Variability of fungi detection in the respiratory tract of racehorses according to the sampling site
Résumé
The potential involvement of fungi in equine asthma was previously investigated based on tracheal wash (TW) rather than bronchoalveolar lavage (BAL). The aim of this preliminary study was to characterise fungi from airways of racehorses using culture and cytology. Our objectives were to compare both techniques and to evaluate if cytology of a pooled BAL (BALP) was representative of corresponding left (BALL) and right (BALR) samples. TW, followed by BAL from each lung separately, were collected through a guarded catheter and a videoendoscope on 30 Standardbred racehorses in training. Fungal culture was positive (i.e. ≥1 mould) for 28/30 (93%) TW and 4/30 (13%) BALP. All horses with a positive BALPculture were also positive on TW. Agreement between BALP culture and cytology (1 both, 3 culture and 6 cytology only) was poor (κ=0.015). TW fungal culture was overly positive and most likely of poor clinical significance. Fungi were identified by cytology in 22/30 (73%) TW and in 8/30 (27%) BALR, 7/30 (23%) BALL, and 7/30 (23%) BALP. Despite a similar prevalence of fungi detection in BALL and BALR, there was a large disparity between lungs on cytology (3 BALL+Rpositive, 4 BALL and 5 BALR only), and detecting fungi by cytology on either BALP or BALL + BALR was found not to be equivalent (P=0.025 by McNemar Test; Se=0.58; Sp=1.00). Pooling BAL is a highly specific while fairly sensitive alternative to combined cytology of individual left/right samples for mould/fungi detection.
Origine : Fichiers produits par l'(les) auteur(s)