An integrated epidemiologic and radiographic algorithm for canine urocystolith mineral type prediction

Research involved 2 databases. One database (occurrence frequency) comprise d the age, breed, gender and urocystolith mineral type (pure chemical types only) from 2041 canine patients submitted to the Minnesota Urolith Center. The other database (imaging) comprised the maximum size, surface (rough, s mooth, and smooth with blunt tips), shape (faceted, irregular, jackstone, o void, and round) and internal architecture (lucent center, random-nonunifor m, and uniform) from 434 canine patients imaged in a urinary bladder phanto m. The imaging database was a partial subset of the occurrence frequency da tabase. Imaging techniques simulated were survey radiography and double con trast cystography. The databases were compared using multivariate analysis techniques. Equations were developed to use clinically-relevant characteris tics (age, breed, gender, maximum size, surface, shape, and internal archit ecture) to predict urocystolith mineral types. The goal was to assess the a ccuracy of the various techniques in predicting the urocystolith mineral ty pes. The combination of signalment (age, breed, gender) and simulated surve y radiographic findings does not improve mineral type prediction accuracy ( average across all mineral types is 69.9%) beyond that achievable with sign alment alone (average across all mineral types is 69.8%). However, the comb ination of signalment and double contrast cystography does improve mineral type prediction accuracy (average across all mineral types is 75.3%). For c omparison, mineral type prediction accuracy without signalment from survey radiographs only was 65.7% across all mineral types. The clinical utility o f the algorithm is the option to distinguish urocystolith mineral types req uiring surgical vs. medical treatment.