Rapid direct-from-urine growth detection to triage culture-negative urine: a prospective diagnostic accuracy study of a two-tube nitrate-reduction and bromothymol blue assay (SIBURI/SIBURI+)
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Objectives
Urine culture takes 24–72 h to report, and most specimens are culture-negative. We evaluated whether a 6-h direct-from-urine growth-detection assay can identify culture-negative specimens.
Methods
In a prospective single-centre study at Goa Medical College, 145 consecutive urine specimens from patients with suspected urinary tract infection (UTI) were tested in parallel. Urine (50 µL) was added to SIBURI (nitrate-reduction broth, developed with Griess reagent) and SIBURI+ (Luria-Bertani broth with bromothymol blue) and incubated for 6 h at 37 °C. Any visible colour change against an uninoculated blank was scored growth-positive. The reference standard was quantitative culture (≥10□ CFU/mL of a single organism). Twenty-five specimens reported as contaminated, “SCC” or with ≥ 3 organisms were excluded. Accuracy was estimated with exact 95% confidence intervals. A cell-free nitrite series and ATCC Escherichia coli and Klebsiella pneumoniae dilution series characterised the colour response.
Results
Of 120 analysed specimens, 41 (34.2%) were culture-positive. Sensitivity was 92.7% (95% CI 80.1–98.5), specificity 92.4% (84.2–97.2), negative predictive value (NPV) 96.1% (88.9–99.2) and positive predictive value 86.4% (72.6–94.8). SIBURI alone detected 29/41 culture-positive specimens (70.7%); nine (enterococci, streptococcus, Acinetobacter, Candida and two unspecified) were detected only by SIBURI+. Classifying the excluded specimens as culture-positive gave a sensitivity of 93.9% and an NPV of 94.8%; classifying them as culture-negative lowered specificity to 71.2%. The colour response was biphasic: pink to red from 10 µM to 1 mM nitrite, and yellow above 1.5 mM.
Conclusions
In this single-centre study, a two-tube 6-h assay showed good accuracy against culture, but the lower confidence limits (sensitivity 80.1%, NPV 88.9%) do not yet support omitting culture after a negative result. Positive results need culture confirmation, and larger multicentre validation is required.
