# Establishment of an Inferred Reference Range for Blood Ammonia in Dogs and Cats Using a Point-of-Care Assay

## TL;DR

This study establishes normal blood ammonia ranges in dogs and cats using a quick test, finding that high ammonia levels in dogs are mostly due to liver disease, while in cats, other causes are more common.

## Contribution

The study introduces inferred reference ranges for blood ammonia in dogs and cats using a point-of-care assay.

## Key findings

- Dogs with blood ammonia above 30 µg/dL are likely to have liver disease.
- Cats with blood ammonia above 25 µg/dL show a wider range of disease causes.
- Most dogs and cats had undetectable blood ammonia levels.

## Abstract

Measuring blood ammonia is important in the context of some diseases, particularly when they affect the liver; however, the existing methods of measurement are complicated and subject to artefact. A simple and rapid test has been developed and previously used in dogs and cats, but reference values are not yet available. We used this test in groups of dogs and cats with many different diseases, seeking to differentiate those with normal and abnormal values. Where present, we also attempted to document the cause of ammonia elevation. Inferred reference values were developed for both species, and we found that, while most dogs with high ammonia had liver disease, cats with high ammonia exhibited a much wider variety of causes.

Point-of-care (POC) tests for blood ammonia (BA) measurement have not been well evaluated in veterinary species. This cross-sectional study sought to establish an inferred reference interval for BA using a POC analyser in dogs and cats. Blood ammonia was measured in 175 dogs and 63 cats for which relevant clinical history and laboratory data was available. Reference values were inferred based on comparisons between patients with and without disease pathologies reported to cause BA elevation. Descriptive statistics, Pearson Chi2, and Mann–Whitney U testing were used to assess for associations between clinical parameters and BA concentration. Seventy-one percent (124/175) of dogs and forty-six percent (29/63) of cats had undetectable BA. Following the exclusion of dogs with potential causes of hyperammonaemia, all remaining dogs had BA < 30 µg/dL. With one exception, all dogs with BA > 30 µg/dL had liver disease. All dogs with a clinical suspicion of hepatic encephalopathy (HE) had BA > 40 µg/dL. Following the exclusion of cats with potential causes of hyperammonaemia, all remaining cats had BA < 25 µg/dL. Only 50% of cats with BA > 25 µg/dL had liver disease. All cats with a clinical suspicion of HE had BA > 30 µg/dL. Based on this study population, BA < 30 µg/dL and <25 µg/dL should be considered normal in dogs and cats, respectively. Additionally, dogs with BA > 30 µg/dL are likely to have liver disease, while cats with BA > 25 µg/dL appear to exhibit a wider variety of disease pathologies.