Insulin in the saliva of pigs: Validation of an automated assay and changes at different physiological conditions

Insulin, but not adiponectin, is detectable in equine saliva using an automated, commercial assay

BACKGROUND

The hormones insulin and adiponectin are commonly measured in equids because of their involvement in endocrinopathic laminitis. These are currently measured in serum/plasma, although jugular venipuncture can cause stress in some animals and may be impossible in needle-shy individuals. However, both hormones can be measured in saliva in other species.

OBJECTIVES

To determine whether [insulin] and [total adiponectin] are detectable in equine saliva using automated assays and whether saliva collection is associated with changes in stress indicators. Additionally, the correlation between serum and salivary [insulin] was investigated.

STUDY DESIGN

In vivo experiment.

METHODS

Paired blood and saliva samples were collected from eight adult ponies at multiple time-points (n = 45 paired samples). [Insulin] and [total adiponectin] were measured using automated assays validated for equine serum/plasma. Blink rates and heart rates were determined, using video recordings and a wearable heart rate monitor respectively, to assess the effects of sample collection on stress indicators compared with a control situation without a stressful stimulus.

RESULTS

[Total adiponectin] was undetectable in saliva. However, salivary [insulin] was measurable with acceptable inter-assay (1.3 ± 0.9%) and intra-assay (1.1 ± 0.6%) variability. Blink and heart rates during saliva collection did not differ significantly from those in the control setting. Serum and salivary [insulin] were not significantly correlated.

MAIN LIMITATIONS

Small sample size comprising native UK ponies; potential sampling bias as animals were recruited to the study partly based on their behaviour during blood sampling; saliva collected from unfasted animals.

CONCLUSIONS

Insulin is measurable in equine saliva using an automated assay currently available in the UK, but further validation and the determination of specific diagnostic thresholds are required. Saliva collection was not associated with changes in stress indicators. Further research is therefore needed to determine the potential of equine saliva as a non-invasive alternative to blood for insulin determination.