Skip to main content

Subclinical ketosis: »It hits the high-performance cows«

By Katharina Heil, DLG Mitteilungen

Stefan Freuen follows a strict plan for detecting subclinical ketosis. This farmer from the West Eifel region of Germany owns with two partners 220 dairy cows and followers on their FNS-Milch farm in Weinsheim, Rhineland-Palatinate. »I don’t like relying on hunches«, says Stefan Freuen. Facts and figures are more his thing. Since 2012, this has also applied to his monitoring for ketosis in the herd. The fat and protein results in the milk (LKV) records are initial indicators pointing to animals that might have ketosis. But backing-up such signs with BHB (beta-hydroxybutyrate) blood tests offered this farmer interesting information. With around 30% of animals, the ketosis signal from the fat-protein quotient (> 1.5) agreed with the blood analysis results (BHB value > 1.2 mmol/l blood.) But another 23% flagged-up as ketosis suspects by the fat-protein quotient showed no sign of disease in the blood test. These cows would have been treated for ketosis without this being necessary, states Freuen. However, he sees as much more serious the situation with a third group. For almost half the cows (47%) there was no ketosis warning signal in the milk records while the blood test indicated ketosis presence. »In criminal cases these would be called the unknown, or unreported, incidents«, says Freuen. »This situation is an absolute no-go.«

Strict blood sampling schedule

The BHB count is measured in blood taken from the tail vein. This is the only vein to be used, says the farmer. The udder vein, for instance, would give different results. He also prefers a digital sampling apparatus for numerical rather than colour code results. »Digital sampling is fast with a result available within 10 seconds. It’s simple, can be carried out with in-milk or dry cows and is also good value for money at just 1 to 2€ per test«, confirms the farmer.

Testing between lactation days 3 and 20

»It’s important that the blood tests take place at the same time of day, preferably mornings, and always the same period before feeding starts«, he states. Best is a couple of hours before feeding. With the FNS-Milch herd, this means sampling is between 10 and 10.30 am. Depending on the results, management of the cows is adjusted. The farm vet is always closely involved.

  • In the group with cows in their first 20 days of lactation, cows with no symptoms and BHB levels of < 1.1 mmol/l remain untreated. A further blood sample is taken on lactation day 8.  
  • Animals with BHB level between 1.2 and 2.4 mmol/l get a 6-day course of 250 ml propylene glycol as drench. Treatment is stopped on the seventh day so that the results of a repeat test on day 8 are not distorted. These cows are then routinely retested one more time, even if this takes place after lactation day 20. Only when the sample result is > 1.1 mmol/l is the cow »released into the lactation«, as Stefan Freuen puts it.
  • Cows with readings of > 2.5 mmol/l also receive a 6-day drench course of propylene glycol but to this is added a corticosteroid hormone application and two glucose infusions by the vet. Here too, there’s no treatment on day 7. Again, the follow-up test is on day 8 with this result deciding the next management move and to which of the three above-mentioned groups the animals are to be assigned.


Precise monitoring of every case

Since monitoring began in mid-2012, ketosis cases have reduced overall. 2015 had most cases of subclinical ketosis with 53 animals giving BHB readings of more than 1.1 mmol/l. In 2017 this applied to only 17 animals. In the following year, 29 cows were affected. But in general, the trend is a declining one. Stefan Freuen attributes this to better silage quality. Surprisingly, case numbers listed per quarter indicate markedly fewer cases during the warmer months. On the other hand, cases peaked in each first quarter. »I don’t have a logical explanation for this. I would have expected rather the converse«, he admits.

Looking at incidence with milkers after at least two lactations behind them shows that subclinical ketosis occurs in around one third of the cows. With performance in mind, Freuen asks: Could we look on ketosis as a »desirable disease« - the sign of a good cow? After all, cows with no ketosis produce less milk with average lactation of 12 800 kg while those with subclinical ketosis average 13 000 kg. Also, the insemination index of 2 is lower for ketosis cows compared with 2.5 for the others. For cows with at least four lactations the picture was just as clear. Here, half had been afflicted with subclinical ketosis. The higher the lactation number, the greater the problem, notes farmer Freuen. But here too, there are definite differences in performance, ketosis cows produced an average of 10 400 kg and insemination index of 2.5 compared with 9 000 kg and insemination index of 5 from non-affected cows. »What these figures actually show is that treatment of the disease is a highly profitable action to take«, says Freuen. But he cautions: »Our comparison is not a scientific one. After all, we are not comparing diseased animals with and without treatment, but instead only animals that show symptoms and are treated against those that show no symptoms«.

Subclinical ketosis: a »desirable disease«?

In order to answer this question, Freuen took 6 cows that had symptoms of subclinical ketosis in their fourth or fifth lactations. In the lactation with the disease, the respective animals produced 10 590 kg milk. However, in the previous lactation these cows had already achieved an output of 10 750 kg and, in the subsequent lactation after the disease, even managed 12 000 kg. »If the animals had not been afflicted by ketosis, their performance would have been better. The disease effectively slowed down their increase in milking performance.«

Evaluating cows with and without ketosis based on the effect on production lifetime led to an unexpected result. »Even I was very surprises by this«, admits the farmer. Animals that had subclinical ketosis remained in the herd for an average 4.5 lactations with a lifetime production of 43 260 kg, while the healthy cows in this respect only lasted for just under 3 lactations, producing 25 000 kg milk. Lifetime daily yield was also markedly better for the ketosis cows: 18.8 against 16.3 kg. »All this is evidence that the disease affects the highest performers. It’s therefore very profitable to identify and treat the disease.« A comparison of daily weight gain in the first year shows hardly any difference between the two groups. This means that more extensive, or intensive, feeding of the young cattle has no influence in this respect.

Culling because of ketosis

A glance at individual animal records with their respective »ketosis history« indicates that repeated ketosis incidence in three consecutive lactations is a is a very clear signal for culling. »This again relativises the point about desirable disease«, argues Freuen. A further discovery: cows that make it to lactations 7 - 10 have had no incidence of subclinical ketosis, or only very late on in production life. »100 000 l cows are not the ones that have problems early-on«, he concludes.

Evaluating cow lines in the herd demonstrates clear differences between families. For instance, there is one cow family in the FNS-Milch herd with only 4% of animals ever having been affected by subclinical ketosis while there’s another one with 45.5% of cows, i.e. almost every second animal, having been affected. Stefan Freuen links this situation to maternal heritability of the mitochondria which he sees as particularly decisive for cell metabolic activity. Therefore frequency in family records should be paid attention to in breeding and section.

Treatment strategies

»I have to qualify to a certain extent the comparison, in that our on-farm treatment strategy depended on the intensity of the ketosis«, says Freuen. Along with the cows which had the disease identified and treated in the above-mentioned strategies, cows judged overweight even pre-calving each had a monensin sodium bolus inserted, a treatment extremely contentious in practice because it includes an antibiotic. But the bolus treatment is authorised for ketosis treatment, so was applied to 67 cows on the farm. »The effect was not at all phenomenal«, reckons Freuen. The treated animals produced an average 10 170 kg compared with the other two treatment groups (corticosteroid hormone, glucose and propylene glycol or propylene glycol on its own), with milk production of 10 600 and 10 400 kg. Only the markedly better insemination index pointed to a desirable long-term effect. The bolus treatment is no longer applied because the farmer reckons that the other variants are markedly more effective as well as costing less.


Subclinical ketosis can only be tackled strategically. Included must be a way of individual cow monitoring, preferably with a digital blood test. Treating the disease in this way is very profitable, especially with high production animals. The therapy should be adjusted to match disease intensity. Especially affected are high-production animals in the herd. From the third occurrence of ketosis the danger of having to cull increases. Prevention of subclinical ketosis can be achieved by feeding measures in the dry phase and during the first days after calving as well as through breeding strategies.

This report is based on a presentation at the VLF (Association for Agricultural Education) Dairy Cattle Conference in Bitburg, Germany.