ID02 - Agreement between Collection Techniques for Detection of Histoplasma Antigen in the Urine of Dogs

Abstract:

Background: A commercially available Histoplasma urine antigen EIA (EIA) test is commonly used to support the diagnosis of histoplasmosis in dogs, but a gold standard urine collection technique has yet to be established.

Objective: Evaluate the agreement between three urine collection techniques—sterile cystocentesis (cysto), free catch (FC), and contaminated cystocentesis (contam)—using the EIA in dogs in an endemic region. Animals: 30 dogs suspected of histoplasmosis and 30 temporally matched, healthy dogs.

Methods: In this 12-month prospective study, three urine samples (cysto, FC, contam) were collected within a 24-hour period from each dog. Agreement between collection techniques was assessed using McNemar’s test and Kappa coefficients, with false positive rate (FPR) for all dogs combined calculated using the clinical diagnosis as the reference standard.

Results: Three clinically ill dogs were diagnosed with histoplasmosis. Cysto and contam samples had substantial agreement (Kappa = 0.68, 95% CI 0.39-0.97%), while moderate agreement was found between cysto and FC (Kappa = 0.45, 95% CI 0.07-0.82%) and FC and contam (Kappa = 0.42, 95% CI 0.10-0.75%) samples. The overall FPR was 16% (9/57; 95% CI 6–25%). Cysto samples had the lowest FPR (4%, 2/57; 95% CI 1–12%), followed by FC (7%, 4/57; 95% CI 3–17%), and contam (11%, 6/57; 95% CI 5–21%). Conclusions and Clinical Importance: While all collection techniques demonstrated reasonable agreement, FC samples might yield false positives. Clinicians should confirm low-level positive FC results with sterile cystocentesis to improve diagnostic accuracy and clinical decision making for histoplasmosis in dogs.