SGLT2 Inhibitors in Pre-Op: The euDKA Blind Spot ASCs Must Fix

The Case That Changes How You Screen Forever

Consider a scenario that euglycemic DKA case reports describe again and again. A 58-year-old woman comes in for an elective abdominoplasty. She's not diabetic. Her pre-op bloodwork looks clean. Her glucose in holding is 102 mg/dL. You take her back, do a beautiful case, and send her to PACU. Six hours later — or maybe the next morning, after she's already home — she's in an emergency department with a pH of 7.18, a bicarbonate of 11, and ketones through the roof. Her glucose? Still 110.

That's euglycemic diabetic ketoacidosis. And her cardiologist had started her on empagliflozin three months ago for heart failure with preserved ejection fraction — not diabetes. Nobody asked. Your intake form had a checkbox for "diabetic medications." She checked no.

This is the blind spot. And in 2026, with SGLT2 inhibitor prescriptions expanding into non-diabetic cardiology and nephrology populations, and CMS adding cardiac and vascular procedures to the ASC Covered Procedures List, this scenario is no longer rare. It's a predictable systems failure — and it's one your pre-op workflow either catches or it doesn't.

What ADA 2026 Actually Says — And Why It Matters for Your ASC

The ADA's 2026 Standards of Care in Diabetes (Diabetes Care, Supplement 1) are unambiguous: SGLT2 inhibitors should be held for 3–4 days before elective surgery, and patients who undergo nonelective surgery without completing that hold should be closely monitored postoperatively for euglycemic DKA. That's not a soft recommendation anymore. That's a standard of care.

This matters operationally because a 3–4 day hold requires you to identify the medication at least four days before the scheduled procedure. That means your pre-op intake process — not your day-of nursing assessment, not the anesthesia provider's pre-procedure huddle — has to be doing this work. If your patient confirms their medication list 48 hours before surgery, you've already missed the window for a guideline-compliant hold.

The drugs we're talking about include empagliflozin (Jardiance), dapagliflozin (Farxiga), canagliflozin (Invokana), and ertugliflozin (Steglatro) — as well as combination products like Synjardy, Xigduo, and Glyxambi. A patient who doesn't think of themselves as diabetic may not connect the dots between their heart failure medication and what your intake form is asking.

The Six-Times Risk Nobody Is Talking About Loudly Enough

Here's the number that should be on a slide at every ASC quality committee meeting in the country: postoperative DKA incidence within 30 days is six times higher in SGLT2 inhibitor users compared to non-users. That's not a trend. That's a signal. And it comes from a population-based study cited directly by the Anesthesia Patient Safety Foundation in their June 2025 editorial on euglycemic ketoacidosis in the perioperative period.

A separate analysis, synthesized by Anecrit.com's 2026 perioperative guideline review, puts a finer statistical point on it: perioperative DKA incidence of 1.02 per 1,000 patients in SGLT2i users versus 0.69 per 1,000 in non-users, with an odds ratio of 1.48 (95% CI 1.02–2.15; p=0.037). These are not astronomical absolute numbers — but in the context of a condition that is routinely missed because glucose looks normal, an odds ratio approaching 1.5 is clinically significant and medically preventable.

The APSF is candid about the evidence quality: there is no Class 1 evidence governing perioperative SGLT2i management. We are operating in a guideline landscape that is informed by pharmacology, case series, and expert consensus. That makes institutional protocols — your ASC's own pre-op screening and decision logic — even more important, because you cannot rely on randomized trial data to bail you out if something goes wrong.

Why Euglycemic DKA Is So Dangerous in an ASC Setting

Standard ASC workflows are built around a glucose-centric model of diabetes risk. We check a fasting glucose. We note a hemoglobin A1c if the surgeon ordered one. We look for hyperglycemia as the danger sign. Euglycemic DKA breaks every one of those assumptions.

As the Frontiers in Endocrinology 2026 comprehensive review explains, euglycemic DKA is characterized by severe metabolic acidosis and ketosis in the absence of significant hyperglycemia — which makes it a diagnostic dilemma that can sail straight through a standard PACU workflow without triggering a single alarm. A patient can have a blood glucose of 130 mg/dL, a pH of 7.20, and a bicarbonate of 12 — and your PACU nurse's glucose check will look reassuring.

The physiological mechanism matters here: SGLT2 inhibitors promote glucosuria (glucose excretion in the urine) while simultaneously shifting metabolism toward fat oxidation and ketone production. Surgical stress, fasting, and relative insulin deficiency amplify ketogenesis. The body is producing ketones faster than it can clear them, but because glucose is being dumped renally, serum glucose stays deceptively low. Without a blood gas and a ketone level, you don't see it coming.

In an outpatient ASC, this is particularly dangerous because the patient may be discharged before the full metabolic picture develops. Symptoms — nausea, malaise, decreased appetite — can be attributed to post-anesthesia effects. By the time euDKA is recognized, the patient may be at home, or in an ED that doesn't immediately connect the dots to an elective procedure done hours earlier.

The Expanded Patient Population You're Not Screening For

Here's what's changed in the last two to three years that makes this a 2026 crisis rather than a 2019 footnote: SGLT2 inhibitors have become first-line therapy for heart failure with reduced ejection fraction, heart failure with preserved ejection fraction, and chronic kidney disease — regardless of whether the patient has diabetes. The Frontiers in Endocrinology review uses the word "ubiquitous" — these drugs are now ubiquitous in surgical patients, including those without diabetes.

That means the patient with heart failure getting a hernia repair is on dapagliflozin. The patient with stage 3 CKD coming in for a cosmetic procedure is on empagliflozin. They don't identify as diabetic. They may not even know the class of medication they're taking. And your intake form's diabetes medication checkbox does exactly nothing to surface that risk.

Now layer in what CMS did with the CY 2026 OPPS/ASC final rule. As Becker's ASC Review reported in 2026, CMS added 289 procedures to the ASC Covered Procedures List, including cardiac ablation and vascular procedures. These are precisely the patient populations — cardiac, cardiorenal — most likely to be on SGLT2 inhibitors for non-diabetic indications. The patients coming through your door are getting more complex. Your pre-op intake has to match that complexity.

From a pure business perspective: if your ASC is now doing cardiac ablation procedures billing at $15,000–$20,000 per case, and a preventable euDKA complication sends a patient to the ICU, you're looking at liability exposure that dwarfs the case revenue. For cash-pay cosmetic procedures — abdominoplasties, body contouring — billing $12,000–$25,000 per case, a single adverse outcome can generate a malpractice claim, a social media crisis, and a referral stream that evaporates overnight. The ROI on getting pre-op medication reconciliation right is not subtle.

What the Guidelines Actually Tell You to Do — and Where They Go Silent

The good news is that the major guideline bodies are aligned on the preoperative side. The ADA, and the clinical societies synthesized by Anecrit's 2026 perioperative review, all agree: minimum 3-day hold before elective surgery, with the ADA now formally specifying 3–4 days. That's your actionable preoperative instruction.

The bad news is what comes after surgery. A systematic international review of nine perioperative SGLT2i guidelines (PMC12464960, 2026) found little consensus on when to safely restart SGLT2 inhibitors postoperatively — and most guidelines offer no recommendation at all. When do you restart dapagliflozin after a laparoscopic cholecystectomy? After a rhinoplasty with significant fluid shifts? The literature doesn't give you a clean answer. That's a gap that institution-level protocols have to fill.

The APSF adds important nuance that a blunt "hold all SGLT2i" protocol misses: for non-diabetic heart failure patients, continuing SGLT2 inhibitors may actually be preferable from a cardiac standpoint, and the risk-benefit calculation is different. The APSF's management algorithm explicitly carves out this population. This means your pre-op protocol can't be a one-size checkbox — it requires clinical decision logic that distinguishes diabetic from non-diabetic indications and routes to appropriate consultation when the picture is complicated.

Closing the Gap: What Your ASC Pre-Op Workflow Needs Right Now

VeriOp.ai was built for exactly this failure mode: intake processes designed for a simpler pharmacological era. The medication landscape has changed faster than most paper-based or static EHR intake workflows can track. SGLT2 inhibitors in non-diabetic patients are exactly the kind of evolving risk that falls through the cracks of "do you take any diabetes medications?"

Here is what a compliant, risk-aware SGLT2i pre-op workflow needs to include in 2026:

• Medication-name–level screening — not just drug class. Patients on empagliflozin, dapagliflozin, canagliflozin, ertugliflozin, or combination products must be flagged regardless of the indication listed.
• Early enough in the pre-op timeline to action a 3–4 day hold. That means medication reconciliation completed at minimum 5–7 days before the scheduled procedure date.
• Indication-aware routing — differentiating the diabetic patient from the non-diabetic heart failure or CKD patient, because the clinical decision (hold vs. continue with monitoring) is genuinely different for each.
• Day-of confirmation that the hold was completed, not just ordered.
• PACU awareness flags for any patient who presents without completing the full hold — with a low threshold for blood gas and beta-hydroxybutyrate if nausea, fatigue, or tachycardia appear postoperatively.
• A documented postoperative restart protocol — even if it's simply "defer to prescribing physician for restart timing" — because the absence of any guidance is itself a liability gap.

VeriOp.ai's clinical decision support infrastructure is designed to meet criteria for Clinical Decision Support under the 21st Century Cures Act, meaning it functions as an intelligent layer on top of your existing workflow — surfacing SGLT2i flags, prompting the right questions at the right time, and generating documentation that shows your ASC exercised appropriate clinical diligence. If you want to see how that performs against real patient populations, our validation study has the data.

The underlying principle is simple: the patients coming through ASC doors in 2026 are pharmacologically more complex than they were five years ago. The intake process has to evolve with them. A drug that was once exclusively a diabetes medication is now a cardiology medication, a nephrology medication, and — in your pre-op queue — a medication that demands a specific, time-sensitive, indication-aware protocol. Building that protocol is not optional anymore. The ADA 2026 Standards of Care made it a published standard. The six-times risk data made it a patient safety imperative. And the expanding ASC procedure list made it a 2026 operational priority that every administrator, surgeon, and anesthesia provider in the outpatient space needs to take seriously right now.