Diagnosis of diseases in cattle at early stages is of significance both economically and clinically. Non-invasive diagnostic samples such as breath are preferred since they cause minimum inconvenience or pain to the animals. In this review, different sampling devices, sample preparation techniques, instrumentation, and statistical analysis approaches that have been designed and tested are described and compared in terms of their applicability in the diagnosis of common cattle diseases. The sample preparation techniques used include solid-phase microextraction (SPME), sorbent extraction, and needle trap device (NTD). The collected volatile organic compounds (VOCs) are determined using gas chromatography-mass spectrometry (GC-MS) and the electronic nose (e-nose) technology. The majority of studies are focused on the diagnosis of ketosis and bovine respiratory disease (BRD). The common diseases and potential biomarkers are summarized and discussed. Due to the differences in the number of subjects and the type of animals used in different studies, the results are not consistent. Acetone, although detected in almost all studies and subjects, has elevated concentrations in cattle suffering from ketosis. The results of currently available studies were not indicative of specific biomarkers for BRD, and further investigation is required. The current studies have shortcomings in regards to defining useful VOC profiles, the impact on animal welfare, and the practical application at the producer level. While the presented approaches are promising, more controlled, standardized clinical studies need to be conducted before breath analysis can be routinely performed on cattle.
