Application of the HOMA Index in Diabetic Dogs and Cats: A Systematic Review of Current Evidence

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The HOMA(Homeostasis Model Assessment) index, as a non-invasive mathematical tool for evaluating insulin sensitivity and pancreatic β -cell function, was originally developed for human diabetes medicine. In modern veterinary medicine, finding an efficient and easy-to-use method for assessing metabolic status has become an urgent need. Non-invasive operation reduces the stress of animals and the physiological changes caused by complex operations. It is used in clinical veterinary diagnosis and monitoring of pet animals to assess diabetes in pet animals (especially dogs and cats). It has many advantages such as simple method, low cost, early prediction and accurate and reliable diagnosis, and has a good application prospect. The application of the HOMA index in the fields of veterinary endocrinology and diabetes represents a significant advancement, offering a simple, reliable and efficient method for assessing metabolic status. It can detect metabolic disorders in pet animals at an early stage (before clinical symptoms appear), enhance the level of early diagnosis, guide the optimization of personalized treatment plans, and improve the management of pet diabetes. To improve the prognosis and quality of life of dogs and cats with metabolic diseases, the development of human diabetes HOMA index is based on the principle that there is a dynamic feedback relationship between insulin secretion and insulin sensitivity under fasting equilibrium conditions The two parameters, HOMA-IR (used to quantify the degree of insulin resistance) and HOMA-β (used to estimate the insulin secretion function of pancreatic β cells), can be calculated simply by substituting the plasma glucose concentration and insulin concentration measured under fasting conditions into two simple mathematical formulas (applicable to humans and companion animals - dogs and cats). There are some significant limitations in its application in veterinary medicine, such as the influence of species specificity (age, weight, basal metabolic rate, circadian rhythm of hormone secretion, tissue insulin sensitivity) and external factors (stress and comorbidities) on blood glucose and insulin values, as well as the lack of standardized, specific-specific reference values (thresholds). These factors highlight the improvement of methodology and the necessity of developing confirmatory protocols tailored to the unique physiological characteristics of dogs and cats. Despite these challenges, HOMA still represents a promising approach to advancing the understanding of the pathophysiology of diabetes in veterinary patients. Future longitudinal studies and controlled trials are crucial for confirming its reliability and enhancing its clinical relevance.

Areas that need improvement

In Figure 2, why are there only two ICONS for the four curves and two 200s on the vertical coordinate? What kind of "-cell" is behind "healthy"? This sentence "healthy?" "-cell function" should be at the bottom of the left graph, right?

2. What does HOMA-B (%B) refer to in Figure 3?

Figure 4 is incomplete (the upper part of the figure lacks the effect on the insulin secretion function of pancreatic β cells) and repeats with Table 2.

4. In Figure 7, three 1000s appear on the vertical axis, and the percentages on the horizontal axis indicate nothing.
5. The result section is missing 3.6 and has directly gone from 3.5 to 3.7. 3.1.-3.8. Should it be 3.1-3.8?

 

 

Author Response

1. In Figure 2, why are there only two ICONS for the four curves and two 200s on the vertical coordinate? What kind of "-cell" is behind "healthy"? This sentence "healthy?" "-cell function" should be at the bottom of the left graph, right?

Figure 2
Figure 2 became figure 3 and I completely redid it because there were some errors in the image. I also added a legend

2. What does HOMA-B (%B) refer to in Figure 3?

Figure 3 became figure 4 and I did it entirely because there was an error in the image.

3. Figure 4 is incomplete (the upper part of the figure lacks the effect on the insulin secretion function of pancreatic β cells) and repeats with Table 2.

Figure 4 became figure 5 but I deleted it because it was the same as table 2

4. In Figure 7, three 1000s appear on the vertical axis, and the percentages on the horizontal axis indicat nothing.

I deleted figure 7 because it wasn't very suggestive.

4. The result section is missing 3.6 and has directly gone from 3.5 to 3.7. 3.1.-3.8. Should it be 3.1-3.8?

I solved this problem, it was a numbering error

Reviewer 2 Report

Comments and Suggestions for Authors

A systematic review follows very specific rules. The databases from which the papers were taken must be indicated and the prisma scheme must be applied (https://www.prisma-statement.org/).

The title should also be changed to clearly indicate that it is a systematic review

Author Response

A systematic review follows very specific rules. The databases from which the papers were taken must be indicated and the prisma scheme must be applied (https://www.prisma-statement.org/).

The title should also be changed to clearly indicate that it is a systematic review

We solved these problems, changed the title and rewrote the text applying the PRISMA schema.

Reviewer 3 Report

Comments and Suggestions for Authors

This is an interesting paper on the use of HOMA in companion animals. Thank you for the opportunity to review it. Unfortunately, I do not think it should be published in the present form because there are several limitations.

General comments

First of all, the HOMA models were designed and validated for humans, and is application for other species violates the model´s assumptions, as discussed in a nice review by Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 27:1487–1495, 2004.

The use of HOMA index in veterinary medicine is very controversial due this. Despite being easy to measure there´s no standardized insulin immunoassay for performing this indexes in dogs or cats, and insulin values are not comparable among different methods, making any index using insulin hard to widely trust.

Second and not less important – the misclassification of diabetes in dogs and cats used in the manuscripts creates a immediate need for reviewing the entire text using proper veterinary diabetes classification proposed by the ALIVE Project (https://doi.org/10.1016/j.tvjl.2022.105910).

There are SEVERAL references that seemed not linked with the text, especially in the introduction. It would be advisable to carefully review if the references listed at the end are the same intended over the text indicated by numbers. I´ll emphasize two examples below but there are really SEVERAL others.

For example, refs 23-25 have apparently no correlaction with the text:

“Furthermore, the evaluation of counter-regulatory hormones, such as glucagon and cortisol, is vital to identify endocrine comorbidities, such as hyperadrenocorticism, which can exacerbate insulin resistance and 96 complicate diabetes management, requiring a specific therapeutic approach [23,25].”

23 . Del Prato, S. (2003). Loss of early insulin secretion leads to postprandial hyperglycemia. Diabetologie, 46(Suppl 1), M2-M8.

24 . Durrer, C., Islam, H., Cen, H. H., Moya Garzon, M. D., Lyu, X., McKelvey, S., Singer, J., Batterham, A. M., Long, J. Z., Johnson, J. D., & Little, J. P. (2024). A secondary analysis of indices of hepatic and beta cell function following 12 weeks of carbohydrate and energy restriction vs. free-living control in adults with type 2 diabetes. Journal of Translational Medicine, 22(1), 125.

25. Ellis, B. W., Randall, N. J., Becket, A. J., & Dudley, H. A. F. (1976). Continuous blood sampling and time series analysis. Journal of Medical Engineering & Technology, 2, 195–199.â

Other example:

“In the study of diabetes mellitus (DM) in companion animals, the use of the HOMA (Homeostasis Model Assessment) index represents an essential tool for evaluating insulin sensitivity and pancreatic β-cell function [35,36,37].

 

The reference 35 is a human paper, references 36 and 37 do not even mention HOMA index within the text.

 

The figures 1, 2, 3 and 7 look like made using AI without proper critical review of their content. Figure 3 is with a black background turning it difficult to read, and it seems the information within are inacurade since there´s HOMA B two times.   

 

Table 2 and figure 4 were created based on references that not even mentioned HOMA indexes.

 

At least three papers applying HOMA indexes in dogs and cats published in veterinary literature would have been added to this review:

Leal et al., 2022 - Is Methylprednisolone Acetate-Related Insulin Resistance Preventable in Cats?

Pöppl et al., 2021 - Pyometra-associated insulin resistance assessment by insulin binding assay and tyrosine kinase activity evaluation in canine muscle tissue. 

Pöppl et al., 2009 - Índices de sensibilidade à insulina em fêmeas caninas: efeito do ciclo estral e da piometra (Paper in Portuguese, but with main results reviewed in Pöppl et al., 2024 - Progesterone-Related Diabetes Mellitus in the Bitch: Current Knowledge, the Role of Pyometra, and Relevance in Practice.

 

Specific Coments

Line 22 – Since the models are used to assess insulin resistance and beta cell activity by means of comparing fasting glucose and insulin values, it is potentially unapplicable to any dog or cat on insulin therapy, and in this way have no role in diabetes treatment monitoring.

Key-words: please do not repeat here words already present in the title.

Introduction

The concept of applying the human classification of type-1 for dogs, and type-2 for cats is now outdated, and the preferred terminology refers to insulin-dependent diabetes or insulin-resistant diabetes as proposed by the ESVE ALIVE Project. Please review your introduction in light of the actual classification of diabetes in dogs and cats. Actually, many dogs are insulin-resistant, there´s a lack of evidence for autoimmunity in the majority of diabetic dogs, and for cats, several other causes of insulin resistance and eventual insulin-deficiency are well characterized.

Line 90 -  There´s no commercial kit to measure c-peptide in veterinary medicine, and despite a potential role to distinguish residual insulin secretion ability it is not use to determine therapy, at least in dogs and cats. The same for glucagon. No veterinarian in the world is able to measure glucagon in the clinical practice outside of a research environment due absence of commercial kits for glucagon.

Line 94 – Simple measures of cortisol usually say nothing about hypercortisolism risk (not hyperadrenocorticism – ALIVE terminology) and are not recommended to be measured in any newly diagnosed dog or cat; except in the presence of clinical signs of Cushing´s syndrome, where a dynamic test will be recommended.

Line 119-125 – Dogs and cats are treated most of the times with exogenous insulin and the reliability of any insulin measurement in this scenario is low due to cross-reaction between exogenous and endogenous insulin in the immunoassays.

Line 141 – How was the validity of the insulin kits used in the veterinary studies Validated? There´re several kits on the market for insulin measurement, most have been poorly validated for use in dogs and cats.

Line 147 – What are the necessary adaptations within veterinary context mentioned  in the sentence “For data analysis, the calculation methods of the HOMA index used in various studies were compared, emphasizing the necessary adaptations within the veterinary context”?

Line 155 – The sentence affirms that “the results contribute to optimizing disease management strategies”. Which disease? Maybe is still to early to assume such importance for the results, since the MMs are being described at this point. “The results may contribute to…” would read better.

Line 162 - I´m not aware about any paper in veterinary medicine that considers HOMA indexes “essential tool for evaluating insulin. sensitivity and pancreatic β-cell function”

Line 169 – Accuracy of HOMA indexes are very questionable due to lack of standardization in dogs, despite being better studied in cats, the main author in this Subject (Apleton et al), recognizes the simple measuring of insulin and glucose (fasting insulin index) is to date, the best option to access insulin resistance and beta cell function in the cat.

 

Author Response

General comments

First of all, the HOMA models were designed and validated for humans, and is application for other species violates the model´s assumptions, as discussed in a nice review by Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 27:1487–1495, 2004.

The use of HOMA index in veterinary medicine is very controversial due this. Despite being easy to measure there´s no standardized insulin immunoassay for performing this indexes in dogs or cats, and insulin values are not comparable among different methods, making any index using insulin hard to widely trust.

We thank the reviewer for this important observation and for pointing out the work of Wallace et al., which indeed provides an excellent discussion of the assumptions and limitations of HOMA modeling. We fully agree that HOMA models were originally developed and validated for human studies and that their direct application to other species should be interpreted with caution because of interspecies physiological differences. However, several veterinary studies have explored the potential for using HOMA-derived indices in dogs and cats as surrogate markers for insulin sensitivity and β-cell function, particularly in the absence of more invasive or complex methods such as euglycemic clamp. These include - Ramos J.R. and Castillo V, 2020,  Sarıkaya E.E.,  Gökce H.I. 2024, Strage E.M. et al 2021- who adapted the model to species-specific reference ranges and validated it against clinical or metabolic outcomes in companion animals. In our study, we took these limitations into account by  - for example, using established fasting glucose/insulin reference ranges for dogs/cats, comparing results with clinical signs, etc. Furthermore, there are studies in the veterinary literature that have explored the use of HOMA-IR in cats and dogs. For example, in overweight cats, HOMA-IR and insulin were significantly higher compared to lean cats, and HOMA-IR was positively correlated with body fat percentage (Strage et al., 2021).Furthermore, in obese dogs, a study demonstrated that all dogs with HOMA-IR > 2.4 can be considered insulin-resistant (Ferreira et al., 2022). These data suggest that although HOMA was initially validated for humans, it provides a useful functional measure in practice.

Second and not less important – the misclassification of diabetes in dogs and cats used in the manuscripts creates a immediate need for reviewing the entire text using proper veterinary diabetes classification proposed by the ALIVE Project (https://doi.org/10.1016/j.tvjl.2022.105910).

Response to Reviewer:

We thank the reviewer for this important and constructive comment. We fully agree that using the proper veterinary diabetes classification is crucial to ensure accuracy and avoid misinterpretation. Following the reviewer’s recommendation, we carefully revised the entire manuscript to align with the classification proposed by the ALIVE Project (O’Kell et al., 2022, The Veterinary Journal, https://doi.org/10.1016/j.tvjl.2022.105910).

In the revised version, all instances of diabetes classification in dogs and cats were corrected accordingly:

• Dogs are now consistently referred to as predominantly presenting insulin-deficient diabetes (IDD), resulting from β-cell destruction and requiring lifelong insulin therapy.

• Cats are described as predominantly affected by insulin-resistant diabetes (IRD), usually associated with concurrent β-cell dysfunction, obesity, chronic inflammation, and glucolipotoxicity.

This revision was applied systematically throughout the manuscript, including the Introduction, Results and Discussion (Sections 3.1, 3.2, 3.5, and 3.7), and Conclusions. The terminology has been harmonized, and interpretations of HOMA-IR and HOMA-β values were adjusted to reflect this species-specific classification.

We believe that these modifications significantly improve the clarity, accuracy, and scientific robustness of the manuscript, ensuring full alignment with current veterinary endocrinology standards.

There are SEVERAL references that seemed not linked with the text, especially in the introduction. It would be advisable to carefully review if the references listed at the end are the same intended over the text indicated by numbers. I´ll emphasize two examples below but there are really SEVERAL others.

For example, refs 23-25 have apparently no correlaction with the text:

“Furthermore, the evaluation of counter-regulatory hormones, such as glucagon and cortisol, is vital to identify endocrine comorbidities, such as hyperadrenocorticism, which can exacerbate insulin resistance and 96 complicate diabetes management, requiring a specific therapeutic approach [23,25].”

23 . Del Prato, S. (2003). Loss of early insulin secretion leads to postprandial hyperglycemia. Diabetologie, 46(Suppl 1), M2-M8.

24 . Durrer, C., Islam, H., Cen, H. H., Moya Garzon, M. D., Lyu, X., McKelvey, S., Singer, J., Batterham, A. M., Long, J. Z., Johnson, J. D., & Little, J. P. (2024). A secondary analysis of indices of hepatic and beta cell function following 12 weeks of carbohydrate and energy restriction vs. free-living control in adults with type 2 diabetes. Journal of Translational Medicine, 22(1), 125.

25. Ellis, B. W., Randall, N. J., Becket, A. J., & Dudley, H. A. F. (1976). Continuous blood sampling and time series analysis. Journal of Medical Engineering & Technology, 2, 195–199.â

Other example:

“In the study of diabetes mellitus (DM) in companion animals, the use of the HOMA (Homeostasis Model Assessment) index represents an essential tool for evaluating insulin sensitivity and pancreatic β-cell function [35,36,37].

The reference 35 is a human paper, references 36 and 37 do not even mention HOMA index within the text.

Response to Reviewer:

We thank the reviewer for highlighting these specific examples and for pointing out the need for consistency between the text and the reference list. Following this recommendation:

• The examples cited by the reviewer (refs. 23–25 and 35–37) have been carefully revised. The mismatched or irrelevant references have been corrected and replaced with appropriate sources directly supporting the statements in the text.

• We then extended this verification process to the entire manuscript, ensuring that every in-text citation is accurately matched with the intended reference and that the numbering sequence is fully consistent.

We are confident that all discrepancies have now been resolved, and the revised manuscript reflects a complete alignment between the in-text citations and the reference list.

The figures 1, 2, 3 and 7 look like made using AI without proper critical review of their content. Figure 3 is with a black background turning it difficult to read, and it seems the information within are inacurade since there´s HOMA B two times.   

Table 2 and figure 4 were created based on references that not even mentioned HOMA indexes.

Response to Reviewer:

We thank the reviewer for this valuable comment. In the revised version, all figures in question (Figures 1, 2, 3, and 7) have been completely re-designed with improved clarity, accuracy, and readability. Figures that were considered not sufficiently conclusive have been removed. In addition, Table 2 and Figure 4 have been entirely re-created using only relevant references that explicitly discuss HOMA indices. We are confident that these changes have substantially improved the scientific accuracy and overall quality of the manuscript.

At least three papers applying HOMA indexes in dogs and cats published in veterinary literature would have been added to this review:

Leal et al., 2022 - Is Methylprednisolone Acetate-Related Insulin Resistance Preventable in Cats?

Pöppl et al., 2021 - Pyometra-associated insulin resistance assessment by insulin binding assay and tyrosine kinase activity evaluation in canine muscle tissue. 

Pöppl et al., 2009 - Índices de sensibilidade à insulina em fêmeas caninas: efeito do ciclo estral e da piometra (Paper in Portuguese, but with main results reviewed in Pöppl et al., 2024 - Progesterone-Related Diabetes Mellitus in the Bitch: Current Knowledge, the Role of Pyometra, and Relevance in Practice.

We thank the reviewer for these very relevant suggestions. In the revised manuscript, the recommended studies have been incorporated and discussed to strengthen the review:

• Leal et al. (2022) – highlighting steroid-induced insulin resistance in cats, now cited in Sections 3.1 and 3.7.

• Pöppl et al. (2021) – addressing pyometra-associated insulin resistance in dogs, integrated into the discussion of endocrine comorbidities.

• Pöppl et al. (2009/2024) – describing insulin sensitivity indices in bitches and the role of pyometra, also included in the comparative analysis.

Their addition provides a broader and more comprehensive coverage of the veterinary literature on HOMA indices in dogs and cats.

Specific Coments

Line 22 – Since the models are used to assess insulin resistance and beta cell activity by means of comparing fasting glucose and insulin values, it is potentially unapplicable to any dog or cat on insulin therapy, and in this way have no role in diabetes treatment monitoring.

Response: The sentence has been revised to clearly state that HOMA indices are not applicable in animals receiving exogenous insulin therapy and therefore cannot be used for treatment monitoring in such cases.

Reviewer’s comment: please do not repeat here words already present in the title.

Response: The keywords section has been revised. Repetitive terms were removed and replaced with additional relevant terms not included in the title.

Reviewer’s comment: Introduction

The concept of applying the human classification of type-1 for dogs, and type-2 for cats is now outdated, and the preferred terminology refers to insulin-dependent diabetes or insulin-resistant diabetes as proposed by the ESVE ALIVE Project. Please review your introduction in light of the actual classification of diabetes in dogs and cats. Actually, many dogs are insulin-resistant, there´s a lack of evidence for autoimmunity in the majority of diabetic dogs, and for cats, several other causes of insulin resistance and eventual insulin-deficiency are well characterized.

Response: The introduction was completely revised according to the ESVE ALIVE Project classification. The text now refers to insulin-deficient diabetes (IDD) in dogs and insulin-resistant diabetes (IRD) in cats, avoiding direct extrapolation from human type 1/type 2 diabetes. Additional causes of insulin resistance in dogs and cats were also included.

Reviewer’s comment: Line 90 -  There´s no commercial kit to measure c-peptide in veterinary medicine, and despite a potential role to distinguish residual insulin secretion ability it is not use to determine therapy, at least in dogs and cats. The same for glucagon. No veterinarian in the world is able to measure glucagon in the clinical practice outside of a research environment due absence of commercial kits for glucagon.

Response: Following the reviewer’s observation, all mentions of C-peptide and glucagon have been removed from the manuscript, since these markers are not applicable in veterinary clinical practice due to the absence of validated commercial kits.

Reviewer’s comment: Line 94 – Simple measures of cortisol usually say nothing about hypercortisolism risk (not hyperadrenocorticism – ALIVE terminology) and are not recommended to be measured in any newly diagnosed dog or cat; except in the presence of clinical signs of Cushing´s syndrome, where a dynamic test will be recommended.

Response: The text has been corrected. The revised version specifies that basal cortisol is not recommended in newly diagnosed diabetic dogs and cats, and that only in the presence of clinical signs of hypercortisolism dynamic tests are indicated.

Reviewer’s comment: Line 119-125 – Dogs and cats are treated most of the times with exogenous insulin and the reliability of any insulin measurement in this scenario is low due to cross-reaction between exogenous and endogenous insulin in the immunoassays.

Response: The sentence has been revised to emphasize that insulin measurement is unreliable in animals under insulin therapy due to cross-reactivity between endogenous and exogenous insulin in immunoassays.

Reviewer’s comment: Line 141 – How was the validity of the insulin kits used in the veterinary studies Validated? There´re several kits on the market for insulin measurement, most have been poorly validated for use in dogs and cats.

Response: The text has been modified to indicate that insulin immunoassays used in veterinary studies often lack proper validation for dogs and cats, and this represents an important methodological limitation.

Reviewer’s comment: Line 147 – What are the necessary adaptations within veterinary context mentioned  in the sentence “For data analysis, the calculation methods of the HOMA index used in various studies were compared, emphasizing the necessary adaptations within the veterinary context”?

Response: The sentence was rewritten to specify the necessary adaptations: species-specific reference ranges, validation of immunoassays, adjustment of cut-off values for insulin resistance, and accounting for interspecies physiological differences.

Reviewer’s comment: Line 155 – The sentence affirms that “the results contribute to optimizing disease management strategies”. Which disease? Maybe is still to early to assume such importance for the results, since the MMs are being described at this point. “The results may contribute to…” would read better.

Response: The sentence has been corrected as suggested. It now reads “The results may contribute to optimizing disease management strategies.”

Reviewer’s comment: Line 162 - I´m not aware about any paper in veterinary medicine that considers HOMA indexes “essential tool for evaluating insulin. sensitivity and pancreatic β-cell function”

Response: The wording was modified to avoid overstatement. The revised version states that HOMA indices are a “promising and potentially valuable tool” in veterinary medicine.

Reviewer’s comment: Line 169 – Accuracy of HOMA indexes are very questionable due to lack of standardization in dogs, despite being better studied in cats, the main author in this Subject (Apleton et al), recognizes the simple measuring of insulin and glucose (fasting insulin index) is to date, the best option to access insulin resistance and beta cell function in the cat.

Response: The discussion was revised to include this important perspective. The text now acknowledges that HOMA indices are limited by lack of standardization in dogs and that, according to Appleton et al., the fasting insulin index remains the most reliable option for evaluating insulin resistance and β-cell function in cats.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The paper has improved significantly after the authors' edits. In my opinion, it can be accepted for publication in its current form.