Abstract
Countless times a day veterinarians around the world rely on published reference intervals to determine what is ‘normal’ when evaluating their patients. A test result or physiologic parameter is of little value without appropriate information for interpretation. Typically, that information is in the form of a reference interval or normal range. Reference intervals are used to evaluate physiologic data (such as temperature, pulse, respiration and heart rate) and laboratory data (such as biochemistry analytes). Methods to establish reference intervals can be complicated and expensive to perform, and may be inaccurate or unreproducible.
It is important to remember that no reference interval or normal range is inherently right or wrong. Most laboratory reference intervals refer to the central 95% of the reference population. This means that 5% of the healthy population will have values that fall above or below the established range and may be flagged as ‘abnormal’.
All this raises questions – ‘What is the reference population?’ and ‘What is healthy?’ Clearly, a good definition of ‘healthy’ is critical to establish any normal reference interval or range and is one of the most problematic steps. As well, selection of the reference population has not often been scrutinized in veterinary medicine. For example, in-house laboratory analyzers may recommend the use of universal reference intervals from other sources (such as textbooks). The work of Randolph Baral and colleagues, published in the August 2015 issue of this journal, highlights the importance of appropriately generated reference intervals for clinical decision-making. Their papers showed that the results from in-house analyzers are not directly comparable with each other or with reference laboratories, and therefore the use of universal reference intervals may not be appropriate.1,2
Determination of body temperature is one of the vital assessments performed on virtually every veterinary patient. It will surprise most clinicians to learn that the methods and reference populations used to establish commonly used normal ranges for rectal temperature have not been well described. In a study published in this issue, Julie Levy and colleagues present a newly validated reference interval for rectal temperature for healthy indoor adult cats. 3 The range they suggest is lower than most published normal ranges and will result in a different interpretation of rectal temperature for some patients. The authors point out that using traditional normal ranges could result in overdiagnosis of hypothermia and underdiagnosis of mild pyrexia. It is important to note that the reference population in this study was healthy indoor adult cats. The authors recommend that additional studies are needed to evaluate normal rectal temperature ranges for juvenile and senior cats, as well as cats in different environments and under different conditions, such as following travel to the veterinary clinic.
The beauty of veterinary medicine is that we have colleagues willing to question, to think, and to find the answers!