Introduction

The periparturient period, defined as 3 weeks around the time of parturition, is physiologically stressful and represents a critical time for dairy cattle (Sundrum 2015). During that time, many physiological, nutritional, metabolic, and immunological alterations can occur as the cow transits from a gestational non-lactating state to the period of copious milk production (Sordillo and Raphael 2013). The ability of a dairy cow to withstand these alterations is crucial to optimize herd management since the demands for milk production cannot be met solely by feed intake (Herdt 2000). Cattle unable to safely transit through this period are at a higher risk for metabolic disorders and decreased milk yield (Drackley 1999; Herdt 2000).

Among the commonly reported metabolic disorders that occurred during the transition phase, hyperketonemia is mostly prevalent and provokes a significant economic impact (Opsomer 2015). In that context, a working definition of SCK is that “a condition indicated by abnormal concentrations of circulating ketone bodies in the absence of detectable clinical signs of the disease” (Andersson 1988). In a dairy cattle herd, SCK can lead to serious economic loss involving a decrease in milk yield, decrease reproductive efficiency, and increase the risk of various diseases such as abomasal displacement and clinical ketosis (Andersson 1988; Opsomer 2015). Hence, early diagnosis of SCK can permit proper treatment and help mitigate further economic losses (Enjalbert et al. 2001; Geishauser et al. 2001). Because of its stability in blood, beta-hydroxybutyrate (BHBA) still considered the gold standard test for the diagnosis of SCK under field condition (Herdt 2000).

Insulin resistance (IR) has been defined as a state when a physiological level of insulin produces a lesser biological response than do normal (Kahn 1978). Other investigators have described IR as a condition in which high amount of insulin is required to produce a normal biological response (Berson and Yalow 1970). It has been reported that insulin could play a fundamental role in the physiological adaptation of dairy cows especially around the time of parturition (Zachut et al. 2013). The glucose metabolism in ruminants is often characterized by low peripheral glucose levels and a low insulin response of the peripheral tissues (Hayirli 2006). After calving, cows undergo a period of transient IR to prioritize the insulin-independent uptake of glucose by the mammary gland in order to favor milk production. Hence, the maintenance of blood glucose values within the normal physiologic ranges is of utmost importance during the transition phase (De Koster and Opsomer 2013).

Factors that cause IR in humans are also related to those involved in the development of ruminant hepatic lipidosis and ketosis (Drackley et al. 1992). The mechanisms for IR are not completely understood and seem to be a combination of genetic, cellular, and environmental causes (Ferrannini and Mari 1998). To date, there have been several endeavors describing IR in dairy cows with emphasis on transition or lactation periods (Sano et al. 1993; Sternbauer and Luthman 2002; Terao et al. 2010; Jaakson et al. 2013) or in association with primary clinical ketosis in cows (Xu et al. 2014) or ketotic cows with primary displaced abomasum (Khalphallah et al. 2015). But, it is not fully explicit what IR plays in the etiology of SCK. In line with these considerations, the present study was designed to evaluate the extent of IR in cows with SCK during the transition period.

Materials and methods

Study area and animal population

The present study was conducted on a commercial herd with a stock population of 730 Holstein-Friesian cattle. The farm is located in the city of Ras El-Bar, Damietta Governorate, Egypt. The study has taken place during the period between August and October 2015 where the average annual temperature was 20.2 °C (26 °C in August and 23 °C in October) (http://en.climate-data.org/location/51100/). Cows on the farm had, in general, an average of 305 days normalized milk production, body weight of 550 ± 50 kg, and at 3–7 years of age and where economics played the main role in farm decisions. The animals were apparent healthy, of different parities, and were artificially inseminated. The cows milked twice a day with a record of milk production in an average of 20.0 ± 3.0 kg/head/day in summer season and had a range of body condition score at 3–3.5. The investigated cows were free from external and internal parasites and had no history of metabolic and production disease in that year. All animals were kept under identical conditions of housing and vaccination throughout the study period. The feed for all cows consisted of a base ration fed as a daily total mixed ration as well as corn silage and Alfa hay in summer as an alternative source of roughage. Water supply was offered to all animals ad libitum. Cows were dried off 60 days before the expected time of parturition.

Sampling and the criteria of animal selection

On the farm, there were 170 cows at different stages of lactation. Twenty-four primiparous cows (i.e., cows pregnant with their second calf), at 3–4 years of age, were only located at a late gestation time approximately 3 weeks before the expected time of parturition. For the initial screening of SCK, 10 ml of venous blood were drained from each of the transition cattle (n = 24), through coccygeal venipuncture into a plain tube, i.e., without anticoagulants. Blood was rapidly cooled on crushed ice and was transported to the laboratory to be centrifuged at 1400×g for 10 min to separate blood serum. Only clear non-hemolyzed serum was collected and then aliquoted for estimation of BHBA. Basically, cows with no remarkable clinical signs but having serum BHBA concentrations >1.20 to 2.9 mmol/l were considered to have SCK as previously mentioned by McArt et al. (2012) and Xu et al. (2014). At the meantime, cows with no clinical signs and serum BHBA concentration <1.00 mmol/l were considered to be negative controls. The collected serum samples were further used to estimate the following biochemical variables: non-esterified fatty acid (NEFA), glucose, insulin, and cortisol, according to the standard protocols of the manufacturers. Commercial kits were used for estimating serum BHBA (Ben Biochemical Enterprise, Italy; Cat No. HB8855), glucose (Spinreact, Spain; Cat No. 41011), insulin (Bovine Insulin ELISA Kits, ALPCO, USA; Cat No. 30-AS1011.1), and cortisol (Bovine cortisol ELISA Kit, MyBioSource, USA Cat No. MBS738051), while NEFA was measured chemically according to the method described by Schuster and Pilz (1979). In general, cows were included in the present study in the following conditions: (1) if their previous gestation period was more than 260 days, (2) if they were not treated for ketosis before BHBA testing, and (3) if they have had a previous normal calving altitude and were being clinically healthy following the parturition.

Measurements of insulin sensitivity

Different surrogate indices were used to calculate insulin sensitivity as follows: revised quantitative insulin sensitivity check index (RQUICKI) = 1 / [log (glucose, mg/dl) + log (insulin, μU/ml) + log (NEFA, mmol/l)] (Xu et al. 2014); quantitative insulin sensitivity check index (QUICKI) = 1 / log (glucose, mg/dl) + log (insulin, μU/ml) (Katz et al. 2000); revised quantitative insulin sensitivity check index including BHBA (RQUICKIBHB) = 1 / log (glucose, mg/dl) + log (insulin, μU/ml) + log (NEFA, mmol/l) + log (BHBA, mmol/l) (Balogh et al. 2008; Abuelo et al. 2016); homeostasis model assessment (HOMA) = insulin (μU/ml) × glucose (mmol/l) (Abuelo et al. 2016); log transformation of HOMA (log HOMA) = log (insulin (μU/ml) × glucose (mmol/l) (Abuelo et al. 2016); and reciprocal score of HOMA (HOMA−1) = 1 / (insulin (μU/ml) × glucose (mmol/l) (Abuelo et al. 2016).

Statistical analysis

Data were statistically analyzed by using SPSS statistical software program (SPSS, version 15, USA). Independent samples t test was used to compare variables between ketotic and control groups. The means and standard deviation (SD) for each variable were calculated. At P < 0.05, results were considered statistically significant. Spearman correlation coefficient was also applied to emphasize the correlation among the tested variables and the indices of insulin sensitivity.

Results

An overview of biochemical alterations as well as surrogate indices of insulin sensitivity in dairy cattle with SCK during the transition period is presented in Tables 1 and 2. In the present study, SCK was diagnosed on the basis of serum levels of BHBA. Cows with serum BHBA >1.20 mmol/l were considered to have SCK (n = 20), while cows whose serum BHB < 1.00 mmol/l were served as controls (n = 4). Cows with SCK had no detectable clinical findings during the initial screening. Biochemically, values of serum BHBA, NEFA, and insulin were significantly higher in cows with SCK than those of controls (P < 0.05), while values of serum cortisol and glucose (despite being higher in diseased cows than those of controls) showed no significant changes between the two groups (P = 0.605 and P = 0.269, respectively) (Table 1).

Table 1 Means ± SD of selected metabolic profile in dairy cattle with subclinical ketosis compared with controls during the transition period
Full size table
Table 2 Means ± SD of insulin sensitivity indices in dairy cattle with subclinical ketosis compared with controls during the transition period
Full size table

Indices of insulin sensitivity showed a significant decrease in RQUICKI (P < 0.047) and a significant increase in HOMA (P < 0.007) in diseased cows compared with controls, while other surrogate indices showed no significant alteration between the two groups (Table 2). A positive correlation between BHBA and NEFA (r = 0.44, P < 0.05) was illustrated in Table 3. There were also statistically significant correlations among the estimated metabolic variables and indices of insulin sensitivity; the detailed results about these correlations are presented in Table 3.

Table 3 The correlations among the metabolic variables and indices of insulin sensitivity in dairy cattle with subclinical ketosis during the transition period
Full size table

Discussion

In the present study, cows with SCK showed a statistically significant higher values of serum insulin, BHBA, and NEFA than those of controls, while values of serum glucose and cortisol showed no significant alterations between the two groups. Unlike the results obtained in the present study, some authors have found hypoinsulinemia and hypoglycaemia in several clinical entities including SCK in cows (Tehrani-Sharif et al. 2012), clinical ketosis in cows (Kerestes et al. 2009; Xu et al. 2014), clinical ketosis in buffalo (Teli and Ali 2007), or physiologically during the transition phase (Abuelo et al. 2016). Nevertheless, some authors have found normal glucose concentrations in cows with SCK (Grohn et al. 1983; Sakha et al. 2006, 2007; Forslund et al. 2010). The enormous glucose drain toward the udder and the distinctive features of the transition period in dairy cows make the glucose metabolism an example of how intensive genetic selection can drive metabolism to extremes. However, it becomes apparent that glucose could have a low sensitivity and cannot be a good criterion for the diagnosis of SCK, and it does not appear to be useful for monitoring SCK. Similar findings were previously obtained by Sakha et al. (2006). It has also been stated that the periparturient period in dairy cows could be associated with alterations in insulin action particularly in the peripheral tissues and the extent of insulin secretion prepartum has been found to correlate with higher insulin action (Zachut et al. 2013).

In the current investigation, BHBA was found to correlate significantly with NEFA (r = 0.44, P < 0.05). However, the negative correlation between BHBA and glucose concentrations was not observed here. These findings were coincided with those previously mentioned by Asl et al. (2011), but unlike those obtained by several authors elsewhere (Sakha et al. 2007; Tehrani-Sharif et al. 2012). In some reports, it has been suggested that plasma BHBA concentrations above 1.4 mmol/l combined with plasma glucose values of less than 55 mg/dl could be used to classify cows as having poor energy status (Whitaker et al. 1993).

Our findings showed that serum cortisol despite being higher in cows with SCK than those of controls, its values did not reach statistical significance (P = 0.605). Nonetheless, serum cortisol was found to correlate positively with NEFA (r = 0.63; P ˂ 0.01) and glucose (r = 0.42; P ˂ 0.05). The insignificant elevation of circulating cortisol besides the decrease in the sensitivity of peripheral tissue to insulin (as demonstrated in this study) could result in a state of hyperglycemia. Nevertheless, the obtained results were away from that obtained by Forslund et al. (2010) who stated that cows with ketosis had significantly low levels of cortisol attributed that to the difficulties in synthesizing cortisol when their energy demands increase in peak lactation.

The sensitivity of the peripheral tissue to insulin has been assessed through the calculation of various surrogate indices according to recently published report (Abuelo et al. 2016). The widely accepted gold standard method for assessing peripheral insulin sensitivity is the use of hyperinsulinemic euglycemic clamp test, but it has been found to be time-consuming and being an invasive procedure; hence, it has not been recommended for use either under field conditions or on a large-scale epidemiological investigations (Muniyappa et al. 2008).

In the current investigation, cows with SCK showed a significant low RQUICKI index and high HOMA compared with controls. High HOMA level could indicate an increase in serum values of glucose, insulin, or both, thereby could suggest a low insulin sensitivity and low glucose tolerance, while low RQUICKI and RQUICKIBHB could reflect the high concentration of insulin, glucose, BHBA, and NEFA and hence could suggest a low insulin responsiveness. The obtained findings were in parallel with the recently published reports in periparturient dairy cows (Abuelo et al. 2016). Similarly, some authors have found that values of RQUICKI were lower in ketotic cows than those in controls (Xu et al. 2014). Likewise, indices of log HOMA, HOMA−1, and QUICKI were not significantly changed between the diseased and control group. These findings were a little bit away from the results obtained by several other authors (Abuelo et al. 2014, 2016), probably due to the low number of animals being studied.

Several other researchers have recently highlighted the role of NEFAs in insulin sensitivity of dairy cattle. In that context, an increase of circulating NEFAs has been detected following a fasting period or by the intravenous administration of a tallow infusion and has caused an impairment of insulin-stimulated glucose uptake by insulin-sensitive tissues (Oikawa and Oetzel 2006; Pires et al. 2007; Schoenberg et al. 2012). In the same line, it has been reported that cows with elevated serum values of BHBA (i.e., more than 1.0 mmol/l 2 days before calving until 7 days post calving) were likely to have higher IR in the glucose metabolism at their peripheral tissues than cows with normal values of BHBA during that time (Kelpe et al. 2003). In the same study, a higher concentration of NEFAs was significantly correlated with a lower insulin secretion. These findings could suggest a significant role of NEFAs in the development of IR. Besides NEFAs, other potential (unknown) factors originating from the adipose tissues could probably affect insulin-stimulated glucose uptake.

Conclusion

The results herein demonstrated that dairy cows could be vulnerable to the development of NEB and SCK during the transition phase. The correlation among the metabolic variables and indices of insulin sensitivity could indicate that dairy cows undergo massive metabolic adaptations during the transition period and could imply a potential role of IR in the development of SCK in dairy cows. Of particular interest, the elevated values of serum glucose, cortisol, insulin, NEFA, and BHBA are suggestive of impaired whole-body insulin sensitivity and were consistent with decreased RQUICKI values in these animals. Further studies are needed to use medications or feed supplements to enhance insulin sensitivity and help mitigate the deleterious consequences of bovine ketosis during the transition time.