Development and Validation of Quantitative Greyscale Analyses for the Detection of Aspiration-related Lung...

Existing scales for the detection of penetration and aspiration of material into the airway do not reliably predict pulmonary outcomes like pneumonia. We present here the development and validation of methods to track the short-term and long-term effects of aspiration on the appearance of the lungs in an infant pig model. Six pigs that varied in frequency and severity of aspiration were recorded as they fed on barium milk over two weeks (from 1 to 9 months human equivalent), capturing the entire airway in high-speed videofluoroscopic recordings (125 fps). Dorsoventral views of the lungs were used to collect minimum and mean greyscale measurements in relation to two phenomena: (1) liquid infiltration in the lungs during a feed and (2) liquid accumulation in the lungs due to repeated aspiration and development of inflammation over several days. Measurements were reliable within and across four raters (intra- and inter-rater ICC ≥ 0.67). We measured the darkening of the lungs due to the infiltration of liquid (paired t-test, p ≤ 0.016) and the accumulation of liquid over five days (p ≤ 0.001). Lastly, we compared the lung greyscale of all six pigs to their swallow safety scores (Infant Mammalian Penetration Aspiration Scale). Worsened lung greyscale was weakly correlated with high frequency of aspiration (Pearson's correlation coefficient - 0.16 to -0.21), indicating other factors play a role in the development of negative health outcomes following aspiration. Using these techniques, future longitudinal studies in the infant pig model may shed light on the links between airway protection and lung health.