*Všetky tabuľky, grafy a obrázky, ktoré sú súčasťou článku, nájdete v priloženom PDF súbore na konci štúdie.
Chemical testing of urine samples (reagent strip method), identification and counting of particles is performed routinely to identify and monitor di- seases of the kidney and urinary tract (1). Manual microscopy is extremely time consuming and has poor precision owing to variations in sediment prepara- tion and counting technique (2, 3, 4, 5). Furthermore, visual urine microscopy is notoriously subjective (6). Improvements in automated systems are needed to eliminate labor intensive manual microscopy (7) given the large number of samples in clinical laboratories. Systems currently available to automate manual microscopy are based either on flow cytometry (8) or image-based microscopy analysis (9). Laboratories of Medirex company use automated urine microscopy analyzers iQ200 (Iris Diagnostics, Chastworth, CA) (9).
The iQ200 analyzer uses flow-imaging microscopy provided by high-definition camera and strobe lamp that stops the fluid motion in order to capture the images and detect particles. An Auto-Particle Recognition (APR) software works as trained neural network system for processing of images. APR software enables to sort particles according to size, shape, contrast and texture and thus to classify and quantify cells and formed particles in native uncentrifuged urine. Cells and particles can be sorted into categories and quantified as in table 1. Work station provides the viewing of the images when needed even though the sample has been already discarded.
Precision control is performed daily in the morning and calibration every 28 days. Before the testing of patients, iQ Focus, iQ Negative and iQ Positive (glutaraldehyde-preserved human red blood cells) control samples (Iris diagnostics) are run according to manufacturer’s instructions. The iQ200 requires a minimum of 3 mL of urine for testing and quantifies particles per 2 μL of sample.
Wah et al. (10) have published the evaluation of the analytic performance of the iQ200 based on a comparison with results from manual microscopy.
For the study of iQ200 within-run imprecision, 166 urine samples were analyzed a total of 20 times during the same day for red blood cells (RBCs), white blood cells (WBCs) and squamous epithelial cells (ECs). Under such conditions, 10 % of the samples tested had cell counts outside the 95 % confidence interval of
the regression lines. The correlation study revealed discrepancies due to misclassification, disintegration of cells, abnormal morphological features (ghost and dysmorphic RBCs) or trapping of cells in mucus strands.
Sample assessment It is necessary to emphasize that the precision study by Wah et al. (10) was performed via analyzer set-up as a walk-away system. In order to achieve the highest diagnostic accuracy and to prevent incorrect automatic classification of cells and particles at the laboratories of the Medirex
company, the walk-away capability of the iQ200 is implemented only for samples possessing negative sediment and chemical analysis. After each assessment, analyzer flags samples that need to be reviewed by trained laboratory technician in case of absent chemical analysis or positive results of selected specimen. Review flags are assigned according to cut off values that convert quantitative data provided by the iQ200 analyzer into semiquantitative ranges correlating the image-based microscopy with
qualitative chemical analysis (table 1). Setting the accurate cut off values for a lab’s specific population is essential so there wouldn’t be marked way too many flags or missed clinically important findings.
General approach for image reviewing and cell counts editing often includes comparison of urine sediment and chemical analysis. Chemical analysis based on dry-reagent strip is performed via on-line connection
of analyzer Aution Max AX-4030 (Arkray, Japan) to the iQ200. Chemical analysis provides relevant supplementary information for assessment of cells and particles especially when misclassification or disintegration of cells occurs.
The most often misclassification of specimens occurs due to size and shape similarity as a consequence of morphological changes or high counts of particles. RBCs or/and WBCs are each other often misclassified
when macroscopic hematuria or a pus is present (figure 1 A). Furthermore, oval calcium oxalate crystals (figure 1 B) or single-cell yeast (figure 1 C) can be added to RBCs. On the other hand, class of dysmorphic RBCs (especially acantocytes) can resemble to budding yeast, what subsequently requires the verification with manual microscopy. Renal tubular epithelial cells (figure 1 D) are rare findings that can be incorrectly sorted into WBCs clump. Often present mucus strands are prone to be classified as hyaline casts. On the contrary, APR software provides better identification of pathologic casts than manual counting. Translucent hyaline casts can be often missed by manual counting due to the use of bright-field microscopy.
In order to prevent an overestimation of hyaline casts, manufacturer recommends verification of all casts by manual review of stored images. In general, the quality of stored images is sufficient for ubclassification of casts with exception of cellular casts, in which the cellular detail is not always adequate for differentiation into cell specific subcategories. Small numbers of pathologic casts are often accompanied with higher protein excretion (more than 1+) what can refer to renal failure before it
is manifested with increased serum creatinine level.
Information about urine chemical properties such as pH, specific gravity, nitrites or ascorbic acid concentration can help in classification of the specimen when morphological features had changed. High count of bacteria (gram-negative bacteria are manifested as positive nitrites) often leads to increased pH of urine and thus to disintegration and abnormally shaped cells. Besides, low osmolality (low specific gravity) of urine can change morphology of RBCs subsequently sorted among WBCs. At the same time, chemical dry reagent strip is very prone to interference. Among all, ascorbic acid is the cause of the most important interferences in urine. It is able induce false negative result predominantly for RBCs (hemoglobin) and at higher concentrations for bilirubine, urobilinogen, bacteria (nitrites), WBCs (leukocyte esterase), etc. Furthermore, the very low specificity and sensitivity of dry-reagent strip for G-positive bacteria or lymfocytes connected to absent analysis of urine sediment can also provide false negative results. False positive sediment is another issue resulting from high counts of bacteria or squamous epithelial cells in improperly collected samples.
The number of urine samples received every day requires the use of automated analysis of urine sediment and chemistry. Such approach enables to achieve as objective, precise and accurate results as possible. Furthermore, trained tech- nician can edit and review questionable findings when misclassification occurs. Taking into account such procedure, even fundamental laboratory examination of urine can help to diagnose the patient’s condition fast and accurately.
1. Cheung KA, Falk RJ, Coffman TM, et al. Primer on Kidney Diseases. 4th ed. Philadelphia, PA: Saunders; 2005.
2. Winkel P, Statland BE, Jorgensen K. Urine microscopy, an ill-defined method, examined by
multifactorial technique. Clin Chem. 1974;20:436–439.
3. Elin RJ, Hosseini JM, Kestner J, et al. Comparison of automated and manual methods for urinalysis. Am J Clin Pathol. 1986;86:731–737.
4. Gadeholt H. Quantitative estimation of urinary sediment, with special regard to sources of error. Br Med J. 1964;1:1547–1549.
5. Ferris JA. Comparison and standardization of the urine microscopic examination. Lab Med. 1983;14:659–662.
6. Lusky K, Cap Today. Paths to validating, using urine sediment analyzers [online]. 2015. Avail- able from: <http://www.captodayonline.com/paths-validating-using-urine-sediment-ana- lyzers/>. Accessed May 25, 2016.
7. European Confederation of Laboratory Medicine. European urinalysis guidelines: summary. Scand J Clin Lab Invest Suppl. 2000;60:1–96.
8. Manoni F, et al. Measurement of urine leukocytes by a second generation flow cytome- ter; application in the diagnosis of acute urinary tract infections in adult patients. Riv Med Lab-JLM. 2001:2:19–2
9. Deindorfer FH, Gangwer JR, Laird CW, et al. „The Yellow IRIS“ urinalysis workstation: the first commercial application of „automated intelligent microscopy.“ Clin Chem. 1985;31:1491–1499.
10. Wah DT, Wises PK, Butch AW. Analytic performance of the iQ200 automated urine mi- croscopy analyzer and comparison with manual counts using Fuchs-Rosenthal cell cham- bers. Am J Clin Pathol. 2005;123:290–296.