Associations of Automatically Recorded Body Condition Scores with Measures of Production, Health, and Reproduction

Question

Answers

A large number of references in the Discussion are unreasonable, I suggest revise the Discussion content.

We shortened the introduction and divided it into paragraphs according to the topics.

There are a few typos and grammar error in this manuscript.

Corrected.

Question

Answers

Introduction can be condensed.

 It appears that the overall objective to establish BCS as an indicator of Health and Pregnancy Status was not accomplished.  The paper clearly indicates associations among the traits included, but by only having 2 classes for nearly all traits, the granularity of the relationships is not sufficient to meet the claim that BCS can be used as indicators of those traits, particularly when a)the methodology was inadequately described and b)no consideration was given to the stage of lactation at which the associations were compared.

All classes have been assigned based on our previous publications and according to the data from the herd.

Not all classes were 2. BCS classes were 3 and assigned also accordint to the data of the research:  2.5–3.0 (1 ^st class ), BCS = > 3.0–3.5 (2 ^nd class), BCS = > 3.5–4.0 (3 ^ndclass). Healthy and diseased classes were also 4 classes.

In this research from all herd were investigated cows in middle stage of lactation (period 101–250 days).

There are many grammatical

errors that must be addressed

through a careful edit of the

manuscript.  Some changes are

listed below, but there are likely

others.

Corrected.

2            Title:   Suggest “Associations of Automatically Recorded Body Condition Scores with Measures of Production, Health and Reproduction”.

We corrected title to – “Associations of Automatically Recorded Body Condition Scores with Measures of Production, Health and Reproduction”.

16          Be sure that terms BCS, BHB, LDH and Mp4 do not need to be spelled out in abstract.

We corrected – “Therefore, the objective of this study was to determine the relationship of the automated registered body condition score (BCS) with pregnancy and inline biomarkers such as milk beta-hydroxybutyrate (BHB), milk lactate dehydrogenase (LDH), milk progesterone (mP4), milk yield (MY) in dairy cows”

16          Here and throughout.  Delete “registered”.  Readers will understand “automated body condition scoring”.  The term “registered” denotes something more like “certified

We deleted “registered” .

27-28    Suggest “parity” instead of “lactation”.

We corrected to –“ …parity…”

29          “…parameters,  BCS is associated with pregnancy success…”

We corrected to – “Based on our investigated parameters,  BCS is associated with pregnancy successbecause BCS of the pregnant…”.

34          What is meant by “times of insemination”?  Do you mean number of inseminations or number of days after calving or something else?  Clarify this throughout manuscript.

We corrected to –“ number of inseminations” in whole manuscript.

36          How did you measure the “activity of LDH”?  Or is it just that LDH content was greater?

We corrected to – “We found that the LDH content was greater of cows with the highest BCS > 3.5-4.0…”

49          Introduction is too long.

We shortened the introduction and divided it into paragraphs according to topics

49          Introduction should be broken into proper paragraphs according to topic being discussed.

we shortened the introduction and divided it into paragraphs according to topics

52          “estimates” instead of “measurements”.

Corrected to – “…estimates…”

54          “…autonomously and offers…”

Corrected to – “…autonomously and offers…

55          “lactating” not “lactation”

Corrected to – “…lactating…”

56          “health” not “healthy”

Corrected to – “health”

58          delete “a”

Deleted.

60-74    Condense this.  No need to focus on background of HN when the main topic of the article is BCS.

Corrected.

70          No matching close parenthesis”

We deleted this sentence.

81          “publicly available”

Corrected to – “…public available…

87          What is “a variable economics”?

We corrected to – “The main objective of commercialized dairy farming is not only to produce one calf per year but also to maintain profitability by changing inputs and output“.

88          Reword.

We corrected to – “These inputs need enormous investment in infrastructure, labor, feed, and treatments, among other things”.

97-99    Presumably if the cows had no metabolic illness, the opposite was likely true.

Corrected to – “Those cows that were overweight had at a higher risk of developing metabolic illnesses had a reduced chance of becoming pregnant during their first breeding”.

107-109              Seems to be out of place.

We deleted this sentence.

119-121              Since not in study, is this sentence needed?

We deleted this sentence.

122-133              Is this specifically about BCS for grazing conditions.  The recommendations are likely a bit different for housed herds feeding primarily TMR.

We deleted this sentence.

147        “…and concerns about animal welfare were avoided.” is very subjective.  Did the study follow an approved protocol or is there some evidence this was the case?

We corrected this sentence – “During the study, contact with animals was kept to a minimum, thus avoiding the impact of the trial on animal welfare?

151-157              It is entirely unclear here how data was collected from cows with the DeLaval AMS.  Apparently, there were 597 cows and one robot, so not all cows could have used it.  In the rest of the study, cows were milked 2x.  How was that different when the cows were milked with AMS?

It was mistake. We corrected to – “The real-time analyser Herd Navigator (Lattec I/S, Hillerd, Denmark) was applied in conjunction with a DeLaval milking parlor system to collect data on mP4, MY, BHB, and LDH (DeLaval Inc., Tumba, Sweden)”.

156-158              The info about BCS cameras belongs in next section.

Corrected.

165        If cows were milked 2x, how did they typically have 3 BCS estimates each day?

We corrected to – “As a result, each cow typically had twovisual BCS assessments taken on the same day each week, as well as fourteen automatic BCS measurements”.

168-170              Why not only report what you used instead of what is typically reported?

We corrected to – “Data from the camera system are given as a one-day BCS rolling average that removes the lowest and highest 20% of data prior to averaging, or as daily (AM) and (PM) BCS values”.

187        Not sure what you mean by “an optical milmeter (?)”.

We corrected to – “An optical milk meter”.

204        “synchronized” not ”synced”

Corrected to –“synchronized”.

208        Delete “(:)”

Deleted.

215-239              How were the thresholds for the classes of effects determined?  The numbers in each group differ widely, but may have had a great deal of impact on your conclusions.  It appears this was done simply by arithmetic mean, but why not by percentile for a more balanced number per category?  It seems use of the arithmetic mean (and admittedly for the median) is arbitrary.  Could you use values from literature to dictate cutoffs for which you would likely see differences?

As we mentioned before, the classes were assigned based on our previous publications and according to the data from the herd. Percentile method is used when you need to find the values other than the central values of the data, we believe in this research the data wasn’t large and without any extreme outliers. The raw data were adjusted using company-specified procedures to account for differences in dry stick sets and variations in surrounding humidity. The most extreme outliers were then excluded from the calculations.

Publications:

Inline Reticulorumen pH as an Indicator of Cows

Reproduction and Health Status

Ramūnas Antanaitis 1,*, Vida Juozaitiene 2, Dovile Malašauskiene 1 and Mindaugas Televicius 1. Sensors 2020, 20, 1022; doi:10.3390/s20041022

Relation of Automated Body Condition Scoring System and Inline Biomarkers (Milk Yield, β-Hydroxybutyrate, Lactate Dehydrogenase and Progesterone in Milk) with Cow’s Pregnancy Success Ramūnas Antanaitis ¹,* , Vida Juozaitienė˙² , Dovilė Malašauskienė ¹ , Mindaugas Televičius¹ , Mingaudas Urbutis¹ and Walter Baumgartner ³ . Sensors 2021, 21, 1414. https://doi.org/10.3390/s21041414

Also, nearly all of the categories you selected would be heavily influenced by stage of lactation.  How was stage of lactation accommodate in your models?  

Is not a large enough dataset to perform statistical modelling. In this research work we used: the descriptive statistics, the difference between the observed means and relationship between variables were calculated ( correlation and the backward stepwise multivariate logistic regression).

So for the stage of lactation the cows were in middle lactation and divided into two classes: 1 ^st – primiparous, 2 ^nd – multiparous and difference between the observed means were investigated, for the relationship of lactation with BCS - a correlation coefficient was calculated.

How were multiple records considered?  It is unclear how many total observations were used.  If data was collected over time, wouldn’t some cows become pregnant during the study?  So then when was pregnancy status determined for each cow?  A more complete description of the data is needed.

 We corrected materials and methods section and added information –

The real-time analyser Herd Navigator (Lattec I/S, Hillerd, Denmark) was applied in conjunction with a DeLaval milking parlor system to collect data on mP4, MY, BHB, and LDH (DeLaval Inc., Tumba, Sweden). During the robot milking operation, an inline sampler automatically took a representative sample of several millilitres of milk from each cow. The material was then loaded into the Herd Navigator^TM for further examination. 3D BCS cameras were used to measure BCS (DeLaval body condition scoring BCS, DeLaval International AB, Tumba, Sweden).These systems were used to collect daily averages of data on the following biomarkers for each cow from the day of estrus to 7 days post estrus: mP4, MY, BHB, LDH, and BCS.

We added –

Identification of health status.

Out of 850 fresh milking cows (from 1 till 30 days after calving), we randomly chose 483 clinically healthy, 21 cows with subclinical ketosis, 26 cows with subclinical mastitis and 67 cows with metritis.  

Healthy group (n=483). Cows had no clinical symptoms of disease after calving and had BHB values at or below 1.2 mmol/L for the entire 30-day post-calving period were categorized in this group. This group of cows had an average milk F/P of 1.2.

Subclinical ketosis group (SCG) (n=21). When at least one beta-hydroxybutyrate (BHB) value throughout the 30-day post-partum period was at 1.2 mmol/L, were identified as having SCK. For that particular herd of cows, the milk fat/protein ratio (F/P) was recorded as being >1.2. After calving, they showed no clinical symptoms of any additional illnesses, including metritis, lameness, mastitis, displaced abomasum, dyspepsia with an average rectal temperature of +38.8 °C, or rumen motility of five to six times every three minutes.

Subclinical mastitis group (n=26). SCC was used to identify cases that belonged to the subclinical mastitis group (CM). SCM was identified in cows with an SCC of more than 200,000 cells/mL [16]. SCC was assessed once daily during all studies. A general clinical evaluation revealed that none of the cows showed clinical indications indicative of any disease.

Metritis group (n=67). Every 3 days after calving till day +21, raginal discharge (VD) was assessed for each cow. A gloved hand was inserted into the vaginal canal up to the cervix to remove any discharge present and allow for visual inspection. Based on the scoring system used by Urton et al. [17], the appearance and smell of the VD were assessed and categorizes: putrid (red/brown color, watery, foul smelling). no mucus or clear mucus = 0, cloudy mucus or mucus with flecks of pus = 1, mucopurulent (50% pus present) and foul smelling = 2, and mucopurulent (50% pus present) and foul smelling =3  All cows were with 3 pints. 

236        “calculated using Tukey’s test” is repeated.

Deleted.

237        considered what?

Corrected to – “A probability below 0.05 was considered statistically significant”.

242        In the Table it appears the difference is “0.09” not “0.29”.  If so, the chi-square difference may be significant, but is the difference really practical if the automated BCS is calibrated according to subjectively assigned BCS which cannot be measure to the .1 level, even by trained observers.

Yes, corrected. the difference is “0.09” not “0.29”.  

Yes, we understand that, the difference is too small for the visual assessment, or automated, should we mention about it in the manuscript?

248        Or did reproductive status have a significant impact on mp4?  This is the point, what you have found are associations, not necessarily indicators, much less cause and effect!

According to our data we estimated statistically significant differences for Mp4 in pregnant and non-pregnant group of cows.

Non-pregnant 12.89; Pregnant 23.82.

The differences between means and the backward stepwise multivariate logistic regression showed a statistically significant relation of Mp4 with reproductive status of cows.

We have changed the  effect to relation.

249        Many of these are affected by stage of lactation in addition to pregnancy status, and risk of pregnancy increases with stage of lactation.  It is difficult to draw any conclusions without both being considered.

In methodology section we added missed information –

“3D BCS cameras were used to measure BCS (DeLaval body condition scoring BCS, DeLaval International AB, Tumba, Sweden). These systems were used to collect daily averages of data on the following biomarkers for each cow from the day of estrus to 7 days post estrus: mP4, MY, BHB, LDH, and BCS”

In results section -

“We corrected – “Table 1. Means and standard errors of the mean of biomarkers based on the pregnancy status of cows from the day of estrus to 7 days post estrus.”

In this research cows were investigated in middle stage of lactation  (period 101–250 days). 

We decided to check the associations between investigated traits at the same period of lactation.

277        Figure key is missing BCS 3.

We enlarged the field of legends, corrected.

315        Similarly here, did cows with higher BCS require more inseminations to become pregnant, or did more inseminations put the cow further into lactation where it could begin to increase BCS.

We corrected to – “Table 3. Means and standard errors of the mean of biomarkers registered from the day of estrus to 7 days post estrus based on the number of insemination”

Cows with higher BCS required more inseminations, also these cows where at higher risk of developing metabolic diseases with subsequent lower likelihood of becoming pregnant at first breeding.

318-320              Presumably this is a footnote to Table 3?

Corrected.

322        Reword.

Corrected to – “The BCS was statistically significantly negatively related to the milk yield, lactation (p < 0.001) and milk lactate dehydrogenase (p < 0.05). It was positively related with number of insemination (p < 0.001) and milk progesterone concentration (p < 0.05)”

326        The correlations can simply be listed in the text.  Figure 3 is not needed.

We deleted figure 3.

352        Because the data was not adequately described, it is difficult to know when this difference of (again 0.09 not 0.29) between pregnant and non-pregnant cows was measured.  The difference is practically very small and it is unclear at what point it can be observed.  Is it soon enough after insemination to be of use in decision-making? 

We corrected to –“ According to our results the BCS registered from the day of estrus to 7 days post estrus of the pregnant cows was higher (+0.09 score) compared to the group of non-pregnant cows, also mP4 in pregnant cows was higher (10.93 ng/mL) the MY was lower (−5.26 kg), (p< 0.001) and the LDH was lower (3.45 µmol/min) compared to the group of non-pregnant cows (p < 0.01)”.

Also, we corrected conclusion section –

According to the aim of our study—to determine the associations of automatically recorded body condition scores with measures of production, health, and reproduction (MY, BHB, LDH, and mP4) in dairy cows we found that automated registered BCS can be an indicator of pregnancy success because BCS of the pregnant cows had a higher (+0.29 score) as well as mP4 was (10.93 ng/mL) higher compared to the group of non-pregnant cows at during the insemination. Number of insemination of cows with the highest BCS > 3.5-4.0 were with 42.41% higher compared to the cows with the lowest BCS 2.5-3.0.

Automatically recorded BCS in cows with subclinical mastitis was higher 4.96% compared to the group of healthy cows. The BCS was the highest in a group of cows with mastitis, it was higher 4.96% compared to the group of healthy cows, also the highest statistically significant mean differences of body condition score (9.04%) were estimated between mastitis and metritis groups of cows.

The difference is “0.09” not “0.29”.  

yes, we understand that, the difference is too small for the visual assessment, or automated, should we mention about it in the manuscript?

373        “Breastfeeding”?  Roche’s research was on dairy cows not humans.

Corrected to – “According to Roche et al. [25], the majority of studies on the physiological effects of energy status and energy balance on fertility revealed a positive link between earlier pregnancy attainment and enhanced BCS and reduced BCS loss during early lactation”

398        “has” not “have”

Corrected to “has”.

427        “score may be associated with health…”

We corrected to – “score may be associated with health..”