Diabetic ketoacidosis can and does occur in people with type 2 diabetes, and the rate has been climbing for years. In England, hospital admissions for DKA among adults with type 2 diabetes rose by roughly four percent per year between 1998 and 2013, a trend that has only continued with the broader use of certain medications and dietary patterns.1PubMed. Trends in Hospital Admission for Diabetic Ketoacidosis in Adults With Type 1 and Type 2 Diabetes in England, 1998-2013: A Retrospective Cohort Study The assumption that DKA is a “type 1 problem” leads to delayed diagnoses, missed warning signs, and worse outcomes for the millions of people living with type 2 diabetes who are genuinely at risk.
Why DKA Happens in Type 2 Diabetes
In type 1 diabetes, DKA occurs because the body produces almost no insulin at all. In type 2, the path to DKA is different but arrives at the same dangerous endpoint. The core mechanism is a state of relative insulin deficiency: the body still makes some insulin, but not enough to suppress the breakdown of fat into ketone bodies. When something tips the balance, whether it is a severe infection, a missed dose of medication, or physiological stress, insulin levels drop below the threshold needed to keep fat metabolism in check. The liver then converts free fatty acids into ketones faster than the body can clear them, and blood becomes dangerously acidic.2Practical Diabetes. Diabetic ketoacidosis in type 2 diabetes mellitus
This relative insulin shortfall is the reason DKA in type 2 diabetes tends to show up in particular circumstances rather than out of the blue. A person whose pancreas still produces moderate amounts of insulin may go years without any ketone trouble, only to tip into DKA when a stressor overwhelms their remaining insulin capacity.
The Most Common Triggers
Two triggers dominate the picture: stopping or missing diabetes medications, and infection. A study comparing DKA admissions in people with type 1 and type 2 diabetes found that about 69 percent of DKA episodes in the type 2 group involved discontinuation of medication, while infections were present in roughly 48 percent of type 2 DKA cases, compared to about 22 percent in type 1.3Archives of Internal Medicine. Diabetic Ketoacidosis in Type 1 and Type 2 Diabetes Mellitus: Clinical and Biochemical Differences These numbers overlap because some people face both triggers at once: an illness makes them skip meals and, with them, their oral medications or insulin, creating a perfect storm.
Beyond those two, other recognized triggers include:
- Surgery or trauma: The physiological stress response raises counter-regulatory hormones that oppose insulin.
- New-onset diabetes: Some people discover they have type 2 diabetes only when they land in the emergency department with DKA. In one urban hospital cohort, a quarter of all DKA admissions involved patients who had never been diagnosed with diabetes before, and new-onset presentations were more common in the type 2 group.4Metabolism. Admissions for diabetic ketoacidosis in ethnic minority groups in a city hospital
- Medications: Certain drugs, including corticosteroids and a class of cancer immunotherapy agents, can destabilize blood sugar enough to provoke DKA in someone with type 2 diabetes.
Ketosis-Prone Diabetes
Some people with type 2 diabetes are especially vulnerable to DKA, and the pattern is common enough to have its own name: ketosis-prone diabetes, sometimes called Flatbush diabetes after the Brooklyn neighborhood where it was first systematically described. People with this condition present in full-blown DKA, yet they test negative for the autoimmune antibodies that define type 1 diabetes. They tend to be middle-aged, overweight, and often male.5PubMed Central. Ketosis-Prone Diabetes (Flatbush Diabetes): an Emerging Worldwide Clinically Important Entity
The clinical course is peculiar. After the initial DKA crisis is treated with intensive insulin, many of these patients recover substantial beta-cell function and can eventually stop insulin entirely, managing their blood sugar with diet alone or diet plus oral medications. Their long-term trajectory looks like type 2 diabetes rather than type 1.6PubMed Central. Ketosis-Prone Diabetes (Flatbush Diabetes): an Emerging Worldwide Clinically Important Entity Researchers have developed a classification system that sorts patients who present with DKA based on two questions: do they have islet autoimmunity, and do they have beta-cell functional reserve? The resulting subtypes predict whether the person will need insulin long-term or will achieve remission.7PubMed Central. SYNDROMES OF KETOSIS-PRONE DIABETES
The remission story is genuinely striking. At the time of their DKA episode, patients with ketosis-prone type 2 diabetes show severe impairment of both insulin secretion and insulin action. But once blood sugar normalizes and they recover, their insulin secretion and sensitivity can improve to levels comparable to people with ordinary type 2 diabetes who never had DKA at all.8Endocrine Practice. Diabetic Ketoacidosis: A Common Debut of Diabetes Among African Americans With Type 2 Diabetes One case report documented a patient whose glucose tolerance nearly normalized after recovery from severe DKA, achieving blood-sugar control in the healthy range without any medication.9PubMed Central. A patient with ketosis-prone type 2 diabetes showing nearly normalized glucose tolerance after recovery from severe diabetic ketoacidosis
Who Is at Higher Risk
Ketosis-prone diabetes and type 2 DKA more broadly show clear demographic patterns. African American and Hispanic populations are disproportionately affected. One multiethnic cohort of 321 people with ketosis-prone diabetes found that 44 percent were African American and 40 percent were Hispanic. Hispanic patients were more likely to have preserved beta-cell function after their DKA episode, while African Americans were more likely to have type 2 rather than type 1 diabetes among DKA admissions.10PubMed. Characteristics of ketosis-prone diabetes in a multiethnic indigent community In another hospital study, 97 percent of DKA patients were African American or Hispanic, and about a third had type 2 diabetes.11Metabolism. Admissions for diabetic ketoacidosis in ethnic minority groups in a city hospital
These patterns reflect a combination of genetic susceptibility, differences in access to consistent care, and the likelihood of treatment interruption. Hispanic patients in one study were less likely to develop DKA from treatment non-compliance and more likely to have DKA triggered by an acute illness, suggesting the underlying pathways to crisis differ across groups.12PubMed. Characteristics of ketosis-prone diabetes in a multiethnic indigent community
SGLT2 Inhibitors and Euglycemic DKA
A newer and increasingly recognized trigger for DKA in type 2 diabetes is a class of medications called SGLT2 inhibitors. Drugs in this family, including empagliflozin, dapagliflozin, and canagliflozin, work by causing the kidneys to excrete excess glucose in urine. They are widely prescribed because of their benefits for heart failure and kidney disease in addition to blood sugar control. The problem is that by driving down blood glucose and altering the insulin-to-glucagon balance, they can create conditions favorable to ketone production even when blood sugar readings look reassuringly normal.13PubMed Central. Euglycemic diabetic ketoacidosis induced by SGLT2 inhibitors: possible mechanism and contributing factors
This phenomenon is called euglycemic DKA, meaning DKA with normal or only mildly elevated blood sugar. It is a diagnostic trap. Emergency physicians and intensive care teams rely heavily on high blood glucose as a red flag for DKA. When the glucose level is unremarkable, the diagnosis can be missed or delayed, leading to worse outcomes.14PubMed Central. Euglycemic diabetic ketoacidosis: A missed diagnosis Some hospitals have developed specific protocols for managing euglycemic DKA, typically using a fixed insulin infusion while titrating dextrose-containing fluids to prevent blood sugar from dropping dangerously low during treatment.15PubMed Central. Developing a Protocol for Management of Euglycemic Diabetic Ketoacidosis
Keto Diets and SGLT2 Inhibitors Together
The combination of an SGLT2 inhibitor and a very-low-carbohydrate or ketogenic diet is one of the riskiest setups for euglycemic DKA. The diet itself shifts the body toward ketone production, and the medication amplifies that shift. Case reports describe patients developing full-blown euglycemic DKA after just a single dose of an SGLT2 inhibitor while following a ketogenic diet, and others who developed it within a week of starting the diet while already taking the medication.16AACE Clinical Case Reports. Euglycemic Diabetic Ketoacidosis Caused by SGLT2 Inhibitors and a Ketogenic Diet: A Case Series and Review of Literature
A review of published cases found that over 90 percent of patients who developed euglycemic DKA while on an SGLT2 inhibitor and a ketogenic or very-low-calorie diet were hospitalized within four weeks of starting the diet. The time from starting SGLT2 inhibitor therapy to the DKA episode ranged from one day to a full year, and the vast majority of cases involved people with type 2 diabetes.17PubMed Central. Dietary Changes Leading to Euglycemic Diabetic Ketoacidosis in Sodium-Glucose Cotransporter-2 Inhibitor Users: A Challenge for Primary Care Physicians? If you take an SGLT2 inhibitor, talk to your prescriber before making any drastic dietary changes toward very low carbohydrate intake.
Other Medications That Can Trigger DKA
SGLT2 inhibitors get the most attention, but other drugs can push type 2 diabetes into DKA. Corticosteroids are a well-known culprit. They raise blood glucose and oppose insulin action across the board. A case of steroid-induced DKA in a 52-year-old woman with insulin-dependent type 2 diabetes who was treated with steroids during septic shock highlights how high-stress clinical settings compound the risk.18Journal of the Endocrine Society. MON-635 Case of steroid-induced DKA
Immune checkpoint inhibitors, a class of cancer drugs that work by unleashing the immune system against tumors, can also destroy pancreatic beta cells as collateral damage. One reported case involved a 79-year-old woman with previously well-controlled type 2 diabetes on metformin alone who developed recurrent DKA after her third cycle of pembrolizumab (Keytruda), ultimately becoming insulin-dependent.19Journal of the Endocrine Society. SAT-676 Pembrolizumab Induced Worsening Glycemic Control and DKA in Type 2 Diabetes Mellitus These cases are relatively rare but are becoming more common as checkpoint inhibitors are used more widely in oncology.
DKA in Adolescents With Type 2 Diabetes
As childhood obesity rises, so does type 2 diabetes in young people, and DKA is part of that picture. In a large consortium study, about 11 percent of youth with type 2 diabetes presented with DKA at diagnosis.20PubMed. Presentation of youth with type 2 diabetes in the Pediatric Diabetes Consortium Clinically, these young patients often have features that should point toward type 2 rather than type 1: obesity, a skin finding called acanthosis nigricans that signals insulin resistance, strong family history of type 2 diabetes, and no autoimmune markers. Yet the severity of the DKA itself can look identical to a type 1 presentation, making the initial distinction difficult.
The good news mirrors the adult ketosis-prone diabetes pattern. Case reports of obese Black adolescents who presented with DKA but were later confirmed to have type 2 diabetes showed that insulin could be stopped within months, and glycemic control was maintained on metformin alone. Follow-up testing confirmed adequate beta-cell reserve.21PubMed. Type 2 diabetes presenting as diabetic ketoacidosis in adolescence Getting the diagnosis right matters, because the long-term management strategy for type 2 diabetes is fundamentally different from type 1.
Worse Outcomes When DKA Hits Type 2
Here is something that surprises many people: when DKA does occur in type 2 diabetes, it tends to be more dangerous than DKA in type 1. A national six-year retrospective study found that mortality among type 2 DKA patients was about 0.9 percent, compared to 0.2 percent for type 1. Hospital stays were longer (roughly four days versus three), and hospital charges were higher.22PubMed. Disparities in Outcomes and Health Care Utilization for Diabetic Ketoacidosis Among Patients With Type 1 and Type 2 Diabetes Mellitus: A 6-Year National Retrospective Cohort Study A separate propensity-matched analysis confirmed the pattern, finding that type 2 diabetes carried over twice the adjusted odds of dying from a DKA episode compared to type 1, along with higher rates of septic shock and substantially greater healthcare costs.23PubMed. Comparing patient characteristics and outcomes in type 1 versus type 2 diabetes with diabetic ketoacidosis: a review and a propensity-matched nationwide analysis
The reasons are partly straightforward. People with type 2 diabetes who develop DKA are typically older and carry more comorbidities: heart disease, kidney impairment, obesity. The infections that often trigger their DKA episodes add another layer of severity. And because DKA is less expected in type 2 patients, there may be delays in recognition and treatment initiation that cost precious time.
The Overlap With Hyperosmolar Hyperglycemic State
Type 2 diabetes has its own classic metabolic emergency, hyperosmolar hyperglycemic state (HHS), characterized by extremely high blood sugar and severe dehydration without prominent ketone production. DKA and HHS are often described as opposite ends of a spectrum, but in practice, many type 2 patients present with features of both at once. In one Chinese hospital-based analysis covering six years, about 12 percent of type 2 diabetes patients admitted for a hyperglycemic emergency had combined DKA and HHS.24PubMed Central. Clinical profiles, outcomes and risk factors among type 2 diabetic inpatients with diabetic ketoacidosis and hyperglycemic hyperosmolar state: a hospital-based analysis over a 6-year period This overlap complicates treatment because the fluid management and insulin strategies differ somewhat between the two conditions.25PubMed Central. Overlap of diabetic ketoacidosis and hyperosmolar hyperglycemic state
Perioperative Precautions for SGLT2 Inhibitor Users
Surgery creates exactly the kind of metabolic stress that can push a vulnerable person into DKA. Add an SGLT2 inhibitor to the equation, and the risk of euglycemic DKA in the perioperative period becomes a genuine concern. Current guidance from the FDA, the American Heart Association, and most international bodies recommends stopping most SGLT2 inhibitors at least three days before elective procedures, and ertugliflozin at least four days before, because pharmacokinetic data show that about 72 hours is needed for less than one percent of the drug to remain in circulation.26British Journal of Anaesthesia. Perioperative management of patients taking sodium-glucose cotransporter 2 inhibitors: Society for Perioperative Assessment and Quality Improvement (SPAQI) multidisciplinary consensus statement Prevention strategies also include maintaining basal insulin through the perioperative period and monitoring for ketosis.27PubMed. Perioperative Diabetic Ketoacidosis in Type 2 Diabetes: Risk and Prevention in the Era of SGLT2 Inhibitors
The picture for emergency surgery is less clear-cut, because you cannot wait three days to operate. One large study found that preoperative SGLT2 inhibitor use in patients undergoing emergency surgery was not associated with a higher risk of postoperative DKA compared to non-users, a finding that may eventually lead to more relaxed guidelines around emergency settings.28JAMA Surgery. Preoperative SGLT2 Inhibitor Use and Postoperative Diabetic Ketoacidosis For now, the safest approach for elective procedures remains to stop these drugs on schedule and confirm the medication has been held before proceeding.
Testing for Ketones at Home
If you have type 2 diabetes and take insulin or an SGLT2 inhibitor, knowing how to check for ketones can provide an early warning. Urine ketone strips are cheap and widely available, but they only measure one type of ketone body and cannot tell you exactly how severe the situation is. Point-of-care blood ketone meters, which measure beta-hydroxybutyrate directly, are more informative. In emergency department testing, blood ketone measurement and urine dipsticks were equally sensitive for detecting DKA at about 98 percent, but the blood test was much better at ruling out DKA in people who merely had high blood sugar without acidosis.29PubMed Central. Diagnostic accuracy of point-of-care testing for diabetic ketoacidosis at emergency-department triage: β-hydroxybutyrate versus the urine dipstick For home use, a blood ketone reading above 1.5 mmol/L with symptoms like nausea, vomiting, abdominal pain, or rapid breathing is a signal to seek emergency care promptly.
The challenge with euglycemic DKA is that your glucose meter may read a perfectly normal number while ketones are climbing. This is why checking ketones rather than relying solely on glucose readings matters during illness, fasting, or any period of unusual dietary restriction, especially if you take an SGLT2 inhibitor.
The Financial Weight of Type 2 DKA
Beyond the clinical toll, DKA admissions are expensive. An analysis from the United Arab Emirates found that the average cost per DKA admission was roughly $12,300, but admissions requiring intensive care were about two and a half times more expensive than ward-only stays, averaging around $16,000 compared to about $6,100.30PubMed Central. Identifying the Healthcare Burden of Diabetic Ketoacidosis: Economic Analysis and Mitigation Strategies for Adult Patients in the United Arab Emirates In U.S. data, type 2 DKA admissions already carry higher average hospital charges than type 1 DKA admissions, and the propensity-matched analysis found an adjusted increase of about $6,000 in total hospital costs for type 2 versus type 1.31PubMed. Comparing patient characteristics and outcomes in type 1 versus type 2 diabetes with diabetic ketoacidosis: a review and a propensity-matched nationwide analysis Because type 2 diabetes affects far more people worldwide than type 1, even a small per-capita rate of DKA translates into a massive cumulative burden on health systems.