The Italian External Quality Assessment (EQA) program on urinary sediment by microscopy examination: a 20 years journey

Abstract

Abstract Objectives In spite of the introduction of automated systems for urinary sediment analysis, microscopy examination remains the gold standard, and it is more than ever important to perform it with a good and reliable quality. External Quality Assessment (EQA) programs on urinary sediment are rare. The present paper provides an analysis of results from 2001 to date of the EQA Italian program which involves today 230 laboratories. Methods The program includes four surveys per year. Participants are asked the identification and clinical associations of urinary sediment particles, shown as phase contrast microscopy images in the website of the Center of Biomedical Research (CRB) (2 surveys), and the diagnosis of clinical cases presented by both images and a short clinical history (2 surveys). The results of each survey are then scored and commented. In 20 years, 298 images were presented: 90 cells (9 types), 23 lipids (5 types), 87 casts (21 types), 53 crystals (14 types), 22 microorganisms (5 types), and 23 contaminants (9 types). Moreover, 27 clinical cases, covering a wide spectrum of conditions with different degrees of complexity, were presented to participants. Results Identification: among urinary particle categories, the correct identification rate (obtained for each particle from the sum of correct + partially correct answers) was very high for micro-organisms (mean ± SD: 96.2 ± 3.5%), high for lipids (88.0 ± 11.8%) and crystals (87.0 ± 16.5%) followed, in decreasing order, by cells (82.1 ± 15.9%), casts (81.8 ± 14.8%), and contaminants (76.7 ± 22.1%). Clinical associations (n=67): the rate of correct answers was 93.5 ± 5.7% ranging from 75.0 to 100% for all but one clinical association (i.e., acute glomerulonephritis: 55.4%). Clinical cases: throughout surveys, due to the overall rate of particle misidentification, only 59.8 ± 17.1%, (range 32.5–88.7%) of participants achieved access to clinical diagnosis. Of these, 88.7 ± 10.6% (range 59.9–99.3%) were able to indicate the correct diagnosis. Conclusions Our program can be used as a tool to improve the identification of urine particles and the knowledge of their clinical meaning and to encourage specialists of laboratory medicine to correlate urinary findings with other laboratory data and the clinical history, an aspect that improves the value of the day by day work.