ECG Findings: Hypokalaemia

---

Key ECG Abnormalities

---

The Answer is A - Hypokalaemia

1. T wave flattening - reduced potassium gradient slows phase 3 repolarization
2. U wave prominence - most characteristic sign; U waves are always present normally but become tall and visible when K⁺ < 3.0 mEq/L
3. T-U fusion - at K⁺ < 2.5 mEq/L, T and U waves merge, creating the false appearance of QT prolongation

---

Clinical Pearl

• K⁺ 3.0-3.5: subtle T flattening
• K⁺ 2.5-3.0: prominent U waves
• K⁺ < 2.5: T-U fusion, ST depression, risk of arrhythmias (PVCs, VT/VF)

---

ECG Findings in Hypokalaemia - A Detailed Guide

Visual Reference

---

Progressive ECG Changes by Potassium Level

---

Finding-by-Finding Breakdown

1. T Wave Flattening

• What it is: The T wave (ventricular repolarisation) becomes progressively smaller and flatter
• Why it happens: Low extracellular K⁺ increases the potassium equilibrium potential, slowing phase 3 repolarisation of the action potential
• When it appears: Starts as early as K⁺ < 3.5 mEq/L
• Clinical significance: The first and most sensitive sign; often subtle and easy to miss

---

2. Prominent U Wave (Most Characteristic Finding)

• What it is: A positive deflection appearing after the T wave, best seen in leads V2-V3
• Why it happens: The U wave likely represents late repolarisation of the Purkinje fibres or mid-myocardial (M) cells. With hypokalaemia, this delayed repolarisation is exaggerated and becomes clearly visible
• When it appears: Becomes prominent at K⁺ < 3.0 mEq/L
• Rule of thumb: A U wave is "abnormally prominent" when its amplitude exceeds the T wave in the same lead

---

3. T-U Fusion and Apparent QT Prolongation

• What it is: The T wave and U wave merge into a single broad, biphasic hump - this is actually a QU interval, not true QT prolongation
• Why it matters: This is frequently misread as QT prolongation. The real QT interval (ending at the T wave) may be normal, but the fusion gives a falsely long measurement
• When it appears: K⁺ < 2.5 mEq/L
• Critical threshold: Once the actual QU/QT interval exceeds 500 ms, risk of torsades de pointes (TdP) increases 2-3 fold - this mandates continuous cardiac monitoring and IV potassium replacement

---

4. ST Segment Depression

• What it is: The ST segment dips below the isoelectric baseline
• Why it happens: Altered ventricular repolarisation shifts the ST segment down
• Clinical note: This is a non-specific finding but when combined with T flattening and U waves, it strongly points to hypokalaemia

---

5. Increased P Wave Amplitude / Prolonged PR Interval

• What it is: P waves become taller and broader; the PR interval may lengthen slightly
• Why it happens: Hypokalaemia alters atrial repolarisation and slows conduction through the AV node
• When it appears: More prominent with severe depletion (K⁺ < 2.5 mEq/L)

---

Summary Table

---

Arrhythmia Risk

• Prolongs the action potential duration (hyperpolarised resting membrane)
• Increases automaticity - ectopic pacemakers become more active
• Facilitates re-entry - heterogeneous repolarisation across the myocardium
• Sensitises to digoxin toxicity - digoxin inhibits Na⁺/K⁺-ATPase, and hypokalaemia potentiates this block dramatically

---

Clinical Pearl: The "0.3 Rule"

Morgan & Mikhail's Clinical Anaesthesiology, 7e | Rosen's Emergency Medicine, 10e

---

Screening - Complete Exam Guide

Sources: Park's Textbook of Preventive & Social Medicine | Henry's Clinical Diagnosis & Management by Laboratory Methods | Rosen's Emergency Medicine | Symptom to Diagnosis, 4e

---

PART 1: THE 2×2 CONTINGENCY TABLE (Foundation of Everything)

                    DISEASE +        DISEASE -
TEST POSITIVE    a (True Pos)    b (False Pos)    → a+b (all positives)
TEST NEGATIVE    c (False Neg)   d (True Neg)     → c+d (all negatives)
                    a+c              b+d              N (total)

• a = sick AND test positive ✓ (TRUE POSITIVE)
• b = healthy BUT test positive ✗ (FALSE POSITIVE - "Boy cried wolf")
• c = sick BUT test negative ✗ (FALSE NEGATIVE - "Cancer missed")
• d = healthy AND test negative ✓ (TRUE NEGATIVE)

---

PART 2: VALIDITY (Sensitivity & Specificity)

Sensitivity = a / (a+c)

• "Of all the sick people, how many did the test catch?"
• = True Positive Rate
• Low sensitivity = many false negatives (misses disease)

Specificity = d / (b+d)

• "Of all the healthy people, how many did the test correctly label as healthy?"
• = True Negative Rate
• Low specificity = many false positives (over-diagnoses)

Memory Mnemonics

The Trade-off (Critical Exam Point)

• Lowering the cutoff → sensitivity ↑, specificity ↓ (more people test positive, catch more disease, but more false alarms)
• Raising the cutoff → sensitivity ↓, specificity ↑ (fewer false alarms, but miss more cases)

• Cut-off at 120 mg/dL → high sensitivity (catches all diabetics) but low specificity (many false positives)
• Cut-off at 180 mg/dL → high specificity but misses early cases

---

PART 3: PREDICTIVE VALUES

Positive Predictive Value (PPV) = a / (a+b)

• "If a test is positive, what is the chance the person actually has the disease?"
• Rises with increasing prevalence

Negative Predictive Value (NPV) = d / (c+d)

• "If a test is negative, what is the chance the person truly does not have the disease?"
• Rises with decreasing prevalence

Classic Example from Park's Textbook

---

PART 4: LIKELIHOOD RATIOS (LR)

Positive Likelihood Ratio (LR+)

• "How much more likely is a positive test in a sick person vs a healthy person?"

Negative Likelihood Ratio (LR-)

• "How much more likely is a negative test in a sick person vs a healthy person?"

Interpretation Guide

---

PART 5: ROC CURVE & CUTOFF

• Y-axis = Sensitivity (True Positive Rate)
• X-axis = 1 - Specificity (False Positive Rate)

Key Rules

Choosing the Cutoff

• Move left on the ROC curve (raise cutoff) → higher specificity, lower sensitivity (use for confirmatory tests)
• Move right on the ROC curve (lower cutoff) → higher sensitivity, lower specificity (use for screening)
• Optimal cutoff = the point on the curve closest to the top-left corner (balances both)

---

PART 6: MULTIPLE TESTING (Serial vs Parallel)

Parallel Testing (Simultaneous)

• Both tests done at the same time
• Patient is positive if either test is positive
• INCREASES sensitivity (catches more disease)
• DECREASES specificity (more false positives)
• Use when: You must not miss disease (e.g., blood bank screening for HIV - test every unit with both ELISA and another test)

Serial Testing (Sequential)

• Second test done only if first is positive
• Patient is positive only if both tests are positive
• INCREASES specificity (fewer false positives)
• DECREASES sensitivity (miss some cases)
• Use when: You want to confirm disease before expensive/invasive treatment

Memory Trick

---

PART 7: LEAD TIME & TYPES OF BIAS IN SCREENING

Lead Time

Lead Time Bias

• Screened patients appear to survive longer after diagnosis because the clock starts earlier (at screening), not at symptoms
• But the actual date of death may be the same
• Falsely makes screening look beneficial when it is not

Length Time Bias

• Screening tends to detect slow-growing, indolent tumours (because fast-growing ones progress to symptoms between screening rounds and are missed by the screen)
• These slow-growing cancers have better prognosis anyway
• Falsely makes screened cancers look better prognosis than non-screened ones

Healthy Worker / Volunteer Bias

• People who come for screening are generally healthier, more health-conscious, and higher socioeconomic status
• Screened group appears healthier, making the screening programme look more effective than it is

---

PART 8: TYPES OF SCREENING

1. Mass Screening (Universal)

• Screen the entire population regardless of risk
• Example: Newborn blood spot screening (PKU, hypothyroidism, congenital adrenal hyperplasia) - every baby gets tested
• Use when: Disease is common, test is cheap, and benefit is universal

2. Selective / High-Risk Screening (Targeted)

• Screen only high-risk subgroups
• Example: Colonoscopy screening for colorectal cancer in patients over 45 with family history; BRCA testing only in women with family history of breast/ovarian cancer
• Use when: Disease is rare in the general population but common in a specific group; increases yield

3. Opportunistic Screening (Case Finding)

• Screening done during a routine visit for another reason - the patient presents for something unrelated and is screened incidentally
• Example: Blood pressure measurement during a consultation for sore throat; cholesterol check during diabetes review
• Advantage: No extra effort or cost; reaches people who would not attend a dedicated screening programme

4. Multiphasic Screening

• Multiple screening tests administered at the same time in one visit
• Example: A health check package including BP, blood sugar, lipids, ECG, BMI, and chest X-ray all in one visit
• Advantage: Cost-effective, convenient
• Disadvantage: Multiple testing increases false-positive rate (see below)

5. Syndromic / Sentinel Surveillance Screening

• Ongoing monitoring of a population for early warning signs of an outbreak
• Example: Influenza surveillance - monitoring emergency department visits for "flu-like illness" to detect early pandemic signal

---

PART 9: THE MULTIPLE TESTING PROBLEM

• 1 test: 5% chance of a false positive
• 5 tests: 1 - (0.95)⁵ = 23% chance of at least one false positive
• 10 tests: 1 - (0.95)¹⁰ = 40% chance of at least one false positive
• 20 tests: 1 - (0.95)²⁰ = 64% chance of at least one false positive

---

QUICK REVISION SUMMARY TABLE

---

TOP EXAM TIPS - AVOIDING CONFUSION

• Sensitivity = fixed property of the test
• PPV = depends on who you test (prevalence)
• Tip: A 99% sensitive HIV test in a low-risk population still has terrible PPV. Prevalence is everything for PPV.

• More time living with the diagnosis is not the same as living longer
• Only a reduction in mortality rate (not survival time from diagnosis) proves screening works

• Never get these backwards. Parallel = both nets out = catches more = sensitive.

• You can NEVER improve both at the same time (unless you get a better test)
• This is the whole point of the ROC curve - it shows all the trade-offs

• LR+ must be > 1 (positive test should increase disease probability)
• LR- must be < 1 (negative test should decrease disease probability)