Microscopic urinalysis and automated flow cytometry in a nephrology laboratory.
In their recent report, Ottiger and Huber (1) compared the OF-100 flow cytometer and the KOVA system and suggested an algorithm for the selection of samples for microscopic analysis. They found that urine samples from nephrology patients had higher microscopic review rates. We agree with them that automated systems foster rapid and standardized analysis of formed elements and offer significant labor savings (2-4), but we think that such a study may lead to different results in a laboratory of nephrology, where the prevalence of renal diseases and pathologic findings is higher.
We collected 298 consecutive midstream urine samples from patients with known or suspected renal diseases. The samples were first examined with a Sysmex OF-50 (software version 0.5; TOA Medical Electronics) and then with a phase-contrast microscope (5), according to the European guidelines, at low (x100) and high (x400) magnification, by the same team (one biologist and one nephrologist, who independently analyzed the samples and then compared and discussed the results). The upper reference limits for phase-contrast microscopy used in our laboratory are as follows: erythrocytes, <2/high-power field (HPF); leukocytes, <4/HPF; small-sized hyaline casts, <4 in each slide; hyaline with medium or large diameter, cellular, granular, granulo-hematic, waxy casts, none; cystine crystals, none; other crystals, rare (<10 in each slide). We usually notice a high prevalence of pathologic findings (-70%) in microscopic analysis of urine samples from our patients.
Applying the published algorithm, we estimated that only 93 samples (31%) would not require microscopic confirmation, but 205 samples (69%) would need microscopic review.
The correlations between the results obtained with the OF-50 and the microscope are showed in Table 1 (Cohen K coefficient; SPSS). The concordance in detecting pathologic samples was good for erythrocytes (K = 0.67) and leukocytes (K = 0.72). The OF-50 revealed remarkable flaws with regard to the analysis of crystals and casts, for which the concordance between the two methods is very low (K = 0.043 and 0.08, respectively). These formed elements must be analyzed by microscopy to define their type and/or size. Flags for casts were produced by the instruments for 7% of samples, probably because of interference by squamous epithelial cells or mucous threads and cylindroids, without microscopic detection, and 40% of samples with pathologic casts were not recognized by the OF-50. This represents a serious limitation in a laboratory of nephrology. Moreover, the flow cytometer detects only granular and cellular casts.
We also tried to evaluate whether flow cytometry is useful in the analysis of erythrocyte dysmorphism by applying the Kitasato criteria (6, 7). However, this system relies on erythrocyte volume and size, and the flow cytometer distinguishes erythrocytes on the basis of cellular diameter but cannot recognize dysmorphic erythrocytes with altered shape, such as acanthocytes or codocytes (8). Thus, in our study we observed that of 131 samples with microhematuria by microscopy, 41 (31%) had predominantly dysmorphic erythrocytes according to both methods, whereas 59 (45%) had predominantly normal erythrocytes, but 16 (12%) revealed their dysmorphism only when analyzed by phase-contrast microscopy with a significant presence of altered cell shape (acanthocytes), and 15 (11%) showed their dysmorphism only when analyzed by flow cytometry, probably because of the presence of high amounts of yeasts or erythrocytes with different sizes.
In conclusion, combining the automated and traditional analyses of urinary formed elements in general laboratories-starting with automated cell counting followed by microscopic analysis, which is more specific in revealing morphologic aspects-may be a time-sparing policy. In a laboratory of nephrology, however, where samples have a strong preselection, such an algorithm is not applicable and all samples must be analyzed by phase-contrast microscopy. Nonetheless, the use of the automated procedure may help to save time on red and white cell counts, thus allowing the operators to dedicate more time to the morphologic definitions.
We suggest to the authors to use microscopic review of all samples referred from selected settings with a high prevalence of renal diseases, as those samples may be pathologic even in the absence of review flags from the flow cytometer.
We give special thanks to Dasit s.p.a. Italy, which provided us with a SYSMEX OF-50 (Toa Medical Electronics), technical assistance, reagents, and disposable equipment in support of our study.
(1.) Ottiger C, Huber AR. Quantitative urine particle analysis: integrative approach for the optimal combination of automation with UF-100 and microscopic review with KOVA cell chamber. Clin Chem 2003;49:617-23.
(2.) Ben-Ezra J, Bork L, McPherson RA. Evaluation of the Sysmex UF-100 automated urinalysis analyzer. Clin Chem 1998;44:92-5.
(3.) Delanghe JR, Kouri TT, Huber AR, HannemannPohl K, Guder WG, Lun A, et al. The role of automated urine particle flow cytometry in clinical practice. Clin Chim Acta 2000;301:1-18.
(4.) Lun A, Ziebig R, Priem F, Filler G, Sinha P. Routine workflow for use of urine strips and urine flow cytometer UF-100 in the hospital laboratory. Clin Chem 1999;45:1305-7.
(5.) European Urinalysis Group. European urinalysis guidelines. Scand J Lab Invest 2000;60:1-96.
(6.) Apeland T, Mestad 0, Hetland 0. Assessment of haematuria: automated urine flowmetry vs microscopy. Nephrol Dial Transplant 2001;16: 1615-9.
(7.) Hyodo T, Kumano K, Sakai T. Differential diagnosis between glomerular and nonglomerular haematuria by automated urinary flow cytometer. Kitasato University Kidney Center criteria. Nephron 1999;82:312-23.
(8.) Kohler H, Wandel E, Brunck B. Acanthocyturia-a characteristic marker for glomerular bleeding. Kidney Int 1991;40:115-20.
Massimo Gai *
Giorgina B. Piccoli
Giuseppe P. Segoloni
Laboratory of Nephrology University of Torino 10126 Turin, Italy
* Address correspondence to this author at: Azienda Ospedaliera "San Giovanni Battista" di Torino, U.O.A.D.U. Nefrologia, Dialisi e Trapianto, Corso Bramante 88, 10126 Torino, Italy. Fax 39-0116963158; e-mail massimogai@ katamail.com.
Table 1. Pathologic samples for erythrocytes, leukocytes, casts, and crystals and concordance of methods after analysis of 298 urine samples by microscopy and flow cytometry. RBCs, (a) >2 WBCs, >4, cells/HPF cells/HPF Microscopy, n (%) 131 (44%) 84 (28%) Flow cytometry (UF-50), n (%) 115 (39%) 88 (30%) Concordant results, % 83.9 88.6 K (Cohen) 0.67 0.72 Pathologic Significant casts crystalluria Microscopy, n (%) 153 (51%) 16 (5%) Flow cytometry (UF-50), n (%) 55 (18%) 9 (3%) Concordant results, % 53.1 92.2 K (Cohen) 0.08 0.043 (a) RBC, red blood cell (erythrocyte); WBC, white blood cell (leukocyte).
|Printer friendly Cite/link Email Feedback|
|Author:||Gai, Massimo; Piccoli, Giorgina B.; Segoloni, Giuseppe P.; Lanfranco, Giacomo|
|Article Type:||Letter to the editor|
|Date:||Sep 1, 2003|
|Previous Article:||Is cystatin C a marker of glomerular filtration rate in thyroid dysfunction?|
|Next Article:||Stability of N-terminal pro-brain natriuretic peptide after storage frozen for one year and after multiple freeze-thaw cycles.|