With an increasing awareness that undiagnosed and untreated infections such as Chlamydia trachomatis (CT) can cause serious complications in asymptomatic women, healthcare programs are seeking ways to screen patients more efficiently. Many laboratories are challenged with the task of choosing automated platforms to handle the workload. We conducted two studies to provide some logic for making a decision, and developed protocols to compare four assays and five instruments from prominent diagnostic companies. Choosing an automated laboratory instrument is not an easy task as a lab has to consider both current and future needs. The most transparent considerations are space and costs for equipment, assays, and consumables. Labor costs require an understanding of the amount of hands-on time that is required. Other aspects that need to be considered include testing capacity for the number of tests and the variety of analytes available, work flow, time-to-results, maintenance, and the accuracy of the assays to be put through the instruments.

Our studies compared four batch processing instruments (Abbott m2000, BD Viper XTR, Roche cobas 4800, and Hologic Tigris) and a continuous feed, random access instrument, Hologic Panther. We compared the respective CT assays for sensitivity and specificity in a head-to-head fashion using urines and self-collected vaginal swabs from 575 women. This enabled us to use results from four assays as a patient-infected status of comparison. We spiked part of each urine and vaginal swab with a laboratory strain of CT to determine whether inhibitors of CT amplification in the samples were playing a role in the comparative outcomes. We recruited four sites performing different assays and they served as the sources for instrument comparisons by two investigators conducting two visits of 2 days each to time interactions taking place during runs of 96 and 192 tests.

Eight Factors of Importance

Our studies revealed eight factors of importance in considering automated STI instrumentation. First is the analytical sensitivity of the assays. In our head-to-head comparisons, we found that the assays varied in their abilities to detect infected patients, and that these variances likely were due to differences in analytical sensitivity rather than the inhibitors used in the assays.

Another consideration is the amount of laboratory space required for the instruments. The instruments we compared needed as little as 10 square feet and as much as 25 square feet.

Automation time is an important factor as well, because more automation time provides more walk-away time for other technical laboratory tasks. In our studies, we found that automation time ranged from about 5 hours to just over 3 hours. Separate from but related to automation time is time-to-results, as this also impacts workflow. Does an instrument start generating results even while it is still in its automation process? We found that one instrument finished its automation after 5 hours but started generating its first result at about the same time as another, which completed a batch run at 3.5 hours. A fifth factor labs should consider in choosing automated instruments for STI testing is that total testing capacity and testing different sized batches can impact reagent costs. How many specimens does the instrument allow during initial loading? Does it allow additional reagent kits and samples to be loaded as needed during processing?

Time at the Instrument

In today’s cost-conscious environment, hands-on time at the instrument—which is heavily influenced by instrument design—is another important consideration. We recommend that laboratorians look at whether the instrument has separate units for extraction and amplification, because this requires extra time for return visits, regardless of the batch size. Second runs on some instruments require time for loading samples, reagents, and consumables, whereas with others this takes only a few seconds to feed in additional reaction tubes. On a separate but related note, laboratories also would want to consider the time needed for maintenance. Two features we were unable to compare in our studies include random access to multiple analytes and consumption of reagents and disposables. Random access to multiple analytes is a nice feature, but in our experience only the Panther possessed this feature for multiple STIs. In our published work we didn’t explore consumption of reagents and disposables, but we recently examined this issue and found the differences remarkable. For 192 tests, we used from 240 to 1,504 pipette tips, depending on the instrument. Overall, we recommend that labs consider assay performance and costs for equipment, reagents, labor, and consumables to provide a true assessment of automated instrument value.

References

  1. Ratnam S, Jang D, Gilchrist J, et al. Workflow and maintenance characteristics of five automated laboratory instruments for the diagnosis of sexually transmitted infections. J Clin Microbiol 2014;52:2299–304.
  2. Chernesky M, Jang D, Gilchrist J, et al. Head to head comparison of second generation nucleic acid amplification tests for Chlamydia trachomatis and Neisseria gonorrhoeae on female urines and self-collected vaginal swabs. J Clin Microbiol 2014;52:2305–10.

Max Chernesky, PhD, is a professor emeritus at McMaster University and St. Joseph’s Healthcare Hamilton in Hamilton, Ontario, Canada. These studies were funded by a research grant from Hologic Inc. Dr. Chernesky has no consulting relationships with the four companies involved in these studies.

He has received research funding and speaker honoraria from Abbott, Roche, and Hologic.

+Email: chernesk@gmail.com