Case Report: Diabetic ketoacidosis after co-administration of empagliflozin and probenecid

Sodium-glucose cotransporter-2 (SGLT2) inhibitors are filtered and secreted to their primary site of action in the proximal tubule of the kidney. At this site, SGLT2 inhibitors also reduce renal elimination of ketone bodies, a finding implicated in their propensity to cause ketoacidosis. Many commonly used medications have potential to diminish renal elimination of SGLT2 inhibitors and to compound the effects of SGLT2 inhibitors on renal elimination of ketone bodies by inhibiting tubular secretion of the SGLT2 inhibitor itself and/or ketone bodies. We present a case of severe diabetic ketoacidosis (DKA) in a patient with type 2 diabetes occurring several days after co-prescription of empagliflozin and probenecid. Other than the recent introduction of empagliflozin, no cause for the DKA episode was apparent. A pharmacokinetic interaction between probenecid and empagliflozin, involving organic anion transporter 3 (OAT3), reduces proximal tubular secretion of empagliflozin and increases patient exposure to the drug. Whether or not this phenomenon is sufficient to cause severe DKA is discussed. An alternative explanation as to the DKA aetiology is proposed, wherein probenecid may compound effects of empagliflozin on renal elimination of ketone bodies. We suggest that clinicians exercise caution when prescribing SGLT2 inhibitors alongside pharmacologic inhibitors of, or competitors for, proximal tubular organic anion transporters in patients with diabetes mellitus due to the risk of severe DKA.

7 years ago) presented to our hospital with a one-week history of right leg redness, swelling, and pain (Figure 1).There was a family history of type 1 diabetes mellitus (T1DM) diagnosed at 8 years of age in a maternal first cousin and of T2DM diagnosed in a maternal uncle in his mid-50s.The patient lived alone, was independent in activities of daily living, drank alcohol rarely on social occasions, and was a lifelong non-smoker.On examination, the patient had a heart rate of 118 beats per minute (sinus tachycardia on an electrocardiogram), respiratory rate of 19 breaths per minute, blood pressure of 141/91 mmHg, peripheral oxygen saturations of 100% on room air, and a temperature of 38.8 degrees Celsius.Erythema, swelling, and tenderness of the entire right lower limb was noted.Examination of the cardiovascular, respiratory, and abdominal systems was unremarkable.
Ultrasound imaging confirmed the presence of a right iliofemoral deep vein thrombosis.The blood glycosylated haemoglobin level was 107 mmol/mol.Streptococcus agalactiae was cultured in peripheral blood samples.No evidence of infective endocarditis was present on transthoracic echocardiography.The patient was anticoagulated with enoxaparin 1 mg/kg (90 mg) twice daily subcutaneously initially, and later transitioned to apixaban 5 mg twice daily orally.Metformin therapy was continued at a dose of 850 mg three times daily orally, modified release gliclazide 120 mg once daily orally was stopped, and linagliptin 5 mg once daily orally and empagliflozin 10 mg once daily orally were commenced to improve glycaemic control.
The Streptococcus agalactiae isolate tested susceptible to penicillin, and treatment with intravenous high dose benzylpenicillin 2.4 g four times daily was commenced with good clinical effect.The source of bacteraemia was unclear; however, given the presence of a deep vein thrombosis, this was presumed to be infected.Blood C-reactive protein and neutrophil count values decreased during her inpatient stay (196 to 33 mg/L and 10.6 to 5.7 x10 9 /L, respectively).Prior to discharge, a peripherally inserted central catheter was placed and the antibiotic regimen was switched to intravenous cefazolin (2 g once daily) with oral probenecid (1g once daily) to facilitate a 6-week antibiotic course via the outpatient parenteral antibiotic therapy service.
The patient was readmitted 48 hours later with severe DKA.The patient's infection was well controlled at the point of readmission with DKA.The patient had received the prescribed cefazolin via the outpatient parenteral antibiotic therapy service after hospital discharge.The patient also reported good compliance with prescribed medications, including oral hypoglycaemics, after hospital discharge.Nevertheless, venous blood gas results included: pH 6.87; bicarbonate 3.6 mmol/L; anion gap 30.4 mmol/L; and glucose 15.8 mmol/L.The blood ketone level was 4.3 mmol/L.The patient was treated with a variable rate intravenous insulin infusion, which was prepared by adding 50 units of Actrapid human soluble insulin to 49.5 mL of 0.9% sodium chloride and administered at insulin infusion rates of 0.5 -8mL/hour.Intravenous isotonic sodium bicarbonate 1.26%, balanced crystalloid (Hartmann's

Amendments from Version 1
Thank you to the reviewers for their thorough review of this case report and constructive feedback.In this revised version, we have responded to the reviewer feedback by: • Updating Figure 1 by removing extraneous arrows to improve its clarity; • Discussing mechanisms of diabetic ketoacidosis in the setting of SGLT2 inhibitor therapy; • Highlighting the lack of plasma empagliflozin levels before and after probenecid administration as a limitation of the case report; • Providing more detailed discussion of the pharmacokinetic drug-drug interaction between empagliflozin and probenecid, and whether or not increased systemic exposure to empagliflozin alone would be sufficient explain the severe episode of diabetic ketoacidosis observed; • Proposing an alternative or complementary explanation for the episode of diabetic ketoacidosis arising due to the empagliflozin-probenecid drug-drug interaction, wherein both drugs act to diminish renal elimination of ketone bodies; and • Expanding the conclusions of the article to highlight the potential for ketoacidosis with co-prescription of SGLT2 inhibitors and inhibitors of proximal tubular organic anion transporters, as well as making recommendations for future studies to tease out the mechanisms involved in the generation of ketoacidosis in this context.

Introduction
Sodium-glucose cotransporter-2 (SGLT2) inhibitors are strongly recommended for patients with type 2 diabetes in the latest clinical practice guidelines based on their demonstrated cardiovascular and renal benefits 1 .Nevertheless, these medications increase the risk of diabetic ketoacidosis (DKA) 2,3 , a severe acute complication of diabetes which carries a substantial risk of morbidity and mortality 4 .
SGLT2 inhibitors are both filtered and secreted to their primary site of action in the proximal tubule of the kidney 5 .Although many commonly prescribed medications have the potential to diminish renal elimination of SGLT2 inhibitors by inhibiting their proximal tubular secretion, these interactions are not recognised to increase the risk of DKA nor other side effects of SGLT2 inhibitors and do not influence routine prescription of these medications.We present a case of severe DKA in a patient with type 2 diabetes occurring several days after co-prescription of the SGLT2 inhibitor empagliflozin alongside probenecid, an inhibitor of organic anion transporter 3 (OAT3) 6 , a transporter which plays a key role in the secretion of empagliflozin into the proximal tubule 5 .

Case report
A 54-year-old woman with obesity (body-mass index 39.5 kg/m 2 ) and type 2 diabetes mellitus (T2DM, diagnosed solution), and 10% dextrose with potassium chloride were also administered as per institutional protocol.Subcutaneous insulin therapy was commenced when the DKA resolved: insulin detemir (Levemir ® ) 10 units twice daily and insulin aspart (Fiasp ® ) 8 units three times daily with meals.Metformin 500 mg twice daily orally and linagliptin 5 mg once daily orally were recommenced 48 hours after readmission, with the metformin uptitrated to 1 g twice daily after one week.
At the point of readmission with severe DKA, and after having commenced systemic anticoagulation for the deep venous thrombosis, the patient reported vaginal bleeding and an ulcerated vulval lesion was noted on physical examination.Biopsies later confirmed the presence of a squamous cell carcinoma, which was staged as IVB disease on the basis of inguinal and pelvic sidewall lymphadenopathy and bone metastases present on staging scans.The vulval cancer is likely to have contributed to formation of the deep vein thrombosis as well as serving as a source of entry of Streptococcus agalactiae into the bloodstream 7 .

Discussion
Other than the introduction of empagliflozin 9 days prior, no cause for the severe DKA episode described in this case was apparent.SGLT2 inhibitors are recognised to increase risk of DKA.A meta-analysis of 39 randomised clinical trials highlighted that SGLT2 inhibitors double the risk of DKA in people with T2DM 2 .This risk is higher in elderly patients and those using SGLT2 inhibitors for longer periods of time 2 , neither of which applied in the current case.Whilst SGLT2 inhibitors substantially increase risk of DKA in patients with T1DM 3 , further evaluation for T1DM after readmission with DKA was negative in the present case.Specifically, an islet autoantibody screen including antibodies to zinc transporter 8, islet antigen 2, and glutamic acid decarboxylase was negative and the 2-hour post-prandial urinary C-peptide to creatinine ratio was 3.51 nmol/mmol, indicative of substantial endogenous insulin secretion.
Apart from undiagnosed T1DM, other common precipitants of SGLT2 inhibitor-induced DKA include infection, surgery, prolonged fasting, alcohol intake, acute vascular events (such as acute coronary syndrome or stroke), trauma, and prolonged exercise 8 .Of these, only infection is potentially pertinent to the DKA episode described herein.However, the patient's infection was well controlled at the time of readmission with DKA.Thus, the occurrence of DKA within several days of co-prescription of empagliflozin and probenecid implicates interaction between these medications rather than other factors as the primary cause of the severe DKA observed in this case.
Empagliflozin reaches its primary site of action in the proximal tubule by a combination of glomerular filtration and tubular secretion by transporters including OAT3, which is expressed on the basolateral membrane of the proximal tubule 6 .Probenecid inhibits OAT3 6 , with this property underlying its use alongside cefazolin in the present case.By diminishing renal cefazolin secretion, sufficient circulating cefazolin concentrations are achieved with once daily dosing.
In a pharmacokinetic study in healthy volunteers, co-administration of probenecid and empagliflozin, at the same total daily doses as those in the present case (1 g and 10 mg, respectively), resulted in a 26% increase in peak empagliflozin plasma concentrations and a 53% increase in area under the concentration-time curve 6 .These findings were not felt to be clinically meaningful 6 .However, the impact of this pharmacokinetic interaction between empagliflozin and probenecid has not been studied in patients with T2DM and it is unclear to what extent empagliflozin exposure would increase in the presence of additional OAT3 inhibitors alongside probenecid.
In the present case, co-administration of empagliflozin and probenecid immediately preceded a presentation with DKA with no other apparent cause.Moreover, both benzylpenicillin and cefazolin, antibiotics administered to this patient while she was receiving empagliflozin, are OAT3 inhibitors and likely contributed to enhanced empagliflozin exposure 9,10 .Although not routinely available in clinical practice, the absence of plasma empagliflozin levels before and after administration of probenecid is a limitation of this case report.In the absence of drug levels confirming increased systemic exposure to empagliflozin, we can only associate rather than definitively implicate co-prescription of empagliflozin and probenecid in the aetiology of the severe DKA described herein.
Although a plausible hypothesis, certain lines of evidence argue against increased systemic exposure to empagliflozin alone being sufficient to cause the severe DKA described herein.Even if systemic exposure to empagliflozin increased by 53% with probenecid 6 , this would not be expected to exceed the typical systemic exposure with the higher approved dose of empagliflozin (25 mg daily) -which does not routinely cause ketoacidosis.Although increased glucosuria is implicated in the mechanism of ketoacidosis with SGLT2 inhibitors 8,11 , the clinically approved doses of SGLT2 inhibitors (10 and 25 mg daily in the case of empagliflozin) were selected because they are near the plateau of the dose-response curve for glucosuria and are believed to achieve near-total inhibition of SGLT2 12,13 .Thus, even with increased systemic exposure to empagliflozin due to probenecid, one would not expect a marked increase in the magnitude of glucosuria.Moreover, mice with genetic knockout of OAT3 had a diminished glucosuric response to empagliflozin despite comparable tubular secretion of the drug to wild-type controls 5 .Although the physiology may differ between rodents and humans, it seems unlikely that increased glucosuria resulting in increased ketogenesis was the sole driver of ketoacidosis in the present case.
The severe DKA episode described herein may be better explained by the effects of SGLT2 inhibitors and probenecid on renal excretion of ketone bodies.Phlorizin (a non-selective SGLT1/SGLT2 inhibitor and the chemical structure upon which empagliflozin is based) increased renal tubular reabsorption of acetoacetate in dogs 11,14 .Through mechanisms reviewed in greater detail by Taylor et al., phlorizin would be expected to exert a similar effect on renal tubular handling of β-hydrobutyrate 11 .It is therefore plausible that empagliflozin decreases urinary excretion of ketone bodies, an effect which was compounded by co-administration of probenecid in the present case.Probenecid inhibits the organic anion transporters, OAT1 and OAT3 6,15 .Knockout of the Oat1 gene in mice decreased urinary excretion of the ketone body 3-hydroxybutyrate by approximately 65% and dramatically elevated plasma concentrations of 3-hydroxybutyrate, suggesting that 3-hydroxybutyrate is a physiological substrate of OAT1 16 .The ligand specificity of OAT1 and OAT3 also appears to be overlapping 17 .This raises the possibility that the severe DKA in this patient arose due to reduced urinary excretion of ketone bodies caused by concurrent treatment with probenecid and other inhibitors of renal organic anion transporters, including benzylpenicillin and cefazolin 9,10 , alongside the SGLT2 inhibitor empagliflozin.Of course, other factors implicated in ketoacidosis caused by SGLT2 inhibitors, such as increased ketogenesis as a consequence of glucosuria as well as increased glucagon secretion would have served to exacerbate the ketoacidosis 8,11 .
In conclusion, we suggest that clinicians exercise caution when co-prescribing empagliflozin and probenecid, or indeed other inhibitors of proximal tubular organic anion transporters, which may increase systemic exposure to the SGLT2 inhibitor and/or diminish urinary excretion of ketone bodies resulting in ketoacidosis.A similar drug-drug interaction may be relevant to dapagliflozin, which is also a substrate of OAT3 18 .Ketoacidosis arising in this context may be severe.
As co-prescription of SGLT2 inhibitors alongside inhibitors of proximal tubular organic anion transporters is likely to increase, investigation into the mechanisms of ketoacidosis in this context warrant further study.Studies assessing plasma levels of SGLT2 inhibitors and ketone bodies (β-hydroxybutyrate, acetoacetate, and acetone) before and after co-prescription of SGLT2 inhibitors with inhibitors of proximal tubular organic anion transporters in healthy volunteers and in people with diabetes mellitus would be informative.

Consent
Written informed consent for publication of their clinical details was obtained from the patient.

Francis M Finucane
Galway University Hospitals, University of Galway, Galway, Ireland This is an excellent description of an interesting case which makes an important contribution to the literature about a challenging and still poorly understood clinical problem.The previous reviewer comments are insightful and constructive and the responses are comprehensive.The figure is very clear and helpful.

I would have the following minor comments about the latest version of the case report:
I would specify the ethnicity of the patient at the start as it is always relevant in considering diabetes phenotype, especially in DKA occurring in type 2 diabetes.
An excellent overview of the mechanistic basis for SGLT2i induced DKA was published last year by Fahy et al. (20231 ) and would be worth incorporating and referencing (PMID 37261785).
It would be worth giving some specific examples when it is first mentioned that other drugs can inhibit OAT.
It would be good to provide more clarity and specificity around the epidemiology of SGLT2i DKAsome papers suggesting they do more than just "doubling" the risk of DKA in patients with T2DM.The revised version fully addresses the comments in our review and is a valuable contribution to the literature.

Are enough details provided of any physical examination and diagnostic tests, treatment given and outcomes? Yes
Is sufficient discussion included of the importance of the findings and their relevance to future understanding of disease processes, diagnosis or treatment?Yes Is the case presented with sufficient detail to be useful for other practitioners?

Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: I am a physician-scientist with a special interest in endocrinology and diabetes --especially with respect to drugs to treat diabetes and obesity.Based on my work at Bristol-Myers Squibb where I participated in R&D leading to approval of dapagliflozin, I have specialized expertise related to SGLT2 inhibitors.
We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Prashant Nasa
Max Super Specialty Hospital Saket, New Delhi, India Thank you for presenting this interesting case of DKA with the use of SGLT2 inhibitors, empagliflozin in this case.The case history is very well presented through a graphical abstract.However, the use of two coloured arrows to depict the drugs creates confusion.It would be better to use a single arrow for the start and end of the drug.

○
The conclusion is missing from the article.

○
The mechanisms described for DKA with the use of SGLT-2 inhibitors are multifactorial.I recommend using this review article to discuss this in detail [1].

○
Since the drug levels of empagliflozin were not measured in this case, the causal relationship between empagliflozin and probenecid drug interactions could not be established.
○ I believe this should be included as a limitation.

○
In addition, include a discussion on other possible mechanisms of DKA related to SGLT-2 inhibitors in this case.

Are enough details provided of any physical examination and diagnostic tests, treatment given and outcomes? Yes
Is sufficient discussion included of the importance of the findings and their relevance to future understanding of disease processes, diagnosis or treatment?No

Is the case presented with sufficient detail to be useful for other practitioners? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Intensive Care Medicine I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.Because of the timing of ketoacidosis shortly after addition of probenecid to the therapeutic regimen, the authors hypothesize that a pharmacokinetic drug-drug interaction between empagliflozin and probenecid may have contributed to the pathogenesis of ketoacidosis.
Although based on circumstantial evidence, this is a reasonable hypothesis.Specifically, the authors hypothesize that probenecid decreases organic anion transporter (OAT)-mediated renal excretion of empagliflozin.As discussed below, the literature contains substantial evidence that does not fully support their specific mechanistic hypothesis.It would have been helpful if the authors had provided data on plasma levels of empagliflozin before and after addition of probenecid.However, these analyses are not usually measured in routine clinical practice.In our review of the manuscript, we propose an alternative mechanism that might mediate a pharmacodynamic drug-drug interaction -specifically a possible role for organic anion transporters in mediating renal excretion of ketone bodies.The case report would be strengthened by adding a discussion of this alternative mechanism.
SPECIFIC COMMENTS Pharmacokinetic drug-drug interaction.Mach et al. (Clin Ther, 36, 280, 2014) [reference #6 in the paper by Martin et al.] reported formal drug-drug interaction studies indicating that probenecid increased exposure to empagliflozin by 53% (90% confidence interval, 46-61%).However, the patient was receiving the 10 mg/day dose of empagliflozin.Even if the exposure increased by 61%, the exposure would not be predicted to exceed the range of exposures typically observed in patients receiving the higher approved dose of empagliflozin (25 mg/day).Thus, it seems unlikely that the relatively modest increase in empagliflozin exposure does not seem to be sufficient to account for the patient's ketoacidosis.Furthermore, the approved doses of SGLT2 inhibitors were selected because they are near the plateau of the dose-response curve for glucosuria and are believed to achieve near-total inhibition of SGLT2.Accordingly, a modest increase in drug levels would not necessarily be expected to increase the magnitude of glucosuria.It is also important to note that treatment emergent cases of ketoacidosis have been observed in SGLT2 inhibitors even in the absence of probenecid.Reviewer Expertise: I am a physician-scientist with a special interest in endocrinology and diabetes --especially with respect to drugs to treat diabetes and obesity.Based on my work at Bristol-Myers Squibb where I participated in R&D leading to approval of dapagliflozin, I have specialized expertise related to SGLT2 inhibitors.
We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above.
They are also highlighted in the Abstract of the revised case report.We believe that the case report now more comprehensively captures the potential mechanisms of diabetic ketoacidosis arising in the setting of empagliflozin and probenecid therapy, and recommendations are made for future studies in this field Competing Interests: No competing interests were disclosed.

References 1 .
Fahy Y, Laffey JG, McNicholas B: Sodium-Glucose Cotransporter 2 Inhibitor-induced Ketoacidosis and Non-Ketotic Acidosis: When Is There a Role for Bicarbonate?. Ann Am Thorac Soc.2023; 20 (6): 911-916 PubMed Abstract | Publisher Full Text Is the background of the case's history and progression described in sufficient detail?Yes Are enough details provided of any physical examination and diagnostic tests, treatment given and outcomes?Partly Is sufficient discussion included of the importance of the findings and their relevance to future understanding of disease processes, diagnosis or treatment?Yes Is the case presented with sufficient detail to be useful for other practitioners?Yes Competing Interests: No competing interests were disclosed.Reviewer Expertise: Obesity, type 2 diabetes, ketone metabolism.I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.Reviewer Report 05 April 2024 https://doi.org/10.21956/wellcomeopenres.23373.r75983© 2024 Taylor S et al.This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Simeon I TaylorUniversity of Maryland School of Medicine, Baltimore, USA Zhinous Shahidzadeh Yazdi 1 Medicine, University of Maryland Baltimore, Baltimore, Maryland, USA 2 Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA

©
2023 Taylor S et al.This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Simeon I Taylor University of Maryland School of Medicine, Baltimore, USA Zhinous Shahidzadeh Yazdi 1 Medicine, University of Maryland Baltimore, Baltimore, Maryland, USA 2 Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA Martin et al. have provided an astute description of a thought-provoking case of ketoacidosis in a patient with type 2 diabetes treated with a multi-drug regimen including empagliflozin as well as probenecid.

Peer Review Current Peer Review Status: Version 2
Open Science Framework: CARE checklist for 'Case Report: Diabetic ketoacidosis after co-administration of empagliflozin and probenecid'.https://doi.org/10.17605/OSF.IO/8VJEQ Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).https://doi.org/10.21956/wellcomeopenres.23373.r89611© 2024 Finucane F. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Is the background of the case's history and progression described in sufficient detail? Yes Are enough details provided of any physical examination and diagnostic tests, treatment given and outcomes? Yes Is sufficient discussion included of the importance of the findings and their relevance to future understanding of disease processes, diagnosis or treatment? Yes Is the case presented with sufficient detail to be useful for other practitioners? Yes
48% increase in empagliflozin exposure.Thus, renal mechanisms are not believed to be the principal route of elimination/degradation for SGLT2 inhibitors.Rather, glucuronidation in the liver is believed to be the principal mechanism whereby SGLT2 inhibitors are cleared from the circulation.Delivery of drug to its site of action in the proximal tubule.Fu et al (Am J Physiol Renal, 315, F386, 2018) [reference #5 in the paper byMartin et al.]reported that genetic knockout of Oat3 in mice decreased the effect of empagliflozin (sub-maximally effective doses) upon glucosuria.When administered at a dose of 30 mg/kg, empagliflozin induced similar degrees of glucosuria in both wild type and Oat3 -/-mice.Based on these data in mice, one would predict that an inhibitor of OAT3 would likely not affect the pharmacodynamic response to maximally effective doses of empagliflozin while possibly decreasing the efficacy of sub-maximally effective doses.This seems to be inconsistent with what Martin et al. have hypothesized (i.e., that probenecid increases the efficacy of empagliflozin).Of course, one cannot be certain that data in the mouse are predictive of human pharmacology.3.A possible pharmacodynamic drug-drug interaction.Notwithstanding concerns related to the hypothesized mechanisms mediating the hypothesized drug-drug interaction, Martin et al. have provided thought-provoking circumstantial evidence suggesting that probenecid may indeed have contributed to the pathogenesis of this patient's ketoacidosis.Accordingly, it is important to consider whether probenecid might have triggered other mechanisms that might have contributed to pathogenesis of ketoacidosis.Eraly et al. (J Biol Chem, 281: 5072, 2006) reported that knockout of the Oat1 gene in mice decreased urinary excretion of 3-hydroxybutyrate by ~65% and also caused a dramatic increase in plasma 3hydroxybutyrate levels.Because the ligand specificity of OAT1 and OAT3 are reported to be overlapping(Nigam et al., Physiol Rev, 95, 83, 2013), it is tempting to speculate that probenecid might also impair renal excretion of ketone bodies.(It is noteworthy that acetoacetate and 3-hydroxybutyrate are both organic anions and therefore reasonable candidates to be transported by organic anion transporters.)Furthermore,Cohenetal. (Am  J Physiol, 184: 91, 1956) reported that phlorizin (the prototype SGLT inhibitor and also the chemical structure on which empagliflozin is based) promotes renal tubular reabsorption of acetoacetate.By analogy to phlorizin, it is plausible that empagliflozin might decrease urinary excretion of ketone bodies.We, therefore, hypothesize that probenecid and empagliflozin might synergize to increase plasma ketone body levels by decreasing renal clearance of those metabolities.Of course, empagliflozin-induced glucosuria indirectly triggers increased ketogenesis, which also contributes to the pathogenesis of SGLT2 inhibitor-induced ketoacidosis(Taylor etal, J. Clin Endocrinol Metab, 100, 2849, 2015).Nigam SK, Bush KT, Martovetsky G, Ahn SY, et al.: The organic anion transporter (OAT) family: a systems biology perspective.Physiol Rev. 2015; 95 (1): 83-123 PubMed Abstract | Publisher Full Text 4. Cohen JJ, Berglund F, Lotspeich WD: Renal tubular reabsorption of acetoacetate, inorganic sulfate and inorganic phosphate in the dog as affected by glucose and phlorizin.Am J Physiol.1956; 184 (1): 91-6 PubMed Abstract | Publisher Full Text 5. Taylor SI, Blau JE, Rother KI: SGLT2 Inhibitors May Predispose to Ketoacidosis.J Clin Endocrinol Metab.2015; 100 (8): 2849-52 PubMed Abstract | Publisher Full Text Competing Interests: No competing interests were disclosed.