Thiamine Deficiency in Diabetes: Implications for Diabetic Ketoacidosis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a well-written and comprehensive narrative review addressing an important and under-recognized aspect of DKA. The manuscript is clear, clinically relevant, and well referenced. Only minor refinements are needed:

• Clarify definitions of thiamine deficiency, as thresholds vary across cited studies.
• Improve transitions between major sections (e.g., physiology -> prevalence) for smoother flow.
• Standardize terminology and formatting (TPP/TDP, capitalization, units).
• Briefly note limitations of the narrative review approach, particularly heterogeneity in measurement methods.
• Clarify mechanistic statements (e.g., ATP reduction percentages) regarding their evidence base (in vitro vs. clinical).

Overall, this is a strong manuscript that requires only minor editorial adjustments. I recommend minor revision.

Author Response

Comment 1: This is a well-written and comprehensive narrative review addressing an important and under-recognized aspect of DKA. The manuscript is clear, clinically relevant, and well referenced. Only minor refinements are needed:

 

Response 1: Thank you for your kind words and for reviewing our article. We have addressed individually all of your comments below.

 

Comment 2: Clarify definitions of thiamine deficiency, as thresholds vary across cited studies.

Response 2: Thank you for raising this. It is a major limitation of the literature and was raised by

Reviewer 2 also. We have added a paragraph to the pathophysiology section to explain how thiamine deficiency is reported in the literature and emphasised the lack of standardised universally acceptable thresholds.

Comment 3: Improve transitions between major sections (e.g., physiology -> prevalence) for smoother flow.

Response 3: We have flagged the transitions between the major sections in the last paragraph of the Introduction so that the reader knows what to expect as they go through the manuscript.

Comment 4: Standardize terminology and formatting (TPP/TDP, capitalization, units).

Response 4: We have standardized the terminology to thiamine pyrophosphate (TPP) throughout the manuscript. Capitalization has been addressed. We have reported units as reported in the primary article.

Comment 5: Briefly note limitations of the narrative review approach, particularly heterogeneity in measurement methods.

 

Response 5: Yes there are certainly limitations. We have now added a paragraph to acknowledge these transparently.

 

Comment 6: Clarify mechanistic statements (e.g., ATP reduction percentages) regarding their evidence base (in vitro vs. clinical).

 

Response 6: This was a clinical study with measurements on patient and healthy volunteer blood. This is now clearly mentioned in the manuscript.

 

Comment 7: Overall, this is a strong manuscript that requires only minor editorial adjustments. I recommend minor revision.

 

 Response 7: Thank you once again. We hope you will now find the manuscript suitable for publication.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript has a good clinical motivation and logically builds an argument for the need for RCTs. However, it is also a narrative review without a clear methodology for literature selection, with several overly precise quantifications without sufficient contextualization of the heterogeneity of measurements, definitions of deficit and confounders.

The topic is highly relevant - DKA is still a significant acute complication, and the issue of slow recovery is practical. The manuscript very clearly explains thiamine-dependent enzymes and the relationship to lactate/oxidative stress. Structure is very good, basically from mechanism to clinic, with a clear line. Authors identified research gaps very well.

However, this article is a narrative review without methodology, and that is associated with the risk of selection bias. In the manuscript is not stated how the literature was searched. Please add a short section Methods (Literature search) with databases, period, keywords, inclusion/exclusion criteria if applicable. I would also suggest adding a table of key studies to summarise the results. The text mixes plasma thiamine, whole blood thiamine, thiamine diphosphate (TDP/TPP) and transketolase activity – but without a clear explanation. Please add - what is best for diagnosis, what are the cut-offs, why results differ between studies, and whether plasma is valid in acute critical illness. I am suggesting to subsection Assessment of thiamine status. Explicitly state that prevalences (25–35% in DKA) depend on the method and threshold used.  In the manuscript, you mentioned “ATP production reduced by 40–48%”, “plasma reduced by ~75%”, “16–24 fold increased renal clearance”, “persistent acidosis explained by…”. These numbers cannot be generalised. Please refine language: “reported”, “suggests”, “in small observational cohorts”, “surrogate markers”, and add a short sentence about the limit (n, design, matrix) for each “strong” number. In the conclusion and abstract, the text suggests that “empiric administration warrants serious consideration”, which may sound like a recommendation. It is necessary to more clearly separate the hypothesis from EBM recommendations, state possible risks and practical barriers. To enhance the manuscript, you can add short Clinical practice considerations - who to consider (malnutrition, bariatric history, alcohol, hyperemesis, prolonged fasting), what regimen (e.g. IV in the first 24–48 h), and always with the wording “pending evidence”. Please, change Data Availability to “Not applicable” or “No new data were created”. 

Author Response

Comment 1: The manuscript has a good clinical motivation and logically builds an argument for the need for RCTs. However, it is also a narrative review without a clear methodology for literature selection, with several overly precise quantifications without sufficient contextualization of the heterogeneity of measurements, definitions of deficit and confounders.

 

The topic is highly relevant - DKA is still a significant acute complication, and the issue of slow recovery is practical. The manuscript very clearly explains thiamine-dependent enzymes and the relationship to lactate/oxidative stress. Structure is very good, basically from mechanism to clinic, with a clear line. Authors identified research gaps very well.

 

Response 1: Thank you for reviewing our manuscript. We have addressed all of your comments individually below.

 

Comment 2: However, this article is a narrative review without methodology, and that is associated with the risk of selection bias. In the manuscript is not stated how the literature was searched. Please add a short section Methods (Literature search) with databases, period, keywords, inclusion/exclusion criteria if applicable.

 

Response 2: A methods section has been added to explain the methodology.

 

Comment 3: I would also suggest adding a table of key studies to summarise the results.

 

Response 3: This has been added as per the reviewer’s suggestion. An abbreviated Table 1 has been added in the manuscript with a more detailed table in the supplement.

 

Comment 4: The text mixes plasma thiamine, whole blood thiamine, thiamine diphosphate (TDP/TPP) and transketolase activity – but without a clear explanation. Please add - what is best for diagnosis, what are the cut-offs, why results differ between studies, and whether plasma is valid in acute critical illness. I am suggesting to subsection Assessment of thiamine status.

 

Response 4: Thank you for this important point. This was also raised by other reviewers. We have done as you have suggested and added a section on assessment of thiamine status. We have added a new subsection 3.1 'Assessment of Thiamine Status' that comprehensively addresses: (1) different biomarkers (whole blood thiamine, plasma thiamine, erythrocyte TPP, transketolase activity), (2) diagnostic thresholds used across studies, and (3) the critical limitation of lack of standardization in measurement methodologies.

 

Comment 5: Explicitly state that prevalences (25–35% in DKA) depend on the method and threshold used.  In the manuscript, you mentioned “ATP production reduced by 40–48%”, “plasma reduced by ~75%”, “16–24 fold increased renal clearance”, “persistent acidosis explained by…”. These numbers cannot be generalised. Please refine language: “reported”, “suggests”, “in small observational cohorts”, “surrogate markers”, and add a short sentence about the limit (n, design, matrix) for each “strong” number.

 

Response 5: The proviso regarding prevalence in DKA has been added at the start of the relevant paragraph. While true that specific numbers from single, small studies cannot be generalised, we have reported the findings of individual studies as originally reported by the various authors. We have added this point to the limitations section.

The language suggestions have been addressed, with increased clarity. We have retained the word “reported” in several instances though, because it is the most accurate way of conveying that the manuscript is simply presenting what was reported in a primary study.
Some of the specific key methodological details have now been clarified with the addition of Tables 1 and Supp Table 1- particularly around the methodological details (eg type of study, sample size) of the key studies.

 

Comment 6: In the conclusion and abstract, the text suggests that “empiric administration warrants serious consideration”, which may sound like a recommendation. It is necessary to more clearly separate the hypothesis from EBM recommendations, state possible risks and practical barriers.

 

Response 6: This is a very important point that we thank the reviewer for raising. We have removed the statement from the abstract as it sounds too much like a recommendation, without the evidence to back it up. We feel that the statement regarding the strength of rationale for conducting a trial is reasonable and should be included in the manuscript.

 

Comment 7: To enhance the manuscript, you can add short Clinical practice considerations - who to consider (malnutrition, bariatric history, alcohol, hyperemesis, prolonged fasting), what regimen (e.g. IV in the first 24–48 h), and always with the wording “pending evidence”.

 

Response 7: Yes this is a great idea. We have now added a section which highlights clinical practice considerations that clinicians may employ at the bedside.

 

Comment 8: Please, change Data Availability to “Not applicable” or “No new data were created”. 

 

Response 8: We have done this. Thank you.

 

Reviewer 3 Report

Comments and Suggestions for Authors

In this narrative review, the authors examine the role of thiamine deficiency in diabetes mellitus complicated by diabetic ketoacidosis (DKA), elucidate its molecular mechanisms, and argue for thiamine supplementation as adjunctive therapy in DKA.

               The article draws attention to the high prevalence of thiamine deficiency in patients with diabetes and DKA, as hyperglycemia induces renal thiamine losses and intracellular thiamine deficiency, thereby aggravating lactic acidosis, mitochondrial dysfunction, and cardiovascular and neurological complications. Indeed, in clinical practice, upon presentation to the emergency department, patients are not routinely investigated for serum thiamine levels, and it is not a main therapeutic measure, according to current guidelines. From this point of view, I consider this article very important, and I congratulate the authors on the idea.

Overall, the article meets the criteria for a narrative review. The clinical question is clearly defined: “thiamine deficiency in diabetes/DKA and the rationale for supplementation in DKA. It has a coherent structure, with logical sections: physiology, pathophysiology, prevalence, molecular mechanisms, clinical implications, supplementation, and future research. It covers clinical studies (observational and randomized trials) in diabetes, studies in DKA, and cell and animal models. It integrates the data into a narrative thread. However, it would be good to introduce the methodology for searching and selecting studies, as databases, keywords, search period, inclusion/exclusion criteria, and assessment of the quality of evidence are not described.

The authors have focused on pro-thiamine in this review, but have not sufficiently discussed negative evidence or uncertainties (for example, the mixed results of thiamine trials in sepsis are only briefly mentioned).

Perhaps the language should be revised so that thiamine deficiency does not appear causal in DKA. Before empirically administering thiamine in DKA, RCT studies would be needed.

The presentation of the relationship between thiamine, lactate, and OCR appears across several sections with similar messages, requiring a rethink of the respective fragments.

It would be useful to mention other causes of thiamine deficiency that can coexist with DKA.

It would also be interesting to read a paragraph explaining the relationship of thiamine with other nutrients.

I would also recommend introducing a paragraph before the conclusions in which the limitations of the studies discussed are discussed.

The figures are not very clear; some words are barely seen.

Author Response

Comment 1: In this narrative review, the authors examine the role of thiamine deficiency in diabetes mellitus complicated by diabetic ketoacidosis (DKA), elucidate its molecular mechanisms, and argue for thiamine supplementation as adjunctive therapy in DKA.

 

Response 1: Thank you for comprehensively reviewing our manuscript.

 

Comment 2: The article draws attention to the high prevalence of thiamine deficiency in patients with diabetes and DKA, as hyperglycemia induces renal thiamine losses and intracellular thiamine deficiency, thereby aggravating lactic acidosis, mitochondrial dysfunction, and cardiovascular and neurological complications. Indeed, in clinical practice, upon presentation to the emergency department, patients are not routinely investigated for serum thiamine levels, and it is not a main therapeutic measure, according to current guidelines. From this point of view, I consider this article very important, and I congratulate the authors on the idea.

 

Response 2: Thank you for your kind words. We have addressed all of your comments below.

 

Comment 3: Overall, the article meets the criteria for a narrative review. The clinical question is clearly defined: “thiamine deficiency in diabetes/DKA and the rationale for supplementation in DKA. It has a coherent structure, with logical sections: physiology, pathophysiology, prevalence, molecular mechanisms, clinical implications, supplementation, and future research. It covers clinical studies (observational and randomized trials) in diabetes, studies in DKA, and cell and animal models. It integrates the data into a narrative thread. However, it would be good to introduce the methodology for searching and selecting studies, as databases, keywords, search period, inclusion/exclusion criteria, and assessment of the quality of evidence are not described.

 

Response 3: Thank you for pointing this out. A methods section has been added to assist the reader in understanding how this review was conducted.

 

Comment 4: The authors have focused on pro-thiamine in this review, but have not sufficiently discussed negative evidence or uncertainties (for example, the mixed results of thiamine trials in sepsis are only briefly mentioned).

Response 4: This is certainly true that the article presents a pro-thiamine view. However, this was not an intentional objective. Rather, it is a reflection of the excellent safety profile of thiamine. This has been studied with relatively large systematic reviews showing low rates of very minor adverse effects. Thiamine is truly a medication with absolutely limited adverse effects. There is no biological rationale for harm, and thus it is not really possible to discuss the negative viewpoint. It is however, repeatedly acknowledged throughout the manuscript that adequately powered RCTs are required to confirm the hypothesised benefits of thiamine in DKA. We strongly believe that this is very important- just being safe does not mean that we should encourage profligate prescription of thiamine.

 

Comment 5: Perhaps the language should be revised so that thiamine deficiency does not appear causal in DKA. Before empirically administering thiamine in DKA, RCT studies would be needed.

 

Response 5: Yes these are very important points to explicitly state. We have added a sentence to clarify that the relationship between thiamine deficiency and DKA is not known to causal. Rather, the DKA course can be more complicated in the patient with thiamine deficiency. We have stressed in various places in the manuscript that RCTs are required to confirm the hypothesised benefits.

 

Comment 6: The presentation of the relationship between thiamine, lactate, and OCR appears across several sections with similar messages, requiring a rethink of the respective fragments.

 

Response 6: It is true that we have mentioned this relationship in both 3.2 and 6.2. One of the key clinical features of DKA in patients with thiamine deficiency is the prolonged lactic acidosis, and subsequent delayed recovery of DKA. This is associated with the profound impairment of mitochondrial dysfunction in the face of increased metabolic demand which cannot be met due to insufficient thiamine. Thus we have explained this the pathophysiology section (3.2) and then reiterated the salient details in the DKA specific section 6.2. We strongly feel that this is warranted given the importance of this interaction in the pathophysiology of thiamine deficiency.

 

Comment 7: It would be useful to mention other causes of thiamine deficiency that can coexist with DKA.

 

Response 7: We have added a new section 5.6 that discusses this point.

 

Comment 8: It would also be interesting to read a paragraph explaining the relationship of thiamine with other nutrients.

 

Response 8: While this would certainly be interesting, we are concerned that this is beyond the scope of this manuscript. Thiamine has many interactions with magnesium, phosphate, B2, B3, B6, B12, glutathione, branched chain amino acids, and several other micronutrients. We respectfully propose that, rather than superficial coverage of these interactions within an already quite long manuscript, we focus on thiamine deficiency and its treatment.

 

Comment 9: I would also recommend introducing a paragraph before the conclusions in which the limitations of the studies discussed are discussed.

 

Response 9: Thank you- this has been raised by all the reviewers. We have added a comprehensive section on limitations as suggested.

 

Comment 10: The figures are not very clear; some words are barely seen.

Response 10: We have increased the font sizes and enlarged the images while maintaining the resolution to make the figures clearer.