Now it’s up to microbiology to catch-up on automation

With testing volumes going through the roof, automation is needed more than ever, particularly in underserved geographic areas.

Although clinical chemistry laboratories had a huge advantage in total automation of their workflows in many, yet not all, analysis steps, microbiology had some modest approaches to customize specific devices. Automated blood culture systems and instruments for identification and antimicrobial susceptibility testing (ID/AST) were introduced more than 30 years ago. The major backbone of the workload in the lab – specimen handling, plate inoculation, incubation, and reading – remained manual processes.

About 10 years ago, devices came to the market that allowed specimen inoculation and, later, also plate transportation to smart incubators, which then were able to read plates digitally. This development was accompanied by numerous new technological approaches, but only a few of them really became routine procedures. The idea of a completely automated microbiology workflow, beginning with the arrival of specimens in the lab to the final reporting in the lab information system (LIS), grew in the field of clinical microbiology and evolved rapidly.

While much better than the older methods, automation in the microbiology lab has not really changed that much in the last 30 or so years.

Meanwhile, many procedures in the microbiology lab were automated, including blood culture analysis, automated AST instruments, and recently, the culture workflow starting with specimen inoculation up to final imaging analysis. Some of the current processes for most microbiology labs and the limited automation that is currently in use include:

Continuous blood culture monitoring. Rather than Gram staining and plating what could be hundreds of blood culture bottles, automated continuous monitoring blood culture instruments do a lot of work for the tech. The blood culture bottles contain a fluorescent dye that reacts with the CO2 the bacteria produce. This dye emits a signal; the blood culture instrument reads this signal and alerts the technologist that there is a positive blood culture. The tech then removes the bottle from the instrument, plates the blood culture, makes and reads the Gram stain, enters the results into the LIS, and calls the value to the clinician. These instruments can be interfaced with the LIS, and negative blood cultures can be updated at set intervals. This saves the technologist time and prevents the technologist from performing menial tasks. The technologist can devote more of their time to positive cultures.

Automated microbial ID and sensitivity instrumentation. There are several different types of automated ID and sensitivity instruments on the market. Essentially, they all work the same way. Once the tech makes the appropriate suspension of the bacterial isolate and inoculates the ID-MIC panel, these automated instruments incubate, add all required supplemental reagents, and read all biochemical reactions and MIC values. They allow for much easier setup and quicker identification of the bacteria. These instruments perform all the same biochemical tests that a technologist would have to set up individually. The instrument uses an online database to compute the ID, based on the positive and negative reactions.

Rapid ID molecular instrumentation. Rapid ID molecular instruments are normally used for stat testing, which would include flu, strep A, SARS-CoV-2 (COVID-19), among others. Some platforms have even introduced rapid ID of things like CREs, MRSA, and VREs. This shortens the time of ID from days or weeks in the case of viral cultures.

Automated specimen processors. These analyzers plate the incoming microbiology specimens. The instrument is programmable to set what kind of plates to use for a particular culture and what kind of streaking pattern should be used. These instruments use liquid media into which the patient samples are inoculated. These tubes are loaded onto the instrument either capped or uncapped depending on the instrument. The instrument pipettes a drop of the liquid in the tube onto the preprogrammed agar plates for that particular culture. The instrument has a bar code reader that scans the specimen label to know what kind of culture it is. Once the plates are inoculated, the instrument streaks the plates, moves the plates to the off rack, and sorts them by culture type. The technologist can then remove these plates and place them into the incubator.

Total laboratory automation instruments. These instruments process the specimens, incubate the specimens, read the plates, and store records of past patient histories. They often have an open architecture, meaning the incubator and processing sections can be added and removed as needed. After processing specimens, the instrument then loads the plates into incubators, reads the plates at intervals updating the no-growth plates in the LIS, and alerting the technologist of positive plates that need attention. This type of system records images of the plates allowing the technologist to look at patient histories and compare current bacterial growth to past bacterial growth. Right now, it is not possible to load positive plates directly onto an ID/sensitivity instrument; however, some companies are trying to develop this.

Why has automation in microbiology been slow to catch on? Some of the reasons why automation has not caught on to the same degree as other areas of the lab include quality concerns, lack of specimen standardization, cost, and space.

Microbiology is a technical area. The technologist has to look at the cultures and interpret which bacteria are pathogens and which bacteria are normal floras from a particular site. There are a lot of little nuances.

For example, alpha-hemolytic Strep generally is not worked up in sputum cultures unless it looks like Streptococcus pneumoniae or it is the only bacteria growing in the culture. Coagulase-negative Staph generally is not worked up in any culture unless it is a culture from a sterile site such as blood, body fluid, or tissue, or urine from a woman of child-bearing age. Coagulase-negative Staph may be mentioned in the final report of a wound or sputum culture, depending on how many other organisms are present, but the actual species is unimportant. All these little rules and exceptions would be incredibly difficult to program into an instrument and it is more efficient to have a human looking at the plates.

Other quality concerns include making sure that cultures are pure before loading into an ID/AST instrument. This means that a technologist may have to isolate bacteria onto a separate agar plate before setting up for ID and sensitivity. If a mixed culture is used for ID, the two different bacteria can cross-react with each other and give an erroneous ID.

This is probably one of the biggest roadblocks to implementing automation. Unlike chemistry and hematology where specimens come in a standardized tube, microbiology specimens can come on a swab, in a sterile container, a bottle, a test tube, and even Tupperware-like containers. It is very hard to build an instrument/analyzer that can accommodate all these different containers.

This is probably the other big roadblock. The current instruments/analyzers that are on the market are very expensive, costing several hundred thousand dollars. As it stands right now the only labs where it makes sense to have these types of analyzers are high-volume laboratories.

Automation takes up space. The analyzers need to be in a well-ventilated area, on a level surface, and cannot have a lot of debris around them. This may require remolding and rearranging of laboratory furniture. This is an arduous process and disrupts the technologists and workflow of the laboratory.

Microbiology is one of the more complicated areas of the laboratory and, therefore, it has been very hard to automate. Technology is becoming more affordable, smaller, and more advanced, making it a possibility that microbiology will soon be as automated as chemistry and hematology. Many people may worry that automation will cost jobs; however as it stands now, the laboratory field is experiencing a staffing shortage that is only going to get worse. Technology can help fill in the gaps where there used to be people. Soon the microbiology lab will be completely unrecognizable.

With testing volumes going through the roof, automation is needed more than ever, particularly for small and mid-sized labs that are in underserved geographic areas. Demand for laboratory testing is increasing. Overall testing volumes are expected to increase 10 to 15 percent per year for the next 20 years, due in part to an aging population that will require more healthcare. Additional testing is also being driven by innovations in medicine that continue to expand life expectancy and manage ever-more-complex patients. For example, more patients are receiving indwelling devices which can become infected, increasing the demand for laboratory services. Infection control also continues to drive utilization of laboratory services through patient screening initiatives and increased vigilance to isolate patients colonized with multidrug-resistant pathogens, and to prevent their spread within the healthcare environment. Each of these factors, while improving the quality of healthcare, contributes to increased demand on the laboratory, despite continued labor shortages. Consolidation of laboratories, particularly for microbiology testing, also continues to increase due to cost reductions associated with economies of scale. Larger laboratories have a greater potential to benefit from lab automation than smaller laboratories. The 24-h, 7-day/week (24/7) microbiology laboratory is becoming much more common, and automation that can shorten TAT is being viewed more favorably. The 24/7 microbiology laboratory also allows cultures to be read following a specified incubation rather than waiting for the day shift, a scientifically unnecessary delay which can result in delays in turnaround times.

Clinical laboratories have played a critical role in response to the COVID-19 pandemic, drawing attention to the work laboratory professionals do, as never before. In addition to fulfilling the current testing needs, laboratories have an unprecedented responsibility to prepare for the pandemic’s aftermath.

The pandemic has also put enormous pressure on clinical laboratories – from increased testing volumes and turnaround time to biosafety, workforce planning, and social distancing within the lab.

The problems, such as shortage of laboratory technologists and technicians, historic financial pressures on hospitals and health systems, and physically and mentally exhausted laboratory professionals — which existed before the pandemic — have only been exacerbated. The shortage of staff is considerable when you look at regional labs staffed with smaller teams. These labs have fewer resources than larger labs, but continue to test at two to three times their average volumes as they secure more local employer and provider COVID-19 testing contracts, while the larger labs often capture the major testing contracts for situations such as student body testing at higher education institutions.

Workflow automation, an important breakthrough in the recent history of laboratory diagnostics, integrates multiple diagnostic specialties to one single track to improve efficiency, organization, standardization, quality, and safety of laboratory testing.

The major economic revenue of workflow automation, resulting from merging many diagnostic platforms within a consolidated system, not only encompasses a reduction of manual workforce (especially auxiliary and technical staff) needed for managing high-volume testing, but is also attributable to lower pre-analytical and post-analytical expenditures.

Automation plays a key role in addressing staff shortages while enabling precious resources to focus on high-value clinical tasks, particularly during the COVID-19 pandemic.

Even though molecular testing, which is the gold standard for diagnosing COVID-19, is not included in current workflow automation solutions, automating manual processes for routine tests free up lab resources to do more COVID-19 or other dedicated scientific work.

In labs that process fewer than 5000 tests a day, there are maybe one or two technicians on staff, making workflow automation even more vital. While automation is a reality today for large laboratories, the available solutions are neither total nor wholly automated for small- and medium-sized labs.

COVID-19 has highlighted the need for small- and medium-sized labs to cater to populations that do not have easy or fast access to large laboratories. In the future, the drive to consolidate will have to be balanced against the need to have adequate geographic coverage for testing. This means that small- and medium-sized labs will continue to see a rise in test volumes without an increase in resources, making automation an attractive option.

The lab automation that caters to labs processing high volumes of samples is out of reach for smaller-volume labs due to space requirements and infrastructure constraints.

The right fit for these labs is a solution that offers comprehensive workflow automation that reduces manual steps through pre-analytical, analytical and post-analytical automation, conserving precious human resources to do higher-value clinical tasks while fitting within the space constraints; adapts to meet a mid-volume lab’s space requirements and infrastructure needs with a flexible design; and dynamically calculates route scheduling for rapid and consistent turnaround time (TAT), with short turnaround times (STATs) always prioritized to deliver results faster with intelligent routing.

Just because a lab is not processing a large number of samples should not mean it has to sacrifice the benefits of intelligent laboratory automation and settle for a marginally automated work cell plus solution. The market is ready for workflow automation that can help small and medium-sized labs meet the testing needs of their community.