DKA vs HHS: How These Hyperglycemic Emergencies Differ

Diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS) are both life-threatening diabetes emergencies triggered by insulin problems, but they differ in one critical way: DKA floods the blood with acidic ketone bodies, while HHS drives blood sugar and dehydration to extreme levels without producing significant ketones. The split comes down to how much insulin the body still has available, and the populations they typically strike, the speed of onset, and the treatment priorities differ in ways that matter for survival.

Why One Produces Ketones and the Other Does Not

The fundamental distinction between DKA and HHS sits at the level of insulin. In DKA, insulin deficiency is severe, sometimes near-total. Without enough insulin, the body cannot use glucose for energy and starts breaking down fat at an accelerated rate. The liver converts those free fatty acids into ketone bodies, which accumulate in the blood and cause it to become dangerously acidic. In HHS, there is still some insulin circulating. It is not enough to move glucose into cells efficiently, so blood sugar climbs to extraordinary levels, but it is enough to put the brakes on fat breakdown and ketone production.¹

This is why HHS patients typically present with much higher blood sugar readings than DKA patients. Without ketoacidosis forcing the issue into a crisis sooner, HHS develops more slowly and insidiously. The gradual rise in blood sugar pulls water out of cells through osmotic pressure, producing severe dehydration. By the time an HHS patient reaches the hospital, their blood can be so concentrated that it thickens and flows poorly, putting them at risk for blood clots, strokes, and organ failure. The diagnostic threshold for HHS involves an osmolality of at least 320 mOsm/kg, blood glucose of at least 30 mmol/L (roughly 540 mg/dL), without significant ketones or acidosis.²

DKA, by contrast, tends to hit fast. Someone might go from feeling fine to vomiting, breathing rapidly, and smelling of acetone within hours. Blood sugar is elevated but often not nearly as high as in HHS, sometimes in the 300–500 mg/dL range rather than 600 or above. The real danger is the acid, not the sugar. The body’s attempt to compensate for that acid produces the characteristic deep, labored breathing pattern known as Kussmaul respiration, which you almost never see in HHS.

Who Gets Which

DKA has traditionally been linked to type 1 diabetes, while HHS has been associated with type 2. That framing is broadly correct but not precise. DKA is the classic presentation when type 1 diabetes is first diagnosed, and it remains a recurring threat for people with type 1 who miss insulin doses, get sick, or have insulin pump failures. But DKA absolutely occurs in type 2 diabetes as well, especially during severe illness or physiological stress. A hospital-based study of type 2 diabetes patients found that among 158 admissions for hyperglycemic emergencies, roughly 41% had DKA, about 47% had HHS, and 12% had a combination of both.³ That is a surprisingly even split, and it challenges the old textbook idea that type 2 patients only get HHS.

HHS tends to affect older adults, often people who were barely managing their diabetes or did not know they had it. Dehydration builds up over days to weeks, sometimes worsened by limited fluid intake in elderly people who live alone or have reduced thirst perception. Infections, particularly urinary tract infections and pneumonia, are the most common triggers for both conditions, followed by medication non-adherence and new-onset diabetes.

When Both Happen at Once

DKA and HHS are not mutually exclusive. A meaningful fraction of patients show up with features of both: high ketones and acidosis alongside extreme hyperosmolarity. That mixed picture, sometimes called DKA-HHS overlap, is more common than many clinicians expect. In the same hospital cohort mentioned above, 12% of type 2 diabetes admissions had combined DKA-HHS.⁴

This overlap matters because combined DKA-HHS carries worse outcomes than either condition alone. After adjusting for age, sex, BMI, race, and other health conditions, patients with combined DKA-HHS had roughly 2.7 times the odds of dying in the hospital compared to those with an isolated crisis. The same study found that treatment-related hypoglycemia (blood sugar dropping below 40 mg/dL during insulin therapy) was associated with a nearly five-fold increase in mortality, and severe hypokalemia (potassium dropping to 2.5 mEq/L or below) carried a similar risk.⁵ In other words, the complications of treatment can be as dangerous as the crisis itself.

Euglycemic DKA and Why It Gets Missed

One of the trickiest variants of DKA is the euglycemic form, where the blood is acidotic and full of ketones but blood sugar is near-normal or only mildly elevated. This runs directly against the expectation that DKA means sky-high glucose, and it leads to delayed diagnoses. The most common modern cause of euglycemic DKA is a class of diabetes medications called SGLT2 inhibitors, which are prescribed not just for diabetes but also for heart failure and chronic kidney disease.⁶

The mechanism is layered. SGLT2 inhibitors work by forcing the kidneys to excrete excess glucose in the urine, which keeps blood sugar low. But that glucose-lowering effect can trigger a drop in insulin secretion and a rise in glucagon, shifting the body’s metabolism toward fat burning and ketone production even when blood sugar looks acceptable.⁷ The usual red flag, a dramatically high glucose reading, never appears. Common triggers that push someone on an SGLT2 inhibitor into euglycemic DKA include fasting, dehydration, infection, and the stress of surgery.⁸ This is why many surgical centers now require patients to stop SGLT2 inhibitors several days before elective procedures.

Euglycemic DKA does not have a clean parallel in HHS. The defining feature of HHS is extreme hyperglycemia, so a “euglycemic HHS” would essentially be a contradiction in terms.

How Treatment Overlaps and Diverges

Both DKA and HHS require aggressive intravenous fluids and careful electrolyte monitoring. Beyond that baseline, their treatment priorities differ. In DKA, the central goal is to shut down ketone production, which means insulin. Continuous intravenous insulin is the backbone of DKA management, and it stays running until the ketones clear and the blood’s acid-base balance normalizes. In HHS, the immediate priority is rehydration. These patients are often profoundly volume-depleted, sometimes losing six to nine liters of fluid or more. Giving fluids alone can drop blood sugar substantially, and insulin is introduced more cautiously, since HHS patients are often exquisitely sensitive to it. Overly aggressive insulin in HHS risks crashing blood sugar and potassium levels with dangerous speed.

Potassium replacement is a priority in both conditions, but the dynamics differ slightly. Both DKA and HHS cause total-body potassium depletion even when the initial blood potassium level looks normal or high, because acidosis and insulin deficiency push potassium out of cells into the blood. Once insulin is given and the acid clears, potassium floods back into cells and blood levels can plummet. In a large retrospective study, hypokalemia at or below 3.5 mEq/L occurred in over half of patients during treatment.⁹ Most protocols add potassium to the IV fluids early unless the starting level is already elevated.

Point-of-care ketone meters have become increasingly important for monitoring DKA in real time. Rather than relying solely on urine ketone strips, which are slow and unreliable for tracking resolution, bedside blood ketone measurements allow clinicians to watch the ketone level fall and make faster decisions about transitioning off IV insulin.¹⁰ In HHS, ketone monitoring is less central since ketones are not the problem, but tracking osmolality and fluid balance takes its place.

The Fluid Type Debate in DKA

For decades, normal saline (0.9% sodium chloride) was the default fluid for treating DKA. It is cheap, available everywhere, and works. But a growing body of research has asked whether balanced crystalloids, which more closely resemble the electrolyte composition of blood, might be better. The concern with large volumes of normal saline is that it delivers a heavy chloride load, which can cause a condition called hyperchloremic metabolic acidosis, essentially replacing one type of acid problem with another.

The evidence is mixed but tilting in favor of balanced fluids. A subgroup analysis from a large cluster-randomized trial found that DKA resolved faster in the balanced crystalloid group, with a median time to resolution of about 13 hours compared to roughly 17 hours with saline, and the time on IV insulin was also shorter.¹¹ A separate systematic review of randomized trials similarly found that time to DKA resolution and hospital stay were probably longer in patients receiving saline, though the absolute difference in hospital stay was less than a day.¹²

However, a more recent meta-analysis of 11 trials found no statistically significant difference in time to DKA resolution, though it did confirm that balanced crystalloids led to lower post-resuscitation chloride levels.¹³ The picture right now is that balanced crystalloids probably offer modest advantages and are unlikely to cause harm, but the case for abandoning normal saline entirely is not airtight. Many emergency departments have quietly shifted toward balanced fluids as their default, while others continue using saline when balanced options are not readily stocked.

Why Bicarbonate Does Not Help as Expected

It seems intuitive that giving bicarbonate, a base, to someone with metabolic acidosis would speed recovery. In practice, this does not appear to work. A systematic review and meta-analysis found that bicarbonate therapy did not improve pH levels, did not shorten the time to resolution of acidosis, and did not affect potassium levels. Bicarbonate was actually associated with a marginally longer hospital stay, by roughly 14 hours on average.¹⁴ An earlier study comparing severely acidotic DKA patients who received IV bicarbonate against those who did not found no difference in time to resolution of acidosis (8 hours in both groups) or time to discharge, but patients who received bicarbonate required more insulin and more fluids in the first 24 hours.¹⁵

In children, bicarbonate carries an additional risk. Studies of pediatric DKA have found that bicarbonate treatment was associated with a roughly four-fold increase in the risk of cerebral edema, even after adjusting for how sick the child was on arrival.¹⁶ Cerebral edema is the most feared complication of pediatric DKA, and while its exact cause is debated, the severity of dehydration and acidosis at presentation, along with certain markers like elevated blood urea nitrogen and low carbon dioxide levels, are consistently identified as risk factors.¹⁷ Most guidelines now recommend against routine bicarbonate use in DKA, reserving it only for the rare patient with life-threatening acidosis (pH below about 6.9) who is hemodynamically unstable.

DKA in Pregnancy

Pregnancy shifts the body’s metabolism in ways that make DKA both more likely and harder to spot. The normal hormonal changes of pregnancy increase insulin resistance and lower the threshold at which ketosis begins, meaning pregnant individuals can develop DKA at blood sugar levels that would not be alarming outside pregnancy. Up to 30% of pregnant patients with DKA present with euglycemia, which makes prompt recognition difficult.¹⁸ The acidosis is dangerous not only for the mother but also for the fetus, which is sensitive to the pH and oxygenation changes that accompany DKA.

Treatment follows the same general principles as outside pregnancy, with insulin, fluids, and potassium replacement, but with the addition of continuous fetal monitoring and typically an intensive care setting.¹⁹ The stakes are higher and the timeline for intervention is shorter, making early recognition, even when blood sugar does not look dramatically elevated, essential.

The Financial Weight of Hyperglycemic Emergencies

Beyond the clinical differences, DKA and HHS carry substantial economic burdens, though the distribution of costs is not always intuitive. Among people with type 2 diabetes and severe kidney disease, DKA hospitalizations were associated with higher in-hospital mortality, more organ complications, longer stays, and markedly higher costs compared to HHS admissions.²⁰ In settings without severe kidney disease, the picture can flip: a cost analysis from a tertiary hospital found that HHS incurred the highest median direct medical costs of any hyperglycemic emergency, with diagnostic fees alone making up the largest expense category.²¹ HHS patients tend to be older with more comorbidities, which drives up testing and monitoring costs, and their longer path to dehydration often means prolonged ICU stays.

For patients and families, these costs are not abstract. The same cost analysis found that a single hyperglycemic emergency hospitalization consumed roughly 22% of the average annual family income in that population. That financial hit, combined with the recurrent nature of these crises, makes prevention strategies genuinely cost-effective at both the individual and health-system level.

Preventing Repeat Episodes

A significant proportion of DKA admissions are repeat events in the same patients. The triggers are depressingly consistent: missed insulin doses, inability to afford medication, infections that are not managed quickly, and a lack of clear guidance on what to do when sick. “Sick day rules” are structured instructions for people with diabetes on how to adjust insulin, monitor ketones, stay hydrated, and know when to seek emergency care during illness. An intervention study focusing on intensive sick day education for young people with type 1 diabetes found that structured teaching led to a decrease in total DKA admissions in that group.²²

For HHS prevention, the strategy is different because the patient population is different. Many HHS cases occur in people who did not know they had diabetes or who have cognitive impairment, limited mobility, or inadequate social support. Prevention for HHS looks less like diabetes self-management education and more like ensuring that vulnerable older adults have access to fluids, regular medical check-ups, and someone who will notice if they are becoming confused or dehydrated over a period of days. Both conditions, in the end, are often preventable, but the prevention has to match the person and the circumstances that lead to each crisis.