Ketosis

What it is

After calving, the dairy cow needs far more energy and glucose to produce milk than she can eat: she enters negative energy balance. In ruminants, glucose is not absorbed from the intestine in sufficient amounts; the liver makes it mainly from propionate produced in the rumen (gluconeogenesis). When glucose is short, the cow mobilises body fat and releases non-esterified fatty acids (NEFA), which reach the liver. There they are oxidised for energy; if the supply exceeds the hepatocyte's capacity, oxidation is incomplete and ketone bodies (β-hydroxybutyrate, acetoacetate and acetone) are formed, and the remaining fatty acids are re-esterified and stored as triglycerides: this is fatty liver (hepatic lipidosis).

• Subclinical ketosis (hyperketonaemia): raised blood β-hydroxybutyrate (BHB) without clinical signs; the most widely used threshold is 1.2 mmol/L.
• Clinical ketosis: higher concentrations with loss of appetite and production, or nervous signs.

Two forms are usually described: one caused by insufficient energy supply, in high-producing cows somewhat after the peak of demand, and one in the first two weeks after calving, linked to fatty liver and excessive fat mobilisation in overconditioned cows. Silages rich in butyric acid also supply ketone bodies.

Who it affects

High-producing dairy cows in the first weeks of lactation, most often in the first and second week after calving. Risk is increased by overconditioning at dry-off and calving, low intake during the transition period, postpartum diseases that reduce appetite (hypocalcaemia, metritis, mastitis, lameness) and the number of lactations. It is a herd problem: where there are clinical cases, there are usually many more subclinical ones.

Anatomy involved

• Liver: the central organ of the disease. It makes glucose from propionate and produces ketone bodies. In severe fatty liver the organ enlarges and becomes pale, yellowish and friable, and its gluconeogenic capacity falls, which worsens the ketosis. In cattle the liver lies entirely on the right, under the ribs, so it is biopsied through a right intercostal space.
• Rumen: its fermentation determines the energy supply. Propionate is the main glucose precursor; butyrate is converted to β-hydroxybutyrate in the ruminal wall. Lower intake reduces rumen fill and volatile fatty acid production, and clinical ketosis is accompanied by ruminal hypomotility.

Clinical signs

Subclinical ketosis has no signs, but it reduces milk yield and increases the risk of other diseases. In clinical ketosis:

• Digestive or wasting form: loss of appetite, usually first for concentrate and then for forage; a sharp drop in milk yield; loss of weight and body condition; firm, dry faeces; a hypomotile rumen.
• Smell of acetone on the breath, milk or urine, which not everyone can detect.
• Nervous form, less common: compulsive licking, chewing without feed, salivation, circling, incoordination, head pressing or aggression.

Diagnosis

• Blood β-hydroxybutyrate: the reference test. Hand-held electronic meters allow it to be measured cow-side with good accuracy.
• Ketone bodies in milk or urine with reagent strips or powders: useful for screening, but less accurate than blood measurement.
• Herd monitoring: testing several cows in the first two weeks of lactation to estimate the prevalence of subclinical ketosis; prepartum NEFA indicate the degree of fat mobilisation.
• Fatty liver: definitive diagnosis requires a liver biopsy with measurement of fat content; ultrasound can be suggestive.
• Concurrent diseases (displaced abomasum, metritis, mastitis), which often trigger or sustain ketosis, must be looked for.

Treatment and prevention

• Oral propylene glycol: a glucose precursor; the usual treatment for subclinical ketosis and part of the treatment of clinical ketosis.
• Intravenous glucose: rapid but short-lived effect; combined with propylene glycol in clinical ketosis.
• Correction of concurrent diseases and encouragement of feed intake.
• Prevention:
  • Avoid overconditioning cows at dry-off.
  • Maintain high intake during the transition period: enough feed bunk space, good access to water, gradual diet changes and stress control.
  • Prevent hypocalcaemia and other postpartum diseases.
  • Detect subclinical ketosis with systematic BHB testing of fresh cows.
  • Monensin reduces ketosis, but its use in dairy cows is not authorised in every country.

Drug use in food-producing animals is subject to each country's regulations and to withdrawal periods for milk and meat.

Key studies

• Duffield et al. (2009), cohort of 1010 Holstein cows from 25 farms in Ontario. Serum BHB of 1200 µmol/L or more in the first week after calving was associated with more displaced abomasum (OR 2.60) and metritis (OR 3.35); in the second week, the threshold for displaced abomasum was 1800 µmol/L (OR 6.22). BHB of 1400 µmol/L predicted clinical ketosis (OR 4.25 and 5.98 in the first and second week). There was no association with clinical mastitis. Hyperketonaemia reduced milk at the first test by up to 1.88 kg/d (first week) and 3.3 kg/d (second week). The farms were mostly tie-stall operations.
• McArt et al. (2012), cohort of 1717 cows from 4 herds in New York and Wisconsin, tested 6 times between days 3 and 16 of lactation. 43.2% had at least one BHB of 1.2 to 2.9 mmol/L; incidence and prevalence peaked on day 5 (22.3% and 28.9%). Cows first testing positive on days 3 to 5 had 6.1 times the risk of displaced abomasum, and those first positive on days 3 to 7 had 4.5 times the risk of leaving the herd, a lower probability of conceiving at first insemination and 2.2 kg/d less milk over the first 30 days.
• McArt et al. (2012), randomised trial in the 741 cows with subclinical ketosis from the same cohort: 372 received oral propylene glycol until BHB normalised and 369 received no treatment. Untreated cows had 1.6 times the risk of displaced abomasum and 2.1 times the risk of dying or being sold in the first 30 days. Propylene glycol improved conception at first insemination in 3 of the 4 herds, without changing days to pregnancy.
• Iwersen et al. (2009), evaluation of an electronic whole-blood BHB meter against laboratory analysis. In 196 samples the correlation was 0.95, with 100% sensitivity and specificity at 1400 µmol/L. In a field study with 926 samples from 35 practices, sensitivity was 88% and 96% and specificity 96% and 97% at 1200 and 1400 µmol/L. Milk and urine tests had clearly lower predictive values.
• Duffield et al. (2008), meta-analysis of 16 papers with about 9500 cows. Monensin reduced the risk of ketosis (RR 0.75), displaced abomasum (RR 0.75) and mastitis (RR 0.91), with no effect on milk fever, lameness, dystocia, retained placenta, metritis or fertility. Prolonged prepartum treatment might increase the risk of dystocia and retained placenta.