Introduction
Point-of-care devices, also referred to as patient-side or in-clinic devices, are laboratory equipment and automated instruments used to perform laboratory testing in proximity to patients outside of a commercial or reference laboratory.1 In veterinary clinics, POC instruments include automated instruments such as hematology and clinical chemistry instruments, microscopes, and refractometers.
All POC devices require regular maintenance as part of a quality management plan (see second article in this series), and POC instruments also require calibration and use of QCMs to ensure accurate performance. The purpose of this article is to introduce quality-assurance concepts regarding POC instruments and equipment by reviewing the procedures for maintenance and care of core laboratory equipment. Concepts of total analytic error and evaluation of instrument performance are also discussed.
Maintenance and Care of Core Laboratory Equipment
Microscopes
Microscopes must be cleaned daily when in use according to the manufacturer's instructions. Generally, all hard, nonglass microscope surfaces should be dusted weekly with a dust-trapping cloth. Oil should be removed from all objective lenses immediately after use with lens tissue. If oil is still present, lens tissue soaked in ethanol or a commercial lens cleaning solution should be used to gently wipe the surface of the lens from inside to outside. Any immersion oil that contacts a nonimmersion objective lens (ie, 10× or 40×) must be immediately cleaned away with this method. Regular lens tissue wetted with methanol can be used to remove oil from all other parts of the microscope except the lenses. Stained slides such as blood smears should be examined with the microscope condenser in its highest position and the diaphragm opened (Figure 1). Wet-mount preparations should be examined with the microscope condenser approximately halfway down and the diaphragm half closed. If oil-immersion objective lenses are being used, care should be taken to ensure that the use of coverslips and coverslip-mounting medium (presumably aqueous medium for in-clinic situations) does not cause aberrations in the intended optical-focusing properties of the objective lenses. Because of its close proximity to the coverslip, the 100× objective lens may break the coverslip and become contaminated by the material on the slide. Veterinary clinics may consider having 2 microscopes: a clean microscope for slides with stained, dry content such as blood smears and cytologic preparations and a dirty microscope for slides with wet content such as urine sediment, fecal flotation, skin scrapings, and ear-swab preparations. Microscopes should receive annual maintenance and deep cleaning, which can be arranged through the supplier.
Photographs of a microscope with a fixed condenser (A and B) and one with a flip-out condenser (C and D) to demonstrate microscope settings for different glass-slide preparations. A—For stained slides such as blood smears, the condenser (blue arrow and bracket) should be in its highest position and the diaphragm should be open. Here, the diaphragm is open all the way to the number 100 (red arrow), which corresponds with use of the 100× objective lens. B—For wet preparations such as urine sediment slides, the condenser (blue arrow and bracket) should be approximately halfway down and the diaphragm partially closed. Here, the diaphragm is set to the number 40 (red arrow), which corresponds with use of the 40× objective lens. C—For stained slides, the condenser is in place and flipped up (arrow). D—For wet preparations, the condenser is flipped out (arrow).
Citation: Journal of the American Veterinary Medical Association 258, 7; 10.2460/javma.258.7.725
Stains
Quick stains such as Romanowsky stain are typically used for in-clinic laboratories, and slide staining is optimized when the manufacturer's directions are observed. Stains should be replaced depending on the frequency of use, whether inadvertent evaporation has occurred owing to an insufficient seal between the container and lid, and whether there is evidence of stain contamination by microbes, which can be checked by microscopically examining a drop of stain on a clean slide. Generally, stains should be filtered into a clean, dry jar by means of drip coffee filters if gross contaminants are visible or if stain precipitate or foreign material is noted on slides. Stain jars should be cleaned with dilute bleach solution, rinsed, and dried prior to returning the filtered stain to the jars and then topped up with fresh stain as necessary. Stains should be replaced if microbial contamination or poor slide staining is noted.
Because microbial overgrowth may occur in stain solutions, 2 sets of stains should be kept: one for dirty material such as fecal or ear swab smears and the other for clean material such as blood smears and fine-needle aspirates. A rack may be used to suspend individual slides over a sink during stain application, with excess stain discarded into the sink; however, it should be noted that with this approach, the sink would be unavailable for other purposes during stain application.
Centrifuges
Samples placed in a centrifuge should always be balanced by weight across the centrifuge head. The centrifuge should be cleaned and disinfected every 1 to 2 weeks, and spills should be cleaned as soon as possible. The manufacturer's manual for the centrifuge should list recommended cleaning products; an alcohol-based solution is generally acceptable. Each centrifuge should be annually serviced, including cleaning and inspection of rotors, buckets, and seals; lubrication; and calibration of centrifuge speed by a technician trained in this calibration technique.
Urine dipsticks
Dipsticks sold for veterinary use are generally designed for human urine; the specific gravity, leukocyte, and nitrate test pads are not reliable, and the urobilinogen test pad is not considered useful in veterinary species.2 Dipsticks should not be used if expired and should always be stored in the original container, with the desiccant intact and the lid tightly closed. The information sheet that accompanies each dipstick product will specify the appropriate time interval between dipping and reading of results, and these instructions should be strictly followed.2 Dipstick results should correspond with the clinical examination findings, macroscopic appearance of the urine, and urine sediment results. Urine sediment examination is strongly encouraged as a compulsory part of a routine urinalysis.2
Clinics may consider the additional use of liquid QCM made specifically for dipsticks to supplement their quality assurance of the urinalysis procedure. This material is designed for use on dipsticks used in human medicine and can be obtained from companies selling clinical diagnostic products.
Refractometers
A refractometer should be calibrated each day that it is used. For calibration, a drop of distilled water is placed onto the prism and the specific gravity is determined. The reading should be 1.000; if not, then an adjustment should be made by turning the small screw or knob on the top of the refractometer until the reading is 1.000. The prism surface should be cleaned after each use with a lint-free laboratory wipe (eg, a delicate task wipera) moistened with a small amount of methanol.
Automated POC Instruments
Considerations for selection of an instrument
Considerations when purchasing or leasing in-clinic automated instruments are covered in detail in the second article3 of this series. Briefly, factors that should be considered include ease of use, footprint, financial viability, appropriateness of the instrument for use in veterinary patients, suitability for data management, availability of QC mode, and availability and nature of technical support from the manufacturer or distributor.
Maintenance
A maintenance protocol for each instrument should be drafted in accordance with the manufacturer's recommendations. Instruments may or may not display reminders for maintenance procedures. Generally, cleaning of some parts, in particular fan filters, is required on at least a monthly basis. Hematology instruments may require a wash or enzymatic cleaning every 2 to 4 weeks to remove debris from the system and ensure optimal performance. These routine maintenance procedures are easy to perform and usually well described in the operating manual or on the instrument screen. The distributor of the instrument should perform an annual service that includes more extensive cleaning, replacement of some parts, and a system check; this may or may not be included in a service package, if available. If not included, it should be budgeted for at the start of each financial year and not omitted. An example of an equipment performance log, provided as an appendix to the American Society for Veterinary Clinical Pathology guidelines on quality assurance for POC testing,1 is available elsewhere.4
Calibration and QCMs
In terms of laboratory instruments, a calibrator or standard is a material that contains a known concentration or activity of a substance that is intended to be measured, which is known as a measurand (often used synonymously with analyte). Calibrators are used to readjust instrument test systems to specific known values in a process called calibration. Periodic calibration is required to address instrument bias (inaccuracy), which can occur over time (called drift) and is often reflected by QC results falling outside of set limits. In reference laboratories, calibration may need to be performed daily to monthly, depending on several factors including the instrument itself, measurand, and instrument workload. Recalibration is indicated following any major instrument service or maintenance (to include part replacement) and after a change in reagents. It is important to note that only a few in-clinic instruments require external calibration by the user; instead, the calibration-type adjustments typically come as software updates from the manufacturer or as code changes for new lot numbers of reagents and rotors.
Most in-clinic instruments automatically perform a daily, automatic internal electronic calibration check without the addition of external calibration material. This method monitors the electronic components of the system but cannot assess the function of other components, such as the reagents. This calibration should always be reviewed, and if aberrant results are obtained, the manufacturer should be contacted for assistance. Patient samples should not be analyzed until the error has been resolved.
As opposed to calibrators, QCMs are materials used to mimic a patient sample and to monitor instrument performance on a regular basis. Their use is recommended every day that a patient sample is being run for that test. Such material needs to be run before patient samples are run, and results are used to ensure that the instrument is performing to an acceptable standard. They are available commercially, often from the instrument manufacturer. Typically, several measurands are contained in a given QCM, mimicking hematologic and biochemical samples on which whole panels are run. These QCMs will be available at various levels representing different patient statuses: 1 that contains measurands in amounts expected for a healthy patient (normal level) and a further 1 or 2 that contain measurands in high, low, or a combination of abnormal concentrations or activities (ie, high-level, low-level, and abnormal or pathological controls, respectively).
The results of QCM analysis must be within acceptable ranges for the instrument before patient samples can be analyzed. The manufacturer of the QCM typically provides a target mean and acceptable range for each measurand, which is usually specific to the measurement methods used by the instrument. It is important to note that the acceptable ranges indicated by a manufacturer are not the same as reference intervals used for clinical decision-making; rather, acceptable ranges generally permit an amount of analytic error (ie, wider range) that would be unacceptable for clinical interpretation purposes. The aim of other methods for evaluation of QCM results, discussed later in this article and in the final article5 of this series, is to detect and control analytic errors before they become large enough to adversely impact patient care. As for instrument reagents and any calibrators, the instrument log should be used to document the lot number, date of receipt at the laboratory, expiration date, and start and end dates of use.
Total Analytic Error and Instrument Performance
Total analytic error
No analytic method is error free, and test results for hematologic and biochemical measurands are never entirely a true reflection of the concentration or activity of that measurand in the patient, which itself is only a snapshot of the patient's dynamic physiologic status. The amount of analytic error present in a measurement (TEobs; explained later in this article) varies between analytic methods and measurands. The TEobs has implications for interpretation of assay results. For example, a serum potassium concentration assay may hypothetically have a TEobs of 10%. This means that for any measurement of serum potassium concentration (eg, 5.3 mmol/L), the true result for a given patient lies in the range of the measured result plus or minus 10% (ie, 4.8 to 5.8 mmol/L). If the reference interval for serum potassium concentration is 4.2 to 5.5 mmol/L, the patient in this example could be hyperkalemic or normokalemic. This distinction has important consequences for treatment, and in this example, the test would not be accurate enough to make a clinical decision should the clinician suspect hyperkalemia.
TEa
The percentage of analytic error that is considered acceptable for informed clinical decision-making based on laboratory results is referred to as the TEa. Mathematically derived (typically from clinical decision thresholds and instrument capabilities6), TEa is a conceptual framework for benchmarking results and is not a direct measurement. The measured TEobs would ideally be smaller than the TEa. For veterinary species, TEa values have been proposed for various clinical chemistry and hematology measurands by expert veterinary clinical pathologists and veterinary internists, critical care specialists, and oncologists.6,7 For example, the recommended TEa for assays of serum potassium concentration in dogs, cats, and horses is 5%, which is half of the TEobs of 10% used in the previous example. A selection of TEa values for common measurands recommended by the American Society for Veterinary Clinical Pathology is provided (Table 1), with more extensive lists freely available elsewhere.6,7
Recommended values of TEa for various hematologic and serum or plasma biochemical measurands.5,6
| Measurand | TEa (%) |
|---|---|
| Hematology instrument | |
| RBC count | 10 |
| Hemoglobin | 10 |
| Hct | 10 |
| Mean cell volume | 7 |
| Mean cell hemoglobin concentration | 10 |
| WBC count | 20 |
| Platelet count | 25 |
| Clinical chemistry instrument | |
| Albumin | 15 |
| ALP | 25 |
| Alanine aminotransferase | 25 |
| Aspartate aminotransferase | 30 |
| Bile acids | 20 |
| Calcium | 10 |
| Cholesterol | 20 |
| Chloride | 5 |
| Creatinine | 20 |
| γ-Glutamyltransferase | 20 |
| Glucose | 20 |
| Lactate | 40 |
| Phosphorus | 15 |
| Potassium | 5 |
| Sodium | 5 |
| Total protein | 10 |
| Triglycerides | 25 |
| Urea | 12 |
Initial evaluation of instrument performance and determination of TEobs
As mentioned previously, clinicians need to be aware of the accuracy of results generated by an in-clinic instrument to make appropriate decisions. Instrument performance, in terms of the TEobs of the various measurands assayed by an instrument, should ideally be determined when the instrument is first set up and before any patient samples are run.1 To do this, 2 to 3 levels of QCM (1 normal and 1 or 2 abnormal levels) should be used.1 Each chosen level of QCM should be measured with the most extensive test panel available on the instrument, between 5 and 20 times over 5 to 10 days.8 The greater the number of data points collected, the better the derived estimates will be. Given the practicalities of financial, technical, and time constraints in veterinary practice, 10 measurements would be acceptable and 5 would be the minimum recommended. Additional data may be accrued over time and added to the total number for periodic update of the calculations. As each data point is recorded, the accumulated data can be used to calculate the TEobs.
The TEobs of a given analytic method is calculated as the sum of its imprecision and its bias, both of which speak to the accuracy of the method. In the context of the use of automated hematology and clinical chemistry instruments for diagnostic data generation, analytic precision refers to the degree of repeatability of results.6 For example, if serum calcium concentration were measured in the same sample 10 times over a 3-day period (assuming adequate sample stability) and results for the 10 measurements differed substantially (eg, each falling into different categories used to differentiate health from disease), then this calcium assay would have low precision (high imprecision). Some imprecision is inherent to instruments but can also occur because of operator errors. Mathematically, imprecision is represented by the SD or CV of a data set, with larger values indicating greater imprecision. The imprecision of an assay is particularly important to consider when evaluating results near the upper or lower reference limit or clinical decision-making thresholds and when evaluating serial results for a given patient. It is important to determine whether any observed changes are due to the patient's physiologic status (eg, improvement or worsening) and not due to assay imprecision.
Analytic bias is the difference between the true value of a measurand and the result yielded by the instrument. Such bias may be positive (instrument result higher than the true value) or negative (instrument result lower than the true value). Different instruments may have different degrees of bias, depending on the analytic methods used. Bias can lead to a result appearing pathologically high or low when, in fact, it is not. For example, if a serum calcium concentration assay has a bias of −15%, then all results will be on average 15% lower than the true value, which could lead to a diagnosis of hypocalcemia in a normocalcemic dog or failure to detect hypercalcemia when it truly exists. Some examples of causes of bias include temperature or humidity changes, poor water quality (eg, for instruments requiring an independent water supply), reagent aging, sample hemolysis, incorrect instrument calibration, or instrument malfunction insufficient to prevent results generation.
Bias is classically assessed as the difference between the results for a diagnostic instrument or method of interest and results of a gold-standard method or widely used and validated field method. For in-clinic purposes, bias is typically calculated from the mean of the 5 to 20 QCM results measured with an instrument, compared with the manufacturer-supplied mean for the particular QCM used. The recorded data from the 5 to 20 QCM measurements should be entered in a spreadsheet, with measurements and calculations recorded separately for each of at least 2 QCM levels. After calculation of the mean and SD, imprecision (CV), bias, and TEobs should be calculated with commercially available software or the following formulas:
where the target mean represents the value provided by the manufacturer. A downloadable spreadsheet for these calculations (Supplementary Appendix S1, available at: avmajournals.avma.org/doi/suppl/10.2460/javma.258.7.725) and an example image from the spreadsheet (Appendix 1) are provided.
When the data for serum ALP activity in Appendix 1 were considered, 10 measurements were made and have been recorded in the spreadsheet. The mean of the 10 measurements is 114.5 U/L, whereas the target mean for the QCM is 107.5 U/L. Application of the formula for bias indicates the presence of a positive bias for this method of 6.5%. In other words, the instrument result is 6.5% higher than the true result. The CV or imprecision of the measurements is 7.5%, and the TEobs for a single measurement of serum ALP concentration with the evaluated instrument is 21.5%.
If the calculated bias is > 5% (absolute value) for any measurand, the instrument manufacturer should be contacted and an attempt should be made to reduce the bias to < 5% through calibration and other procedures recommended by the manufacturer. In the example, although the TEobs for serum ALP concentration measurements is less than the TEa, the bias (6.5%) is higher than 5% and the manufacturer of the instrument should be contacted.
The TEobs derived from the QCM measurements should be compared with the TEa (Table 1; Appendix 1). If the TEobs for the instrument is less than the TEa, the performance of the instrument for that measurand is acceptable for clinical interpretation of results.8 In the example, although the TEobs of 21.5% is less than the TEa of 25% and appears to be a large error, it has been deemed unlikely to affect clinical decisions.
If the TEobs is greater than TEa (or if QCM results fall outside the manufacturer's acceptable range), expiry dates of reagents should be checked and performance of regular maintenance procedures should be verified. Further discussion and a flowchart outlining how to troubleshoot these issues in veterinary clinics are provided in the final article5 of this series. If efforts to reduce the TEobs are not successful, assistance should be sought from the manufacturer to further troubleshoot the problem. If it is not possible to improve the TEobs for a particular measurand, ideally that test should not be run on that instrument and another method should be used, such as use of a reference laboratory or replacement of the instrument if the problem persists. Less ideally, if the TEobs is only slightly more than the TEa, the TEa can be relaxed to accommodate the TEobs, but caution should be applied when making clinical decisions on the basis of results for the measurand in question.9
Ongoing Evaluation
As previously stated, QCM measurement should be performed each day that a particular assay is run for patients.1 Many instruments have a special QCM mode, whereby QCM results are recorded and displayed in graphical format. This graphical format (ie, a Levey-Jennings chart) plots QCM results over time against the pre-entered target mean. More details and examples of Levey-Jennings charts are provided in the final article5 of this series. The simplest way to monitor instrument performance when performing ongoing QCM measurements is to first compare each result as it is measured to the manufacturer's target mean and acceptable range (eg, by use of the Levey-Jennings charts). However, such acceptable ranges are based on the average performance of several instruments (of the same model) and do not take the TEa into account. They may fail to allow detection of clinically relevant problems of instrument performance.
A second method to monitor instrument performance is regular entry of instrument QCM results into a self-constructed Levey-Jennings chart or spreadsheet (Supplementary Appendix 1) on each day that a patient test is run. The values for mean, bias, and CV can then be evaluated for the most recent 15 to 20 QC measurements. In the bias calculation, the manufacturer's target mean will need to be adjusted with each new reagent and QCM lot. Therefore, it is recommended that the same reagent lot be used for as many analyses as possible, provided the expiry date has not been reached. The advantage of the methods described here is that clinically relevant decreases in instrument performance will be detected. The disadvantage is that clinically important error may be detected only several days after it first occurs. This is because the first TEobs calculation is possible only after at least 5 daily QCM results have been collected; thus, a clinically relevant analytic error, calculated by use of TEobs, will be detected 5 days into running daily QCM tests at the earliest.
The POC instrument QC data summary (eg, a spreadsheet similar to Supplementary Appendix S1 containing information on initial and ongoing instrument performance) should be readily accessible for all team members interpreting the results.
Clinical Bottom Line
Paying attention to and setting up formal procedures for regular maintenance and monitoring of analytic equipment and processes are costly in terms of time and money, and such costs and procedural requirements must be considered when conducting POC testing. These costs, however, need to be weighed against the considerable risks associated with producing inaccurate results, which include poor clinical decisions, adverse effects on patient health, costs of repeated testing of patient samples, and adverse reputational and legal consequences for the clinic. Just as a standard best practice for performing surgery is the use of autoclaved instruments, the laboratory QC measures described herein should become routine protections for patient safety. To facilitate implementation, a checklist is provided (Appendix 2; Supplementary Appendix S2, available at: avmajournals.avma.org/doi/suppl/10.2460/javma.258.7.725). These measures are achievable with commitment and practice and with help that is available from clinical pathologists and professional technologists or laboratorians with expertise in quality assurance.
Acknowledgments
No third-party funding or support was received in connection with this study or the writing or publication of the manuscript. The authors declare that there were no conflicts of interest.
Footnotes
Kimwipes, Kimberly-Clark Professional, Irving, Tex.
Abbreviations
| ALP | Alkaline phosphatase |
| CV | Coefficient of variation |
| POC | Point of care |
| QC | Quality control |
| QCM | Quality control material |
| TEa | Total allowable analytic error |
| TEobs | Total observed analytic error |
References
- 1. ↑
Flatland B, Freeman KP, Vap LM, et al. ASVCP guidelines: quality assurance for point-of-care testing in veterinary medicine. Vet Clin Pathol 2013;42:405–423. Available at: cdn.ymaws.com/www.asvcp.org/resource/resmgr/QALS/Other_Publications/ASVCP_POCT_QA_Guideline_May_.pdf. Accessed Aug 21, 2020.
- 2. ↑
Arnold JE, Camus MS, Freeman KP, et al. ASVCP guidelines: principles of quality assurance and standards for veterinary clinical pathology (version 3.0). Vet Clin Pathol 2019;48:542–618. Available at www.asvcp.org/page/QALS_Guidelines. Accessed Aug 21, 2020.
- 3. ↑
Hooijberg EH, Freeman KP, Cook JR. Quality management for in-clinic laboratories: facilities, instrumentation, health and safety, training, and improvement opportunities. J Am Vet Med Assoc 2021;258:273–278.
- 4. ↑
Freeman KP. Appendix—example forms and logs. Available at: cdn.ymaws.com/www.asvcp.org/resource/resmgr/qals/poct_example_forms_and_logs.pdf. Accessed Aug 21, 2020.
- 5. ↑
Freeman KP, Cook JR, Hooijberg EH. Quality management for in-clinic laboratories: introduction to statistical quality control. J Am Vet Med Assoc 2021;258:733–739.
- 6. ↑
Nabity MB, Harr KE, Camus MS, et al. ASVCP guidelines: allowable total error hematology. Vet Clin Pathol 2018;47:9–21.
- 7. ↑
Harr KE, Flatland B, Nabity M, et al. ASVCP guidelines: allowable total error guidelines for biochemistry. Vet Clin Pathol 2013;42:424–436.
- 8. ↑
Rishniw M, Pion PD, Maher T. The quality of veterinary in-clinic and reference laboratory biochemical testing. Vet Clin Pathol 2012;41:92–109.
- 9. ↑
Lester S, Harr KE, Rishniw M, et al. Current quality assurance concepts and considerations for QC of in-clinic biochemistry testing. J Am Vet Med Assoc 2013;242:182–192.
Appendix 1
Example of a statistical QC chart for a hypothetical QCM lot number, as completed via the interactive spreadsheet available in Supplementary Appendix S1.
| Measurement No. | Albumin (g/L) | ALP (U/L) | ALT (U/L) | Calcium (mmol/L) | Creatinine (μmol/L) | Glucose (mmol/L) | Phosphorus (mmol/L) | Total protein (g/L) | Urea (mmol/L) |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 30 | 109 | 90 | 2.77 | 193 | 6.56 | 1.59 | 58 | 13.3 |
| 2 | 30 | 117 | 92 | 2.76 | 198 | 6.56 | 1.58 | 58 | 13.8 |
| 3 | 31 | 121 | 97 | 2.77 | 187 | 6.58 | 1.56 | 57 | 15 |
| 4 | 31 | 114 | 93 | 2.81 | 199 | 6.51 | 1.57 | 60 | 13.5 |
| 5 | 31 | 115 | 94 | 2.75 | 198 | 6.64 | 1.59 | 60 | 13.6 |
| 6 | 30 | 114 | 106 | 2.79 | 188 | 6.58 | 1.57 | 61 | 12.8 |
| 7 | 31 | 122 | 103 | 2.77 | 193 | 6.66 | 1.55 | 60 | 13.9 |
| 8 | 31 | 93 | 110 | 2.71 | 189 | 6.61 | 1.58 | 61 | 13.9 |
| 9 | 31 | 121 | 100 | 2.76 | 201 | 6.73 | 1.53 | 60 | 13.7 |
| 10 | 31 | 119 | 99 | 2.73 | 192 | 6.51 | 1.56 | 60 | 13.6 |
| Mean | 30.7 | 114.5 | 98.4 | 2.762 | 193.8 | 6.594 | 1.568 | 59.5 | 13.71 |
| SD | 0.5 | 8.6 | 6.5 | 0.0 | 5.0 | 0.1 | 0.0 | 1.4 | 0.6 |
| CV (%)* | 1.6 | 7.5 | 6.6 | 1.0 | 2.6 | 1.0 | 1.2 | 2.3 | 4.1 |
| Manufacturer's target mean | 30.2 | 107.5 | 93.8 | 2.77 | 197.9 | 6.8 | 1.56 | 56.9 | 13.8 |
| Bias (%)* | 1.7 | 6.5 | 4.9 | –0.3 | –2.1 | –3.0 | 0.5 | 4.6 | –0.7 |
| TEobs (%)* | 4.8 | 21.5 | 18.1 | 2.3 | 7.2 | 5.1 | 2.9 | 9.1 | 8.8 |
| Recommended TEa† (%) | 15 | 25 | 25 | 10 | 20 | 20 | 15 | 10 | 12 |
| Is TEobs < TEa? | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
Appendix 2
Checklist for equipment maintenance and instrument performance for in-clinic laboratories.
| Recommendation | Compliant? | Additional comments by auditor |
|---|---|---|
| Microscopes | ||
| Lenses and body cleaned of oil daily; lens-cleaning tissue used for lenses | □ Yes □ No | |
| Dusted weekly | □ Yes □ No | |
| Professional maintenance and cleaning performed annually | □ Yes □ No | |
| Stains | ||
| Stain filtered or replaced on detection of precipitates, gross contaminants, foreign material, or microorganisms | □ Yes □ No | |
| Two separate sets of stain available for clean and dirty material | □ Yes □ No | |
| Centrifuges | ||
| Cleaned and disinfected once weekly (preferably) or once every 2 weeks (minimum) | □ Yes □ No | |
| Annual maintenance performed | □ Yes □ No | |
| Urine dipsticks | ||
| Not used past expiration date | □ Yes □ No | |
| Kept in tightly closed container away from heat and light | □ Yes □ No | |
| Refractometers | ||
| Cleaned after each use | □ Yes □ No | |
| Calibrated daily | □ Yes □ No | |
| POC instruments | ||
| Purchase includes consideration of ease of use, amount of space needed, financial viability, appropriateness of the instruments or assay for use in veterinary patients, data management, and availability of technical support | □ Yes □ No | |
| Maintenance procedures are scheduled and performed in accordance with manufacturer's recommendations; a maintenance log is kept for each instrument | □ Yes □ No | |
| Annual service performed for each instrument | □ Yes □ No | |
| Calibration and QCMs | ||
| Instrument calibration procedures performed at intervals recommended by the manufacturer | □ Yes □ No | |
| If calibration fails, testing of patient samples ceases and manufacturer is informed | □ Yes □ No | |
| At least 1 level of QCM assayed every day that the instrument is used (optimum) or weekly (minimum) | □ Yes □ No | |
| Total analytic error and instrument performance | ||
| QCM results fall within the range provided by the manufacturer (minimum); otherwise, an attempt should be made to troubleshoot the problem or contact the manufacturer and testing of patient samples ceases | □ Yes □ No | |
| Instrument performance initially evaluated by calculation of the CV, bias, and TEobs from at least 5 measurements/QCM level | □ Yes □ No | |
| TEobs < TEa for all measurands | □ Yes □ No | |
| No in-clinic testing of measurands with TEobs > TEa | □ Yes □ No | |
| Summary QC data for POC instruments recorded and readily accessible for all team members interpreting the results | □ Yes □ No |
