Microfiltration Technology: A Sensible Approach to Automating Sample Preparation

In the world of laboratories, the discerning eye of an efficiency expert must wince while witnessing sample preparation. Nowhere else in the laboratory is there such an erratic work flow.

Laboratorians measure, heat, mix, centrifuge, and extract fragile biological samples in preparation for accurate analyses. Incessant deadlines attempt to maximize output, but the batching protocol for plasma and serum samples fractures any illusion of automation.

Over 330,000 specimens are processed per year by a typical hospital or independent laboratory,1 with 85% requiring some form of preanalytical preparation.2 Current methods require a medical technician to make more than 60 decisions while preparing a sample. The extrapolated conclusion here is that sample preparation—in a high-work-flow laboratory—is a by-product of 16.8 million decision points yearly, at zero tolerance for error.

Laboratories today are investing in large, complex laboratory automation systems that emulate manual techniques. Much of this technology does not appear to improve or streamline the sample preparation process. If we are serious about laboratory automation, we must dissect the process of sample preparation, locate the inherent flaws, fixate on delays, and reinvent the technology until the word “efficiency” can be honestly reapplied to the entire laboratory.3

Centrifugation as a sample preparation technology

There is no better place to begin our critical examination of sample preparation than to look at centrifugation. Centrifugation is, by far, the most common form of plasma and serum separation (70–80% of blood specimens are centrifuged).4 The technology’s primary drawbacks are the subjugation of living organic matter to 10–15 min of immense pressure—amounting to 1000–5000 g forces—and the delays.

Why, then, should we be surprised that after a resultant sample is analyzed, measured, and interpreted, the analysis must be rejected? After all, we have created the harshest of conditions for cells, large proteins, and molecules. Therefore, it should not shock us that some results of subsequent analyses mandate reprocessing.

Present-day methods require large ancillary equipment investments for already shrinking floor space, and entail additional planning to calculate loading times, spin-up times, spin-down times, and unloading times. In addition, centrifuging exacerbates batching, further delaying the entire process.

Microfiltration: a reliable, alternative technology

Is there a logical alternative to centrifugation? Microfiltration gently separates living cells, thereby ensuring the integrity of the specimen and, more importantly, completely eliminating centrifugation.

Through specially constructed disposable filter cells approximating the length of a baby’s finger, microfiltration has evolved beyond an original concept of cross-flow and/or tangential-flow filtration.6,7

Basic microfiltration system

The fundamental microfiltration system from Bio/Data Corporation (Horsham, PA) comprises three primary components: 1) disposable filter cell, 2) filtration module whose design is determined by the back-end application, and 3) proprietary software that manages and controls the microfiltration process. The filtration module processes the filter cell (Figure 1). The filter cell’s patented architecture highlights the intricate realities of microfluidics therein. A whole-blood specimen is precisely manipulated between two reservoirs over a microporous membrane and then along a specially designed, smoothly sculpted flow channel.

Figure 1 - A filter cell (left) awaits injection of a specimen before cellular and fibrin solids are removed from the processed plasma (or serum) in the filtration station (right) during microfiltration. This image was photographed inside the Microsample Coagulation Analyzer™ model MCA-210 (Bio/Data).

A back-and-forth motion continually cleans the membrane pores. Optical sensors control the sample movement until virtually cell-free plasma is determined to be present. All the preanalytical variables are radically reduced, as 400 μL of plasma for analytical testing is produced from 2 mL of whole blood in 45 sec. The software ensures precision in the amount of the deliverable sample.

Technicians’ involvement is minimized, as is their exposure to blood-borne pathogens. The laboratory manager now observes a single operation in which the plasma or serum is extracted. In addition, the primary sample tube is never uncapped. (Microfiltration also eliminates the sample pumps and rinse fluids required in most analyzers.)

The filter cell

A plastic, disposable microfiltration filter cell comprises three or four elements:

  1. Reservoir, to contain the sample—The reservoir consists of two equal-volume chambers that, when assembled with the base, are connected via a horizontal flow channel.
  2. Microporous membrane, 0.6 μm pore size, positioned between the reservoir and base with a flow channel above and a collection grid below.
  3. Base, which collects and delivers the filtrate for processing or testing.
  4. Cover sheet attached below the collection area to the bottom of the base (optional, dependent upon design).

Filter cells are intended to reside in an analyzer, whose design varies by analysis type, e.g., optical (density, color, or fluorescence), electrical (conductance or impedance), or mechanical (magnetic bead or bar). Design also depends upon whether test reagents are introduced into the filter cell.

System operation

Two optical paths in each of the filter cell’s two reservoir sections are continuously monitored by detection circuitry arrays in the filtration module. The two paths correspond to the lower and upper levels of each reservoir, and the results from the optical arrays engage and disengage the system’s variable filtration motion, and determine when the proper volume of plasma or serum is achieved (Figure 2a).

Figure 2 - Schematics depict how the optics from a filtration module guide the back-and-forth movement of a specimen across a microporous membrane: a) basic filter cell, b) introduction of a specimen into the cell, and c) operation during microfiltration.

When both optical paths are clear, no blood is present. If the paths in one reservoir are blocked, a fluid is in the chamber. A filter cell is moved into position to initiate operation, and the optical system checks to confirm two paths in one reservoir chamber are blocked and two paths in the other chamber are clear (Figure 2b).

A pneumatic system gently pressurizes the side of the reservoir whose upper path is shown to be blocked. The pressure eases the blood down a passageway in the bottom of the filled reservoir chamber into the flow channel, across the microporous membrane, and up through a passageway in the bottom of the empty chamber.

The pressure differential causes plasma to pass through the membrane, as cells flow across the membrane surface. The pressure continues to be applied until the lower optical path indicates the chamber is clear, at which time the pressure is relieved.

The system then checks the opposite chamber to ensure its lower path is blocked. Once confirmed, pneumatic pressure is applied to that reservoir until optics confirm its lower path has cleared. Thus, an alternating forward-reverse flow ensues. Once the optical sensors indicate a predetermined sample volume has been produced, the plasma or serum is promptly removed for analysis (Figure 2c).8

Plasma quality

The indiscernible attributes of native plasma are maintained in microfiltration, because the separation method is less traumatic to the cellular matter in blood. The filtration pressure—about 129 mm of mercury—is comparable to normal blood pressures of 120–140 mm of mercury. Cells are not subject to extreme forces or pressures and do not excrete activating materials into the plasma.

The effectiveness of plasma extraction may be gauged upon how many residual platelets—the smallest cells in blood—reside in the sample. For an example of microfiltration’s effectiveness, blood was drawn from three people and each specimen was placed into a sodium citrate anticoagulant. The specimens were then apportioned into batches of 20 and 16 for processing by microfiltration and centrifugation, respectively.

The 20 specimen tubes were processed sequentially for 30 sec each, utilizing the MCA-310 microfiltration coagulation analyzer (Bio/Data) with the filter cells. The other 16 specimens were collectively spun using a centrifuge at 2500 × g for 15 min.9

Platelet counts in the centrifuged samples ranged from 4490 to 8000, well within the Clinical Laboratory Standards Institute’s tolerance of 10,000 platelets per cubic millimeter in “platelet-poor plasma.”9 By comparison, microfiltration-prepared samples were so low in platelets that final results had to be determined manually. The samples studied contained fewer than 200 platelets per cubic millimeter, 0.5% of centrifugation’s allowance (Table 1);10 thus, samples processed by this technology can be described as “virtually platelet-free plasma.”

Summary

Clinical managers are constantly evaluating, reviewing, and reevaluating how their protocols expedite or slow down the work flow. The constant emphasis on better methods of sample preparation is a self-admission of the challenges faced by laboratories utilizing centrifugation. Microfiltration rapidly separates plasma or serum from whole blood and easily integrates into automated configurations to complement standard laboratory testing.

The advantages of microfiltration are:

  1. Simplification of the sample preparation process, enhancing automation.
  2. Elimination of centrifugation.
  3. Major reduction of time required to obtain plasma or serum.
  4. Elimination of interferences from cellular materials.
  5. Suitability of samples for all existing test methods.

References

  1. Wolman, D.M.; Kalfoglou, A.L.; LeRoy, L. Medicare Laboratory Payment Policy: Now and in the Future. National Academy Press: Washington, DC, 2000; Table 2.1.
  2. Zakowski, J.; Powell, D. The future of automation in clinical laboratories. IVD Technol. Jul 1999, 5(4), 48–57.
  3. Valenstein, P.; Souers, R.; Wilkinson, D.S. Staffing benchmarks for clinical laboratories. Arch. Pathol. Lab. Med. Apr 2005, 467–73.
  4. Holman, J.W.; Mifflin, T.E.; Felder, R.A.; Demers,
    L.M. Evaluation of an automated preanalytical robotic workstation at two academic health centers. Clin. Chem.2002, 48(3), 540–8.
  5. Pfister, M. Laboratory automation—industry transformation or dead end. HBS Quarterly (London), Mar 2004, 7–9.
  6. Van Reis, R.; Zydney, A. Membrane separations in biotechnology. Curr. Opin. Biotechnol.2001, 12, 208–11.
  7. Murkes, J.; Carlsson, C.G. Crossflow Filtration Theory and Practice. John Wiley & Sons, New York, NY, 1988.
  8. Coville, W.E.; Trolio, W.M. Preanalytical specimen preparation. J. Assoc. Lab. Automat. Mar 2000, 5(1), 72–80.
  9. Procedures for the Handling and Processing of Blood Specimens; Approved Guideline, 2nd ed. Clinical Laboratory Standards Institute (CLSI, formerly National Committee for Clinical Laboratory Standards), H18-A2, Oct 1999, 19(21).
  10. Laurich, M. Microfiltration Platelet Counts. Product Development Study Report, Bio/Data, Mar 1998.

Mr. Coville is Manager of Product Development, and Mr. Loika is Corporate Communications Manager, Bio/Data Corporation, 155 Gibraltar Rd., Horsham, PA 19044-0347, U.S.A.; tel.: 215-441-4000; fax: 215-443-8820; e-mail: [email protected].