ABSTRACT:
Diagnosis of bacteriuria in veterinary medicine is commonly based on unstained urinary sediment evaluation. Nonetheless, amorphous particles can be confused with bacteria. This study aims to investigate whether the stained sediment increases the sensitivity and specificity of bacteriuria detection. One hundred urine samples were collected, with 60 obtained through voided specimens and 40 through catheterization, from dogs of various breeds, genders, and ages. Additionally, a 1 ml aliquot from each sample was subjected to quantitative bacteriological culture, serving as the gold standard test for comparison with the sediment analysis. Comparing to the bacteriology culture, the stained sediment of urine collected by catheterization exhibited a relative sensitivity of 60% and specificity of 70%, while to voided specimens showed a sensitivity of 93.3% and specificity of 57.8%. Unstained sediment presented a sensitivity of 100% and specificity of 10% to urine collected by catheterization, and 93.3% and 4.4% to voided samples, respectively. Both stained and unstained methods demonstrated negative predictive values > 60%, but the positive predictive value of unstained sediment ranged to 24.6% (voided samples) to 27% (catheterization) indicating a lower probability of true positives. Both techniques showed a low correlation coefficient (kappa) indicating that they cannot replace the gold standard method for confirming bacteriuria. However, Gram staining of urinary sediment improved the differentiation between amorphous substances and bacteriuria in canine urine samples, suggesting its potential applicability in laboratory routines.
Key words:
urinalysis; stained urinary sediment; bacterial culture; bacteriuria; Gram
Thumbnail
