DKA and HHS: 2024–2026 Advances in Diagnosis and Management

Diabetic ketoacidosis (DKA) and the hyperglycemic hyperosmolar state (HHS) remain the two most dangerous acute metabolic crises in diabetes care—yet our clinical approach is evolving rapidly, driven by new consensus guidance, a rising incidence in many settings, and novel complications linked to modern therapies and technologies. (scirp.org)

Background / Overview​

DKA and HHS occupy opposite ends of a shared pathophysiologic spectrum: DKA is defined by insulin deficiency with ketosis and an anion-gap metabolic acidosis, while HHS is dominated by extreme hyperglycemia, marked hyperosmolarity and dehydration with little or no ketoacidosis. Both are medical emergencies that demand urgent, protocolized care—fluid resuscitation, careful insulin administration, and proactive electrolyte management—plus rapid identification and treatment of the precipitating cause. (scirp.org)
Recent, high‑profile consensus work and evidence reviews issued in 2024–2025 have clarified diagnostic thresholds, emphasized the practical value of bedside beta‑hydroxybutyrate measurement, and endorsed broader use of subcutaneous insulin regimens for selected, nonsevere DKA presentations. Those changes aim to make care safer, more resource‑sensitive, and more easily standardized across hospitals and wards. Clinicians and health systems must interpret these updates in the context of local resources, patient mix, and the demographic shifts that are altering the clinical epidemiology of hyperglycemic crises.

---

What the recent review adds: key takeaways from the SCIRP paper​

The review "Current Concepts in Diabetic Ketoacidosis and Hyperglycemic Hyperosmolar State" (Journal of Diabetes Mellitus, 2026) synthesizes contemporary literature on epidemiology, pathophysiology, diagnosis, and treatment and highlights several clinically important trends:

• DKA continues to be the commonest hyperglycemic emergency in younger or insulin‑dependent patients, whereas HHS is more frequent in older adults with type 2 diabetes and carries higher mortality. (scirp.org)
• Laboratory thresholds remain central to diagnosis: classically, DKA is associated with blood pH < 7.3, bicarbonate < 18 mEq/L and glucose ≥ 250 mg/dL; HHS typically involves glucose > 600 mg/dL and osmolarity > 320 mOsm/kg (though consensus groups have refined these cutoffs). (scirp.org)
• Management fundamentals—rapid intravascular volume restoration, insulin to suppress ketogenesis, potassium repletion guided by serial labs, and treatment of the precipitant—are unchanged, but operational details (e.g., insulin dosing, fluid rates, when to add dextrose, when to consider subcutaneous insulin) have been refined by recent consensus recommendations. (scirp.org)

The review also discloses that authors used generative AI tools (GPT‑5 and Microsoft Copilot 365) to optimize search strategies and prose. They state AI did not alter technical accuracy and that co‑authors performed final review. This transparent disclosure is appropriate, but it underscores a broader methodological point: when AI assists literature searches, reviewers and readers should demand reproducible search strings and clear statements about the nature and extent of AI contributions. (scirp.org)

---

Epidemiology: a shifting landscape​

Hospitalizations for hyperglycemic crises have not been static. The SCIRP review summarizes rising DKA incidence figures in U.S. inpatient data (for example, a progression from ~32 to ~61.6 cases per 10,000 admissions in the 2000s–2010s in cited datasets) and emphasizes that DKA remains concentrated in younger people and those with type 1 diabetes, while HHS predominates in older adults with type 2 disease and carries a substantially higher case mortality. (scirp.org)
These trends align with broader national surveillance showing growing diabetes prevalence, larger prediabetes pools, and increasing healthcare utilization for metabolic complications. In the United States, national estimates put the total number of people with diabetes in the tens of millions (CDC updated profiles show roughly 38–40 million depending on the reporting year) and near‑universal pressures on emergency services and inpatient units where acute metabolic decompensation is treated. Surveillance methods and case definitions vary across studies—so numeric trends should be interpreted with methodologic caution.

---

Pathophysiology distilled: why these emergencies look different​

Both DKA and HHS result from relative or absolute insulin deficiency combined with excess counter‑regulatory hormones (glucagon, catecholamines, cortisol, growth hormone). The difference lies in the metabolic response:

• In DKA, low insulin levels permit unchecked lipolysis and hepatic ketogenesis, producing high circulating ketone bodies (notably β‑hydroxybutyrate) and an anion‑gap metabolic acidosis. Inflammatory cytokines and oxidative stress appear to amplify the metabolic derangement and its vascular complications. (scirp.org)
• In HHS, partial insulin action is often sufficient to blunt overt ketogenesis but not to prevent extreme hyperglycemia and osmotic diuresis. The resulting dehydration, hyperosmolarity and concentration of blood solutes produce neurologic compromise and a higher risk of thrombosis and multisystem organ dysfunction. (scirp.org)

Understanding which physiology predominates at presentation drives the treatment focus: restore perfusion and renal function in HHS; suppress ketogenesis and correct acidosis in DKA—while recognizing that mixed pictures are common and treatment must be individualized.

---

Diagnosis: the role of beta‑hydroxybutyrate and updated criteria​

A major, pragmatic recommendation emerging from 2024–2025 consensus work is to incorporate quantitative bedside β‑hydroxybutyrate into routine DKA assessment. Quantitative ketone measurement enables more accurate diagnosis, staging and resolution‑decision making, and reduces reliance on less‑sensitive urine dipsticks or indirect criteria. The consensus document provides explicit cutoffs for severity (e.g., β‑hydroxybutyrate 3–6 mmol/L for mild–moderate and >6 mmol/L for severe DKA in some frameworks) and recommends that clinicians use clinical judgment in combination with these measures.
Practical diagnostic checklist clinicians should use on admission:

• Confirm hyperglycemia (note that SGLT2‑associated euglycemic DKA can present with lower glucose).
• Measure β‑hydroxybutyrate quantitatively if available.
• Obtain venous blood gases (or venous pH) and a basic electrolyte panel including corrected sodium, potassium, BUN/creatinine, and calculated effective serum osmolarity. (scirp.org)
• Assess mental status and hemodynamic stability to triage level of care (ward vs. ICU).

Caveat: some historical diagnostic cutoffs (anion gap thresholds, reliance on urine ketone strips) are being deemphasized in favor of direct ketone measurement and a more integrated severity assessment; local lab availability will shape real‑world practice.

---

Management fundamentals: fluids, insulin, potassium—what’s new and what hasn’t changed​

The SCIRP review reiterates the timeless pillars of management while aligning with consensus pragmatism:

• Initial fluid resuscitation should aim to restore intravascular volume quickly—commonly 500–1,000 mL of isotonic saline in the first hour or two—then titrate based on corrected sodium, urine output and hemodynamics. Balanced crystalloids are an acceptable alternative where clinically indicated. Avoid overly rapid correction of osmolarity to reduce neurological risk, especially in HHS. (scirp.org)
• Insulin therapy remains central: typical IV insulin infusion regimens for DKA use a 0.1 U/kg bolus followed by 0.1 U/kg/h infusion; in HHS, lower infusion rates (e.g., 0.05 U/kg/h) may be appropriate once fluids have begun to correct hyperosmolarity. Importantly, begin insulin only after ensuring serum potassium is >3.3 mEq/L. (scirp.org)
• Potassium must be managed proactively. Total body potassium is depleted in most patients despite normal or high serum concentrations at presentation; replacement is typically required once levels are below 5.0 mEq/L, and insulin should be withheld if potassium .3 mEq/L. (scirp.org)

What’s changed or emphasized recently:

• Quantitative β‑hydroxybutyrate is recommended for diagnosis and to guide resolution.
• For selected patients with mild‑to‑moderate DKA who are hemodynamically stable and without major comorbidities, subcutaneous rapid‑acting insulin protocols administered in monitored ward settings are now an accepted alternative to mandatory ICU admission and continuous IV insulin—this aims to reduce resource use without compromising safety when strict criteria and monitoring are in place. The consensus stresses careful patient selection and robust monitoring protocols.
• Avoid routine bicarbonate: bicarbonate therapy has no proven mortality benefit and can introduce harm (hypokalemia, reduced tissue oxygen uptake, theoretical risk for cerebral edema); reserve bicarbonate for profound acidosis (pH ≤ 7.0) and treat cautiously. (scirp.org)

---

Step-by-step practical protocol (high‑level)​

• Triage: determine DKA vs. HHS vs. mixed picture; assess airway, breathing, circulation, mental status.
• Initial labs: BMP, venous blood gas, quantitative β‑hydroxybutyrate, serum osmolality, CBC, and tests for precipitant (blood cultures, chest X‑ray, urinalysis). (scirp.org)
• Fluid resuscitation: isotonic crystalloid 500–1,000 mL in first hour; reassess and adjust. Add dextrose when glucose ≤ 250 mg/dL to continue insulin without hypoglycemia. (scirp.org)
• Potassium: verify serum K; if .3 mEq/L, delay insulin and replace potassium; if 3.3–5.0 mEq/L, give replacement to maintain 4–5 mEq/L. (scirp.org)
• Insulin: IV infusion 0.1 U/kg/h after K is safe (or subcutaneous rapid‑acting protocols for eligible mild cases).
• Monitor: glucose hourly (capillary), electrolytes and β‑hydroxybutyrate every 2–4 hours, fluid balance and mental status closely.
• Transition to maintenance: initiate basal‑bolus subcutaneous insulin 1–2 hours before stopping IV insulin; ensure consistent nutrition plan and outpatient follow‑up.

---

Special topics clinicians must watch​

SGLT2 inhibitors and euglycemic DKA​

SGLT2 inhibitors have transformed diabetes care with cardiorenal benefits, but they introduce a diagnostic and management hazard: euglycemic DKA, where substantial ketoacidosis occurs despite only modest hyperglycemia. Recognition requires high clinical suspicion, routine ketone testing in symptomatic patients on SGLT2i, and rapid cessation of the offending drug. Management follows DKA principles but may require relatively greater dextrose to allow higher insulin dosing while avoiding hypoglycemia; outcomes and optimal insulin/dextrose strategies remain areas of active study.

Neurologic complications and children​

Children with DKA are at risk for cerebral edema, particularly when fluid and osmolar corrections are rapid. Pediatric fluid strategies and careful osmolality change limits (and early neurology consultation when mental status declines) are essential. SCIRP and consensus summaries reiterate conservative correction targets and vigilant neurologic monitoring in pediatric DKA. (scirp.org)

Thrombosis risk in HHS​

HHS creates a hypercoagulable milieu; prophylactic anticoagulation during hospitalization is commonly recommended unless contraindicated, and clinicians should weigh bleeding risk carefully in elderly or comorbid patients. (scirp.org)

---

The technology angle: AI, devices and data in acute care​

Two technology threads intersect with hyperglycemic crises management:

• Diagnostic and therapeutic devices: bedside β‑hydroxybutyrate meters, point‑of‑care electrolytes, continuous glucose monitors (CGMs) and integrated pump systems have the potential to make inpatient management more precise and patient‑centered. CGMs, when used with appropriate hospital protocols, can reduce fingerstick burden and support tighter glucose tracking—though regulatory, interoperability, and reimbursement issues remain barriers for routine inpatient CGM use.
• AI and evidence synthesis: the SCIRP authors disclosed using GPT‑5 and Copilot 365 to refine search strategies and manuscript text. That practice is increasingly common and can improve efficiency, but it raises reproducibility and auditability questions: what exact prompts and search strings were used, and how were AI‑suggested changes validated? High‑quality reviews should publish reproducible methods and be explicit about the human checks that preserve scientific integrity. Clinicians relying on AI‑assisted syntheses should seek the raw search strategy or supplementary materials when possible. (scirp.org)

---

Critical analysis: strengths, limitations and risks​

Strengths

• The SCIRP review is concise, clinically oriented, and aligns with contemporary consensus recommendations; it synthesizes management principles in a usable bedside format. Its explicit disclosure of AI use demonstrates methodological transparency often missing in the literature. (scirp.org)
• Recent consensus guidance (2024) has moved the field forward pragmatically: encouraging quantitative ketone measurements, validating ward‑based subcutaneous insulin for selected DKA cases, and simplifying treatment focus to fluids, insulin and potassium—changes likely to improve resource allocation without compromising safety when implemented appropriately.

Limitations and risks

• Epidemiologic claims are sensitive to dataset and case‑definition choice. The SCIRP review cites rising incidence figures that align with some national datasets, but surveillance methods vary; readers should interpret absolute rates with caution and consult primary surveillance sources (e.g., CDC, national registries) when planning services or policy. (scirp.org)
• The adoption of subcutaneous insulin protocols outside ICUs depends heavily on hospital monitoring capacity and nursing ratios. Mismatched implementation—i.e., applying ward‑based protocols without rigorous monitoring—could delay recognition of deterioration. Systems must pair new protocols with training, checklists, and escalation pathways.
• The expanding use of SGLT2 inhibitors increases the incidence of euglycemic DKA and diagnostic delay. Healthcare providers and patients need explicit education: temporarily stop SGLT2i during acute illness, surgical procedures, or when oral intake is limited; have a low threshold for ketone testing in symptomatic patients on these agents. Evidence is still evolving about optimal dextrose/insulin balances in these cases.
• The use of generative AI in manuscript preparation—while disclosed—requires reproducible methods. Reviews that rely on AI for search optimization should publish search strings and prompt details as supplementary materials to allow independent verification. Where such transparency is absent, readers should treat AI‑assisted methodological claims with cautious scrutiny. (scirp.org)

---

Practical recommendations for clinicians and systems​

• Standardize local DKA/HHS order sets to reflect consensus priorities: include point‑of‑care β‑hydroxybutyrate when available, clearly defined potassium thresholds for insulin initiation, and explicit fluid/dextrose sequences.
• Develop strict eligibility criteria and monitoring protocols before adopting ward‑based subcutaneous insulin pathways for DKA. Training, nursing capacity, and scripted escalation criteria are essential.
• Update medication reconciliation and perioperative pathways to flag SGLT2 inhibitors; implement standardized instructions to stop SGLT2i before surgery or during acute illness, and educate patients about euglycemic DKA warning signs.
• Leverage technology responsibly: promote point‑of‑care ketone testing, consider inpatient CGM pilots with defined governance, and demand reproducibility when consuming AI‑assisted evidence syntheses. (scirp.org)

---

Conclusion​

DKA and HHS remain high‑stakes clinical emergencies—but in 2024–2026 the field has moved toward more pragmatic, evidence‑based, and resource‑sensitive care. Quantitative β‑hydroxybutyrate, clearer diagnostic frameworks, and selective use of subcutaneous insulin regimens represent practical advances that can improve patient flow and conserve ICU resources when implemented with appropriate safeguards. At the same time, new therapeutics (notably SGLT2 inhibitors) and the increasing use of AI in research and clinical workflows introduce new diagnostic and governance challenges that clinicians must manage proactively.
For practicing clinicians and hospital leaders, the immediate priorities are straightforward: update local protocols to reflect contemporary consensus, ensure staff education on SGLT2‑related euglycemic DKA, build robust monitoring for any non‑ICU DKA pathways, and insist on methodological transparency when reviews report AI‑assisted literature searches. When those pieces are in place, the clinical community can translate updated knowledge into safer, more efficient care for patients who present with these life‑threatening metabolic emergencies. (scirp.org)

---

Source: SCIRP Open Access Current Concepts in Diabetic Ketoacidosis and Hyperglycemic Hyperosmolar State