The coming of age of microbiology lab automation

In today's competitive healthcare industry, automation is at the forefront when it comes to streamlining operations and advancing patient care. Automation has also gained a foothold in the laboratory, generating significant gains in productivity and changing many outdated operations. However, these gains did not come easily or without opposition. Even as analyzers began to infiltrate laboratory processes and preanalytics and as line systems hit the market, many people thought their use would be restricted to academic institutions or large reference labs. A common initial reaction was to see them as bulky, costly, and impersonal.

But with laboratories under pressure to deliver accurate results quickly and with fewer resources, the landscape has changed. Full automation has now become standard in chemistry and hematology labs. Laboratories that have followed the trend toward automated systems often find that technologists are more likely to embrace innovation, which leads to improvements in quality and recouping of investments. The shift to automation has provided an opportunity for creating additional platforms to accelerate analyte identification and save time. More-accurate information is the product of automated solutions.

Today's laboratory challenges, such as the shortage of skilled laborers and increase in work volume, make the argument for automation quite compelling. But making that case to the microbiology lab has been a hard-fought battle. Until recently, few people believed that automation could become an aspect of daily practice for microbiologists. After all, microbiology has historically been considered one of the most “manual” and complex of scientific disciplines. For many, automating these processes seemed nearly impossible and was thought to carry too high a price tag.

The Case for Automation

Microbiology laboratories today are experiencing a collective attitude shift similar to that which chemistry and hematology labs went through years ago. Full automation is no longer a distant and perhaps dreaded possibility, but is implemented in labs across the globe. The reasons for this are as complex as microbiology itself.

Workload. Microbiology labs are facing growing pressure from a variety of sources. Chiefly, the increase in patients with chronic illnesses, the prevalence of healthcare-associated infections (HAIs), and the emerging drug resistance of bacteria, as well as an aging population, have caused more samples and requests for analysis to pour into laboratories. Along with this, the microbiology laboratory itself has been evolving as it incorporates a more diverse range of technologies, such as immunoassay and molecular diagnostics, to increase its ability to identify samples faster and produce more-accurate susceptibility results.

For instance, the Kaiser Permanente laboratory in North Hollywood, CA, runs 1.1 million tests a year with an average of 99,000 tests per month. Exacerbating the volume demand, skilled technicians are becoming more difficult to find and also expensive. Industry experts estimate that for every seven clinical laboratory scientists facing retirement, only two enter the field.¹ Additionally, quality demands have lowered the tolerance for errors and increased the need for improved traceability and reporting in microbiology labs.

[5]

Figure 1. (click to enlarge [5]) Labor demand by process step.

Efficiency. Routine tasks such as plate streaking can occupy much of a microbiologist's time, contributing to lab inefficiencies and slow response times (see Figure 1 [5]). Time dedicated to routine could be better spent interpreting results of identification/antibiotic-susceptibility testing (ID/AST) to help quickly deliver the crucial information that clinicians need to make timely and accurate diagnoses.

Even if an abundance of microbiologists did exist, they would still not be able to overcome the laws of nature. Laboratory conditions limit the clinical productivity of microbiologists using only manual processes. If a colony requires 24 hours to culture using conventional laboratory means (e.g., an agar plate), then a result cannot be obtained for immediate effectiveness. The physician can use such a test more as a confirmatory measure than as a diagnostic tool.

Automating the microbiology laboratory is not easy. The effort is challenged by the variety of specimens, volumes, containers, and processing methodologies involved. Microbiologists work with blood, urine, and a host of other body fluids: sputum and aspirates; fungal and wound-site collections; scrapings and tissue samples; spinal fluids; and stool specimens, to name a few. Samples may come in tubes, on swabs, or in variously shaped bottles and containers. Without automation and standardization as part of sample collection and handling, the chance that misidentification or cross-contamination of specimens will occur increases.

Manufacturers have been challengd to develop a universal preanalytical system for microbiology that can complete the sample preparation required. From bar coding to standardizing the containers, when standardized preanalytical systems are in place to handle such tasks, microbiologists are able to work more efficiently and effectively.

Better Results. Many laboratory tasks can be specified but are repetitive, resulting in higher strain indices and variances. But inconsistent, or improper, techniques can affect quality and, ultimately, patient outcomes. And delayed or incorrect diagnoses can result in improper treatment, greater patient length of stay (LOS), adverse drug events, medication errors, and even loss of life. Automation can ensure the presence of key laboratory performance attributes, such as consistency and quality, at every process stage: preanalytical, analytical, and postanalytical.

Relying on bar coding, robotics, and computerization rather than manual transcription significantly reduces data loss and errors. As this technology advances, the accuracy of data transfer and the thoroughness of information regarding the patient, specimen, plate, and more will continue to increase. Automation also reduces the likelihood of plate information and patient identification being duplicated or transposed, providing for more-accurate diagnosis and better patient care.

Utilization of Resources. Automated solutions free technologists from most repetitive manual processing labor and give them more time to focus on complex tasks that require their professional skills and experience. Often, laboratories can build an automated system in order to take on greater volumes and expand services, adding value to the institution's bottom line. Modular designs allow scalability, while advanced systems with intelligent interfaces ensure that data are accurately collected and stored throughout the laboratory process. Integration of systems increases efficiency, maximizes output, and minimizes delays in delivery of results.

Microbiology Automation Today

Time to result is becoming ever more important with the tremendous proliferation of HAIs. The hospital microbiology lab provides a front line against the growing threat of community-acquired and nosocomial infections. According to the Centers for Disease Control and Prevention, HAIs account for an estimated 1.7 million infections and 99,000 associated deaths annually in U.S. hospitals alone.² Of these infections, 32% are associated with the urinary tract, 22% with the surgical site, 15% with the lung (pneumonia), and 14% with the bloodstream.²

The financial costs associated with this problem are high. The estimated average cost of a non-life-threatening nosocomial infection is about $15,000, while a serious bloodstream infection costs an additional $57,000 on average.³ Delivering the appropriate treatment more expeditiously can reduce these costs, as well as others less tangible, resulting in improved antibiotic management, fewer side effects, decreased hospital mortality, shortened LOS, and better outcomes.⁴

The Analytical Phase. To increase the speed and relevance of test results, automated systems have been developed to assist with the analytical phase of the microbiology laboratory work flow. The authors' own company, bioMérieux Inc. (Marcy l'Etoile, France) offers a number of instruments dedicated to helping microbiologists in this regard, particularly with ID/AST work flow. The Vidas system has a routine menu that includes more than 80 parameters in immunochemistry and infectious disease, BacT/Alert 3D is a small-footprint microbial detection automation instrument, and DiversiLab, a tool for uncovering hidden reservoirs of resistance in hospitals, provides DNA fingerprinting and analysis of both bacteria and fungi.

Nucleic acid extraction and amplification is another tool set that should become a standard for routine testing. BioMérieux's NucliSens easyMAG and easyQ platforms use this technology to identify viral targets.

Testing menus continue to expand in response to demand. Representative of this is the automated test for the detection of Clostridium difficile from bioMérieux, VIDAS C. difficile Toxin A&B.

All of these systems can be integrated with the laboratory information system (LIS) to alert lab personnel the moment results are available. Often, the results can be delivered electronically to physicians, allowing them to react more quickly. Whether results are positive or negative, clinicians receive the critical data needed to make an accurate diagnosis and determine the appropriate course of treatment. Patients in turn receive higher-quality care delivered more quickly, and administrative costs are reduced.

Moving beyond plates and cultures, a wider range of tools are available for getting to the source of infection. Doctors thus are more aware of what they are dealing with and can better treat their patients. Phages is an example of recent technological advances in immunoassay platforms that allow assays to target a pathogen or virus of interest exclusively. This capability dramatically increases the accuracy, sensitivity, speed, and specificity of medical testing.

[6]

Figure 2. (click to enlarge [6]) bioMérieux's Full Microbiology Lab Automation Bench.

Preanalytics. Not only the analytical stage of the microbiology work flow can be automated; so can the preanalytical phase. It is the goal of bioMérieux to automate the entire microbiology workbench (see Figure 2 [6]). This strategy is based on the conviction that full microbiology laboratory automation is necessary for streamlining operations. Preanalytic work flows tend to be mundane and time-consuming, yet maintenance of specimen integrity throughout is essential. Misidentification, improper handling, delayed entries, or inadequate storage can ruin a sample and lead to inaccurate results.

Developing an automated universal microbiology specimen collection system has been a challenge. Microbiology system manufacturers continue to search for methods that will work with different specimens and various collection devices. While developers pursue this Holy Grail, they are also considering other possibilities. Some manufacturers have collaborated by consolidating research and development efforts in order to produce solutions more quickly. For example, bioMérieux has partnered with several companies individually to automate key microbiology lab processes.

Figure 3. bioMérieux's Previ Isola Automated Pre-Pour Media Streaker.

One such process is plate streaking, which is the most labor-intensive task in the microbiology laboratory. BioMérieux signed a licensing agreement with LabTech Systems Ltd. (Kent Town, Australia) to use that company's patented robotic instrument in its own Previ Isola automated prepour media streaker (see Figure 3).

Figure 4. Plates streaked by bioMérieux's Previ Isola Automated Pre-Pour Media Streaker.

The device automates routine agar plate processing, managing biplates as well as up to five different types of media at one time. The system uses a comb-like mechanism and plate rotation to streak plates the same way every time (see Figure 4). Streaking is consistent, without the variation typically seen when more than one microbiologist handles the task. Inconsistent streaking can lessen the success of a culture, as well as delay the next process step if not completed in a timely fashion. But good consistency optimizes quality, traceability, and turnaround.

An average laboratory that streaks roughly 1000 specimens a day can save about one full-time employee's worth of labor using the system. The instrument also spares the technologists from about a thousand streaking motions a day, thus reducing the potential for occurrences of repetitive-motion disorder.

Figure 5. bioMérieux's Previ Color Gram Automated Gram Stainer

The Previ Color Gram automated Gram-stainer is the fruit of a partnership between bioMérieux and Wescor Inc. (Logan, UT) (see Figure 5). The former has incorporated the latter's automated spray technology into its cytocentrifugation Gram-staining device. The cytocentrifuge's rotor uses centrifugal force and three patented-design chambers to sediment cells onto the slide. Suspension fluid is simultaneously absorbed into an absorption pad as cells come into contact with the microscope slide with the single or dual standard-volume chambers. A high-volume chamber retains the excess fluid from a suspended sample. The system produces ready-to-use slides. With an eight-slide capacity and designed to be technologist-independent, it requires no technical expertise of the user performing the staining process.

The system's hands-off design saves time and promotes safety. By reducing the amount of reagent used, it saves money and minimizes chemical waste.

Urinalysis. Another step that has been automated for the microbiology laboratory is urinalysis, a process that, despite high volumes, is still based primarily on manual microscopic cell enumeration. Kaiser Permanente estimates that it performs 38,000 urine cultures a month, more than a third of its total testing volume. Urinary screening is one of the most frequently performed analyses in laboratories. It requires accurate detection and enumeration of the bacteria and particles present in a urine specimen and rapid confirmation as to whether the patient could have a urinary tract infection or an orientation toward renal pathology. What has been needed most is a time-saving automated instrument that produces results efficiently.

Figure 6. The UFI-1000i, which was developed through a partnership between bioMérieux and Sysmex Corp.

Developed through a partnership between bioMérieux and Sysmex Corp. (Kobe, Japan), the UFI-1000i instrument uses advanced flow cytometry technology with hydrodynamic focusing, specific fluorescent dyes for bacteria and sediment, and high-definition reproducible measurement signals for size, structure, and fluorescence to analyze the physical and chemical properties of a urine sample (see Figure 6). A continuous loading function provides immediate sample processing or series testing for up to 50 specimens. Quantitative results are delivered in approximately 1 minute. Manual entry is minimized; the instrument includes a bar code reader that identifies samples and reagents and also a bidirectional LIS connection. Results are easy to interpret, and flags alert users to abnormal results.

The device reduces not only the need for manual labor but also unnecessary cultures, owing to the negative predictive value of automated urinalysis in microbiology. Studies of microurinalysis automation conducted by bioMérieux have determined that a technologist using such a system can improve overall bench efficiency by 25%.

The Integration of Postanalytics

The increased efficiency made possible by automation should continue once testing is completed. Postanalytics represent another opportunity for improvements in work flow. Bidirectional communication that eliminates manual entry requirements ensures the integrity of data while eliminating paper waste. Information must flow from one system to another in a rational way as the specimen moves through the laboratory and hospital system.

Automation solutions should therefore be developed with an eye toward integration. Manufacturers and laboratories both sometimes can neglect system integrity and develop or purchase, respectively, 10 different solutions that cannot immediately be integrated. Not only does this increase the likelihood of lab errors through a reliance on manual entry, but it also has consequences for the laboratory work environment. Systems integrated into a single workstation save the desktop or counter space that otherwise would be allocated to individual computer stations. Maximum consolidation is best. Many of the systems on the market are capable of integration with an LIS.

Integrated systems make the information available also for trending, budgeting, regulatory requirements, and other needs. Middleware solutions add rules-checking and decision-making capabilities. The Observa system from bioMérieux, for example, consolidates results from the company's BacT/Alert and Vitek 2 instruments so that the data are easily transmitted to the LIS or the physician. A system on Vitek validates results and suggests resistance mechanisms. Therapy and infection management systems provide similar assistance, incorporating patient profiles and parameters with a database in order to recommend a management regime for treatment by medication.

Conclusion

Just as microbiology now follows chemistry and hematology into a more fully automated world, it will also follow those disciplines into a more structured one. Clinical laboratories considering and then adding automation in many cases also embrace process improvement techniques such as lean operations and Six Sigma. Manufacturers now provide organizational auditing services to supplement their systems offering. Through analysis of work flow, configuration, and other aspects of its organization, a lab can decide intelligently which processes to automate and which work flows to modify in order to increase efficiency and output quality. Also, it can explore appropriate training and service methods for improving performance. Training for new systems is instrumental whether the user is a new hire or an experienced laboratorian.

Microbiology is going beyond plates and cultures. Molecular approaches to microbiology and immunoassays are prominent among the greater number of tools now available to identify sources of infections. Through the development and implementation of advanced optics and tactile robotic systems, microbiologists will continue to be able to handle a wide variety of lab functions more and more efficiently, so that more-accurate results will be obtained more quickly.

Customer demands will continue to drive development toward the eventual automation of the microbiology bench in its entirety, or close to it. Once the broader needs are met, manufacturers will likely turn to niche products. Right now, however, microbiology is a niche in itself.

Demands for greater efficiency, higher accuracy, and speedier performance that are coming from various quarters will inevitably lead to microbiology automation becoming more universal. Some countries may see full automation within two years; others will need 10 years. But the full automation of microbiology will come just as it came for other types of clinical laboratories before. In the end, fully automated microbiology labs will bring the benefits of improving patient care while saving time and money.

Thierry Bernard is corporate vice president, commercial operations, at bioMérieux Inc. (Marcy l'Etoile, France). He be reached at thierry.bernard @er.biomerieux.com [7]		Jean-Louis Tissier is vice president, microbial infection, at bioMérieux Inc. He can be reached at louis.tissier@ eu.biomerieux.com [8].