10.1 Diabetes Mellitus and Glycemic Control in Cardiac Care

Why Diabetes Is a Cardiac Diagnosis

Roughly one in three patients admitted with an acute coronary syndrome (ACS) carries a diagnosis of diabetes mellitus, and another large fraction has undiagnosed hyperglycemia discovered on admission. The CMC test plan lists diabetes under Non-Cardiovascular Conditions, but on the exam it is almost never tested as pure endocrinology — it is tested as a modifier of cardiac presentation, cardiac risk, and cardiac drug therapy.

Diabetes has historically been described as a coronary risk equivalent: the untreated 10-year event rate approximates that of a patient who has already had a myocardial infarction (MI). Contemporary guidelines have softened the absolute equivalence — risk varies with duration, albuminuria, and glycemic control — but the practical implication is unchanged. A patient with diabetes and any ischemic symptom is a high-risk patient until proven otherwise, and diabetes is a risk enhancer that pushes borderline and intermediate ASCVD risk toward statin therapy.

Atypical and silent ischemia

Cardiac autonomic neuropathy (CAN) is the mechanism behind the most heavily tested clinical pearl in this area. Damage to the afferent sympathetic fibers that carry anginal pain blunts or abolishes the classic substernal pressure. The diabetic patient with an evolving infarct is far more likely to present with:

• dyspnea or new orthostatic intolerance rather than chest pain
• nausea, vomiting, or epigastric discomfort mistaken for gastroenteritis
• profound fatigue, confusion, or a new fall in an older adult
• unexplained hyperglycemia or new DKA as the only signal of infarction

CAN also produces a resting tachycardia of 90 to 100 beats per minute, a fixed heart rate that fails to vary with respiration or position, orthostatic hypotension without a compensatory heart rate rise, and blunted heart rate response during exercise testing. Practically: a diabetic patient whose heart rate does not accelerate appropriately during a stress test has an abnormal chronotropic response, not a "good baseline." Silent ischemia means the nurse must rely on the monitor, the 12-lead, and troponin trends rather than on the patient's symptom report. Continuous ST-segment monitoring in the correct lead has more value in this population than in almost any other.

Worse outcomes and revascularization strategy

Diabetic patients have more diffuse, distal, multivessel disease with smaller reference vessel diameters, more calcification, more left main involvement, and a higher burden of microvascular disease. Consequences the CMC exam expects you to know:

• higher rates of in-stent restenosis and target-vessel revascularization after percutaneous coronary intervention (PCI)
• higher rates of heart failure, cardiogenic shock, and death after MI
• CABG is preferred over multivessel PCI in most diabetic patients with multivessel disease. The FREEDOM trial demonstrated lower death and MI with coronary artery bypass grafting compared with drug-eluting stents; ACC/AHA revascularization guidance gives CABG a stronger recommendation in diabetics with multivessel disease and suitable anatomy, particularly with a left internal mammary artery graft to the left anterior descending.
• diabetic cardiomyopathy: diastolic dysfunction and heart failure with preserved ejection fraction develop independently of epicardial coronary disease.

Inpatient Glycemic Control in the Cardiac ICU

The target, and the trials that set it

The current American Diabetes Association Standards of Care target for most critically ill hospitalized patients is a glucose of 140 to 180 mg/dL, with therapy initiated at a persistent glucose above 180 mg/dL. A tighter range of 110 to 140 mg/dL is acceptable only in selected patients if it can be achieved without significant hypoglycemia.

Know the history, because it is a favorite distractor. A landmark single-center surgical ICU trial reported a mortality benefit from intensive insulin therapy targeting 80 to 110 mg/dL, and tight control briefly became standard. The multicenter NICE-SUGAR trial then randomized more than 6,000 ICU patients and found higher 90-day mortality with the 81 to 108 mg/dL target compared with a target at or below 180 mg/dL, driven substantially by severe hypoglycemia. Tight glycemic control in critical illness is superseded practice. If an option says "titrate to 80 to 110 mg/dL," it is wrong.

Intravenous insulin infusion and the transition off it

Continuous IV regular insulin is the preferred approach for critically ill patients, patients on vasopressors, patients with unstable perfusion and unreliable subcutaneous absorption, and hyperglycemic emergencies. Practice points:

• Use a validated, nurse-driven or computerized titration protocol, not ad hoc dosing. Point-of-care glucose hourly until stable for 3 to 4 consecutive readings, then every 2 hours.
• Verify glucose with a laboratory or blood gas value when perfusion is poor, when the patient is on high-dose vasopressors, or when the point-of-care value conflicts with the clinical picture — capillary samples are unreliable in shock, edema, and severe anemia.
• Anticipate rising requirements with corticosteroids, dextrose-containing infusions, enteral or parenteral nutrition, and vasopressors; anticipate falling requirements when nutrition is held for a procedure. Never stop the tube feeding and leave the insulin infusion running unchanged.

Transition to subcutaneous therapy is the single most common source of iatrogenic hyperglycemia and DKA relapse. Administer the first dose of basal insulin 2 to 4 hours before discontinuing the drip (glargine or detemir need roughly 2 to 4 hours to take effect; a rapid-acting analog needs 1 to 2 hours of overlap). Estimate the 24-hour requirement from the average stable infusion rate over the previous 6 to 8 hours, multiply by 24, and give roughly 60 to 80 percent of that total as the subcutaneous dose, split about half basal and half prandial.

Basal-bolus versus sliding scale

A basal-bolus-correction regimen — a long-acting basal insulin, scheduled prandial coverage matched to nutrition, and a correction scale — is the standard for non-critically-ill inpatients. Sliding-scale insulin as the sole regimen is explicitly discouraged. It is purely reactive, it treats hyperglycemia only after it occurs, it produces oscillating glucose, and randomized inpatient data showed better control with basal-bolus and no excess hypoglycemia. If the patient is NPO, continue basal insulin (often at 50 to 80 percent of the usual dose) and hold prandial insulin — do not hold everything.

Hypoglycemia and the beta-blocker problem

Level 1 hypoglycemia is a glucose below 70 mg/dL; level 2 (clinically significant) is below 54 mg/dL; level 3 is severe hypoglycemia with altered mental status requiring assistance. Nearly every cardiac patient is on a beta blocker, which blunts the adrenergic warning signs — tremor, palpitations, tachycardia, anxiety — that normally alert the patient and the nurse. Diaphoresis is cholinergically mediated and is preserved, so unexplained sweating in a beta-blocked diabetic patient is hypoglycemia until a glucose disproves it. Non-selective agents can also impair hepatic glycogenolysis and prolong recovery.

Treat with 15 to 20 g of oral carbohydrate if the patient can swallow safely, or dextrose 50% 25 mL IV (12.5 g) if there is IV access and the patient is obtunded, or glucagon 1 mg IM/SC if there is no access. Recheck in 15 minutes and repeat until above 70 mg/dL, then give a complex carbohydrate and protein. Hypoglycemia is itself arrhythmogenic — catecholamine surge, QT prolongation, and hypokalemia — and in ACS it is independently associated with worse outcomes.

Hyperglycemic Emergencies

Precipitants are the same list a cardiac nurse should already own: infection (most common), missed insulin, myocardial infarction, stroke, pancreatitis, corticosteroids, and SGLT2 inhibitors.

The treatment sequence

1. Fluids first. Isotonic 0.9% sodium chloride at 15 to 20 mL/kg in the first hour (about 1 to 1.5 L in an average adult), then 250 to 500 mL/hr. After the first hour, use the corrected sodium to choose the next fluid: if corrected sodium is normal or high, switch to 0.45% sodium chloride; if it is low, continue 0.9%. Balanced crystalloids are an acceptable alternative and may close the gap faster with less hyperchloremic acidosis.
2. Check potassium before insulin. This is the highest-yield rule in the section. Total-body potassium is depleted by 3 to 5 mEq/kg even when the serum value looks normal or high, because acidosis and insulin deficiency drive potassium out of cells. Insulin drives it straight back in.
   • Potassium below 3.3 mEq/L: hold insulin and replace potassium at 20 to 30 mEq/hr until above 3.3. Starting insulin here can precipitate fatal arrhythmia or respiratory muscle weakness.
   • Potassium 3.3 to 5.2 mEq/L: add 20 to 30 mEq to each liter of fluid and start insulin.
   • Potassium above 5.2 mEq/L: withhold potassium, start insulin, and recheck every 2 hours.
3. Insulin. Regular insulin 0.1 unit/kg IV bolus then 0.1 unit/kg/hr, or 0.14 unit/kg/hr infusion with no bolus. Target a fall of 50 to 75 mg/dL per hour. When glucose reaches about 200 mg/dL in DKA (250 to 300 mg/dL in HHS), add dextrose to the infusion and reduce the insulin rate — you continue insulin to close the gap, not to lower the glucose.
4. Endpoint. Per the 2024 ADA/EASD/AACE/JBDS/DTS consensus, DKA is resolved when the plasma ketone falls below 0.6 mmol/L and the venous pH is 7.3 or above, or bicarbonate 18 mmol/L or above — not when the glucose normalizes. The consensus explicitly states the anion gap should not be used as a resolution criterion, because the large volumes of 0.9% sodium chloride used in resuscitation produce a hyperchloremic acidosis that holds the gap open after the ketosis has cleared. Overlap subcutaneous basal insulin 1 to 2 hours before stopping the drip. Stopping the infusion at a normal glucose with an open gap is the classic relapse mechanism.
5. Adjuncts. Bicarbonate only if pH is below 6.9. Phosphate replacement only if the level falls below 1.0 mg/dL or there is cardiac dysfunction, respiratory depression, or hemolysis — cardiac patients are exactly the group in whom this matters, but routine repletion causes hypocalcemia.

Cardiac-specific cautions

In a patient with heart failure with reduced ejection fraction, end-stage renal disease, or a recent infarct, the standard 15 to 20 mL/kg first-hour bolus can produce flash pulmonary edema. Give smaller aliquots (250 to 500 mL) with reassessment after each, use point-of-care ultrasound or filling pressures to guide, watch for a rising oxygen requirement and new crackles, and consider that dialysis-dependent patients may need very little volume at all. Correct potassium and glucose with continuous ECG monitoring — the combination of shifting potassium, magnesium, and phosphate is highly arrhythmogenic. Cerebral edema is chiefly a pediatric complication but argues against overly rapid osmolar correction in any patient.

Euglycemic DKA and SGLT2 inhibitors

Sodium-glucose cotransporter-2 (SGLT2) inhibitors — empagliflozin, dapagliflozin, canagliflozin, ertugliflozin — cause glycosuria and can produce a wide-anion-gap ketoacidosis with a glucose below 250 mg/dL. This is a trap in cardiac care because SGLT2 inhibitors are now foundational heart-failure therapy given to patients who may not even have diabetes. Suspect it in any acidotic patient on one of these drugs, and measure serum beta-hydroxybutyrate rather than urine ketones (the nitroprusside urine test detects acetoacetate and underestimates beta-hydroxybutyrate). Treatment is the same insulin-plus-dextrose approach, often requiring dextrose from the outset. Hold SGLT2 inhibitors 3 days before major or elective surgery (4 days for ertugliflozin) and during acute illness with poor oral intake.

Antidiabetic Drugs Through a Cardiac Lens

Long-term glycemic targets and discharge teaching

A general A1c target below 7 percent applies to most adults; a value of 7 percent corresponds to an estimated average glucose of about 154 mg/dL. Individualize upward to below 8 percent for patients with limited life expectancy, advanced complications, extensive comorbidity, hypoglycemia unawareness, or a long history of difficult control — pushing an 80-year-old with three-vessel disease and CKD to an A1c of 6.5 percent buys hypoglycemia, not longevity. Check A1c on admission for any cardiac patient with hyperglycemia and no recent value; it distinguishes stress hyperglycemia from previously undiagnosed diabetes and shapes the discharge regimen.

Nursing education at discharge should cover sick-day rules (never omit basal insulin, check glucose and ketones every 4 hours, maintain hydration), hypoglycemia recognition despite beta blockade, glucose meter or continuous glucose monitor technique, insulin injection site rotation, statin adherence, a blood-pressure goal below 130/80 mm Hg, foot inspection, and — the piece most often omitted — that SGLT2 inhibitors and GLP-1 agonists are being continued for cardiac benefit even if the glucose looks fine, so they should not be stopped by an outside prescriber.