Results
Descriptive analyses
During the data collection period, a total of 737 samples were received at the diagnostic laboratory. Data control was performed and duplicates were removed, as only the first sample from a horse was included. Finally, six hundred and four BAL samples were included, originating from 70 different veterinary practices around Denmark. Forty-seven different horse breeds were represented, with warmblood (n=108) and Icelandic horse (n=89) being the two most common breeds. The mean age of the total population was 10.5 (±4.5) years (Table 1). Bronchoalveolar lavage differential cell counts were done for all horse samples included (Table 2).
Clinical signs
An anamnesis of poor performance without concomitant coughing was not associated with TMS (p=0.060). Horses without an anamnesis of coughing and poor performance, and horses with only poor performance had significantly lower levels of BAL neutrophils compared to the group of horses with coughing (p=0.009 and p=0.001, respectively) and the group of horses with both poor performance and coughing (p=0.005 and p<0.001, respectively). No differences were found between horses with clinical signs coughing or poor performance and the percentage of BAL mast cells (p=0.129).
Tracheal mucus score
A TMS was available for 448 horses, and a significant association between TMS and asthma types was found (p<0.001). The largest contributions to the differences between the observed and expected values in the Chi-square test were due to horses with normal BAL cytology having a TMS of 1 (14.8%), while horses with SEA had a higher frequency of TMS score 4 (15.7%) and score 5 (14%). The TMS plotted against the different BAL cytology diagnosis in shown in figure 1.
The further specification of horses diagnosed with MEA and differentiated into MEA neutrophilic, MEA mastocytic, or MEA mixed diagnoses revealed significant differences in TMS (p<0.001) (Table 2). Horses with normal BAL cytology had a higher frequency of lower TMS scores compared to horse with SEA, with a contribution of horses with normal BAL cytology /TMS scores 0-1 of 13.8% vs. SEA/TMS score 4-5 of 26.7%. Amongst the differentiated asthma types, the MEA mastocytic accounted for the highest proportion of differences in TMS scores with a total contribution of 9%, followed by MEA neutrophilic 6%, MEA eosinophilic 3% and MEA mixed 1.6%. No significant association between TMS and BAL mast cell count was found.
BAL volume aspirated
The BAL volume aspirated was available for 540 horses. No significant difference was found for the group of horses with normal BAL cytology values compared with the group of horses with MEA (p=0.076), but significantly higher compared to the group of horses with SEA (50.0% vs. 40.0%, p=0.035) (table 2).
No significant difference between the BAL volume aspirated and the age of the horse was found, neither when horses were sub-divided into age groups (p=0.139).
Housing
Information on housing (single box or loose-housed in groups) was collected for 523 horses. The Chi- square test revealed dependency between housing and EA groups (p=0.011). Of the horses in loose-housed groups 28% had normal BAL cytology, 34% had MEA and 38% had SEA. Of the horses housed in single boxes, 19% had normal BAL cytology, 49% had MEA and 32% had SEA diagnosis (Figure 2). Information on roughage fed (divided into dry hay n=186, soaked hay n =56 or haylage n=254) was available for 496 horses, the Chi-square test revealed dependency between roughage fed and EA groups (p<0.001). Of the horses with the SEA diagnosis, 56% were fed hay, 14% were fed soaked hay and 30% were fed haylage at the point of examination. For the horses with the MEA diagnosis, 27% were fed hay, 10% were fed soaked hay and 63% were fed haylage.