Bronchoalveolar lavage fluid in Standardbred racehorses: influence of unilateral/bilateral profiles and cut-off values on lower airway disease diagnosis.

The aim of this study was to determine whether the lung side being sampled would significantly influence bronchoalveolar lavage (BAL) cytological profiles and subsequent diagnosis in Standardbred racehorses. One hundred and thirty-eight French Trotters in active training and racing were included in a prospective observational study. BAL was performed using videoendoscopy in both right and left lungs during summer meetings in 2011 (64 horses) and 2012 (74 horses). Cytological data performed 24h later from right and left lungs were compared and specifically used to classify horses as affected with exercise-induced pulmonary haemorrhage (EIPH), inflammatory airway disease (IAD), or were 'controls'. For IAD, cytological definition was based on two different cut off values. Neutrophil percentages, haemosiderophage percentages and the haemosiderophage/macrophage (H/M) ratios were significantly higher in the right compared to the left lung. Measures of intra-class correlation coefficients revealed a fair agreement between left and right lungs for percentages of mast cells, eosinophils, and for the H/M ratio, and a moderate agreement for neutrophil percentages. Fair to moderate agreements were observed between left and right lungs for the diagnosis of IAD and/or EIPH based on kappa coefficients. When sampling one lung only, the risk of incorrectly classifying a horse as a 'control' increased with the use of the restraint cut-off values for IAD. As BAL from one lung is not representative of the other lung in the same horse, both lungs should be sampled for a better assessment of lung cellularity and for a precise diagnosis of lower airway diseases.