Ketosis has garnered a lot of interest in the last few years. Recent research has pointed to the role of ketosis as a gateway condition for transition cow health events and loss of milk production.

Fundamentally, ketosis is a state of energy imbalance, where energy demand exceeds energy absorption from the diet and energy production in the body. Most animals, including cattle, store energy as fat to meet these periods of negative energy balance.

The mobilization of fat as a strategic energy reserve is appropriate and purposeful during these times, but when the cow exceeds her ability to properly metabolize the mobilized fat, a disease called ketosis develops.

Ketosis is defined as the accumulation of detrimentally high systemic ketone bodies. Their presence results in reduced milk production, weight loss, reduced appetite, dull coat and fever. Diagnosis is often determined by a sweet smell from the cow’s breath.

One extreme clinical presentation is nervous ketosis. In this metabolic state, ketones are high enough to cause neurological signs such as aggressive behavior, licking, excessive salivation, chewing, bellowing and incoordination.

Click here to or on the image above to view it at full size in a new window.

For every cow that demonstrates ketosis symptoms, there are many cows that are subclinical or have elevated ketones without outwardly visible signs. These subclinical cows are the real money drain to the dairy since there are exponentially more subclinically ketotic animals than animals showing visible signs.

The clinical signs of ketosis are also associated with many different diseases, but since the greatest financial impact to the dairy are ketotic animals without clinical signs, it is appropriate to utilize tests to identify subclinically ketotic cows.

There are three ketone bodies that can be formed during energy production – acetate, acetoacetate and beta-hydroxybutyrate (BHB). As concentrations of ketone bodies increase, animals that are clinically or subclinically ketotic accrue measurable ketones in bodily fluids. The three most common fluids evaluated in the dairy industry are milk, urine and blood.

Testing milk

Milk is certainly the fluid that is the most abundant and easiest to sample on dairies. One of the tests used on milk is a powder that tests for acetoacetate. Powder tests for acetoacetate have a lower ability to identify sick animals, ranging from finding 41 to 89 percent of truly subclinically ketotic cows (sensitivity) depending on the brand.

This means the powder test may not be finding 60 percent of the subclinically ketotic animals. The value of a test is not only its ability to identify sick animals, but also to correctly determine healthy animals as negative, which is termed the specificity of the test.

When using the powder to check the milk, it is generally assumed that if the test is negative the animal does not have ketosis, since it has about a 96 to 99 percent ability to correctly identify healthy cows from the animals without ketosis (specificity).

When there is a low number of subclinically ketotic cows (prevalence), this statement is generally true. But as the number of subclinical cows in the herd increases (increased prevalence), this statement becomes less true as the animals with subclinical ketosis are missed and called negative by the powder milk tests.

Powder milk tests for ketosis are generally deemed as having limited value for identifying subclinical ketosis given the test performance (test characteristics) and probably shouldn’t be used unless there are no other options available or unless the testing frequency is high (daily) or the prevalence of subclinical ketosis is very low (5 percent or less).

Milk can also be tested with chemical strips for BHB. The chemical strips are better than powder tests in finding subclinical ketosis, identifying 73 to 91 percent (sensitivity) of subclinical animals.

This means that up to 27 percent of test-negative animals could have subclinical ketosis but were missed. The BHB milk strips are not good at identifying which animals do not have subclinical ketosis, identifying 56 to 90 percent (specificity) of animals without subclinical ketosis correctly.

Depending on how much subclinical ketosis actually exists on your dairy, that means 85 to 37 percent of the animals you treat for subclinical ketosis using this test will actually be normal, adding significant time and cost. With these test characteristics, this test has limited value as well. The most appropriate way to use it would be in a herd where the prevalence of ketosis is high, such as greater than 30 percent.

Testing urine

There are also commercial tests available to detect acetoacetate in urine. These tests generally find 90 to 100 percent (sensitivity) of subclinically ketotic animals and correctly identify negative animals 59 to 85 percent (specificity) of the time. The ability to find subclinically ketotic animals is better using urine acetoacetate, however it still is wanting in being able to correctly identify animals without subclinical ketosis.

Using “trace” as the cut off, 84 to 42 percent of the animals you treat will be normal. Again, this limits the value of the test to situations where the prevalence is higher.

Another limitation of this test is the inability to obtain a urine sample. In one reported study, only about 50 percent of cows could be induced to urinate.

A significant limitation of all the aforementioned tests is the subjectivity of evaluating the color change used to determine positive cows. Many people are color blind or have poor ability to distinguish minor color changes, which probably explains some of the dramatic variation in reported test characteristics.

Testing blood

Blood can also be tested for the concentration of BHBs. Historically, this test has been either an enzymatic or photometric assay performed in a laboratory, and has been considered the “gold standard” against which all other tests are evaluated.

The most commonly used cowside meter for blood BHBs has reported the ability to find subclinical cows 90 to 100 percent (sensitivity) of the time and correctly identifies healthy animals 98 to 100 percent (specificity) of the time.

With these test characteristics, blood BHBs are quickly becoming the standard for subclinical ketosis identification cowside when it is convenient to obtain blood samples. Limitations for blood BHB include having a facility to easily acquire blood samples, as well as the cost and availability of strips to use with the meters.

Since there are no perfect tests, multiple testing strategies may be employed to achieve a desired end goal not possible by using only one test.

One option is to test in series to improve the certainty that a test-positive animal truly has the disease. This is the strategy historically employed in the federal tuberculosis eradication program.

Testing in series improves the certainty of the test-positive animal of having ketosis, and it will also reduce the cost of using a very good test, such as the cowside blood BHB meter, as animals are screened with a cheaper sensitive test such as a urine acetoacetate or milk BHB test, which have less ability to correctly identify nonketotic animals that test positive from healthy animals (false positives).

Once an animal is screened to reveal a positive urine or milk BHB test, it could be confirmed with a blood meter BHB test, which would in turn reduce the number and associated costs of blood BHB strips used.

This kind of testing scheme comes at a cost though, as it will reduce the ability to find all ketotic animals. Some of the truly subclinically ketotic animals will be incorrectly identified as false negative by either test.

Any testing strategy will depend on the prevalence of ketosis in the herd, the facilities that allow for sample collection and ultimately, the goals of the dairy – all of which should be done in close consultation with your herd veterinarian.

Gerald Poppy has a veterinary degree and an MBA. He is the president of Fermented Nutrition.