quantitation of turbinate atrophy in pigs to measure the severity of induced atrophic rhinitis

The two-fold purpose of this study was to establish a useful image analysis technique for quantitation of turbinate atrophy and to determine an optimum bacterial dose for inducing atrophic rhinitis (AR). Two morphometric analysis methods were compared to determine a turbinate area ratio (TAR) and a turbinate perimeter ratio (TPR); the ratios of turbinate area to total nostril area and of turbinate perimeter to total nostril perimeter, respectively. Our first image analysis method differed from Collins et al (1) in that we used direct image capture (digitalization) via a video camera and a Macintosh microcomputer, rather than photographs and a digitizer tablet. The tracing techniques were the same as those used by Collins et al. The second morphometric method was modified from the first by exclusion of dorsal turbinate when tracing the nostril area and directly tracing only the ventral turbinate to get a turbinate measurement without subtracting. Area and perimeter ratios, for both methods, were compared to conventional visual snout scores, ventral measurements, and to each other. The results of the two image analysis methods correlated well, both with each other and with the visual scores. Doses of Pasteurella multocida (Pm) at a constant level, and Bordetella bronchiseptica (Bb) at various concentrations, were administered to 36 Hampshire-Duroc F1 SPF pigs to determine the best dose and frequency for inducing AR. Although the dose selection may have been somewhat affected by the pre-existing presence of Bb, the optimal dose per naris in this study was 2 mL Bb at 10(7) cfu/mL combined with 2 mL Pm at 10(9) cfu/mL inoculum. The frequency of administration (1 x or 2 x) did not greatly affect results. Turbinate area ratio was the best tool for quantitating gross morphological turbinate changes associated with atrophic rhinitis in this study. Our simplified modification of Collins et al image analysis method (exclusion of dorsal turbinates and direct measurement of ventral turbinates) correlated well with visual scores, and, when compared to Collins et al method, required less data manipulation and labour.