Effects of a blend of essential oils in milk replacer on performance, rumen fermentation, blood parameters and health scores of dairy heifers

The objective of this study was to evaluate how the inclusion of a blend of essential oils in milk replacer (MR) affects different outcomes of dairy heifers. The outcomes evaluated: feed intake, performance, body development, blood cells and metabolites, insulin-like growth factor-1 (IGF-1), rumen fermentation, fecal scores and respiratory scores. All outcomes were evaluated during pre-weaning (4 – 60 d of age), and carryover effects on post-weaning (61 – 90 d of age) periods. The experimental units utilized were 29 newborn Holstein × Gyr crossbred dairy heifers, with genetic composition of 5/8 or more Holstein and 3/8 or less Gyr and body weight (BW) at birth of 32.2 ± 5.2 kg. Experimental units were randomly assigned to either a control group (CON, n = 15) or a treatment group, consisting of supplementation of a blend of essential oils (BEO, n = 14) with1 g/d/calf (Apex Calf, Adisseo, China). During the pre-weaning phase, all heifers were fed 5 L of MR/d reconstituted at 15% (dry matter basis), divided into two equal meals. Water and starter were provided ad libitum. During the post-weaning, animals received a maximum 3 kg of starter/d, and ad libitum corn silage, divided into two meals. The outcomes feed intake, fecal and respiratory scores were evaluated daily. BW was measured every three days, while body development was recorded weekly. Blood samples were collected on 0, 30 and 60 d of age for total blood cell count, weekly to determinate ß-hydroxybutyrate, urea and glucose, and biweekly for IGF-1. Ruminal parameters (pH, volatile fatty acids, ammonia-N and acetate:proprionate proportion - C2:C3) were measured each 14 days. A randomized complete block design with an interaction between treatment and week was the experimental method of choice to test the hypothesis of effect of BEO on all outcomes. An ANOVA procedure was used for continuous outcomes and a non-parametric test was used for the ordered categorical outcomes both adopting a C.I. = 95%. Results indicated that there was not enough evidence to accept the alternative hypothesis of effect of BEO in MR on: feed intake, performance, body development and blood metabolites during both pre-weaning and post-weaning periods. However, results indicated that the inclusion of BEO in MR significantly affects the proportion of C2:C3 during pre and post-weaning (P ≤ 0.05). Similarly, the effect is significant for basophil (P ≤ 0.001) and platelet (P ≤ 0.05) counts in pre-weaning. The interaction between week and treatment was also significant for lymphocytes (P ≤ 0.001), revealing a long-term treatment immunological effect. Lastly, the effect on fecal scores was also significant (P ≤ 0.05) during pre-weaning, with lower values for BEO. BEO contributed for ruminal manipulation on pre-weaning and carry over effect on post-weaning; immunity improvement and a decrease morbidity of neonatal diarrhea in pre-weaning phase.

141 box. They were then mixed with a 10 mL of milk, homogenized and incorporated in 142 0.49 L of MR (0.5 g/calf at morning meal and 0.5 g/calf at afternoon meal) to ensure 143 total ingestion of the product. After ingesting 0.5 L MR with 0.5 g of the blend of 144 essential oils, the rest of the meal was given. 145

285
The analysis of each outcome was performed independently of all others using 286 linear mixed models (package: nlme). Each independent outcome was modeled as a 287 function of the following fixed effects: treatment, experimental week, the interaction 288 between treatment and week. Birth weight and serum Brix refractometer were tested as 289 a covariate but did not improve statistical significance. Therefore they were eliminated 290 from the model. The genetic composition of the animal was included as a blocking 291 effect. The effect of heifer within treatment was included in the models to account for 292 individual variability.

293
The continuous outcomes such as intakes, performance, ruminal and blood 294 parameters were analyzed with ANOVA. A 95% Confidence Interval was adopted to 295 accept or deny the null hypothesis and P-values were produced with a Fisher test. In 296 order to meet the required assumptions of this model, all outcomes were tested for 13 297 normality, and variable transformation was applied to milk replacer intakes to meet that 298 assumption.

299
The categorical outcomes fecal and respiratory scores were analyzed using a non-300 parametric aligned rank transformation test, implemented in the R package ARTool. A 301 95% Confidence Interval was also adopted for the non-parametric tests.

302
303 Results and discussion 304 Intake and heifer performance 305 Most studies evaluating essential oils or a supplement with BEO to dairy calves, 306 feed the additive in the starter, to benefit rumen development and accelerate growth.
307 However, the intake of starter in the firsts weeks of age is small (23). Due to the calf's 308 limited capability of ingesting large solid feed amounts in the first days of life, the 309 supplement intake could be compromised during pre-weaning period and possibly mask 310 any effects. Therefore, in this trial, it was decided to offer BEO in the liquid diet.

311
The tested BEO has a strong aroma. Nevertheless, due to the way it was offered, 312 no rejection of the mixture BEO and MR was observed. The supplemented heifers 313 consumed liquid diet equally to the control treatment, with no refusal and good 314 acceptance (Table 2). Some essential oils have different acceptability by animals.  Table 2. Pre and post-weaning milk replacer (MR) intake, starter intake, total dry 322 matter intake (DM), total crude protein intake (CP), total gross energy and water 323 intake of heifers (n = 29) of control (CON) and supplemented with blend essential 324 oils (BEO) in milk replacer during pre-weaning.    (Table 3) 355 during pre and post-weaning. As was also observed for intake and ADG, a week effect 16 356 (P ≤ 0.05) was detected in all variables due to normal animal growth.  Table 4). Since there were no differences between treatments during pre-375 weaning, the carryover effect may not be assumed to be the answer for this difference.
376 Although no differences on intake were observed, heifers' ingestion behavior might 377 justify the difference in post-weaning pH. However, this parameter was not evaluated. 378 Nevertheless, previous studies with essential oils supplementation also found low pH 379 values for pre-weaning calves, but no essential oils treatment effect (16,25,27).  458 Since essential oils can increase of insulin sensitivity, not finding glucose differences 459 between treatments does not mean that there were no changes in glucose pathway.
460 Therefore, further investigations over theses aspects is needed.

461
All blood cells count were within age and specie normality. According to Benesi

515
The average age for diarrhea occurrence was 12.2 ± 3.6 d for BEO and 13.6 ± 3.8 516 d for CON with no statistical difference. Diarrhea incidence on pre-weaning in BEO 517 treatment was 85% against 93% for CON treatment with no statistical difference. Fecal 518 score was different between treatments (P ≤ 0.05), with lower values for BEO, and 519 changed through time (P ≤ 0.05, Table 7). Days with diarrhea and days with severe 520 diarrhea were not different within treatments (Table 7). Three animals of each group 521 were treated for diarrhea with anti-inflammatories, and the treatment duration was of 522 1.6 ± 0.57 d of treatment for BEO and 3.0 ± 1 d for CON. It is noteworthy that this 523 treatment was done outside the total cell count.
524 525 Table 7. Pre and post-weaning mean values of fecal score, health score, days with 526 health score above 4, days with fever, days with diarrhea, days with severe 527 diarrhea of control heifers (CON) and heifers supplemented with blend of essential 528 oils (BEO) in milk replacer during pre-weaning.