Diabetes mellitus & glycemic management

SECTION 1

Historical Context & Motivation

Few diseases have shaped modern medicine as profoundly as diabetes mellitus. The term itself derives from the Greek diabetes ("to pass through") and the Latin mellitus ("sweet"), reflecting the ancient observation that the urine of affected individuals attracted insects. Ancient Egyptian papyri from 1500 BCE describe a condition of excessive urination and wasting, and Indian physicians of the same era noted the sweet taste of diabetic urine. For millennia, diabetes remained a uniformly fatal diagnosis, with no effective treatment and only rudimentary understanding of its pathogenesis. The story of how diabetes progressed from an inevitably lethal disease to one that can be managed—and its acute crises averted—is a landmark narrative in the history of endocrinology.

1889

Pancreatic Origin Identified

Oskar Minkowski and Joseph von Mering demonstrated that pancreatectomy in dogs produced diabetes, establishing the pancreas as the organ responsible for glucose regulation.

1921

Discovery of Insulin

Frederick Banting and Charles Best isolated insulin from canine pancreatic extracts at the University of Toronto. By January 1922, Leonard Thompson became the first human patient to receive insulin injections, transforming type 1 diabetes from a death sentence into a manageable condition.

1959

Type 1 vs. Type 2 Classification

Solomon Berson and Rosalyn Yalow developed the radioimmunoassay for insulin, enabling clinicians to distinguish between insulin-deficient (type 1) and insulin-resistant (type 2) forms of diabetes—a distinction that remains foundational.

1993

DCCT Proves Tight Control Matters

The Diabetes Control and Complications Trial (DCCT) demonstrated that intensive glycemic control in type 1 diabetes dramatically reduced microvascular complications, establishing HbA1c as a central therapeutic target and ushering in the era of evidence-based glycemic management.

2008–Present

Modern Pharmacotherapy Revolution

The EMPA-REG OUTCOME and LEADER trials demonstrated cardiovascular and renal benefits of SGLT2 inhibitors and GLP-1 receptor agonists, fundamentally changing the treatment algorithm for type 2 diabetes beyond glucose-lowering alone.

Today, diabetes affects over 537 million adults globally and remains one of the leading causes of cardiovascular death, end-stage renal disease, blindness, and non-traumatic limb amputation. The central clinical challenge—and the focus of this lesson—is threefold: How do we accurately diagnose diabetes? How do we recognize and manage life-threatening hyperglycemic emergencies such as diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS)? And how do we achieve safe glycemic control in the inpatient setting?

SECTION 2

Core Principles & Definitions

Before diving into diagnostic criteria and management algorithms, it is essential to establish the foundational concepts that underlie diabetes classification, pathophysiology, and glycemic targets. Diabetes mellitus is not a single disease but a heterogeneous group of metabolic disorders characterized by chronic hyperglycemia resulting from defects in insulin secretion, insulin action, or both. Understanding the interplay between these two mechanisms is the key to rationalizing both chronic management and acute crisis intervention.

Type 1 Diabetes Mellitus

Autoimmune destruction of pancreatic β-cells leads to absolute insulin deficiency. Patients are prone to DKA and require lifelong exogenous insulin. Anti-GAD, anti-IA2, and anti-ZnT8 antibodies confirm autoimmune etiology.

Type 2 Diabetes Mellitus

A spectrum of insulin resistance with progressive β-cell dysfunction. Accounts for ~90–95% of diabetes cases. Initially managed with lifestyle modification and oral agents; many patients eventually require insulin.

Gestational Diabetes Mellitus (GDM)

Glucose intolerance first recognized during pregnancy, typically in the second or third trimester. Driven by placental hormones (human placental lactogen) that induce insulin resistance. Confers long-term risk of type 2 diabetes in the mother.

HbA1c as a Glycemic Marker

Glycated hemoglobin reflects average blood glucose over 2–3 months (the lifespan of a red blood cell). An HbA1c ≥ 6.5% is diagnostic of diabetes. It is the primary metric for monitoring long-term glycemic control.

Anion-Gap Metabolic Acidosis in DKA

In the absence of insulin, unrestrained lipolysis generates free fatty acids that undergo hepatic β-oxidation to ketone bodies (β-hydroxybutyrate, acetoacetate). Their accumulation causes a high anion-gap metabolic acidosis—the hallmark of DKA.

✦ KEY TAKEAWAY

Think of insulin as the key and the insulin receptor on muscle and adipose tissue as the lock. In type 1 diabetes, the key factory (β-cells) is destroyed—there are no keys at all, so glucose accumulates outside the cells. In type 2 diabetes, the locks are rusty (insulin resistance)—the keys exist, but they cannot open the doors efficiently. Eventually the factory wears out too, producing fewer keys. DKA arises when the body, starved of keys, resorts to burning fat for fuel and generates acidic ketone by-products. HHS occurs when there are just enough keys to prevent ketosis but not enough to prevent massive hyperglycemia and osmotic diuresis.

SECTION 3

Visual Explanation — Diagnostic Criteria & Classification

The ADA recognizes four diagnostic pathways for diabetes mellitus. Note that a random glucose ≥ 200 mg/dL with classic symptoms does not require confirmatory retesting. The prediabetes range (green box) identifies patients at high risk who benefit from lifestyle intervention and possibly metformin.

The diagram above illustrates the four recognized diagnostic pathways according to the American Diabetes Association (ADA). A clinician may use any single criterion—fasting plasma glucose (FPG), 2-hour oral glucose tolerance test (OGTT), or HbA1c—provided the result is confirmed on a separate day. The exception is a random plasma glucose ≥ 200 mg/dL in a patient exhibiting classic symptoms of hyperglycemia (polyuria, polydipsia, unexplained weight loss), which is diagnostic without confirmation. It is important to recognize conditions that can falsely alter HbA1c: hemoglobin variants (HbS, HbC), hemolytic anemias, and chronic kidney disease may produce discordant results, in which case FPG or OGTT should be used instead.

💡 Clinical Pearl

The ADA recommends screening all adults aged ≥ 35, or earlier if overweight/obese with additional risk factors (family history, high-risk ethnicity, history of GDM, PCOS, hypertension, dyslipidemia, or HbA1c ≥ 5.7%). Screen every 3 years if results are normal.

SECTION 4

Pathophysiology — DKA vs. HHS

The two major hyperglycemic emergencies—diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS)—share the common thread of insulin deficiency but diverge in their pathophysiologic trajectories. In DKA, the insulin deficiency is typically absolute (or near-absolute), which unleashes counter-regulatory hormones (glucagon, cortisol, catecholamines, growth hormone). Glucagon drives hepatic gluconeogenesis and glycogenolysis, raising blood glucose, while simultaneously promoting lipolysis and hepatic ketogenesis. The resulting accumulation of β-hydroxybutyrate and acetoacetate produces a high anion-gap metabolic acidosis. In HHS, residual insulin secretion is sufficient to suppress ketogenesis but inadequate to control hyperglycemia, leading to profound osmotic diuresis, severe dehydration, and serum glucose levels that often exceed 600 mg/dL with serum osmolality > 320 mOsm/kg.

ANION GAP

AG = Na⁺ − (Cl⁻ + HCO₃⁻)

Normal AG ≈ 12 ± 4 mEq/L. In DKA, the AG is typically > 20 mEq/L due to ketoacid accumulation. Always calculate the delta-delta (ΔAG / ΔHCO₃⁻) to identify mixed acid-base disorders.

EFFECTIVE SERUM OSMOLALITY

Osm_eff = 2 × Na⁺ + Glucose/18

Effective osmolality excludes BUN (which crosses cell membranes freely). In HHS, Osm_eff > 320 mOsm/kg is diagnostic. Values > 340 correlate with obtundation and coma.

CORRECTED SODIUM

Na⁺_corrected = Na⁺_measured + 1.6 × ((Glucose − 100) / 100)

Hyperglycemia causes osmotic water shift from intracellular to extracellular space, diluting serum sodium. The corrected sodium reveals the patient's true sodium status and guides fluid management.

Key differentiating features of DKA vs. HHS

Feature	DKA	HHS
Typical diabetes type	Type 1 (can occur in type 2)	Type 2
Glucose	> 250 mg/dL (often 300–800)	> 600 mg/dL (often > 1000)
pH	< 7.30 (mild 7.25–7.30; mod 7.00–7.24; severe < 7.00)	> 7.30
Serum HCO₃⁻	< 18 mEq/L	> 18 mEq/L
Anion gap	Elevated (> 12)	Normal or mildly elevated
Ketones	Positive (serum β-hydroxybutyrate > 3 mmol/L)	Absent or trace
Osmolality	Variable (usually < 320)	> 320 mOsm/kg
Mental status	Alert → stupor (varies with severity)	Stupor → coma (correlates with osmolality)
Mortality	< 1% with appropriate management	5–20% (higher due to age, comorbidities)

⚠️ High-Yield Board Point

Euglycemic DKA (glucose < 250 mg/dL) can occur with SGLT2 inhibitor use, pregnancy, starvation, or heavy alcohol intake. Do not rule out DKA solely based on a normal glucose—always check the anion gap and ketones in the appropriate clinical context.

SECTION 5

Management of DKA, HHS & Chronic Diabetes

Acute Management: DKA & HHS

The management of both DKA and HHS centers on four pillars: aggressive intravenous fluid resuscitation, insulin therapy, electrolyte replacement (especially potassium), and identification and treatment of the precipitating cause. The mnemonic for common DKA/HHS precipitants is the "5 I's": Infection, Ischemia (MI/stroke), Intoxication, Insulin non-compliance, and Initial presentation of new-onset diabetes.

The DKA management algorithm emphasizes the critical sequence: fluids first, then potassium assessment before insulin. The most common and dangerous error is initiating insulin when K⁺ < 3.3 mEq/L, which can precipitate fatal cardiac arrhythmias. Resolution is defined by normalization of the anion gap—not merely by glucose reduction.

Chronic Management of Type 2 Diabetes

The foundation of type 2 diabetes management begins with lifestyle modification (dietary changes, 150 min/week moderate-intensity exercise, and 5–7% weight loss) combined with metformin as first-line pharmacotherapy. Metformin reduces hepatic glucose production, improves peripheral insulin sensitivity, and has a favorable weight and cardiovascular profile. Beyond metformin, second-line agent selection is now driven by comorbidities rather than glucose-lowering efficacy alone. Patients with atherosclerotic cardiovascular disease (ASCVD) should receive a GLP-1 receptor agonist (liraglutide, semaglutide, dulaglutide) with proven cardiovascular benefit. Those with heart failure or CKD should receive an SGLT2 inhibitor (empagliflozin, dapagliflozin, canagliflozin), which reduces hospitalizations for HF and slows CKD progression. When additional glucose-lowering is needed, options include DPP-4 inhibitors, thiazolidinediones, sulfonylureas, or insulin—tailored to the patient's glycemic target, hypoglycemia risk, weight concerns, and cost.

HbA1c target < 7% for most non-pregnant adults (ADA recommendation). More stringent targets (< 6.5%) may be appropriate in younger patients with short disease duration and no CVD. More relaxed targets (< 8%) are recommended in older adults, those with limited life expectancy, or those prone to severe hypoglycemia.
Insulin therapy in type 2 DM is indicated when HbA1c remains above target despite triple oral/injectable therapy or when glucose toxicity is prominent (HbA1c > 10%, glucose > 300 mg/dL, or symptomatic hyperglycemia at diagnosis). Basal insulin (glargine, detemir, degludec) is typically started at 10 units or 0.1–0.2 U/kg/day and titrated every 2–3 days.

SECTION 6

Worked Example — DKA Recognition & Initial Management

Case: 22-year-old with type 1 DM presents with nausea, vomiting, and abdominal pain × 12 hours

Step 1 — Gather Initial DataVitals: HR 118, BP 92/58, RR 28 (deep, Kussmaul breathing), T 38.2°C. Fingerstick glucose: 425 mg/dL. The patient reports running out of insulin 2 days ago. Labs are sent: BMP, VBG, serum β-hydroxybutyrate, CBC, UA, lipase, and blood cultures (febrile). On exam, the patient has dry mucous membranes, poor skin turgor, and a fruity odor on the breath.

Clinical picture consistent with DKA: hyperglycemia + signs of dehydration + Kussmaul breathing + fruity breath

Step 2 — Confirm DKA with Lab ResultsLabs return: Na⁺ 131, K⁺ 5.4, Cl⁻ 98, HCO₃⁻ 8, BUN 32, Cr 1.6, Glucose 438 mg/dL. VBG pH 7.12, pCO₂ 18 mmHg. Serum β-hydroxybutyrate 6.8 mmol/L. Calculate the anion gap: AG = 131 − (98 + 8) = 25 mEq/L. Calculate corrected Na⁺: 131 + 1.6 × ((438 − 100) / 100) = 131 + 5.4 = 136.4 mEq/L (eunatremic after correction, so use NS initially).

Severe DKA confirmed: pH 7.12, HCO₃⁻ 8, AG 25, β-OHB 6.8

Step 3 — Initiate IV FluidsBegin 0.9% normal saline (NS) at 1 L/hr for the first hour. The patient is estimated to have a 6–7 L fluid deficit. After the first hour, continue NS at 250–500 mL/hr. Monitor urine output closely (target > 0.5 mL/kg/hr).

Step 4 — Assess Potassium Before Starting InsulinK⁺ is 5.4 mEq/L. Although this appears elevated, remember that total body potassium is invariably depleted in DKA; the measured K⁺ is falsely elevated because acidosis drives K⁺ out of cells, and insulin deficiency prevents cellular uptake. Since K⁺ > 5.3, hold potassium replacement for now but recheck every 2 hours—K⁺ will drop rapidly once insulin is started.

K⁺ > 3.3 → safe to start insulin. Hold K⁺ supplementation; recheck q2h.

Step 5 — Start Insulin Drip & Ongoing MonitoringAdminister regular insulin IV bolus 0.1 U/kg (patient weighs 70 kg → 7 units), then start continuous infusion at 0.1 U/kg/hr (7 U/hr). Monitor hourly point-of-care glucose. Goal: glucose should drop 50–75 mg/dL per hour. When glucose reaches 200–250 mg/dL, switch IVF to D5½NS and reduce insulin infusion to 0.02–0.05 U/kg/hr to prevent hypoglycemia while continuing to close the anion gap. Do NOT stop insulin until all resolution criteria are met (pH > 7.30, HCO₃⁻ ≥ 15, AG < 12, glucose < 200, patient eating). When transitioning to subcutaneous insulin, administer the first SC dose 1–2 hours before discontinuing the drip.

Insulin drip 7 U/hr started. Glucose dropping at target rate. AG closing over 12 hours. Transition to SC insulin after resolution criteria met.

SECTION 7

Inpatient Glycemic Management

Hyperglycemia in hospitalized patients—whether from known diabetes, stress hyperglycemia, or steroid-induced glucose elevation—is independently associated with increased mortality, surgical site infections, length of stay, and ICU readmission. Achieving safe inpatient glycemic control requires understanding the distinction between ICU and non-ICU settings, the concept of basal-bolus-correction insulin protocols, and the dangers of both hyperglycemia and iatrogenic hypoglycemia.

Inpatient glycemic targets and regimen by clinical setting

Setting	Glycemic Target	Preferred Regimen
ICU (critically ill)	140–180 mg/dL (ADA/AACE). Avoid targets < 110 mg/dL (NICE-SUGAR showed increased mortality).	Continuous IV insulin infusion (regular insulin). Hourly glucose monitoring. Nurse-driven protocol.
Non-ICU (general ward)	Pre-meal < 140 mg/dL; random < 180 mg/dL. Avoid glucose < 70 mg/dL.	Basal-bolus SC insulin (preferred over sliding-scale-only). Basal (glargine/detemir) + rapid-acting (lispro/aspart) with meals + correction scale.
NPO patients (non-ICU)	Same targets. Adjust for lack of carbohydrate intake.	Basal insulin only (at ~50% of total daily dose). Hold prandial/correction. Add D5 IV if needed.

⚠️ Why Sliding Scale Alone Is Harmful

Sliding scale insulin (SSI) without basal insulin is a reactive approach—it treats hyperglycemia after it has already occurred rather than preventing it. Studies (RABBIT-2 trial) demonstrated that basal-bolus insulin achieved significantly better glycemic control than SSI alone in general surgery patients with type 2 DM, with no increase in hypoglycemia. SSI monotherapy should be avoided; it is acceptable only as a correction factor on top of a scheduled basal-bolus regimen.

✦ KEY TAKEAWAY

Think of inpatient insulin management like climate control in a building. Basal insulin is the thermostat—it maintains a steady baseline temperature (glucose). Prandial (bolus) insulin handles the heat spikes from opening doors (meals). And the correction scale is a manual override when the temperature unexpectedly drifts too high. Relying only on the override (sliding scale alone) means you are constantly playing catch-up in a building without a functioning thermostat—temperatures will swing wildly, and occupants (patients) will suffer.

When calculating the total daily dose (TDD) of insulin for a hospitalized patient, a reasonable starting estimate is 0.4–0.5 U/kg/day for insulin-naïve patients (reduce to 0.2–0.3 U/kg/day if elderly, CKD, or at risk for hypoglycemia). Divide the TDD: 50% as basal insulin and 50% divided equally among three pre-meal prandial doses. Add a correction scale for glucose above target, typically adding 1 unit for every 30–50 mg/dL above 150 mg/dL (sensitivity factor). Crucially, always hold oral hypoglycemics (especially metformin) in acutely ill inpatients due to risks of lactic acidosis, contrast nephropathy, and unpredictable absorption.

SECTION 8

Connections to Advanced Endocrine & Critical Care Topics

The principles of glycemic management extend well beyond the boundaries of diabetes alone. Understanding how diabetes intersects with critical care medicine, perioperative management, and cardiovascular risk reduction is essential for advanced clinical practice and board readiness.

Bridging foundational and advanced concepts in diabetes management

Foundational Concept (This Lesson)	Advanced Application
DKA management with IV insulin and potassium monitoring	Cerebral edema prevention in pediatric DKA: avoid overly aggressive fluid resuscitation and rapid glucose correction; target glucose drop ≤ 50–75 mg/dL/hr
SGLT2 inhibitors in type 2 DM	SGLT2i cardio-renal protection extends to non-diabetic HFrEF and CKD (DAPA-HF, DAPA-CKD trials). Counsel on risk of euglycemic DKA, genital mycotic infections, and Fournier gangrene
GLP-1 RA cardiovascular benefit	Semaglutide (oral and injectable) now approved for obesity (Wegovy) and has emerging data for MASH/NAFLD. Tirzepatide (dual GIP/GLP-1 RA) represents the next generation of incretin therapy
Inpatient basal-bolus insulin protocols	Steroid-induced hyperglycemia: anticipate postprandial-predominant spikes with prednisone (peak at 8–12 hrs) and adjust NPH or prandial insulin accordingly. Dexamethasone causes more sustained hyperglycemia → increase basal insulin
HbA1c as a diagnostic and monitoring tool	Continuous glucose monitoring (CGM) and time-in-range (TIR: 70–180 mg/dL > 70% of the time) are emerging as complementary and possibly superior metrics to HbA1c for assessing glycemic variability

Looking forward, the field of diabetes management is rapidly evolving. Closed-loop insulin delivery systems ("artificial pancreas") are increasingly available for type 1 DM, and dual agonists like tirzepatide are redefining glycemic and weight outcomes in type 2 DM. For Step 2 preparation, ensure you can confidently manage the acute crises (DKA and HHS), select appropriate pharmacotherapy based on comorbidity profiles, and implement safe inpatient insulin protocols—these represent the highest-yield clinical scenarios for board examinations.

SECTION 9

Practice Problems

PROBLEM 1 — CONCEPTUAL

A 45-year-old obese man has a fasting plasma glucose of 118 mg/dL and an HbA1c of 6.1%. How do you classify his glycemic status, and what is the most appropriate next step in management?

PROBLEM 2 — BASIC CALCULATION

A patient with DKA presents with Na⁺ 128 mEq/L, K⁺ 4.0 mEq/L, Cl⁻ 92 mEq/L, HCO₃⁻ 10 mEq/L, and glucose 520 mg/dL. Calculate the anion gap and the corrected sodium. What do these values tell you?

PROBLEM 3 — INTERMEDIATE

A 68-year-old woman with type 2 DM, HbA1c 8.4% on metformin 2000 mg/day, has a history of heart failure with reduced ejection fraction (LVEF 35%) and stage 3b CKD (eGFR 38 mL/min). What second-line agent would you add, and why? Would you continue metformin?

PROBLEM 4 — APPLIED

An 80 kg insulin-naïve patient with type 2 DM is admitted for elective hip replacement. His HbA1c is 9.2% and home medications include metformin and glipizide. Morning glucose is 285 mg/dL. Design an inpatient insulin regimen, including calculations for TDD, basal, prandial, and correction doses.

PROBLEM 5 — CRITICAL THINKING

A 55-year-old woman with type 2 DM on empagliflozin and metformin presents with nausea, vomiting, and abdominal pain. Her glucose is 165 mg/dL, but her VBG shows pH 7.18, HCO₃⁻ 12 mEq/L, anion gap 22, and serum β-hydroxybutyrate 5.2 mmol/L. What is the diagnosis? Why is this presentation atypical, and how does it change management? What should be done with the SGLT2 inhibitor?

SUMMARY

Diabetes Mellitus & Glycemic Management — Summary

Diabetes mellitus is diagnosed by any one of four criteria: FPG ≥ 126 mg/dL, 2-hr OGTT ≥ 200 mg/dL, HbA1c ≥ 6.5%, or random glucose ≥ 200 with symptoms. Type 1 DM results from autoimmune β-cell destruction (absolute insulin deficiency), while type 2 DM involves progressive insulin resistance and β-cell failure. First-line treatment for type 2 is metformin plus lifestyle modification, with second-line agent selection guided by comorbidities: GLP-1 RAs for ASCVD and SGLT2 inhibitors for HF/CKD.

The hyperglycemic emergencies—DKA (high AG metabolic acidosis, ketosis, glucose > 250) and HHS (profound hyperglycemia > 600, osmolality > 320, minimal ketosis)—are managed with IV fluids, insulin (after confirming K⁺ ≥ 3.3), and potassium replacement. DKA resolution requires normalization of the anion gap, not merely glucose correction. In the inpatient setting, basal-bolus insulin is preferred over sliding-scale-only regimens, targeting 140–180 mg/dL in the ICU and pre-meal < 140, random < 180 on the general ward. Remember the critical exceptions: euglycemic DKA with SGLT2 inhibitors, false HbA1c in hemoglobinopathies, and the importance of holding oral hypoglycemics in acutely ill hospitalized patients.