Diabetes: Beyond Metformin Part 2

Part 2 of the SGLT2 inhibitor breakdown. The emerging liver fat evidence. The safety profile of Jardiance, Farxiga, and Invokana. And the single most important medication-coaching conversation in the entire drug series for patients combining these drugs with low-carbohydrate or ketogenic dietary approaches.

Part 1 of this SGLT2 conversation covered the underrated cardiovascular and kidney protection data, the mechanism (60-90 grams of glucose excreted in urine daily, imposing a 240-320 calorie deficit), the compensatory hyperphagia that partially erases the drug’s caloric impact, and the foundational coaching framework: proper nutrition and exercise are the primary intervention, with SGLT2 inhibitors as bridges to better health rather than replacements for the work.

Part 2 covers two topics that deserve their own focused treatment: the emerging liver fat data (genuinely interesting but more nuanced than headline results suggest) and the critical safety conversation for patients combining these drugs with low-carbohydrate or ketogenic dietary approaches.

If you’re on an SGLT2 inhibitor AND pursuing low-carb or keto, the second half of this post is essential reading. The combination is not contraindicated, but the safety considerations are real and underappreciated in the standard prescribing conversation.

The same precision standards from Post 1 (metformin) and Part 1 of this post apply throughout. Where evidence is more nuanced than initial framing might suggest, I’m naming it. Where findings vary across the SGLT2 class, I’m distinguishing them. Where my coaching framework differs from standard positions, I’m calling that out.

The liver fat data

This is the part of the SGLT2 story most relevant to the diabetes series’s underlying framework — because liver fat accumulation is the upstream mechanism driving type 2 diabetes in the first place. The evidence here is genuinely interesting but more measured in its current state than the early trial data alone might suggest.

The E-LIFT trial (Diabetes Care, 2018) randomized 50 patients with type 2 diabetes and NAFLD to empagliflozin 10 mg daily or standard treatment for 20 weeks. Liver fat was measured by MRI-PDFF (proton density fat fraction), the gold-standard non-invasive measurement.

Empagliflozin group: liver fat dropped from 16.2% to 11.3% within the group (a 30% within-group relative reduction, p<0.0001)
Control group: liver fat dropped from 16.4% to 15.5% (not statistically significant, p=0.057)
Between-group difference: -4.0 percentage points in absolute MRI-PDFF (p<0.0001)
ALT also improved significantly in the empagliflozin group (-10.9 IU/L, p=0.005)

That magnitude of liver fat reduction over 20 weeks is meaningful. It’s not curative — patients still had elevated liver fat at the end of treatment — but it’s a measurable improvement that metformin has not consistently demonstrated in trials with biopsy or MRI-PDFF endpoints.

Cheung et al, Hepatology 2024 (80:916-927) extended the empagliflozin/liver fat finding into a non-diabetic population. This was a randomized, double-blind, placebo-controlled trial in 98 non-diabetic patients with MRI-PDFF-confirmed MASLD, treated with empagliflozin 10 mg daily or placebo for 52 weeks. The empagliflozin group showed a statistically significant reduction in hepatic fat content compared to placebo, with a mean between-group difference of -1.06 percent (p=0.025). Median MRI-PDFF reduction was -2.49 percent on empagliflozin versus -1.43 percent on placebo. Notably, the resolution of hepatic steatosis (the harder clinical endpoint) was 44.9 percent on empagliflozin versus 28.6 percent on placebo — a trend in the favorable direction but NOT statistically significant (p=0.094). A 2025 commentary on this trial (El-Kassas et al, Hepatobiliary Surgery and Nutrition) described the magnitude of hepatic fat reduction as “modest” while noting the trial provides a valuable signal warranting further investigation. The honest read: empagliflozin reduces liver fat in non-diabetic MASLD patients, but the effect size is modest in absolute terms and the impact on hepatic steatosis resolution did not reach statistical significance in a 1-year trial.

The 2025-2026 meta-analysis picture is more nuanced. Several recent systematic reviews and meta-analyses have synthesized the SGLT2 + liver fat data, and the most current conclusions are more measured than individual trial results suggest:

A 2025 meta-analysis in Biomolecules and Biomedicine (Ge et al, April 2025) confirmed that SGLT2 inhibitors reduce liver fat content, but noted heterogeneity across studies.
A Frontiers in Medicine meta-analysis (16 RCTs, 11,300 subjects, published online late 2025, formally 2026) concluded that SGLT2 inhibitors “may delay the progression of liver fibrosis in patients with MASLD, particularly in improving serologic parameters” — note the more cautious “may delay” framing.
A 2025 meta-analysis specifically in non-diabetic NAFLD/MASLD populations (Abbas et al, Journal of Diabetes & Metabolic Disorders) concluded that “effects on hepatic steatosis resolution and fibrosis progression remain inconclusive” in non-diabetic patients.
A 2026 Bayesian network meta-analysis (Gomez et al, JGH Open) actually found that dapagliflozin showed stronger improvements in non-invasive markers of steatosis and fibrosis (CAP and LSM) than other SGLT2 inhibitors in MASLD — a finding that suggests the class effect on liver fat is not uniform across drugs.

The honest reading: the SGLT2 inhibitor class does appear to reduce liver fat, particularly in patients with type 2 diabetes. The effect appears to extend to non-diabetic patients with MASLD as well, though the most recent meta-analyses describe this effect as not yet conclusively established for steatosis resolution or fibrosis improvement at the histological level. Effect sizes vary across drugs within the class, with some recent evidence suggesting dapagliflozin may have stronger effects on non-invasive markers than empagliflozin specifically.

To be clear, none of the SGLT2 inhibitors are currently FDA-approved for fatty liver disease, and the AASLD does not yet position any of them as recommended therapy for MASH. But the evidence base is building, and the trial data — combined with the cardiovascular and kidney outcomes covered in Part 1 — makes a stronger case for this class addressing some of the underlying mechanism of metabolic disease than older agents have managed.

Definitely worth engaging in conversation with your doctors about.

My coaching perspective on liver fat: the SGLT2 evidence is suggestive enough to consider for patients with both T2D and MASLD, particularly as part of a comprehensive treatment plan that includes the dietary intervention that has actually been shown to reverse fatty liver. The 10-15% body weight reduction documented in DiRECT and ReTUNE — achieved through sustained dietary change — remains the most effective intervention for clearing liver fat. SGLT2 inhibitors may compound that effect. They do not replace the dietary work that drives most of the liver fat improvement in the first place!

I personally feel doctors need to be better educated about proper nutrition and ingredient repercussions for health. Most are not.

The safety profile

SGLT2 inhibitors have a generally favorable safety profile, but there are real adverse effects worth understanding, and one specific interaction that is critically important for the patient population I coach.

Genital mycotic infections are the most common adverse event — yeast infections, particularly in women, and balanitis in uncircumcised men. The mechanism is straightforward: the drug puts sugar in urine, and that creates a favorable environment for fungal growth. Pooled meta-analyses show a 3- to 4-fold class-wide increase in genital mycotic infection risk versus placebo, with the magnitude varying by drug: empagliflozin shows approximately 3.7-3.8 fold increase versus placebo, dapagliflozin approximately 4.3-5.0 fold, and canagliflozin the highest at approximately 4.9-5.2 fold. Real-world studies comparing SGLT2 inhibitors to other diabetes medications (rather than placebo) show somewhat lower effect sizes around 2.5-fold. Most cases are mild and treatable with standard antifungal therapy, and the recommendation from cardiology and nephrology consensus statements is that uncomplicated infections do not require drug discontinuation. For patients prone to recurrent yeast infections, this is worth knowing and managing proactively.

In absolute terms: a large clinical trial meta-analysis (113 RCTs, more than 105,000 adults) found genital tract infections in 4.51% of patients on SGLT2 inhibitors versus 1.01% on placebo. Real-world UK primary care data showed 8.1% incidence within 1 year on SGLT2 inhibitors versus 1.8% on DPP-4 inhibitors. Women are approximately 3.6 times more likely than men to develop these infections (HR 3.64), and the strongest single predictor is a prior history of genital infection within the last year (HR 4.38). Most cases occur within the first 30 days of starting therapy.

Fournier’s gangrene — the rare but serious perineum complication worth knowing about. This deserves a specific call-out separate from the common yeast infections above, because it’s a different category of risk entirely and it carries an active FDA warning on all SGLT2 inhibitor prescribing information since August 2018.

Fournier’s gangrene is a rapid, life-threatening necrotizing infection of the external genitalia, perineum, and perianal region. It requires immediate surgical debridement and broad-spectrum antibiotics. In the general population, the background incidence is approximately 1.6 per 100,000 men overall and 3.3 per 100,000 men aged 50-79. When it occurs, mortality runs around 20-40 percent in average cases, and up to 88 percent in severe cases — most of that mortality driven by sepsis rather than the infection itself.

The original FDA pharmacovigilance signal came from a 2019 case series (Bersoff-Matcha et al, Annals of Internal Medicine) identifying 55 cases in SGLT2 inhibitor users between March 2013 and January 2019. The patients ranged in age from 33 to 87 years. Time from drug initiation to event ranged widely (5 days to 49 months, median 4.5 months). The cases included both sexes (39 men and 16 women), important because Fournier’s gangrene is historically thought of as primarily a male condition. All patients required surgery. Three patients died. Updated FDA Adverse Event Reporting System data through 2021 has identified 542 total cases across all SGLT2 inhibitors, with empagliflozin associated with the highest number (232 cases). Originally, cases were not reported with ertugliflozin, but additional ertugliflozin cases were identified in 2019, so the entire class now has reported cases.

However — and this is the important precision point — more recent controlled real-world studies have provided a substantially more measured picture of the absolute risk than the original FDA pharmacovigilance signal suggested. A 2025 retrospective cohort study (April 2025, ScienceDirect) analyzed Fournier’s gangrene incidence within 1 year of starting various diabetes medications. After propensity matching for confounders, the absolute incidence rates were:

Biguanides (metformin): 1.10 per 10,000 starts (lowest)
SGLT2 inhibitors: 1.16 per 10,000 starts (second lowest)
Other diabetes medications: intermediate
Insulin users: 1.71 per 10,000 starts (highest)

This finding suggests the original FDA signal may have been confounded by SGLT2 users being disproportionately male and obese — both independent risk factors for Fournier’s gangrene. A 2025 large matched cohort study using the TriNetX database (Hertel et al, Frontiers in Pharmacology, December 2025) involving 192,245 SGLT2 inhibitor users provided additional real-world evidence consistent with this more measured interpretation. And a meta-analysis of randomized controlled trials (2019) found no statistically significant difference between SGLT2 inhibitors and comparator therapies in Fournier’s gangrene incidence.

The honest synthesis: the FDA warning remains in place and is appropriate. Patients on SGLT2 inhibitors should know the warning signs, and any unusual genital or perineal symptoms with systemic features require immediate evaluation. But the absolute risk is genuinely low — about 1.16 per 10,000 starts — and the most current controlled data suggests SGLT2 inhibitors are not among the higher-risk diabetes medications for this complication when properly compared to alternatives. Insulin and several older diabetes drugs carry equal or higher absolute risk, despite carrying no equivalent FDA warning.

The warning signs to know: pain, tenderness, redness, or swelling in the genital, perineal, or perianal area, combined with fever or general malaise. That combination — particularly if symptoms are progressing quickly — is a surgical emergency requiring immediate medical evaluation, not a wait-and-see situation. Do not assume it is “just another yeast infection.” Severity progressing rapidly with systemic symptoms (fever, feeling unwell) is the distinguishing feature.

For most patients on these drugs, this complication will never be a concern. The point of including it here is to make sure the warning signs are known, not to create alarm about a rare event with absolute risk comparable to other widely-prescribed diabetes medications. Recognizing the difference between a routine genital infection (mild, treatable with antifungal cream, no systemic symptoms) and the early presentation of Fournier’s gangrene (severe pain, rapid progression, fever) is what allows for timely treatment in the rare case when it matters most.

Urinary tract infections occur at slightly increased rates but the increase is modest. Hydration and standard UTI prevention measures generally manage this well.

Volume depletion and hypotension can occur, especially in older patients, those on diuretics, or those with already-low blood pressure. SGLT2 inhibitors have a mild diuretic effect, and combining them with loop diuretics requires monitoring.

A note on the canagliflozin amputation history. Canagliflozin previously carried an FDA boxed warning for lower-extremity amputation risk based on the 2017 CANVAS trial, which showed approximately twofold higher amputation rates in canagliflozin users versus placebo. That boxed warning was removed by the FDA in August 2020 after additional real-world data did not replicate the signal at the magnitude initially observed. Current evidence suggests the risk may be modestly elevated but not at the level that initially prompted the warning. Patients with significant peripheral vascular disease or prior amputation history should still have this discussed with their prescribing physician.

The kidney function paradox. SGLT2 inhibitors cause an initial drop in eGFR within the first few weeks of treatment — typically 3-6 mL/min/1.73m² in pooled trial data, with the nadir occurring within the first 2-4 weeks of therapy. This is expected, hemodynamic in origin, and partially reverses over the subsequent 6-8 weeks. The long-term effect is preservation of kidney function. Consensus guidance from cardiology and nephrology specialties is that SGLT2 inhibitors should not be discontinued unless eGFR drops by more than 30%, but the initial dip sometimes leads to inappropriate discontinuation by clinicians who don’t recognize the pattern as expected.

The interaction that matters most for low-carb patients — euglycemic DKA

This deserves its own focused attention because it is genuinely dangerous, it’s underappreciated in the standard prescribing conversation, and it’s directly relevant to the audience I work with.

Diabetic ketoacidosis is normally a complication of type 1 diabetes, presenting with very high blood glucose, ketones in the urine and blood, dehydration, and acidosis. Euglycemic DKA is the same metabolic emergency without the elevated glucose — blood sugar can be under 250 mg/dL while the patient is in serious ketoacidosis. It’s harder to recognize, more often missed, and equally life-threatening.

SGLT2 inhibitors substantially increase the risk of euglycemic DKA. A review of FDA Adverse Event Reporting System cases of SGLT2-associated ketoacidosis (Blau et al) found that 71 percent of reported cases presented as euglycemic. Real-world cohort studies have shown somewhat lower proportions in the 57-58% range, but the dominant pattern is consistent: a majority of SGLT2-related DKA cases present without the marked hyperglycemia that normally signals DKA. The absolute risk of DKA on an SGLT2 inhibitor in general T2D populations is low — less than 1 percent per year, with most real-world cohort studies showing rates in the range of 1 to 3 per 1,000 person-years. But the relative risk is increased roughly 3-fold compared to DPP-4 inhibitors in population cohort studies, with FDA Adverse Event Reporting data suggesting up to 7-fold increased risk versus DPP-4 inhibitor therapy. The risk is elevated further by anything that pushes the body toward ketosis: dehydration, infection, illness, prolonged fasting, surgery, excessive alcohol use, and — critically — very low-carbohydrate or ketogenic diets.

The specific data on the keto + SGLT2 combination is now available and worth knowing precisely. A February 2025 analysis published in Diabetes, Obesity and Metabolism examined a telemedicine clinic specializing in ketogenic diets for type 2 diabetes management. Among 50,404 patient-years of follow-up from patients with T2D on ketogenic dietary approaches, the DKA incidence was 1.01 per 1,000 person-years overall. Among the 12,763 patient-years of follow-up specifically from patients also taking SGLT2 inhibitors, DKA incidence was 2.90 per 1,000 person-years — approximately 3-fold higher specifically in the keto-plus-SGLT2 combination. That study comes from a clinic that does this combination intentionally and with medical supervision; the rate in patients combining the two without medical supervision is likely meaningfully higher.

There are now multiple published case reports and case series of euglycemic DKA in patients taking SGLT2 inhibitors who simultaneously pursued ketogenic or very-low-carbohydrate diets. The combination creates a perfect storm: the drug is pushing glucose out through urine, the low-carb diet is pushing the body toward fat oxidation and ketone production, insulin levels are low, and the safety brake that normally prevents pathological ketosis can fail. Patients can develop life-threatening acidosis with normal blood sugar readings, which delays diagnosis and treatment.

This matters enormously for patients in the diabetes-reversal space, because both interventions — SGLT2 inhibitors AND low-carb dietary approaches — are increasingly used together. The combination is not contraindicated by guidelines, but it does require explicit medical management and patient awareness. From the FDA label for Jardiance: urinary glucose excretion persists for 3 days after discontinuing the drug, and post-marketing reports have documented ketoacidosis and glucosuria lasting up to 2 weeks after discontinuation. The drug’s effects don’t disappear the day you stop taking it.

The practical implications for patients combining SGLT2 inhibitors with low-carb dietary approaches:

Discuss the combination with your prescribing physician explicitly. Many patients are doing this combination without telling their doctor what their diet looks like, which removes the safety net.
Be aware of euglycemic DKA symptoms: nausea, vomiting, abdominal pain, fatigue, difficulty breathing, fruity breath — even with normal blood sugar readings.
Hold the SGLT2 inhibitor before any planned surgery per FDA label guidance: at least 3 days for empagliflozin, dapagliflozin, and canagliflozin; at least 4 days for ertugliflozin due to its longer half-life. The 2026 ADA Standards of Care echoes this 3-4 day range. A 2025 study published in JAMA Surgery (Dixit et al) challenged the strict withholding paradigm based on a nationwide analysis of 34,000+ patients undergoing emergency surgery, but the current standard recommendation for elective surgery remains the FDA-labeled withholding period.
Hold the SGLT2 inhibitor during acute illness, especially with vomiting, diarrhea, or inability to eat.
Consider checking urine or blood ketones periodically, especially during deeper carbohydrate restriction.
If symptoms develop, this is a medical emergency requiring immediate evaluation, even if your glucose meter shows normal numbers.

This is the safety information that gets glossed over in the standard prescribing conversation. For my coaching practice specifically, this is one of the most important medication-coaching intersections to get right. The combination can be safe with appropriate medical management — but it requires the management, not just the assumption that everything is fine because blood sugar looks normal.

Who genuinely benefits from SGLT2 inhibitors

Working through who derives the most clinical benefit from this drug class, with the coaching framework intact — nutrition and exercise as the primary intervention, drugs as bridges:

Patients with type 2 diabetes and established cardiovascular disease. This is the population EMPA-REG OUTCOME was designed for, and the benefits with empagliflozin specifically are substantial in this population. For patients in this category, SGLT2 inhibitors — particularly empagliflozin given its specific FDA indication for reducing cardiovascular mortality — have a stronger cardiovascular evidence base than metformin and should be part of the conversation alongside the lifestyle interventions that address the underlying metabolic disease.

Patients with heart failure, regardless of diabetes status. Both HFrEF and HFpEF populations benefit, and the heart failure hospitalization reduction is the most consistent class-wide effect in the SGLT2 literature. For patients with HFpEF specifically — a condition that had no effective treatments before this drug class — empagliflozin represents the first real pharmacological advance in decades.

Patients with chronic kidney disease, regardless of diabetes status. EMPA-KIDNEY established meaningful kidney protection in a broad population, and KDIGO 2024 guidelines now position SGLT2 inhibitors as 1A first-line therapy alongside ACE inhibitors and ARBs for patients with T2D and CKD. The kidney protective effect appears to be one of the most consistent class-wide findings.

Patients with type 2 diabetes plus fatty liver disease (MASH/MASLD). While not formally indicated for liver disease, the liver fat reduction data is strong enough that for patients with both T2D and MASLD, SGLT2 inhibitors are a reasonable consideration as part of a comprehensive treatment plan. Recent meta-analyses suggest the effect may not be uniform across all SGLT2 drugs — some evidence suggests dapagliflozin may have particularly favorable non-invasive marker effects for liver disease. This is an area where the specific drug choice within the class may matter, and the decision should be made with the prescribing physician based on individual factors. Dietary intervention remains the primary intervention for fatty liver.

Patients with type 2 diabetes who haven’t tolerated other agents. The insulin-independent mechanism makes this class useful for patients with declining beta cell function who can’t get sufficient glucose lowering from metformin alone.

The patient populations where SGLT2 inhibitors are less appropriate include those on very low-carbohydrate diets without close medical supervision (DKA risk, as discussed), patients with recurrent UTIs or mycotic infections, patients with severely reduced kidney function below the labeled threshold, and patients with type 1 diabetes (though this is a developing area with ongoing research).

The framework — combining SGLT2 inhibitors with lifestyle change

The foundational principle from Part 1 applies just as much here: proper nutrition and exercise are the primary intervention. SGLT2 inhibitors are bridges to better health, not replacements for the lifestyle work that actually reverses metabolic disease.

For patients combining SGLT2 inhibitors with dietary intervention, particularly low-carb or ketogenic approaches, the practical framework looks like this:

Coordinate with your physician explicitly. Don’t pursue low-carb or keto in parallel with an SGLT2 inhibitor without informing the prescribing physician about the dietary approach. The euDKA risk requires medical management, and your physician can’t manage what they don’t know.

Hydration becomes more important. The combination of pharmacologic glucose loss through urine and ketogenic state both push the body toward dehydration. Adequate hydration is part of the safety framework, not just a general wellness recommendation.

Consider ketone monitoring. Periodic urine or blood ketone testing during deeper carbohydrate restriction gives you and your physician early warning of dangerous ketosis levels. This is standard practice for patients with type 1 diabetes; it should be considered for SGLT2 + low-carb patients as well.

Know the warning signs. Nausea, vomiting, abdominal pain, fatigue, difficulty breathing, fruity breath — even with normal blood sugar. If these develop, seek immediate medical evaluation.

The medication may need to be paused. During acute illness, before planned surgery, during prolonged fasting — these are circumstances where the drug should be temporarily held. Coordinate with your physician on the specific protocol.

As lifestyle change drives improvement, the medication conversation evolves. For patients without established cardiovascular or kidney disease, the goal of the coaching framework is to use the medication as a bridge while lifestyle work addresses the underlying disease — with the possibility of medication tapering as the underlying state resolves, under medical supervision. For patients with established cardiovascular or kidney disease, the medication may legitimately be a long-term part of the treatment plan even after the lifestyle work has resolved the metabolic component of the disease.

Where this leaves us — Part 2

The SGLT2 inhibitor class is doing more than most patients on these drugs realize. The liver fat data is interesting and building. The cardiovascular and kidney outcomes from Part 1 are substantial. And the safety considerations — particularly the euglycemic DKA risk in combination with low-carb dietary approaches — are real and require proper medical management.

The coaching framework remains constant: proper nutrition and exercise are the primary intervention. The medication is the bridge. The lifestyle work is the destination. The June 2026 CKM Syndrome Guideline endorses exactly this framework. The compensatory hyperphagia data from Part 1 explains biologically why drugs alone don’t deliver what their calorie-deficit math would predict. The patients who get the most benefit from this drug class — including the liver fat benefits and the cardiovascular and kidney protection — are the ones who pair it with intentional dietary work, resistance training, sleep, and stress management.

For patients pursuing low-carb or ketogenic dietary approaches alongside an SGLT2 inhibitor: the combination can be safe with appropriate medical management, but the safety considerations require explicit conversation with your prescribing physician, awareness of euDKA symptoms, and the practical safety protocol outlined above. This is not a casual combination. Done well, it can be part of a comprehensive remission protocol. Done without medical coordination, it can be life-threatening.

The next two posts in this series cover the drug class that has dominated the cultural conversation in metabolic medicine for the last several years — the GLP-1 receptor agonists. Semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro, Zepbound). Post 3 will be the honest accounting of what these drugs actually do, including the August 2025 FDA approval of semaglutide for fatty liver disease (MASH). Post 4 will be the honest accounting of what they cost — the lean mass loss problem, the discontinuation rebound, and the smart-use framework for patients on these drugs.

A coaching note before closing

One thing I want to be clear about as we close this two-part SGLT2 conversation. Every drug has potential side effects. There is no pharmaceutical that comes without a trade-off. As a coach, my position has been consistent: medications used to improve a negative health situation should always be accompanied by the lifestyle changes that address the underlying disease. Proper nutrition. Consistent exercise. The work that resolves the upstream cause rather than just managing the downstream symptoms. The medication may be the bridge. The lifestyle change is the destination. That framing applies to metformin, SGLT2 inhibitors, GLP-1s, and every other drug class this series will cover. Drugs are tools. They are not a substitute for the work.

A note on what this is and isn’t

This series is educational content built from peer-reviewed research, clinical trial data, and my work as a nutrition and fitness coach. It is not medical advice. I am not a physician. The information here is designed to give you the framework and the questions to bring to your own medical team — not to replace that team.

Do not stop or adjust any SGLT2 inhibitor based on what you’ve read here. These drugs are often prescribed for serious cardiovascular, renal, or hepatic indications, and discontinuation should be coordinated with your prescribing physician.

If you’re combining an SGLT2 inhibitor with a low-carbohydrate or ketogenic diet, talk to your physician explicitly about the euglycemic DKA risk. This is the single most important safety conversation in this post. The combination is not contraindicated, but it requires explicit management. Know the symptoms. Hold the drug per FDA guidance for surgery and illness. Consider ketone monitoring.

For planned surgery or anesthesia, hold the SGLT2 inhibitor for at least 3 days beforehand for empagliflozin, dapagliflozin, and canagliflozin — and at least 4 days beforehand for ertugliflozin (per the FDA labels and 2026 ADA Standards of Care). Inform your surgical team that you’ve been taking it.

Find a physician who works with you on the framework. If you’ve been on an SGLT2 inhibitor for years without anyone discussing the safety considerations, the lifestyle interventions that compound the drug’s benefits, or the long-term cardiovascular and renal monitoring that should accompany this drug class — those conversations are worth having.

Individual results vary. Trial data describes populations. You are an individual with your own medical history, comorbidities, and clinical context. The general framework in this post will not be the right framework for every reader. Use your judgment. Talk to your team. Ask the questions.

The goal of this work has always been informed patients, not stopped medications. For SGLT2 inhibitors specifically, the more accurate framing is informed patients who understand what the drug is actually accomplishing, what the safety considerations require, AND what the lifestyle work needs to look like alongside the medication.

The medication is the bridge. Nutrition and exercise are the destination. That framework drives everything in this practice.

You have the framework. The work is yours.