Previous Lactation Risk Factors Associated with Hyperketonemia in the First Week Postpartum in Dairy Cows: A Retrospective Analysis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

General Comments:

This article reports about the relevance of previous lactation risk factors to the occurrence of hyperketonemia. It is interesting because the study supplies many effective data for feeding management. However, authors should add the details in some items as the followings.

ABSTRACT, INTRODUCTION AND MATERIALS AND METHODS

1. Why not diagnosed subclinical ketosis instead of hyperketonemia? Have they not checked for clinical symptoms at the time of blood sampling?

MATERIALS AND METHODS

1. The sizes of 7 commercial farms should be described (for example, the number of cows).
2. L73: “no ethical approval” is described. What kind of purpose was the sampling carried out?
3. Was sampling time standardized across farms?
4. Ketone body measurements are said to be more reliable from 3 days after calving onwards. Why did the authors include measurement results from 1 day after delivery?
5. L83: “with [7] and [8]”. The names of authors (…… et al) and so on should be described before reference no.

RESULTS

1. Table 2 should be placed before the figures (first in the results).
2. The notes state that the parentheses in Table 2 indicate SD, but practically they are expressed as %.
3. In Table 2, the total percentages for some items are 100.1% or 99.9%, so there are some places where they do not add up to 100%. Please check it out.
4. The estimates in Figure 1 do not match those in the text. What do the estimates indicate?

Author Response

Reviewer: 1
General comment
This article reports about the relevance of previous lactation risk factors to the occurrence of hyperketonemia. It is interesting because the study supplies many effective data for feeding management. However, authors should add the details in some items as the followings.
Thank you for your comments.
ABSTRACT, INTRODUCTION AND MATERIALS AND METHODS
Comment 1: Why not diagnosed subclinical ketosis instead of hyperketonemia? Have they not checked for clinical symptoms at the time of blood sampling?
Response 1: Thanks for your question. We should clarify that subclinical ketosis and hyperketonemia refer to the same condition—elevated blood BHB concentrations (>1.2 mmol/L) in the absence of clinical signs. Indeed, we did not assess clinical signs during sampling, as our focus was specifically on cows without clinical signs. For this reason, we used the term hyperketonemia.
MATERIALS AND METHOD
Comment 1: The sizes of 7 commercial farms should be described (for example, the number of cows).
Response 1: Thanks for this suggestion. A descriptive table was added to the text.
Comment 2: L73: “no ethical approval” is described. What kind of purpose was the sampling carried out?
Response 2: The samples were collected as part of different research projects aimed at describing the prevalence of hyperketonemia during the first and second weeks postpartum, as well as studies focused on the development and validation of algorithms to predict hyperketonemia in the first week postpartum. We have added this information to the text.
Comment 3: Was sampling time standardized across farms?
Response 3: Yes, all blood samples were collected during the first week postpartum. Cows were sampled in the morning after milking. This information has been added to the text.
Comment 4: Ketone body measurements are said to be more reliable from 3 days after calving onwards. Why did the authors include measurement results from 1 day after delivery?
Response 4: We collected samples once per week during the first 7 days postpartum. Most of the current literature lacks data before 3 DIM, as research teams often choose not to sample cows this early. However, this strategy does not imply that hyperketonemia is absent immediately after calving. It is important to highlight that no cows at 1DIM was enrolled in the study.
2
Comment 5: L83: “with [7] and [8]”. The names of authors (…… et al) and so on should be described before reference no
Response 5: Done.
RESULTS
Comment 1: Table 2 should be placed before the figures (first in the results).
Response 1: Thanks for this suggestion
Comment 2: The notes state that the parentheses in Table 2 indicate SD, but practically they are expressed as %.
Response 2: Thanks for catching this. You are correct—Table 2 is expressed as percentages. We have removed the SD statements from the table footer.
Comment 3: In Table 2, the total percentages for some items are 100.1% or 99.9%, so there are some places where they do not add up to 100%. Please check it out.
Response 3: We edited the total percentages in these items and highlighted them for clarification.
Comment 4: The estimates in Figure 1 do not match those in the text. What do the estimates indicate?
Response 4: The estimates reflect associations between the predictors and log-transformed BHB concentrations. To improve interpretability in the text, we exponentiated these estimates to convert them back to the original scale and expressed the results as percentage changes in BHB concentration. This clarification was added to the Materials and Methods section under Statistical Analysis (Lines 133–135) to avoid confusion.

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Colleagues,

       thank you for this study about HYK. I have the folloing suggestions and recommendation:

1. HYK is only possible at a NEB and therefore, HYK is - as you have shown - a metabolic diseases related to milk production. It ist known since the the study of Buttchereit et al. (2011; J. Dairy Sci. 94, 1585-1591) that milk yield is negatively correlated  with the negative energy balance. It means the degree of NEB is related to milk production. It further means that selection for more milk increases the risk of HYK. Oikonomou et al. (2008) have demonstrated a relatively high heritability of BHB in early lactation which again means that HYK has a genetic background. This problem should be addressed (J. Dairy Sci. 91, 4323-4332). Furthermore, Semianer et al. (1991) have published a genetic correlation between MY and ketosis (J. Dairy Sci. 74, 5348-62). 
2.  Blood  samples were taken from day 1 - 7 postpartum.  This is a wide range and probably include immediately postpartum cows with later HYK+ values. Please mention this problem.
3. Contradictory data about HYK+ very often exist. If that is the case please look at the time of blood sampling. 

Author Response

Reviewer: 2
General comments
Thank you for this study about HYK. I have the following suggestions and recommendation:
HYK is only possible at a NEB and therefore, HYK is - as you have shown - a metabolic diseases related to milk production. It is known since the study of Buttchereit et al. (2011; J. Dairy Sci. 94, 1585-1591) that milk yield is negatively correlated with the negative energy balance. It means the degree of NEB is related to milk production. It further means that selection for more milk increases the risk of HYK. Oikonomou et al. (2008) have demonstrated a relatively high heritability of BHB in early lactation which again means that HYK has a genetic background. This problem should be addressed (J. Dairy Sci. 91, 4323-4332). Furthermore, Semianer et al. (1991) have published a genetic correlation between MY and ketosis (J. Dairy Sci. 74, 5348-62).
Response: Thank you for this valuable input. We have incorporated information addressing the genetic background of hyperketonemia, including its association with negative energy balance, milk yield, and heritability of BHB concentrations.
Blood samples were taken from day 1 - 7 postpartum. This is a wide range and probably include immediately postpartum cows with later HYK+ values. Please mention this problem. Contradictory data about HYK+ very often exist. If that is the case, please look at the time of blood sampling.
Response: The wide range of blood sampling days observed in this study is due to our sampling strategy, which involved visiting each farm once per week. This approach allowed us to include a large number of cows but inevitably resulted in sampling animals at different DIM within their first week of lactation. Importantly, this strategy reflected the farm management practices at the time of the study and was intentionally chosen to capture a realistic cross-section of early lactation cows. Additionally, most existing studies do not include data before 3 DIM, as research teams often avoid sampling cows too soon after calving. For clarification, we have acknowledged this sampling approach and its implications as a study limitation.

Reviewer 3 Report

Comments and Suggestions for Authors

Since authors had blood samples, a suggestion to add sth really new would be the measurement of NEFA at first week postpartum.

Also the herd details and comparison could reveal interesting results and add significant scientific value to this manuscript. Herd A with 53 cows has very favourable production and health indices in comparison with e.g. herd G (700 cows).

Author Response

Reviewer 3:
Comment 1: Since authors had blood samples, a suggestion to add the really new would be the measurement of NEFA at first week postpartum.
Response 1: We appreciate this suggestion and fully agree that the addition of NEFA—and potentially inflammatory markers—would have significantly enhanced the depth and quality of the dataset. In fact, our original study design included plans to measure NEFA. However, due to budgetary constraints, we were unable to pursue this analysis, as the cost of processing over 2,300 samples would have exceeded our available resources.
Nonetheless, we believe that the use of BHB, which can be measured reliably and inexpensively on-farm using hand-held meters, provides valuable and practical insights for herd-level management. Importantly, our focus on parameters that can be obtained through routine on-farm data collection ensures that the findings can be directly translated into management practices without adding financial burden to producers.
Comment 2: Also, the herd details and comparison could reveal interesting results and add significant scientific value to this manuscript. Herd A with 53 cows has very favorable production and health indices in comparison with e.g., herd G (700 cows).
Response 2: Thank you for this insightful comment. We agree that differences in herd-level characteristics, such as size, productivity, and metabolic health, are of interest and could enrich the interpretation of our findings. While a detailed comparative analysis between herds was beyond the scope of this study, we have now incorporated a paragraph in the Discussion that highlights these differences and explores potential interpretations. Additionally, herd was included in our models as a random intercept to account for underlying differences in management, genetics, and production levels. We appreciate your suggestion to bring more attention to these herd-level factors.

Reviewer 4 Report

Comments and Suggestions for Authors

Independent review

Article type: research article.

Title: Previous lactation risk factors associated with hyperketonemia 2 in the first week postpartum in dairy cows: a retrospective analysis

The study conducted by Mahmoud H. Emam et al. evaluated the risk factors associated with elevated beta-hydroxybutyrate (BHB) concentrations and the incidence of hyperketonemia during the first week postpartum in high-yielding dairy cows. The research highlights two significant risk factors: an extended dry period exceeding 60 days and a previous lactation with more than 130 days open, both of which were found to correlate strongly with increased BHB levels and the emergence of hyperketonemia in these dairy cows. Furthermore, the study revealed a quantifiable relationship between milk production and the risk of postpartum hyperketonemia, indicating that for each additional 1000 kg of milk produced in the previous lactation, there is an increased likelihood of developing this metabolic disorder after calving. These findings underscore the importance of managing dry periods and monitoring lactation length to mitigate the risks associated with hyperketonemia, ensuring the health and productivity of high-yielding dairy herds.

The study seems to have strong evidence for suporting the conclusions.

However my concerns are:

Comment Materials and Methods: It is difficult to accept that there are no differences in the management practices across the seven farms. Each farm operates in unique conditions that may significantly influence its management strategies. Consequently, I believe this section should delve deeper into these specifics, examining how each farm's unique circumstances shape its operational decisions and management practices. This would provide a more comprehensive understanding of the factors at play in each farm.

Line 83: Please use the names of the authors in this context in association with the citation number;

In Herd D, the average production is 9,770 kg of milk over a 305-day period. This figure suggests that the farm operates at a medium level of productivity compared to industry standards for high-yielding dairy cows, as the other 6 farms fit in. This factor may significantly influence the results of this study.

Line 89: Could you clarify why the univariate analysis was not included in the statistical evaluation to identify a potential association (P<0.25) between BHB, HKT, and the dependent variable from the previous lactation? This analysis is important for understanding the relationships between these variables and may warrant further exploration in a multivariate analysis. Maybe you could include it in your study along with the Multivariable linear mixed model.

Please explain your reasoning.

Suggestions:

Line 134: Descriptive results can be moved to the Materials and Methods section.

Line 248: Table 2 can be moved to the Materials and Methods section.

Line 256: standard deviation

Line 274: periods

Line 281: the periparturient

Line 286: an excessively

Line 311: the high concentration

Line 319: please use mostly that largely

Line 322: et al

Line 350: of more

 

Author Response

Reviewer 4
General comment 1: The study conducted by Mahmoud H. Emam et al. evaluated the risk factors associated with elevated beta-hydroxybutyrate (BHB) concentrations and the incidence of hyperketonemia during the first week postpartum in high-yielding dairy cows. The research highlights two significant risk factors: an extended dry period exceeding 60 days and a previous lactation with more than 130 days open, both of which were found to correlate strongly with increased BHB levels and the emergence of hyperketonemia in these dairy cows. Furthermore, the study revealed a quantifiable relationship between milk production and the risk of postpartum hyperketonemia, indicating that for each additional 1000 kg of milk produced in the previous lactation, there is an increased likelihood of developing this metabolic disorder after calving. These findings underscore the importance of managing dry periods and monitoring lactation length to mitigate the risks associated with hyperketonemia, ensuring the health and productivity of high-yielding dairy herds. The study seems to have strong evidence for supporting the conclusions.
4
Response: Thank you for your kind words about our work. We appreciate your time and effort in helping us improve the manuscript.
General comment 2: It is difficult to accept that there are no differences in the management practices across the seven farms. Each farm operates in unique conditions that may significantly influence its management strategies. Consequently, I believe this section should delve deeper into these specifics, examining how each farm's unique circumstances shape its operational decisions and management practices. This would provide a more comprehensive understanding of the factors at play in each farm.
Response: Farm was introduced into our models as a random intercept to avoid any variation in the results. however, we added differences in the management practices across the seven farms as a study limitations.
Specific comments
Comment 1: Line 83: Please use the names of the authors in this context in association with the citation number
Response 1: Done and highlighted.
Comment 2: In Herd D, the average production is 9,770 kg of milk over a 305-day period. This figure suggests that the farm operates at a medium level of productivity compared to industry standards for high-yielding dairy cows, as the other 6 farms fit in. This factor may significantly influence the results of this study.
Response 2: Thank you for raising this important point. We agree that differences in herd productivity—such as those noted for Herd D—can influence metabolic outcomes and warrant consideration. To account for such variability, we included herd as a random intercept in our multivariable models. This approach adjusts for unmeasured, herd-specific factors (e.g., management practices, genetics, environment) and enables us to focus on fixed effects of interest across the dataset. Additionally, we have expanded the Discussion section (Lines 354–367) to acknowledge the observed differences in herd-level performance and discuss their potential implications. We appreciate your suggestion to highlight this more explicitly in the manuscript.
Comment 3: Line 89: Could you clarify why the univariate analysis was not included in the statistical evaluation to identify a potential association (P<0.25) between BHB, HKT, and the dependent variable from the previous lactation? This analysis is important for understanding the relationships between these variables and may warrant further exploration in a multivariate analysis. Maybe you could include it in your study along with the Multivariable linear mixed model.
Response 3: Thank you for this suggestion. We agree that univariable analyses can offer valuable preliminary insights into the relationships among variables. In fact, we did conduct univariable analyses prior to building the multivariable models, and all predictors included in the final models were selected based on their significant associations in those preliminary analyses (P < 0.25). This approach is already described in the Statistical Analysis section, and we have now highlighted it for clarity.
Additionally, we believe that the relationships of interest are adequately represented through our descriptive tables and multivariable model outputs, which effectively summarize key patterns in the data. While univariable results were not shown separately, they informed the structure of our multivariable analyses and model selection.
Comment 4: Line 134: Descriptive results can be moved to the Materials and Methods section and Line 248: Table 2 can be moved to the Materials and Methods section.
5
Response 4: Thanks for your suggestion but according to the journal policy all results and tables should be in the results section not in Materials and Methods. Thanks for your understanding
Suggestions
1. Line 256: standard deviation
Edited (line 168).
2. Line 274: periods
Done (line 274)
3. Line 281: the periparturient
Done (line 280)
4. Line 286: an excessively
Done (line 285)
5. Line 311: the high concentration
Done (line 310)
6. Line 319: please use mostly that largely
Done (line 318)
7. Line 322: et al
Done (line 221)
8. Line 350: of more
Done (line 349)

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The publication has significantly been improved, but I am still missing a perspective for the future. What should be done? An increase of milk is obviously a risk factor for HYK. 

Author Response

Please see the attachment for details.

Author Response File: Author Response.pdf