Perioperative

Management of

Patients on SGLT-2

inhibitors

Jennifer Iyengar

Clinical Assistant Professor

Michigan Medicine

SGLT-2 inhibitors

SGLT2 is expressed in the proximal tubule of the kidney and

mediates reabsorption of glucose from the tubular lumen.

In individuals without DM, the filtered glucose load is less than the

maximal glucose transport capacity (∼180 g/day).

SGLT2 inhibitors block the reabsorption of filtered glucose from the

tubular lumen and thereby promote the renal excretion of glucose

SGLT-2 inhibitor

Indications

Glycemic

control in T2DM

Reduce major

cardiovascular

events in T2DM

with

established

cardiovascular

disease

Improve

cardiovascular

outcomes in

patients with

HFrEF and

HFpEF

Decrease the

risk of kidney

disease

progression in

patients with

T2DM and CKD

SGLT-2 inhibitors

Currently Approved

•Canaglifozin (Invokana)

•Dapagliflozin (Farxiga)

•Empagliflozin (Jardiance)

•Ertugliflozin (Steglatro)

•Bexagliflozin (Brenzavvy)

SGLT-2 inhibitors – Side Effects

• Urinary frequency

• UTI and pyelonephritis

• Fournier's gangrene (Necrotizing fasciitis of the perineum)

• Dehydration/hypotension/AKI

• Diabetic ketoacidosis

https://youtu.be/I4GfJHNaa

MQ?si=v9Av_pipO4NmqpTq

Raiten et al. Journal of Cardiothoracic and Vascular

Anesthesia, Volume 38, Issue 1, 2024, Pages 57-66,

https://youtu.be/I4GfJHNaa

MQ?si=v9Av_pipO4NmqpTq

El-Remessy AB. Diabetic Ketoacidosis Management: Updates and Challenges for Specific Patient

Population. Endocrines. 2022; 3(4):801-812. https://doi.org/10.3390/endocrines3040066

Euglycemic DKA

BG <200 mg/dL, ketonemia (serum β-

hydroxybutyrate≥3.0 mmol/L), and at

least one of the following criteria to

define EDKA:

1.Arterial pH≤7.3

2.Serum bicarbonate ≤18 meq/L

3.Anion gap >10.

Chow et al. BMJ Open Diabetes Research and Care 2023;11:e003666.

Euglycemic DKA

BG <200 mg/dL, ketonemia (serum β-

hydroxybutyrate≥3.0 mmol/L), and at

least one of the following criteria to

define EDKA:

1.Arterial pH≤7.3

2.Serum bicarbonate ≤18 meq/L

3.Anion gap >10.

Chow et al. BMJ Open Diabetes Research and Care 2023;11:e003666.

 Population-based cohort study of adults started

on SGLT2i use from 2013 to 2017

 Primary objective was to identify potential

predictors of diabetic ketoacidosis

 N=111,442 adults

 DKA

o N=475 (0.4%) Inpatient or Outpatient

o N=192 (0.2%) Inpatient

Society of General Internal Medicine 2021

Risk factors identified in SGLT2i-

associated DKA

Problem

Statement

Patients who present for surgery without

discontinuing their SGLT2i, whether due to

emergency circumstances or simply

forgetting to pause the medication, need

further risk assessment and monitoring

perioperatively given eDKA concerns.

Guideline

Scope

Adult patients >18 years

Both with and without diabetes

Undergoing surgery or

procedures at Michigan Medicine

across all surgical sites including

both elective and

urgent/emergent cases.

Guideline

Team

Anesthesia

Representatives

Ross Blank

David Garcia

Ellen Janke

Sathish Kumar

Pharmacy

Representatives

Simona Butler

Scott Ciarkowski

Endocrine

Representatives

Firdhous Abdul

Kather

Lynn Ang

Nazanene

Esfandiari

Jennifer Iyengar

Sima Saberi

Guideline Development - Our Goal

The overall goal is to mitigate risk of eDKA in high-risk patients

while avoiding unnecessary cancellations of low-risk patients.

Help surgical and anesthesia teams stratify which

patients/scenarios are high risk for development of euglycemic

diabetic ketoacidosis (eDKA) and advise appropriate diagnosis,

monitoring, and treatment strategies

Elective Procedures

SGLT-2 INHIBITORS

HELD APPROPRIATELY ->

PROCEED

ADVISE THE PATIENT

TO RESUME SGLT-2

INHIBITORS ONCE

THE DIET IS BACK TO

NORMAL.

Elective Procedures

IF IMAGING ONLY (CT,

MRI, PET) -> PROCEED

ALL ENDOSCOPIC

PROCEDURES ->

PROCEED

LOW-RISK PROCEDURES

BASED ON CLINICAL

JUDGMENT -> PROCEED

ADVISE THE PATIENT TO

RESUME SGLT-2

INHIBITORS ONCE THE

DIET IS BACK TO

NORMAL.

Elective

Procedures

Patient does not have diabetes -> proceed

• outpatient procedure -> Provide

instructions on signs and symptoms

of diabetic ketoacidosis upon

discharge. Hold medication until the

diet is back to normal.

• inpatient procedure -> Check VBG

in PACU/ICU. Follow the post-op

algorithm.

Elective

Procedures

Check VBG

If AG > 12 and bicarb ≤ 18, refer to ER for eDKA management.

If AG > 12 and bicarb is > 18, consider other causes of the

anion gap. If otherwise clinically stable resume diet. Provide

instructions on signs and symptoms of diabetic ketoacidosis

upon discharge

High-risk Criteria

 Surgery/procedure duration > 3-4 hours

 If patient not anticipated to resume oral intake post-op

 Diagnosis of T1D/LADA/diabetes due to pancreatitis or pancreatic surgery

 H/o DKA if known

 Pre-op HbA1c > 10%

 ASA status > 3

 Excessive alcohol intake

Person with DM & high-risk criteria -> postpone

Elective Procedures

DM, no high-risk criteria -> Check preop VBG

-- If AG ≤ 12 -> proceed

re-check VBG Q1-2 hr during the procedure, if AG

becomes >12, follow the blue box

. If AG stays ≤ 12

during the procedure, check VBG in PACU/ICU and

follow the post-op algorithm.

-- If AG > 12 -> postpone

If AG > 12 and bicarb ≤ 18, refer to ER for eDKA

management.

If AG > 12 and bicarb is > 18, consider other causes

of the anion gap. If otherwise clinically stable resume

diet. Provide instructions on signs and symptoms of

diabetic ketoacidosis upon discharge.

If you have a gap…

If AG > 12, look at bicarb.

If bicarb ≤ 18,

• Check stat serum ketones (beta-hydroxybutyrate)

• Start standard anesthesia insulin infusion protocol.

• 5% dextrose fluid at 50 ml/hr if BG 150-250 mg/dl*

• 10% dextrose fluid at 50 ml/hr if BG < 150 mg/dl*

• Monitor glucose checks Q1hr and VBG Q1-2hr while intra-op.

• Once in PACU/ICU, can transition to standard DKA protocol insulin drip with glucose

and VBG/ BMP monitoring per protocol until the anion gap closes. If serum ketones

come back negative, consider alternative causes of anion gap acidosis, and

discontinue insulin drip if clinically appropriate.

• Consult endocrine post-op: when the patient is admitted to the unit or the floor.

If bicarb > 18, Consider alternative causes of anion gap, and continue checking VBG Q1-2

hr. Consider sending ketones or starting an insulin drip if the etiology remains unclear.

* If appropriate for volume status/clinical scenario.

Urgent &

Emergent

Procedures

All urgent/emergent cases-> proceed

Check stat VBG:

If AG > 12  follow the blue box below

If AG ≤ 12  proceed with case and

monitor VBG Q1-2hr, if AG becomes > 12

then follow the blue box. If AG remains

< 12 during the procedure, follow the

post-op algorithm.

If you have a gap…

If AG > 12, look at bicarb.

If bicarb ≤ 18,

• Check stat serum ketones (beta-hydroxybutyrate)

• Start standard anesthesia insulin infusion protocol.

• 5% dextrose fluid at 50 ml/hr if BG 150-250 mg/dl*

• 10% dextrose fluid at 50 ml/hr if BG < 150 mg/dl*

• Monitor glucose checks Q1hr and VBG Q1-2hr while intra-op.

• Once in PACU/ICU, can transition to standard DKA protocol insulin drip with glucose

and VBG/ BMP monitoring per protocol until the anion gap closes. If serum ketones

come back negative, consider alternative causes of anion gap acidosis, and

discontinue insulin drip if clinically appropriate.

• Consult endocrine post-op: when the patient is admitted to the unit or the floor.

If bicarb > 18, Consider alternative causes of anion gap, and continue checking VBG Q1-2

hr. Consider sending ketones or starting an insulin drip if the etiology remains unclear.

* If appropriate for volume status/clinical scenario.

Post-operative management of patients who did not hold SGLT-2 inhibitors at least 3 days

(4 days for ertugliflozin) undergoing emergent/urgent or elective surgery/procedure

Planned Admission

Outpatient procedure

Surgery/procedure

 Check VBG on arrival to PACU/ICU

AG > 12

 eDKA unlikely

 Evaluate other etiologies of anion gap/

metabolic acidosis

 Check BMP/VBG Q6H until able to tolerate

oral intake **

 Consult endocrine post-op*** if clinical

suspicion remains high for eDKA

 Discharge

 Provide instructions on signs and

symptoms of diabetic ketoacidosis

Yes

Check serum ketone (beta-hydroxybutyrate)

Positive

Negative

 Oral intake* or enteral/parental

nutrition is tolerated

 Do not discharge

 Check VBG

Yes

AG ≤ 12, Bicarb > 18

 eDKA unlikely

 Monitor BMP/VBG Q6H until

able to tolerate oral intake*

o Start DKA protocol insulin drip

o Monitor glucose checks per protocol

o Do not discontinue insulin drip until the anion gap

is closed

o If the patient is on insulin as an outpatient, the drip

should not be discontinued without transitioning to

the appropriate subcutaneous insulin regimen

o Consult endocrine post-op ***

* Includes a clear liquid

diet as long as it contains

carbs.

** If there is prolonged

fasting post-op, space

out VBG to Q12-24 hror

sooner based on clinical

judgment. Stop checking

once the patient is 3-4

days after the last SGLT2

inhibitor use.

***When the patient is

admitted to the unit or

the floor.

Guideline Development - Our Goal

The overall goal is to mitigate risk of eDKA in high-risk patients

while avoiding unnecessary cancellations of low-risk patients.

Help surgical and anesthesia teams stratify which

patients/scenarios are high risk for development of euglycemic

diabetic ketoacidosis (eDKA) and advise appropriate diagnosis,

monitoring, and treatment strategies

Plan – Do – Check - Act

Baseline survey

October 2024

Guideline approved

November 2023

Follow-up survey

April 2024

Follow-up Survey (n=46)

How familiar are you with the Michigan Medicine Perioperative SGLT-2i

Guideline?

Not at all

Fairly

Slightly

Somewhat

Completely

Follow-up Survey (n=46)

How many times have you used the Michigan Medicine Guideline to help

with decision-making for a patient on an SGLT-2 inhibitor?

Never

3 or More

Once

Twice

Baseline Survey (n=55)

How many times in the last 3 months have you had a case canceled due to

a patient not properly holding their SGLT-2i

Never

3 or More

Once

Twice

Baseline Survey (n=55)

How many times have you proceeded with a case despite a patient not

properly holding their SGLT-2 inhibitor prior to surgery/procedure?

Never

3 or More

Once

Twice

Baseline Survey (n=55)

How confident are you in your ability to recognize/diagnose euglycemic

diabetic ketoacidosis (eDKA)?

Not at all

Fairly

Slightly

Somewhat

Completely

Baseline Survey (n=55)

How confident are you in your ability to treat euglycemic diabetic

ketoacidosis (eDKA)?

Not at all

Fairly

Slightly

Somewhat

Completely

Follow-up Survey

Before After Significant?

At least one case

cancellation

55% 22% p = <0.01

At least one case

proceeded

65% 76% p = 0.17

Felt “completely” or

“fairly” confident in dx

eDKA

33% 33% p = 0.91

Felt “completely” or

“fairly” confident in

eDKA management

22% 28% p = 0.28

Feedback

“I disagree with the guideline. The FDA very clearly states sedation cases or elective generals

should not proceed. We are putting patients at risk and favoring RVU generation over patient

safety.”

“I am comfortable monitoring/managing these patients in the OR. I am more concerned about

surgical teams managing or continuing to monitor on the floor. Surgical awareness seems low

from my discussions and interactions.”

“Challenging with ambulatory surgery center environment resources” and “We are unable to

follow the guidelines at EAA as we do not have a pharmacy to prepare insulin infusion,, etc.”

“The new changes are very helpful, thank you!”

University of

Pennsylvania

Raiten et al, Journal of Cardiothoracic and Vascular

Anesthesia, Volume 38, Issue 1, 2024,

Possible Guideline Validation Strategies

Evaluate SGLT-2i case

cancellation rates pre-

and post- guideline.

Evaluate the number

of elective cases

where the guideline

recommended

proceeding where

eDKA occurred.

Evaluate rates of eDKA

among

urgent/emergent

cases where SGLT-2i

could not be held and

assess risk

factors/contributors.

Evaluate all cases of

DKA/eDKA to

understand how many

involved SGLT2i.

Evaluate effect of

guidelines on costs

Thank You

QUESTIONS, COMMENTS,

CONCERNS?

JENNIFER IYENGAR

JMACD@UMICH.EDU