Validity of pulmonary function indices derived from the volumetric capnogram in horses with recurrent airway obstruction (RAO).

This is a study examining how accurately pulmonary function indices derived from the volumetric capnogram can diagnose different degrees of recurrent airway obstruction (RAO) in horses. The research found that while some indices might indicate a difference between healthy horses and those with subclinical or minor RAO, none deliver an accuracy rate of over 90% when compared to a traditional clinical examination.

Research Objective

• The main aim of this research was to assess the reliability of pulmonary function indices obtained from the volumetric capnogram as a diagnostic tool for various stages of RAO in warmblood horses. This involved evaluating the sensitivity, specificity, area under the receiver operating curve (AUC), optimal cut-off values, and predictive values of these pulmonary function indices. The authors conducted this research on 63 warmblood horses.

Key Findings

• The study found that no single index had a specificity and sensitivity of over 90% in differentiating between varieties of RAO as judged against the traditional clinical examination. Therefore, none of the indices deliver what might be considered a highly accurate diagnosis.
• However, two of the indices – namely A(1)/A(2) and VD(alv)/ VT(alv) – displayed relatively higher sensitivity and specificity (AUC=0.55 and 0.53 respectively). This suggests that these indices might be useful in differentiating between healthy horses and those affected by subclinical or minor RAO.
• On the other hand, the indices VD(phys)/ VT and VD(Bohr)/ VT were found to be more effective in differentiating between horses suffering from subclinical or minor RAO and those presenting with moderate to severe RAO (AUC=0.71 and 0.70 respectively).
• Despite these findings, the authors cautioned against the adoption of cut-off values optimised for both maximal sensitivity and specificity. They asserted this approach would mistakenly classify a significant number of horses.

Conclusions

• In conclusion, the research underscores the lack of a highly accurate single index for diagnoses of different RAO degrees in horses when using the volumetric capnogram. While some indices displayed relative strength in differentiating between various health statuses relating to RAO, none achieved a benchmark of 90% accuracy.
• The authors advise caution about using cut-off values optimised for both maximum sensitivity and specificity, pointing to the likelihood of incorrect classifications. This means that while the volumetric capnogram can hold some diagnostic value, it cannot replace more traditional clinical examinations in assessing RAO in horses.