HHS vs DKA: How to Differentiate the Two Diabetic Crises

Hyperglycemic hyperosmolar state (HHS) and diabetic ketoacidosis (DKA) are both dangerous diabetes emergencies driven by too little insulin and dangerously high blood sugar, but they diverge in one critical way: DKA produces a flood of acidic ketones that drops your blood pH, while HHS instead pushes blood sugar and concentration (osmolality) to extreme levels without significant ketone buildup. That single biochemical difference cascades into distinct symptoms, different typical patient profiles, and treatment protocols that share a skeleton but diverge in important details. The picture is further complicated by the fact that up to a third of patients show features of both conditions at the same time.

Why One Produces Ketones and the Other Does Not

The core distinction between DKA and HHS comes down to how much insulin your body still makes. In DKA, insulin is essentially absent. Without it, fat cells release fatty acids at a high rate, and the liver converts those fatty acids into ketone bodies, which are acidic. The acid accumulates, blood pH drops, and the cascade of symptoms that defines DKA takes hold. In HHS, a small amount of residual insulin remains in circulation. That trace of insulin is not enough to keep blood sugar under control, but it is enough to keep fat breakdown in check, which prevents the runaway ketone production seen in DKA.¹ This is why DKA is traditionally linked to type 1 diabetes, where the insulin-producing cells are destroyed, and HHS to type 2 diabetes, where the cells still produce some insulin but not nearly enough. The association is not absolute, though. People with type 2 diabetes can develop DKA, and mixed presentations are common.

How Doctors Tell Them Apart at the Bedside

Clinicians use a handful of lab values to separate DKA from HHS when a patient arrives in crisis. The diagnostic thresholds differ between U.S. and U.K. guidelines, but the general framework is similar.

DKA is defined by three features: elevated blood sugar (usually above 250 mg/dL, though sometimes lower), the presence of ketones, and a drop in blood pH below 7.3 or bicarbonate below 18 mEq/L. HHS, by contrast, is defined by very high blood sugar (above 600 mg/dL), elevated serum osmolality above 320 mOsm/kg, and the relative absence of significant ketones or acidosis. U.K. guidelines from the Joint British Diabetes Societies specify that HHS features include marked dehydration, glucose at or above 30 mmol/L (about 540 mg/dL), osmolality at or above 320 mOsm/kg, without significant ketonaemia (3.0 mmol/L or less) and a pH above 7.3 with bicarbonate at or above 15 mmol/L.²

In practice, these cutoffs describe opposite ends of a spectrum. DKA patients are “acid but not as concentrated,” and HHS patients are “concentrated but not acidic.” The blood sugar in HHS tends to be much higher than in DKA because the condition develops more slowly, often over days to weeks, while the body is still producing just enough insulin to prevent ketosis but not enough to stop glucose from climbing relentlessly.

The Overlap Problem

One of the most clinically important and underappreciated facts about these two conditions is that they frequently show up together. Up to one-third of patients present with mixed features of both DKA and HHS, meaning they have the high osmolality and extreme hyperglycemia of HHS alongside the acidosis and ketones of DKA.³ This overlap syndrome is not a mild middle ground; it is the most dangerous version. After adjusting for factors like age, sex, BMI, race, and comorbidities, patients with combined DKA-HHS had roughly 2.7 times the odds of dying in-hospital compared with patients who had one condition alone.⁴ The overlap pattern also complicates treatment, since protocols designed for pure DKA or pure HHS may not address both the osmolality crisis and the metabolic acidosis at the same time.

Symptoms and How They Differ in Real Life

Both conditions cause extreme thirst, frequent urination, weakness, and dehydration, but the timelines and dominant symptoms feel different to the person experiencing them. DKA tends to come on within hours to a day or two. The hallmark symptoms are nausea, vomiting, abdominal pain, and a distinctive rapid deep breathing pattern as the body tries to blow off carbon dioxide to compensate for the acid load. You may smell a fruity odor on the breath, which comes from the exhaled ketones. Mental status changes can occur but are not always prominent early on.

HHS develops more insidiously, often over a week or longer. Because the insulin deficiency is partial rather than total, the dramatic vomiting and abdominal pain of DKA are often absent. Instead, the slow climb in blood sugar draws water out of cells through osmotic shifts, leading to profound dehydration. The water deficit in HHS is typically much larger than in DKA. Altered consciousness is the defining neurological feature of HHS, and research suggests that serum osmolality, rather than blood sugar or ketones per se, is the key factor determining how alert or confused the patient is.⁵ Many patients with HHS present drowsy or obtunded, and in severe cases they can be comatose. This neurological presentation often leads families and first responders to suspect stroke rather than a diabetes emergency, which can delay appropriate treatment.

Who Gets Which Crisis

DKA classically strikes younger patients with type 1 diabetes, though it is increasingly seen in type 2 diabetes as well. Common triggers include missed insulin doses, new-onset type 1 diabetes (especially in children and young adults), infections, and physiological stress from illness or surgery. HHS tends to affect older adults with type 2 diabetes, often people who have limited access to fluids or who cannot drink enough to keep up with the massive urinary losses. Nursing home residents and elderly people living alone are at particular risk because they may not recognize or respond to mounting thirst. Infections, particularly pneumonia and urinary tract infections, are the most common precipitants for both crises. Certain medications, including corticosteroids and some antipsychotics, can also push blood sugar high enough to trigger either condition.

An important observation from studies comparing the two populations is that when both groups are hospitalized and compared head to head, DKA patients tend to be younger and have fewer comorbidities but can still face high mortality. In diabetes patients with heart failure, for instance, those admitted with DKA had roughly three times the adjusted odds of in-hospital death compared to those with HHS, likely because the acid burden places additional strain on an already compromised cardiovascular system.⁶

Mortality and Why HHS Is Often Deadlier

On its own, HHS historically carries a reported mortality rate between 10 and 20%, roughly ten times the mortality seen in DKA alone.⁷ Several factors explain this gap. HHS patients are typically older and have more underlying health problems. The degree of dehydration is usually more severe, and the slow onset means the condition has often progressed significantly before anyone seeks help. The profound changes in mental status also complicate recovery and increase the risk of aspiration, falls, and other secondary events. As noted above, the overlap syndrome is the deadliest combination, with mortality around 8% in one large analysis compared with roughly 3% for isolated DKA and 5% for isolated HHS.⁸

Treatment Similarities and Key Differences

The three pillars of treatment for both crises are the same: aggressive intravenous fluids, insulin, and electrolyte replacement. But the emphasis and timing differ. In DKA, the acidosis is the immediate life threat, so insulin is started promptly to shut down ketone production. In HHS, the dehydration and hyperosmolality are the primary dangers, so fluid resuscitation typically takes priority and insulin may be introduced more gradually. U.K. and U.S. guidelines differ on some specifics of this sequencing.⁹ Despite the severity of HHS, no prospective randomized trials have been conducted specifically for its treatment; management has largely been extrapolated from DKA research.¹⁰

One active area of research concerns which type of IV fluid works best. Normal saline (0.9% sodium chloride) has been the traditional choice, but balanced crystalloid solutions like Ringer’s lactate are gaining interest. A subgroup analysis from a cluster randomized trial found that patients with DKA who received balanced crystalloids resolved their DKA about four hours faster (median 13 hours vs. roughly 17 hours) and came off insulin drips sooner compared to those receiving normal saline.¹¹ A systematic review and meta-analysis similarly found that time to DKA resolution was probably longer in the saline group by about three and a half hours, with hospital stays close to one day longer.¹² However, a more recent meta-analysis of randomized controlled trials reached a less definitive conclusion, finding no statistically significant difference in time to DKA resolution between the two fluid types, though balanced crystalloids did result in lower chloride levels after resuscitation.¹³ The evidence is leaning toward balanced fluids but is not yet settled, and normal saline remains the standard in most emergency departments.

Complications That Differ Between the Two

Each condition carries its own set of feared complications. In DKA, particularly in children, cerebral edema is the most dreaded outcome. It occurs when treatment lowers blood sugar and osmolality too rapidly, causing fluid to shift into the brain. Research shows the swelling is primarily driven by the rapid reduction of plasma glucose and osmolality rather than by the acidosis itself.¹⁴ Certain factors at presentation increase the risk: lower initial carbon dioxide levels and higher blood urea nitrogen concentrations have been linked to cerebral edema, and treatment with bicarbonate was associated with roughly a fourfold increase in risk in one landmark study.¹⁵ This is one reason why bicarbonate administration in DKA is handled cautiously and is generally reserved for the most severe acidosis.

In HHS, the primary threats beyond the neurological effects of hyperosmolality include thromboembolic events. The severe dehydration and high blood viscosity in HHS have long been thought to raise clotting risk. However, a nationwide study in Taiwan found that after adjustment for confounders, patients admitted with HHS did not have a statistically significant increase in venous thromboembolism risk compared to those admitted without HHS.¹⁶ Some guidelines still recommend prophylactic anticoagulation for HHS patients, but the evidence base for this is thinner than many clinicians assume. Acute kidney injury from profound dehydration and rhabdomyolysis from immobility during altered consciousness are additional risks that are more pronounced in HHS than in typical DKA.

Euglycemic DKA and Why It Matters

One variant of DKA that blurs the usual diagnostic picture is euglycemic diabetic ketoacidosis (EDKA), where the blood sugar is only mildly elevated or even near-normal despite the presence of dangerous ketoacidosis. This has become more common with the rise of SGLT-2 inhibitor medications, which lower blood sugar by causing the kidneys to excrete glucose into the urine. The renal clearance of glucose in EDKA is roughly twice as high as in conventional DKA, which keeps blood sugar deceptively low while ketones build up unchecked.¹⁷

The danger of EDKA is that both patients and clinicians may be misled by the relatively normal glucose reading and prematurely rule out a metabolic crisis. The classic osmotic symptoms of DKA, like intense thirst and excessive urination, are also muted by the SGLT-2 inhibitor’s effect. Symptoms may include nausea, vomiting, fatigue, and shortness of breath, but they can come on more gradually than in classic DKA. A high index of suspicion is critical: anyone on an SGLT-2 inhibitor who develops unexplained nausea, vomiting, or breathing difficulty should have ketones checked even if their blood sugar looks reassuring.

Measuring Ketones to Differentiate the Two

Because the presence or absence of ketones is the defining line between DKA and HHS, how you measure ketones matters. Traditional urine dipstick tests detect acetoacetate, while point-of-care blood meters measure beta-hydroxybutyrate, which is the predominant ketone body during active DKA. Research has found a good correlation between the two tests at low ketone levels, but a poor correlation at high values, meaning that either test can rule out ketosis but the blood ketone test is more accurate at confirming ketoacidosis.¹⁸ A systematic review found evidence that blood beta-hydroxybutyrate testing reduces emergency department visits, hospitalizations, and recovery time compared to urine testing.¹⁹ For people with type 1 diabetes managing sick days at home, a blood ketone meter is a meaningfully better tool for catching DKA early.

Readmission and What Predicts It

Surviving the acute crisis is only part of the story. Readmission rates for both DKA and HHS are troublingly high. In one study, about 15% of patients were readmitted for severe hyperglycemia, and 46% were readmitted for any cause.²⁰ The readmission rate for hyperglycemia was similar between the DKA and HHS groups, but the predictors were different. In the DKA group, younger age, smoking, hypoglycemia episodes, higher osmolality, and hyperthyroidism were significant predictors of readmission. In the HHS group, the strongest predictor was a higher hemoglobin A1c level, suggesting that chronic poor glucose control is the dominant risk factor for HHS recurrence.

These differing risk profiles highlight why a one-size-fits-all discharge plan does not work. Research has proposed an “ABCD” scoring system for identifying high-risk patients: age, behavioral health comorbidities, insurance coverage, and drug or alcohol abuse. Patients scoring high on multiple factors are candidates for more intensive follow-up, including pre-discharge diabetes education, coordination with psychiatric providers, attention to medication affordability, and closer outpatient follow-up.²¹ For DKA specifically, many recurrent episodes are driven by insulin omission, whether because of cost, mental health barriers, or disordered eating behaviors that involve manipulating insulin doses. For HHS, inadequate access to fluids and social isolation in elderly patients are recurring themes.

When SGLT-2 Inhibitors Complicate Both Conditions

The growing use of SGLT-2 inhibitors across diabetes and heart failure has added a new layer of complexity to the DKA-versus-HHS question. These drugs lower blood glucose by blocking glucose reabsorption in the kidneys, but they also shift metabolism toward using fat for fuel. In conditions of physiological stress, that shift can tip into ketone overproduction. Because the glucose is being actively excreted, the blood sugar may stay below the traditional DKA threshold, producing the euglycemic DKA described earlier. The onset of symptoms can be more gradual than in classic DKA, and the absence of the dramatic blood sugar readings that usually trigger alarm means diagnosis can be delayed.²² Guidelines now recommend holding SGLT-2 inhibitors before planned surgery, during acute illness, and whenever a patient is unable to eat or drink normally. Patients taking these drugs should have access to a blood ketone meter and know when to use it, particularly during illnesses that cause vomiting or reduced food intake.