are derived. For example, an instrument that can process a new colorimetric test every 15 seconds puts through 240 tests per hour. If the instrument is figured hypothetically to run three electrolytes on a separate electrode system, then the throughput would be 960 tests per hour. But the manufacturer may be putting together tests that the laboratory does not frequently perform in combination.

A useful common denominator is the number of photometric tests per hour, which can be calculated as 3,600 divided by the pipetting time in seconds. If an instrument loads a new test every 10 seconds, it can do 360 tests per hour. For multichannel instruments that may run a full profile on some patients and fewer tests on others, specimen cycle time--the time between starting tests on each new specimen--may be a more valid index of instrument capacity.

Specimen and reagent volume specifications are key points in connection with pediatric and other kinds of micro-sample testing. The volumes also affect reagent costs. Most modern instruments use total sample-reagent volumes in the range of 300 to 400 l, but the specific numbers may be smaller or larger.

While the number and range of available tests are important, how many tests are on the system at one time may be more critical. This is usually limited by the number of positions on the reagent rack or carousel. What if your usual admission profile or Stat repertory includes 15 tests and the analyzer can hold only 12? You may have to change reagent trays frequently and do two runs per patient.

Storage and stability of reagents also are factors to consider if you plan to run Stat or off-shift tests.

And where would you get the reagents? If an instrument is not locked into only the manufacturer's reagents, you have the flexibility to develop your own test methods or shop around for reagents.

How many calibrators the instrument needs may affect the range of accuracy and linearity, and the frequency of standardization will surely affect how much time you spend running calibrators. Some instruments use only a single-point calibrator or two or three calibrators. Others use up to five or six, adding to reagent cost and procedure time. Check out claims. An instrument may remain stable for just a few days although its manufacturer indicates calibration is needed only once a week.

A good kinetic photometric system, taking multiple readings over a period of time and calculating reaction rate from these readings, is important for enzyme assays and critical for certain tests such as EMIT drug assays. A small change in measured reaction rate translates into a large change in calculated drug concentration in these assays, especially at the high end of the standard curve. A minimum of 8 to 10 readings are necessary to insure valid reaction rates.

Approaches vary on electrolyte measurements. Most systems use ion-selective electrodes for sodium and potassium, but some perform the measurement on diluted specimen, and others read directly from undiluted specimen. The indirect and direct readings should be considered different methods, and the results may not correlate with each other.

In addition to the technical data, consider the instrument's market history. A new analyzer may look promising, but unexpected problems often come up: Instruments in hospital use less than a year are still unproven and should be evaluated with extra caution.

A substantial market base and nearby service make good vendor support more likely. Price information on the system itself and on such major options as electrolyte or data-handling modules, which some users will need, should be readily available from sales representatives. Per-test reagent cost information varies widely with the test and may be confusing. CK is a common, relatively expensive test that can serve as a basis for comparison. The cost of disposable rotors or other cuvettes--as much as 10 cents per test on some systems--should certainly be noted and taken into account.

With this basic information in hand, you can narrow the field to two or three systems, having climinated those that are too slow, too limited in capability, or more complex and expensive than your needs. Negative impressions from other users or from seeing analyzers demonstrated may further limit the choices.

Ask manufacturers of the candidate instruments for user references. These should be institutions as similar as possible to yours, making it likely that they use the system as you will. They should also have had the instrument in place long enough, at least a few months, to have gained reasonable experience with it.

Check the experience of users.

Contact the chemistry supervisor or whoever has the most direct experience with the instrument. If that person is unavailable, don't settle for someone else. Call at another time or leave a message.

Figure III shows what to ask users. Basic hospital demographics and information on how the instrument is utilized are straight-forward and should take only a few minutes to obtain. Ask what other instruments they considered and why they selected the one they did. Besides giving information on the instrument they have, this may also shed light on the pros and cons of other systems.

Start-up experience can be revealing. Did the analyzer work well right out of the box, or did they have to spend weeks or months getting everything to work? What about daily maintenance? Did the vendor's claim of 15 minutes each morning turn out to be more like an hour or two?

Day-to-day precision on certain critical tests is one of the most important criteria to look at. Almost any system can give acceptable results for such analytes as glucose or BUN, but certain tests are more troublesome. Those listed in Figure III are, in my experience, particularly susceptible to problems. They are good examples to use in evaluating an instrument's performance.

The clinically significant variation in calcium is close to the analytic precision of many systems. A good analyzer should achieve a coefficient of variation of 2 per cent or less on this test. Creatinine, bilirubin, and AST often have large CVs in the medically important upper-normal to minimally elevated range. A CV of 7 per cent in this range is clearly preferable to one of 12 per cent.

Reagent instability is a frequent problem with creatinine and uric acid assays. Some general questions to users about the various chemistries may elicit valuable information. Perhaps they found that the uric acid reagent supposedly stable for 24 hours was good only for six, or that they have to spend an hour a week running a special cleaning routine to keep the phosphoruses within range. Calcium, phosphorus, and iron are especially vulnerable to carryover contamination in systems with reusable cuvettes. Ask specifically about such experience with analyzers of this type.

To evaluate instrument reliability and service support, ask users about downtime and how promptly and effectively problems were remedied. Such reports may also identify frequent problem areas. The final two items in Figure III are global questions, inviting users to state what they like best and least in their analyzers.

By this time you should have enough information to make a final choice. But if no system does what you want or has the endorsement of satisfied users, or if no clear choice exists between a couple of leading contenders, sit back and wait a few months. By then some initial problems with an instrument may have been worked out, or more users may have experience with the system. A decision as big as the purchase of a new instrument should not be taken lightly or hurried.

On the other hand, if the evaluation process leads to a system that looks good for your lab, you are ready to negotiate a price and sign on the dotted line.

Photo: Figure I Defining needs and selection criteria

Photo: Figure II Gathering data on a chemistry analyzer

Photo: Figure III What to ask other users
COPYRIGHT 1986 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1986 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Horvitz, Richard A.
Publication:Medical Laboratory Observer
Date:Jul 1, 1986
Previous Article:A cost-effective emergency room laboratory.
Next Article:MBO monitored by computer.

Related Articles
Creating a calibration data base.
A single-instrument approach to TDM toxicology.
Instrument sales growing despite Washington.
A microcomputer test-costing program.
Reequipping the lab: a brisk pace of renewal.
Do automated instruments reduce job satisfaction?
Pitfalls to avoid in bringing a new analyzer on line.
Preparing the lab for the year 2001: workstation consolidation.
Integrated automated task targeted analyzer systems: an antidote to front-end processing. (Lab Management).

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters