ICU · Equipment, physics & clinical measurement
ECG & Electrical Safety
Also known as ECG · Electrocardiogram · Electrical safety · Microshock · Macroshock · Line isolation monitor · Equipotential earthing · CF equipment
ECG and electrical safety for the ICU First Part: how the 12-lead ECG records cardiac potentials (Einthoven's triangle, limb and precordial leads, electrical axis), the signal path and filtering, and the principles of electrical safety - macroshock vs the much lower microshock threshold when current reaches the heart through a central line, and the protective measures (isolation, equipotential earthing, CF equipment).
On this page & tools
Your progress
Saved locally on this device.
8 MCQs with explanations
Target exams
Overview
The ECG records the small electrical potentials (about 1 mV) generated by cardiac depolarisation and repolarisation, via electrodes on the body surface. The same monitoring wires and conductive patient connections create the central electrical-safety hazard of the ICU: a tiny current delivered directly to the heart can be lethal.[1]


The cardiac conduction system — where the ECG signal comes from
Cardiac electrical activity begins in the sinoatrial (SA) node at the junction of the right atrium and superior vena cava. The impulse spreads through atrial myocardium (and the Bachmann bundle to the left atrium), depolarising both atria and producing the P wave. It reaches the atrioventricular (AV) node in the floor of the right atrium, where conduction is deliberately slowed (the PR segment corresponds to this AV-nodal delay, ~0.12-0.20 s) — long enough to allow complete atrial emptying before ventricular contraction. From the AV node the impulse enters the bundle of His, then divides into the right bundle branch (to the right ventricle) and the left bundle branch (further dividing into anterior and posterior fascicles to the left ventricle), and finally ramifies through the Purkinje network to depolarise ventricular myocardium rapidly and synchronously, producing the QRS complex (~0.06-0.10 s). Ventricular repolarisation then produces the T wave, and (in some patients) the atrial repolarisation wave is buried within the QRS.[1][1]
Each cycle therefore generates a predictable sequence of waves and segments on the surface ECG. Knowing which depolarising/repolarising event each deflection represents is the foundation of ECG interpretation in the ICU. [1]
The ECG waves, segments, and intervals — what each represents
| Deflection | Event | Normal duration / morphology |
|---|---|---|
| P wave | Atrial depolarisation (right then left atrium) | < 0.12 s, upright in I/II/aVF; inverted in aVR |
| PR interval | Onset of P to onset of QRS (AV-nodal conduction time) | 0.12-0.20 s (3-5 small squares) |
| PR segment | Isoelectric delay as impulse traverses the AV node and bundle of His | Baseline |
| QRS complex | Ventricular depolarisation (Q, R, S waves) | 0.06-0.10 s (1.5-2.5 small squares) |
| ST segment | Ventricular depolarisation complete, before repolarisation begins | Isoelectric at baseline; elevation/depression is ischaemia until proven otherwise |
| T wave | Ventricular repolarisation | Upright, asymmetric (upstroke slower than downstroke) |
| QT interval | Onset of QRS to end of T (ventricular depolarisation + repolarisation) | < 0.44 s; rate-corrected QTc < 0.44 s (men), < 0.46 s (women) |
| U wave | (Probable) Purkinje or papillary-muscle repolarisation | Small, follows T; prominent in hypokalaemia |
The cardiac action potential and the surface ECG
The surface ECG is the algebraic sum, at any instant, of all myocardial cellular action potentials. A working ventricular myocyte action potential has five phases: rapid upstroke (phase 0, sodium influx), brief early repolarisation (phase 1), the plateau (phase 2, calcium influx balanced by potassium efflux — the long refractory period that prevents tetany), repolarisation (phase 3, potassium efflux), and the resting membrane potential (phase 4). The QRS corresponds to phase 0 (depolarisation), the isoelectric ST segment to phase 2 (the plateau, during which all cells are at roughly the same potential so the net vector is near zero), and the T wave to phase 3 (repolarisation).[1]
This is why ST-segment change is the electrocardiographic signature of ischaemia: injury to the myocardium shifts the cells' resting and plateau potentials, creating a voltage gradient between injured and normal tissue — a direct-current injury current that displaces the ST segment up or down between beats. Diffuse subendocardial injury produces ST depression; transmural (epicardial) injury produces ST elevation. The QT interval reflects total ventricular action-potential duration; anything that prolongs it (hypocalcaemia, hypokalaemia, hypomagnesaemia, class IA/III drugs) prolongs repolarisation and predisposes to early after-depolarisations and torsades de pointes. [1]
The 12-lead ECG
- Limb leads from the right arm, left arm, and left leg electrodes form Einthoven's triangle: lead I (LA-RA), lead II (LL-RA), and lead III (LL-LA). The augmented leads aVR, aVL, and aVF are derived from the same electrodes. The right leg electrode grounds the circuit and reduces interference.[1]
- Precordial leads V1-V6 are placed across the chest wall, each a different spatial view of the heart; together the 12 leads give twelve projections of the same cardiac electrical event.[1]
- The electrical axis is the net direction of depolarisation (normal about -30 to +90 degrees), assessed from leads I and aVF. Lead II most closely follows the normal depolarisation vector and is the usual monitoring lead.[1]
Limb electrode placement (the four electrodes)
A standard 12-lead ECG uses ten electrodes but only four limb electrodes, with the remaining six being precordial chest electrodes. The limb electrodes are placed as follows:[1]
- RA (right arm) — on the right arm, just below the lateral clavicle or on the right wrist. In practice, electrodes may be placed on the shoulders, upper arms, or anywhere along the limb without changing the vector (the torso is electrically equivalent to the limb tip in a limb lead).
- LA (left arm) — mirror image of RA on the left side.
- LL (left leg) — on the left leg or left lower abdomen/iliac crest.
- RL (right leg) — the reference/ground electrode. It drives a driven right-leg circuit that actively cancels common-mode mains interference; it is NOT a sensing electrode, and an incorrect RL placement (or a disconnected RL) produces 50/60 Hz mains noise. [1]
Einthoven's triangle and the law
The three standard limb leads (I, II, III) form Einthoven's equilateral triangle, with the heart conceptually at the centre: [1]
- Lead I = potential at LA minus RA = LA - RA (views the lateral wall, axis 0 degrees).
- Lead II = potential at LL minus RA = LL - RA (views the inferior wall, axis +60 degrees).
- Lead III = potential at LL minus LA = LL - LA (views the inferior wall, axis +120 degrees). [1]
Einthoven's law is the algebraic identity that follows: lead II = lead I + lead III. This is the basis for verifying electrode misconnection — if lead I + lead III does not equal lead II, the electrodes are placed or connected incorrectly.[1][1]
Augmented limb leads and the hexaxial reference system
The three augmented leads (aVR, aVL, aVF) are derived by comparing one limb electrode to a "Wilson central terminal" formed by averaging the other two. Because the average is augmented (multiplied by 1.5) to compensate for the divisor, they are called "augmented". They fill in the angles between the standard limb leads, giving six views of the frontal (coronal) plane at 30-degree intervals: [1]
- aVR — positive pole toward the right shoulder, axis -150 degrees (or +210). Normally aVR is NEGATIVE (the depolarisation vector moves away from the right shoulder).
- aVL — positive pole toward the left shoulder, axis -30 degrees. Views the lateral wall.
- aVF — positive pole toward the feet, axis +90 degrees. Views the inferior wall. [1]
Together leads I, II, III, aVR, aVL, aVF form the hexaxial reference system — six axes radiating through the heart in the frontal plane at 30-degree intervals, used to calculate the mean frontal QRS axis.[1]
Precordial (chest) lead placement — the horizontal plane
The six precordial (V) leads are placed in a single horizontal plane around the left hemithorax, each positive electrode connected to the Wilson central terminal. Each views a different part of the left ventricle in the horizontal (transverse) plane: [1]
- V1 — 4th intercostal space, RIGHT sternal border (septal; small R, deep S normally).
- V2 — 4th intercostal space, LEFT sternal border.
- V3 — midway between V2 and V4.
- V4 — 5th intercostal space, midclavicular line (apex; tallest R).
- V5 — same level as V4, anterior axillary line (lateral).
- V6 — same level as V4, midaxillary line (lateral). [1]
V1-V2 are septal, V3-V4 are anterior, V5-V6 are lateral. The R wave normally progresses from a small R/deep S in V1 to a dominant R in V4-V6 (poor R-wave progression suggests old anterior ischaemia, LVH, or LBBB).[1]
The 12 leads as twelve simultaneous views of one beat
Only one cardiac cycle generates the ECG — all twelve leads are different spatial projections (vectors) of that same electrical event, viewed from twelve different angles. Limb leads view the frontal plane; precordial leads view the horizontal plane. This is why a single lead (the monitor's lead II) gives rhythm but a full 12-lead is needed for ischaemia, axis, chamber enlargement, and conduction defects — these are spatial diagnoses that require multiple simultaneous projections. [1]
Why lead II is the monitoring lead
The mean depolarisation vector (from SA node toward the apex/inferior-left) runs roughly parallel to lead II (+60 degrees), so lead II produces the largest, most consistently positive P-QRS-T complex — ideal for rhythm monitoring and pacemaker spike detection. Lead V1 is also useful because its biphasic R-S complex best reveals atrial activity (P waves) during broad-complex tachycardias. For ischaemia surveillance, however, lead II alone is insufficient — ischaemia is a regional, spatial diagnosis and requires the full 12 leads.[1]
Signal path and filtering
- The signal flows: electrodes → leads → a differential amplifier (high input impedance, common-mode rejection to cancel mains interference) → filters → analogue-to-digital conversion → display.[1]
- Filters: a high-pass filter (about 0.05-0.5 Hz) removes baseline drift from breathing and movement; a low-pass filter (about 40-150 Hz) removes electromyographic artefact; a notch filter removes mains interference (50 Hz in Australia, 60 Hz in the US).[1]
- Diagnostic bandwidth (0.05-150 Hz) preserves ST segments for ischaemia; monitoring bandwidth (0.5-40 Hz) is narrower to reduce artefact at the cost of ST fidelity.[1]
Systematic ECG interpretation — the ICU method
A disciplined, reproducible approach prevents the eye from anchoring on the most obvious abnormality and missing others. Every ECG is read in the same fixed order. The most widely taught sequence is Rate, Rhythm, Axis, Intervals, Segments, Waves (with chamber size and a final global summary added in some schemes).[1]
Systematic ECG interpretation — Rate, Rhythm, Axis, Intervals, Segments, Waves
- RATE. For a regular rhythm, divide 300 by the number of large squares between two consecutive R waves (300 / large squares = bpm). Alternatively, count the number of QRS complexes in a 10-second rhythm strip (50 large squares at 25 mm/s) and multiply by 6. Normal sinus rate is 60-100 bpm; bradycardia < 60, tachycardia > 100. If the rhythm is irregular, the 10-second strip method (multiply QRS count by 6) is the only reliable estimate.
- RHYTHM. Is it regular or irregular? Is there a P wave before every QRS and a QRS after every P? Is the PR interval constant? The four big questions: (a) Sinus rhythm = upright P in I/II/aVF, inverted in aVR, constant PR 0.12-0.20 s; (b) atrial activity present or absent; (c) atrial and ventricular rates equal or dissociated; (d) wide (> 0.12 s) or narrow QRS.
- AXIS (frontal QRS axis). Use the quadrant method on leads I and aVF: both positive = normal axis (-30 to +90); I positive, aVF negative = left axis (-30 to -90, check aVL — left axis if QRS upright in aVL); I negative, aVF positive = right axis (+90 to +180); both negative = extreme/northwest axis (-90 to -180). A more precise method uses the lead with the smallest (isoelectric, equiphasic) QRS — the axis is perpendicular to it, in the direction of the most positive of the two leads at right angles. Normal axis is -30 to +90 degrees.
- INTERVALS. Measure PR (0.12-0.20 s), QRS duration (0.06-0.10 s), and QT — but always correct for rate with Bazett's formula: QTc = QT / sqrt(RR in seconds) (normal < 0.44 s men, < 0.46 s women). A prolonged QTc > 0.50 s carries a high torsades risk. Compare to baseline QTc for the patient, especially after drugs.
- SEGMENTS. The PR segment is the baseline. The ST segment is isoelectric normally; elevation or depression > 1 mm in the limb leads or > 1-2 mm in the precordial leads (measured at the J point, 0.04 s after the QRS) is ischaemia until proven otherwise. Look in each territory: inferior (II, III, aVF), lateral (I, aVL, V5, V6), anterior (V1-V4), septal (V1-V2), posterior (tall R in V1 with ST depression in V1-V3 — mirror image of an anterior STEMI).
- WAVES — morphology. P-wave morphology (bifid = left atrial enlargement, "P mitrale"; tall peaked = right atrial enlargement, "P pulmonale"); QRS (pathological Q > 0.04 s and > 1/3 R amplitude in any lead except III and aVR suggests old infarction; tall R in V1 = posterior MI, RVH, or RBBB); T waves (peaked = hyperkalaemia or hyperacute ischaemia; flattened/inverted = ischaemia, electrolyte, drug); U waves = hypokalaemia, bradycardia, CNS disease.
- CHAMBER SIZE (additional step). Voltage criteria for left ventricular hypertrophy (Sokolow-Lyon: SV1 + RV5 or RV6 > 35 mm; Cornell: R aVL + S V3 > 28 mm men / 20 mm women); left atrial enlargement (P-terminal force in V1 > 0.04 mm·s); right-axis deviation + tall R in V1 for RVH.
- GLOBAL SUMMARY. State the rhythm, axis, intervals, ischaemia, and any chamber/enlargement/conduction abnormality in one sentence — this is the report. Never report a fragment; the ECG is read whole.
Determining the electrical axis — two methods
Two methods to determine the frontal QRS axis
| Method | How | Best for | Pitfall |
|---|---|---|---|
| Quadrant method (leads I + aVF) | I and aVF both positive = normal axis; I +ve, aVF -ve = left axis; I -ve, aVF +ve = right axis; both -ve = extreme/northwest | Quick bedside triage | Imprecise — does not separate normal from left axis at the -30 degree boundary; check aVL (upright QRS in aVL confirms left axis) |
| Isoelectric lead method | Find the limb lead with the smallest (equiphasic) QRS — the axis is perpendicular to it (90 degrees away), in the direction of the most positive of the two leads at right angles to the isoelectric one | Precise axis calculation | Fails if no lead is clearly equiphasic; harder with low-voltage QRS |
Causes of axis deviation
- Left axis deviation (< -30 degrees): left anterior fascicular block (most common), LVH, inferior MI, pacing, WPW (left-sided pathway).
- Right axis deviation (> +90 degrees): right ventricular hypertrophy, left posterior fascicular block, lateral MI, pulmonary embolism (acute right heart strain), dextrocardia, WPW (right-sided pathway), normal in children and tall thin young adults.
- Extreme/northwest axis (-90 to -180 degrees): ventricular rhythm, severe RVH, or limb-lead misplacement. [1]
Axis determination — worked examples
The isoelectric-lead method is the most precise bedside technique. The frontal QRS axis lies at 90 degrees to the most equiphasic limb lead; the polarity (which of the two perpendicular leads is positive) decides on which side of the circle the axis falls. [1]
Worked frontal-axis calculations using the isoelectric-lead method
| Most equiphasic lead | Perpendicular leads | If most positive lead is... | Approximate axis |
|---|---|---|---|
| aVF (+90 degrees) | I (0) and aVL (-30) | lead I | 0 degrees (normal) |
| aVL (-30 degrees) | II (+60) and III (+120) | lead II | +60 degrees (normal) |
| Lead III (+120 degrees) | aVR (-150) and I (0) | lead I | -30 degrees (borderline left) |
| Lead aVR (-150 degrees) | III (+120) and -aVL (+150) | aVF | +120 degrees (right axis) |
| Lead I (0 degrees) | aVL (-30) and aVF (+90) | aVF | +90 degrees (borderline right) |
| Lead II (+60 degrees) | aVL (-30) and aVF (+90) | aVL | -30 degrees (borderline left) |
A practical shortcut: the lead with the tallest positive QRS points toward the axis; the lead with the deepest negative QRS points 180 degrees away. If lead II is the tallest and aVR the most negative, the axis lies near +60 degrees (normal). If lead I is tallest and aVF isoelectric, the axis lies near 0 degrees. [1]
Determining the frontal QRS axis in 4 steps
- Inspect the six limb leads (I, II, III, aVR, aVL, aVF). Identify the lead with the most equiphasic (smallest net-area) QRS complex.
- The axis is perpendicular to that lead (90 degrees to it). List the two limb leads that lie at right angles to the isoelectric one.
- Polarity decides direction. Whichever of the two perpendicular leads is the most positive indicates the side of the axis. If both are positive the axis lies between them.
- Cross-check with lead aVR. A negative aVR is normal (axis between -30 and +90); a positive aVR is abnormal and indicates extreme/leftward axis or limb-lead reversal.
Electrode misplacement and limb-lead reversal
Limb-electrode reversal produces recognisable patterns and is a frequent cause of an "abnormal" ECG in an otherwise well patient. [1]
Limb-lead reversal patterns and how to recognise them
| Swap | Hallmark ECG finding | Why it happens | Pitfall |
|---|---|---|---|
| RA-LA reversed | Negative lead I (P and QRS inverted in I); aVR and aVL swapped; lead II becomes lead III and vice versa | Lead I = LA - RA becomes negative | Mimics dextrocardia; can mask lateral ischaemia. The single most useful rule: lead I should always be positive |
| RA-LL reversed | Lead II looks like old lead III; aVF and aVR swap; bizarre "inverted" limb-lead pattern | The right-arm electrode is at the left leg | Hard to spot; look for a very flat or inverted lead II with a normal lead I |
| LA-LL reversed | Lead III inverted; aVL and aVF swap; lead II nearly isoelectric | Lead III = LL - LA becomes negative | Often missed; the clue is an inverted lead III with normal I and II |
| RA-RL reversed | Grossly distorted tracing, very low voltage, dominant 50/60 Hz mains artefact | Right leg is the driven reference electrode; losing it destroys common-mode rejection | Looks like machine failure; re-check all four limb electrodes |
Dextrocardia versus RA-LA reversal is resolved by the precordial leads: dextrocardia gives reverse R-wave progression (dominant R in V1, small R in V6), whereas simple RA-LA reversal preserves normal chest-lead progression. Lead I is positive in every normally-sited heart — an inverted lead I is reversal or dextrocardia until proven otherwise. [1]
ECG artefact recognition in the ICU
The ICU ECG is recorded from a moving, shivering, electrically-connected patient; artefact is the rule, not the exception. Misreading artefact as arrhythmia causes inappropriate investigation and treatment. [1]
Common ICU ECG artefacts and their signatures
| Artefact | Appearance | Cause | How to resolve |
|---|---|---|---|
| 50/60 Hz mains (electrical interference) | Fine, regular, high-frequency oscillation of the baseline | Poor electrode contact; faulty lead; patient near mains cable; loss of the driven right-leg electrode | Reapply electrodes; check RL connection; reposition leads away from power cables; enable notch filter (monitoring only) |
| Muscle tremor (electromyographic) | Irregular, high-frequency "fuzz" obscuring the baseline | Shivering, agitation, Parkinsonism, cold | Warm the patient; sedate; treat shivering; select a cleaner lead |
| Baseline wander | Slow undulation of the baseline with respiration | Respiratory movement; poor electrode contact on the chest | Reposition chest electrodes off bony prominences; enable high-pass filter |
| Loose electrode / lead | Sudden abrupt jumps, flat segments, random spikes | Detached electrode, dried gel, lead tugged | Reapply electrodes with fresh gel; secure leads; check continuity |
| Pacemaker artefact (spike) unrecognised | Narrow vertical deflection preceding a broad QRS; sometimes oversensing/undersensing | Unipolar or bipolar pacemaker | Confirm with a paced-rhythm lead (II or V1); distinguish spike from artefact |
| Body movement / patient handling | Broad, irregular swings | Turning, physiotherapy, percussion, CPR | Pause movement during recording; use a rhythm strip with the patient still |
Distinguishing true arrhythmia from artefact on a monitor strip
- Correlate the tracing with the patient. A "ventricular tachycardia" on the screen in a patient who is sitting up talking is artefact until proven otherwise. Check the pulse, perfusion, and conscious state — artefact does not change these.
- Inspect all monitored leads, not one. Artefact is often confined to a single lead (a faulty electrode); a true arrhythmia appears in every lead simultaneously.
- Examine the baseline before and after the "arrhythmia". True VT has a consistent morphology and a defined onset/offset; artefact usually shows abrupt baseline disruption and underlying normal QRS complexes can often be "marched out" through the noise.
- Reproduce or remove the cause. Turn off suspect equipment (fluid warmer, haemofilter, electric bed), reapply electrodes, warm/shush the patient, and repeat. If the "arrhythmia" vanishes with electrode change, it was artefact.
- When still uncertain, obtain a 12-lead. A high-quality diagnostic-bandwidth 12-lead resolves the majority of ambiguous rhythm strips.
Common ICU ECG patterns I — myocardial ischaemia
Ischaemia injures myocardium, and the injury current displaces the ST segment. The pattern depends on whether injury is transmural (epicardial) — producing ST elevation — or subendocardial — producing ST depression.[1]
STEMI — ST-elevation myocardial infarction
Diagnostic criteria (Fourth Universal Definition of MI): ST elevation measured at the J point in two contiguous leads, with:
- > 1 mm in any lead EXCEPT V2-V3, AND
- In V2-V3: > 2 mm in men > 40 years, > 1.5 mm in women and men < 40 years (age and sex adjusted because the normal ST in V2-V3 is higher).[1]
The localising value of the territory is critical for the catheter-lab referral: [1]
STEMI localisation by involved leads
| Territory | Leads with ST elevation | Culprit artery |
|---|---|---|
| Inferior | II, III, aVF | Right coronary artery (RCA); if ST elevation in III > II and V1 — proximal RCA / RV infarct |
| Anterior | V1-V4 | Left anterior descending (LAD) |
| Lateral | I, aVL, V5, V6 | Circumflex or distal LAD |
| Septal | V1-V2 | LAD septal perforator |
| Posterior | Tall R in V1, ST depression V1-V3, upright T in V1 (mirror image); confirm with V7-V9 | RCA or circumflex |
| Right ventricular | ST elevation in V4R (right-sided V4) with inferior STEMI | Proximal RCA — avoid nitrates, preload-dependent |
Other ischaemic patterns
- ST depression > 1 mm, horizontal or downsloping: subendocardial ischaemia (NSTEMI), reciprocal change to a STEMI (e.g., ST depression in I, aVL with inferior STEMI — actually confirms the STEMI), or demand ischaemia (tachycardia, sepsis, hypotension, hypoxia — diffuse).
- T wave inversion: symmetrical, deep (> 3 mm) T-wave inversion in contiguous leads suggests ischaemia (especially in V1-V4 — anterior ischaemia; or in Wellens syndrome — deeply inverted or biphasic T in V2-V3 with critical proximal LAD stenosis, pain-free at the time of the ECG).
- Hyperacute T waves: broad-based, tall, asymmetric T waves in the territory of an evolving STEMI — the earliest sign, appearing within minutes, before ST elevation. Easily mistaken for hyperkalaemia — but hyperkalaemic T waves are narrow-based and peaked, and diffuse.
- Pathological Q waves: > 0.04 s wide and > 1/3 the R-wave amplitude — represent transmural scar from an old infarction (hours to days old).
- De Winter T waves: upsloping ST depression with tall, symmetric T waves in the anterior leads — a STEMI-equivalent of proximal LAD occlusion. [1]
Ischaemia in the presence of LBBB — Sgarbossa criteria
Diagnosing STEMI in LBBB is difficult because LBBB itself causes ST/T changes (appropriate discordance: the ST/T vector is opposite the QRS vector). The Sgarbossa criteria (1996) identify the inappropriate concordant changes that indicate superimposed infarction. Three criteria (any one is specific, criterion 3 alone is poorly specific without the modified Smith amendment):[3]
- Concordant ST elevation > 1 mm in any lead (ST in the SAME direction as the QRS).
- Concordant ST depression > 1 mm in V1-V3 (discordant ST elevation is normal in V1-V3 in LBBB; depression is abnormal).
- Excessive discordant ST elevation — ratio of ST elevation / QRS amplitude > 0.25 (Smith-modified Sgarbossa, improving sensitivity).
Common ICU ECG patterns II — electrolyte effects
Electrolyte disturbances produce stereotyped, predictable ECG changes because they directly alter myocyte action-potential morphology.[5]
Electrolyte disturbances and their ECG signatures
| Electrolyte | ECG progression | Mechanism |
|---|---|---|
| Hyperkalaemia | (1) Peaked, narrow-based T waves ("tenting") and shortened QT (K > 5.5); (2) loss of P waves, PR prolongation (K ~6.5); (3) widened QRS (K ~7-8); (4) sine wave — sine-wave QRS-T fusion (K > 8); (5) asystole / VF / PEA (K > 9-10) | Hyperkalaemia speeds terminal repolarisation (peaked T) and slows phase 0 depolarisation (wide QRS, loss of atrial conduction) |
| Hypokalaemia | (1) Flattened or inverted T waves; (2) prominent U waves (best seen in V2-V3); (3) ST depression; (4) prolonged QU/QT; (5) premature ventricular ectopics, torsades, VT/VF | Low K prolongs repolarisation and increases automaticity |
| Hypercalcaemia | Short QT (short ST segment; QTc < 0.35 s); rarely arrhythmias | High Ca shortens phase 2 plateau |
| Hypocalcaemia | Long QT (prolonged ST segment with normal T wave; QTc often > 0.50 s); torsades | Low Ca prolongs phase 2 |
| Hypomagnesaemia | Often coexists with hypokalaemia — prolonged QT, prominent U waves, torsades | Mg is the cofactor for Na-K-ATPase; deficiency mimics hypokalaemia |
Pearl: Hyperkalaemia progresses predictably and can arrest at any stage — peaked T waves are an early warning, but the sine wave and asystole can develop rapidly. Treat with calcium gluconate (membrane stabilisation, immediate), insulin-dextrose (intracellular K shift, ~30 min), salbutamol (additional intracellular shift), and bicarbonate if acidotic; remove K with resonium or (definitively) dialysis. The ECG is more urgent than the biochemistry — treat the ECG, not the number.[5]
Common ICU ECG patterns III — drug effects
Common ICU drugs and their ECG signatures
| Drug | Therapeutic effect | Toxic effect |
|---|---|---|
| Digoxin | "Dig effect" at therapeutic level — sagging ("scooped") downsloping ST depression in lateral leads (I, aVL, V4-V6), T-wave flattening/inversion, shortened QT, mild PR prolongation — NOT toxicity | Digoxin toxicity — almost ANY arrhythmia: atrial tachycardia with block, bidirectional VT (pathognomonic), bradycardia, AV block, premature ventricular beats. Toxicity risk rises sharply > 2 ng/mL and with hypokalaemia (K and digoxin compete for the same site on the Na-K-ATPase — hypokalaemia potentiates digoxin toxicity) |
| Quinidine and class IA (procainamide, disopyramide) | Prolonged QT and QRS, prominent U waves at therapeutic dose | Marked QT prolongation, torsades de pointes ("quinidine syncope"), widened QRS, T-wave inversion. Quinidine also widens the QRS — at high levels this mimics VT |
| Amiodarone (class III) | Prolonged QT (after weeks-months), bradycardia, mild T-wave changes | Marked QT prolongation and torsades (rare but real); pulmonary fibrosis, thyroid dysfunction are non-ECG toxicities |
| Sotalol | Combined beta-blocker + class III — prolonged QT, bradycardia | QT prolongation and torsades — dose-dependent; the commonest cause of drug-induced torsades |
| Beta-blockers, calcium-channel blockers, ivabradine | Bradycardia, AV block | High-degree AV block, junctional bradycardia, asystole in overdose; CCB overdose causes hypotension with bradycardia |
| Tricyclic antidepressants | — | Sodium-channel blockade: widened QRS, tall R in aVR (R/S > 0.3 in aVR), prolonged QT, right-axis deviation of the terminal 40 ms. QRS > 100 ms predicts arrhythmia risk; treat with IV bicarbonate |
| Phenothiazines, antipsychotics, antihistamines, fluoroquinolones, methadone | — | QT prolongation and torsades risk — the "long-QT drug" list |
Pacemaker ECG analysis in the ICU
The paced ECG is common in the ICU (temporary transvenous, transcutaneous, epicardial, or permanent devices). The pacing spike (a narrow vertical deflection representing current delivery) is the defining feature; analysing whether the pacemaker is capturing and sensing correctly is the core skill. [1]
Pacemaker function — capture and sensing
| Function | Definition | ECG evidence of normal function | ECG evidence of failure |
|---|---|---|---|
| Capture | Each pacing impulse depolarises the chamber | A pacing spike followed immediately by a wide paced QRS (ventricular) or a paced P wave (atrial) | Spike with no following depolarisation ("failure to capture") — leads displaced, threshold too low, battery failure, exit block from fibrosis |
| Sensing | The pacemaker detects intrinsic cardiac activity and inhibits appropriately | No pacing spike when an intrinsic beat occurs within the sensing window | Pacing spikes falling on T waves or random points ("failure to sense / undersensing") — risk of R-on-T and VF; oversensing senses noise and inappropriately inhibits |
| Output | The pacemaker delivers an adequate current | Consistent spike amplitude; consistent capture | Intermittent or absent spikes — battery depletion, lead fracture |
Pacemaker modes (NBG / NASPE code) and their ECG appearance
| Mode | Code meaning | Atrial paced? | Ventricular paced? | Typical ECG appearance |
|---|---|---|---|---|
| VVI | Ventricular paced, sensed, inhibited | No | Yes (only when no intrinsic QRS) | Spike before wide QRS only when the ventricular rate falls below the set rate; intrinsic beats are sensed and inhibit |
| AAI | Atrial paced, sensed, inhibited | Yes | No | Spike before a P wave; used when AV conduction is intact (sick sinus syndrome) |
| DDD | Dual-chamber paced, sensed, with triggered/inhibited response | Yes | Yes | Spike before P and/or QRS; maintains AV synchrony; the most physiological mode |
| VOO / DOO | Asynchronous (fixed-rate) pacing | — | Yes | Spikes at a fixed rate regardless of intrinsic activity — used as a "safety" mode in magnet application or electromagnetic interference; risk of R-on-T |
The magnet response is examined: applying a magnet over a permanent pacemaker switches it to an asynchronous fixed-rate mode (usually DOO/VOO at ~85-100 bpm), which both tests capture and prevents inhibition by electromagnetic interference (e.g., during surgery or diathermy). [1]
Troubleshooting a pacing problem on the ICU monitor
- Identify the pacemaker, mode, and rate settings. Temporary boxes display mode and output/sensitivity directly; for permanent devices check the pacemaker card.
- Look for capture. Is every spike followed by a depolarisation? If not, increase the output (mA) — if capture returns, threshold rise (lead maturation, fibrosis, electrolyte shift, MI at the lead tip) is the cause. Persistent failure needs lead repositioning.
- Look for sensing. Are spikes appearing where they should not (on intrinsic QRS/T)? Decrease sensitivity (increase the mV threshold). Are spikes inappropriately absent (oversensing noise)? Increase sensitivity.
- Check the patient. Hyperkalaemia, severe acidosis, hypoxaemia, and myocardial ischaemia raise the capture threshold acutely — correct these before assuming lead failure.
- Verify hardware. Battery depletion (permanent), lead fracture/displacement, loose connection on the temporary box — all produce intermittent or absent spikes. Chest X-ray confirms lead position and continuity.
- Provide back-up pacing. If transvenous pacing fails, apply transcutaneous pads (with analgesia/sedation) while arranging definitive fixation.
Common ICU ECG patterns IV — arrhythmia recognition
The ICU monitors continuous ECG precisely because arrhythmias are common and often haemodynamically significant. Recognising the rhythm from a rhythm strip depends on a small set of discriminating questions.[2][4]
Narrow-complex tachycardias (QRS < 0.12 s)
- Sinus tachycardia: regular, normal P-wave morphology before every QRS, rate < 220 (age-adjusted: 220 - age). Reacts to the cause (fever, hypovolaemia, pain, anxiety, thyrotoxicosis, PE).
- Atrial fibrillation (AF): IRREGULARLY IRREGULAR, no discrete P waves, fibrillatory baseline. Commonest arrhythmia in the ICU, often triggered by sepsis, electrolyte disturbance, hypoxia, or sympathetic surge. Treat rate with beta-blocker or amiodarone; anticoagulate per CHA2DS2-VASc.[4]
- Atrial flutter: REGULAR, sawtooth flutter waves (best in inferior leads II, III, aVF), typically atrial rate ~300 with 2:1 conduction = ventricular rate 150. Carotid sinus massage or adenosine can unmask the flutter waves by transiently increasing AV block.
- AV nodal re-entry tachycardia (AVNRT) and AV re-entry tachycardia (AVRT, WPW): regular, narrow, rapid (150-250), P waves buried in or after the QRS. Terminated by vagal manoeuvres or adenosine. Do NOT give AV-nodal blockers (adenosine, beta-blocker, diltiazem, digoxin) to AF with WPW — they can accelerate conduction down the accessory pathway and precipitate VF.
- Multifocal atrial tachycardia (MAT): irregularly irregular, three or more distinct P-wave morphologies — seen in COPD and hypokalaemia.
Broad-complex tachycardias (QRS > 0.12 s)
Distinguishing ventricular tachycardia (VT) from SVT with aberrancy (e.g., bundle branch block) is the central diagnostic problem. If in doubt in a haemodynamically unstable patient, treat as VT (synchronised cardioversion). When stable, apply these rules: [1]
VT versus SVT with aberrancy — discriminating features
| Feature | Favouring VT | Favouring SVT with aberrancy |
|---|---|---|
| AV dissociation (independent P and QRS, capture/fusion beats) | Diagnostic of VT | Absent |
| Concordance (all precordial leads same direction — all positive or all negative) | Strongly favours VT | Rare in SVT |
| Fusion beats (QRS a hybrid of a sinus and a ventricular beat) | Diagnostic of VT | Absent |
| Extreme axis (northwest, -90 to -180) | Favours VT | Rare |
| Initial R wave in aVR (large R) | Favours VT (Vereckei aVR criterion) | Absent |
| QRS > 0.16 s, notching/slurring | Favours VT | Usually narrower (< 0.14 s) with typical RBBB/LBBB morphology |
| Brugada algorithm | All four steps absent = VT | Any step present = SVT |
Polymorphic VT (varying QRS morphology, often torsades de pointes — twisting around baseline): torsades is associated with prolonged QT — stop any QT-prolonging drug, correct electrolytes, give IV magnesium 2 g. Monomorphic VT in a structurally abnormal heart is usually scar-mediated re-entry — amiodarone or cardioversion. [1]
Bradycardias and conduction blocks
Atrioventricular blocks — recognition and implications
| Block | ECG features | Significance |
|---|---|---|
| First degree | PR > 0.20 s, every P followed by a QRS | Usually benign; observe; common with AV-nodal drugs and age |
| Second degree, Mobitz I (Wenckebach) | Progressive PR prolongation until a P is not conducted, then cycle resets | Usually at the AV node; often benign and asymptomatic; monitor; rarely needs pacing |
| Second degree, Mobitz II | Constant PR, sudden non-conducted P (no preceding PR prolongation); often wide QRS | Below the AV node (His-Purkinje); high risk of progression to complete block — pacing often required |
| 2:1 AV block | One P not conducted for every one conducted — could be Mobitz I or II | Decide by QRS width and PR; if uncertain, treat as Mobitz II (high-grade block) |
| Third degree (complete heart block) | Atrial and ventricular rates independent, no constant PR, escape rhythm narrow (junctional, ~40-50) or wide (ventricular, ~20-40) | Permanent pacing required; temporary transvenous pacing if symptomatic or unstable |
Bundle branch blocks
- Right bundle branch block (RBBB): QRS > 0.12 s with rsR' ("rabbit ears") in V1-V2 and deep slurred S in V5-V6. RBBB is common, often benign, and does not by itself preclude ischaemia diagnosis (use Sgarbossa-like reasoning only for LBBB). A new RBBB in a sick patient may indicate acute right-heart strain (e.g., PE) or acute ischaemia.
- Left bundle branch block (LBBB): QRS > 0.12 s with broad notched R in V5-V6, deep broad S in V1-V2, and appropriate discordant ST-T changes. Diagnosing STEMI in LBBB requires the Sgarbossa criteria — never read ST elevation in LBBB as automatically normal or automatically ischaemic.[3]
- Bifascicular block (RBBB + left anterior or posterior fascicular block) in the setting of acute MI carries a high risk of progression to complete heart block and is an indication for prophylactic pacing.
ECG patterns suggesting acute pulmonary embolism
Massive PE produces acute right-heart strain. The classic (sinus tachycardia) is most common; the specific changes are: S1Q3T3 (deep S in I, Q in III, inverted T in III), right-axis deviation, T-wave inversion in V1-V4, new RBBB, and right precordial ST changes. None are sensitive, but in context they support the diagnosis. Right-heart strain pattern + haemodynamic compromise + risk factors = investigate and treat aggressively. [1]
Pulseless arrest rhythms
- Ventricular fibrillation (VF): chaotic, irregular, no identifiable QRS — immediately defibrillate.
- Pulseless ventricular tachycardia (pVT): broad complex, regular, no pulse — immediately defibrillate.
- Asystole and pulseless electrical activity (PEA): organise the arrest search for reversible causes (4 Hs and 4 Ts — Hypoxia, Hypovolaemia, Hypo/hyperkalaemia, Hypothermia; Thrombosis coronary, Thrombosis pulmonary, Tension pneumothorax, Tamponade). [1]
Approach to a tachyarrhythmia in the ICU
- ASSESS the patient, not the strip first. Airway, breathing, circulation; pulse, blood pressure, perfusion, conscious state. An unstable patient (hypotension, shock, ischaemic chest pain, acute heart failure, altered consciousness) with a tachyarrhythmia needs immediate synchronised cardioversion — do not delay for a perfect rhythm diagnosis.
- DETERMINE QRS width. Narrow (< 0.12 s) = atrial/AV-nodal origin; broad (> 0.12 s) = ventricular OR SVT with aberrancy. If broad and unstable, treat as VT.
- IF NARROW and regular: consider sinus tachycardia (treat the cause), atrial flutter, AVNRT/AVRT. Vagal manoeuvres or adenosine (6 mg then 12 mg IV) both diagnose (by transiently blocking AV node — flutter waves emerge) and terminate AVNRT/AVRT.
- IF NARROW and irregular: atrial fibrillation (rate control with beta-blocker or diltiazem; consider amiodarone if structurally abnormal heart), or MAT (treat the underlying COPD/hypokalaemia).
- IF BROAD and regular: treat as VT if unstable — synchronised cardioversion. If stable, give IV amiodarone 300 mg and seek a cardiology opinion; apply the VT-vs-SVT discriminators. AVOID verapamil in a broad-complex tachycardia — if it is actually VT, verapamil can cause haemodynamic collapse.
- IF BROAD and irregular: polymorphic VT — if pulseless, defibrillate; if torsades (long QT), stop QT-prolonging drugs, give IV magnesium 2 g, correct K/Mg/Ca.
STEMI-equivalents and high-risk ischaemic patterns
Several ECG patterns carry the prognostic and management implication of a STEMI (proximal epicardial occlusion requiring emergent reperfusion) without meeting the formal ST-elevation criteria. Recognising them is essential. [1]
STEMI-equivalent and occlusion-MI (OMI) patterns
| Pattern | ECG signature | Pathology | Implication |
|---|---|---|---|
| Wellens syndrome | Deeply inverted or biphasic T waves in V2-V3 (often during a pain-free interval), preserved R-wave progression, no/q minimal ST elevation | Critical proximal LAD stenosis | High risk of anterior MI; refer for urgent catheterisation; do NOT stress-test (provokes occlusion) |
| de Winter T waves | Upsloping ST depression (1-2 mm) in the precordial leads with tall, symmetric, positive T waves and no ST elevation; persistent | Acute proximal LAD occlusion | STEMI-equivalent; activate the catheter lab |
| Hyperacute T waves | Broad-based, tall, asymmetric, "bulky" T waves in a vascular territory, appearing within minutes of occlusion, before ST elevation | Earliest transmural ischaemia | Harbinger of evolving STEMI; obtain serial ECGs and prepare for reperfusion |
| Posterior MI | Tall R in V1-V2, ST depression V1-V3, upright T in V1 (mirror image); horizontal ST depression V1-V3 with dominant R | Posterior wall occlusion (RCA or circumflex) | Confirm with posterior leads V7-V9 (ST elevation > 0.5 mm); treat as STEMI |
| Right ventricular infarction | ST elevation in V4R (right-sided V4) with an inferior STEMI; ST elevation in III > II | Proximal RCA occlusion | Avoid nitrates and preload-reducing agents; volume-load; the RV is preload-dependent |
| Left bundle branch block (new) | New or presumed-new LBBB with ischaemic symptoms | Possible underlying anterior occlusion | Apply Smith-modified Sgarbossa criteria; historically STEMI-equivalent for reperfusion |
| Left main / multivessel ischaemia | Diffuse ST depression in seven or more leads with ST elevation in aVR > 1 mm (aVR sign) | Left main or proximal three-vessel disease | High-risk finding; urgent catheterisation; poor prognosis |
The aVR sign (ST elevation in lead aVR > ST elevation in V1) indicates left main or proximal LAD occlusion and carries a high in-hospital mortality — it is one of the most easily missed high-risk patterns. [1]
Common ICU ECG patterns V — the non-ischaemic mimics
Several conditions produce ST changes that mimic ischaemia and must be distinguished to avoid unnecessary catheter-lab activation: [1]
- Early repolarisation: diffuse, concave ST elevation (especially in young, male, athletic patients), prominent J-point notching, no reciprocal change. Benign, but must be distinguished from pericarditis and STEMI.
- Acute pericarditis: diffuse, concave ST elevation in most leads (not territory-localised), PR depression (the giveaway, reciprocal to aVR ST elevation), no reciprocal change. Evolves through stages: diffuse ST elevation/PR depression → ST normalisation with T flattening → diffuse T inversion → normalisation.
- LV aneurysm: persistent ST elevation in the same territory as an old Q-wave MI, weeks-to-years after the infarction — represents chronic paradoxical bulge, not acute ischaemia.
- Brugada pattern: coved ST elevation in V1-V3 (type 1) with right bundle branch morphology — a channelopathy predisposing to sudden death; not ischaemic.
- Paced rhythm: the pacing spike followed by a broad paced QRS causes secondary ST-T changes that can mask or mimic ischaemia. Use the Sgarbossa-style approach for paced rhythms (modified Sgarbossa for paced rhythm).
- Hypothermia: Osborne (J) waves — a positive deflection at the J point; bradycardia, prolonged QT, shivering artefact. Not ischaemic. [1]
Electrical safety - macroshock vs microshock
- Macroshock is current applied to the body surface. Rough thresholds: perception about 1 mA, pain and let-go threshold around 10-15 mA, sustained tetany above 20 mA, and ventricular fibrillation around 100 mA. Dry skin resistance (about 100 kilo-ohms) limits the current, which is why a low-impedance contact (wet skin, a defect) is dangerous.[1]
- Microshock is current delivered directly to the heart through a low-impedance conductive path - a saline-filled central venous catheter, a pulmonary-artery catheter, or a pacing wire. The fibrillation threshold is about 10 microamperes (0.01 mA), roughly a thousand times lower than for macroshock. This is the defining electrical hazard of the ICU.[1]
Protective measures
- Isolating transformer and line-isolation monitor (LIM): the patient-side circuit is isolated from earth, so a single fault cannot drive current through the patient to ground. The LIM alarms if the isolation is breached.[1]
- Equipotential earthing (equipotential bonding): all conductive surfaces in the area are connected so no potential difference exists between them, eliminating current flow through a patient touching two surfaces. Required in cardiac-protected treatment areas.[1]
- Equipment classification: Class I (earthed chassis), Class II (double-insulated, no earth), Class III (battery / safety extra-low voltage).[1]
- Applied-part classification: type B (body contact), BF (body, floating), and CF (cardiac floating) - the highest grade of isolation, permitting direct cardiac connection (microshock-safe).[1]
Macroshock — body-surface contact in detail
When mains current reaches the body surface, the magnitude of current (and therefore the harm) depends on Ohm's law: I = V / R, where R is dominated by skin resistance (typically ~1000 ohm wet to ~100,000 ohm dry). A 240 V mains contact through dry skin drives ~2.4 mA (perception only); through wet skin the same voltage drives ~240 mA — well into the VF range.[1]
Macroshock — physiological effects by current magnitude (60 Hz AC)
| Current | Effect | Significance |
|---|---|---|
| 0.5-1 mA | Threshold of perception (tingling) | First awareness |
| 1-5 mA | Mild shock, unpleasant but tolerable | Maximum "harmless" current |
| 10-15 mA | Pain; "let-go" threshold — sustained muscle tetany prevents release of the source | Above let-go, the victim cannot voluntarily break contact |
| 20-50 mA | Sustained tetany of respiratory muscles → asphyxia if prolonged | Paralysing shock |
| 100-200 mA | Ventricular fibrillation | Lethal; the classic electrocution range |
| > 1000 mA | Sustained tetany of the myocardium; if shock ends, the heart may resume in sinus (defibrillation) | Internal and surface burns dominate |
Skin resistance is the main protection. Wet skin, a skin defect, a needle, or any breach of the stratum corneum drops resistance by orders of magnitude and converts a trivial contact into a dangerous one. This is the reason that the electrode gel on an ECG electrode or a transcutaneous pacing pad lowers skin resistance — useful for sensing/pacing, but a reminder that patients with multiple conductive electrode contacts are more susceptible to macroshock than dry-skinned individuals.[1]
Microshock — the defining ICU hazard in detail
Microshock is current delivered DIRECTLY to the myocardium through a conductive intracardiac path. The fibrillation threshold is only about 10 microamperes (0.01 mA) — roughly ten thousand times lower than the macroshock VF threshold. The conductive path that creates this hazard is one of three structures found routinely in ICU patients:[1][1]
- A saline-filled central venous catheter — saline conducts electricity, so a catheter whose tip lies in or near the right atrium is effectively a wire to the heart. Even the catheter PORTS are a hazard if a connected transducer or fluid warmer has a leakage fault.
- A pulmonary-arterty (Swan-Ganz) catheter — directly in the right heart and proximal pulmonary artery.
- A transvenous or epicardial pacing wire — a direct electrical connection to the endocardium/epicardium, by design the most sensitive microshock path. [1]
A patient with any of these becomes a microshock candidate, and a leakage current from any connected equipment (ECG, pressure transducer, fluid warmer, haemofilter, monitor) can, in theory, fibrillate the heart. This is why cardiac-protected treatment areas, isolated power supplies, equipotential earthing, and CF-rated equipment exist: to guarantee that no leakage current can reach 10 microamperes through any patient-connected pathway.[1][1]
Protective measure 1 — the isolated power supply and line-isolation monitor
In an isolated (unearthed) power supply, the secondary winding of an isolating transformer is NOT referenced to earth. A patient touching one side of the isolated circuit and an earthed surface therefore completes NO circuit — no current flows. The first fault simply charges the isolated system to earth potential without delivering current through the patient. A SECOND fault, on the other side, would create an earthed circuit and a shock hazard.[1][1]
The line-isolation monitor (LIM) continuously measures the degree of isolation between the supply and earth. The LIM does NOT detect current flowing through the patient — it detects the POTENTIAL for such current, by intentionally injecting a small test current and measuring the resulting leakage. The alarm threshold is typically 5 mA (the assumption: 5 mA is below the perception threshold for macroshock and well below the microshock threshold — so a circuit capable of leaking 5 mA is approaching danger).[1]
When the LIM alarms: the isolation has been breached (one side of the isolated circuit has developed a low-impedance path to earth — typically through a piece of faulty equipment). The system now behaves like an earthed circuit, and a second fault could deliver a shock. The correct response is to identify and disconnect the faulty device (unplug the most recently connected equipment first), not to silence the alarm. The patient is NOT yet being shocked — but the system is no longer protected. [1]
Protective measure 2 — equipotential bonding (cardiac-protected treatment areas)

In a cardiac-protected treatment area (defined under AS/NZS 3003 in Australia/NZ, and equivalent standards elsewhere), every conductive surface within the patient's reach and the protective earth of the electrical installation are bonded together into a single equipotential node, so that no potential difference can exist between any two points the patient could touch simultaneously.[1]
This eliminates the macroshock path "patient touches two earthed surfaces at different potentials" — a hazard when a fault on one piece of equipment raises its chassis potential relative to another. A dedicated equipotential bonding busbar (EPB) is provided, with clearly marked green/yellow terminals, to which movable equipment can be bonded via an equipotential lead. [1]
AS/NZS 3003 also defines body-protected treatment areas (lower-risk, requiring isolated supply or RCD but NOT full equipotential bonding) and cardiac-protected treatment areas (the highest grade — required wherever the heart may be directly contacted, e.g., cardiac catheter labs, ICU, CCU, operating theatres for cardiac surgery).[1]
Treatment area classifications (AS/NZS 3003 and equivalents)
| Area | Risk | Requirements |
|---|---|---|
| Cardiac-protected | Direct cardiac contact possible (ICU, CCU, cardiac theatre, cath lab) | Isolated power supply + LIM + equipotential bonding of all conductive surfaces; leakage kept below the microshock threshold |
| Body-protected | No direct cardiac contact but skin/mucosal contact (general ICU bay without cardiac lines, endoscopy, dialysis) | Isolated supply OR residual-current device (RCD); not necessarily equipotential bonding |
| General (unclassified) | No applied part (office, corridor) | Standard installation; standard leakage limits |
Protective measure 3 — equipment and applied-part classifications (IEC 60601)
IEC 60601-1 classifies medical electrical equipment by how it is protected against electric shock.[1]
Equipment class (means of mains protection)
| Class | Construction | Earth |
|---|---|---|
| Class I | Earthed metal chassis; basic insulation + protective earth as the fault-protection | Yes — relies on earth + fuse/RCD |
| Class II | Double-insulated enclosure; no protective earth needed | No |
| Class III | Supplied by safety extra-low voltage (SELV, typically < 60 V DC) — intrinsically safe | No |
Applied-part type (degree of patient isolation, IEC 60601-1)
| Type | Maximum patient leakage (normal / single fault) | Intended contact | Floating? |
|---|---|---|---|
| Type B | 100 µA / 500 µA | Body surface, non-cardiac | May be earth-referenced |
| Type BF | 100 µA / 500 µA | Body surface, floating (galvanic isolation from earth) | Yes — isolated |
| Type CF | 10 µA / 50 µA | Direct cardiac connection (pacing wire, intracardiac catheter) | Yes — highest isolation; microshock-safe |
The CF (cardiac floating) type is the only applied-part classification rated for direct cardiac connection. Its leakage-current limit of 10 microamperes in normal condition and 50 microamperes in single-fault condition matches the microshock threshold — by design, CF equipment cannot deliver a microshock-level current. Equipment with an intracardiac connection (a transvenous pacing wire, certain pulmonary-artery catheters with electrical connections, intracardiac ECG leads) MUST be CF-rated. Using non-CF equipment with an intracardiac line is a never-event.[1]
Protective measure 4 — leakage current limits in practice
Leakage current is the unintended current that flows from the live parts of a device to earth or to the patient, through stray capacitance, imperfect insulation, or filter capacitors. Some leakage is unavoidable; the standards set the maximum tolerable leakage for each equipment and applied-part type so that the leakage is well below the relevant shock threshold:[1]
- Type CF (cardiac): 10 µA normal, 50 µA single fault — below the 10 µA microshock VF threshold under normal operation, with margin to the threshold even when one fault has occurred.
- Type BF (body, floating): 100 µA normal, 500 µA single fault — below the macroshock perception threshold.
- Type B (body): 100 µA normal, 500 µA single fault.
- Patient auxiliary currents (current intended to flow between electrodes, e.g., impedance respiration) have separate, lower limits. [1]
The dual-margin philosophy — well below threshold in normal operation, still below threshold with ONE fault — is the basis of the "two-fault" safety design: a single fault must never deliver a dangerous current; only a (much less likely) double fault can. [1]
Responding to a line-isolation monitor alarm in ICU
- DO NOT silence and ignore the alarm. The LIM alarm means the isolated circuit has been breached — the system is now no safer than an earthed circuit, and a second fault could deliver a shock. This is particularly dangerous for a patient with an intracardiac line (microshock candidate).
- PRESERVE patient-critical support first. Do not disconnect life-sustaining equipment (ventilator, balloon pump, vasoactive infusions) abruptly. Identify which device is the offender by process of elimination.
- UNPLUG the most recently connected device first and observe whether the alarm resolves. Often the offender is the last piece of equipment added — a fluid warmer, a new transducer, a temporary monitor, a faulty charger.
- CONTINUE eliminating — unplug one device at a time, working backward, until the alarm clears. The device that clears the alarm is the faulty one — remove it from service, tag it as faulty, send for biomedical engineering inspection.
- IF NO DEVICE can be found, escalate to biomedical engineering and electrical maintenance; the fault may be in the fixed installation (the isolating transformer or its wiring).
- DOCUMENT the event and the offending device; report via the institutional incident system. A recurring alarm on the same outlet is a fixed-installation problem.
Defibrillation physics — energy, impedance, and waveform
Defibrillation works by delivering a brief, high-energy current that depolarises a critical mass of myocardium simultaneously, abolishing the re-entry circuits sustaining VF and allowing the SA node to resume. [1]
Monophasic versus biphasic defibrillation waveforms
| Feature | Monophasic damped sinusoidal (MDS) | Biphasic truncated exponential (BTE) |
|---|---|---|
| Current direction | One direction through the myocardium | Reverses partway through the shock |
| Typical energy | 200, 300, 360 J escalating | 120-200 J (manufacturer-specific; often 150 J biphasic) |
| Efficacy at lower energy | Lower — needs higher joules | Higher — terminates VF at roughly half the energy of monophasic |
| Myocardial injury | Greater | Less |
| First-shock efficacy for VF | ~60-90 per cent | > 90 per cent |
| Current practice | Largely obsolete | Standard on all modern defibrillators |
Transthoracic impedance is the resistance the chest offers to current flow — typically 70-80 ohms, but highly variable. Higher impedance means less current reaches the myocardium for a given energy. Factors that reduce impedance and improve shock success: [1]
- Firm pad pressure (≈ 8 kg of force) — displaces air and improves contact.
- Adequate conductive gel — but never gel bridging between the two pads, which shunts current across the chest wall and away from the heart.
- Large adult pads (lower impedance than paediatric pads).
- Exhalation (reduces lung air volume, a high-resistance medium).
- Sequential shocks — impedance falls ~8 per cent with each successive shock, so an escalating or repeat strategy can succeed where the first fails. [1]
Defibrillation pad/paddle placement and why it matters
| Position | Where | Indication / advantage |
|---|---|---|
| Anterolateral (sternal-apex) | One pad below the right clavicle, one in the V6 position (midaxillary line) | Default; quick; standard |
| Anteroposterior | One over the precordium (V3), one on the back (left infrascapular) | Lower impedance; preferred for elective cardioversion; better current vector through the heart |
| Avoidance of midline overlap | Pads must be separated by at least the pad width | Overlap shunts current across the chest wall and away from the myocardium |
| Avoidance of implanted devices | Place pads at least 8 cm from a pacemaker/ICD generator | Prevents device damage and current shunting through the device pocket |
The ICD (implantable cardioverter-defibrillator) delivers an internal shock (~30 J) directly to the endocardium through a lead, bypassing transthoracic impedance entirely — far more efficient than external defibrillation. When an ICD fires inappropriately or fails to terminate an arrhythmia, external pads can be applied (posterior to the device) and external shocks delivered. [1]
Diathermy / electrosurgery and the ICU patient
Surgical diathermy (electrosurgery) uses high-frequency (300 kHz to 3 MHz) alternating current to cut and coagulate tissue. The same high-frequency current is a serious hazard to the ICU patient with a pacemaker/ICD, a central line, or any intracardiac conductor. [1]
Diathermy hazard pathways in the ICU patient
| Pathway | Mechanism | Consequence |
|---|---|---|
| Pacemaker/ICD interference | High-frequency current is sensed as intrinsic cardiac activity; the device may inhibit (asystole in pacemaker-dependent patient), track at high rates, or deliver inappropriate shocks | Asystole, VT, inappropriate ICD shocks; switch pacemaker to asynchronous mode (magnet) before diathermy |
| Current diversion via the return electrode | A faulty or misplaced return ("patient") plate concentrates current at a small area | Diathermy burn at the plate site |
| Current via a central line / pacing wire | The dispersive current seeks any low-impedance path to earth; an intracardiac conductor is one | Microshock-range current to the myocardium; VF |
| Capacitive coupling to ECG leads | High-frequency current couples capacitively to monitoring leads | ECG artefact; potential burns at electrode sites |
Safe use of diathermy in a patient with a pacemaker or ICD
- Assess the indication. Use bipolar diathermy (current confined between forceps tips, no return plate) whenever possible — far safer than monopolar near a device.
- Keep the current path away from the device. For monopolar diathermy, place the return plate so the device and its leads are OUTSIDE the path between the active electrode and the plate.
- Use short, intermittent bursts rather than sustained activation, and use the lowest effective power.
- Have a magnet ready and continuously monitor the pulse (plethysmography or arterial line) — pacemaker inhibition may produce asystole; ICDs may deliver inappropriate shocks during diathermy.
- Reprogram / interrogate afterwards. Temporary asynchronous mode is reversed by removing the magnet, but the device should be interrogated post-procedure to exclude reprogramming or damage.
Residual current device (RCD) versus isolated power supply (IPS)
Both RCDs and IPS protect against electric shock, but by fundamentally different mechanisms — a frequent First-Part examination question. [1]
RCD versus isolated power supply (IPS) — mechanism and behaviour
| Feature | Residual current device (RCD) | Isolated power supply (IPS) + line-isolation monitor (LIM) |
|---|---|---|
| Mechanism | Detects imbalance between live and neutral currents (a current "leaking" to earth) and trips the supply | Secondary of an isolating transformer is not referenced to earth; a single fault cannot complete a circuit through the patient |
| Response to first fault | Trips — disconnects power (and life-support equipment) immediately | Does NOT trip; alarms via the LIM; supply continues — continuity of care preserved |
| Continuity of supply | Lost on first fault | Preserved on first fault |
| Suitability | Body-protected areas, domestic, general wards | Cardiac-protected areas (ICU, theatre, cath lab) where both safety and supply continuity are required |
| Microshock protection | Trip threshold (~5-30 mA) far above the microshock threshold (10 \u00b5A) — does NOT protect against microshock | Combined with CF equipment and equipotential bonding, keeps leakage below 10 \u00b5A — DOES protect against microshock |
| Testing | Press-to-test button periodically | LIM continuously monitors; periodic leakage testing by biomedical engineering |
The defining contrast: an RCD trades supply continuity for safety (it disconnects), whereas an IPS trades complexity for continuity (it isolates and alarms but keeps the power on). In an ICU, where disconnecting a ventilator or balloon pump is itself a hazard, the IPS is the correct choice. [1]
Defibrillation safety — the special case of high-energy discharge
Defibrillation delivers a brief, high-energy (150-360 joule biphasic) shock across the chest to depolarise the entire myocardium simultaneously and abort VF. Two safety points are examinable.[1]
- Oxygen-rich atmosphere fires: a defibrillation spark in a high-O2 environment (oxygen mask, high-flow nasal cannula, NIV mask leaking) ignites facial or airway fires and scorches drapes. Remove or displace oxygen sources (move the mask 1 m away, turn down HFNC, disconnect the NIV circuit) before delivering a shock. Do not lean across the patient holding an oxygen mask.
- Operator and bystander shock risk: "Clear!" — ensure no staff member is in contact with the patient or the bed. Wet gel on the chest reduces macroshock skin resistance, so a bystander touching the bed during a shock can receive a dangerous current. Conductive gel must not bridge the two paddles (a "bridging" gel path dissipates the energy across the chest wall and not through the myocardium). [1]
Exam practice — SAQs
SAQ — Line-isolation monitor alarm in a patient with an intracardiac catheter
10 minutes · 10 marks
A 64-year-old man is day 3 in ICU after an out-of-hospital cardiac arrest. He has a right internal jugular triple-lumen central venous catheter whose tip lies in the right atrium, a radial arterial line, a transvenous pacing wire in the right ventricle for intermittent complete heart block, and is on a norepinephrine infusion. He is connected to the bedside ECG monitor, a pressure transducer, and a fluid warmer. The line-isolation monitor (LIM) at the bedside suddenly alarms. The nurse asks whether to silence it.
SAQ — Safe defibrillation of ventricular fibrillation in an oxygen-rich environment with an ICD in situ
10 minutes · 10 marks
A 72-year-old man in ICU day 2 after a non-ST-elevation myocardial infarction suddenly loses consciousness. The monitor shows coarse ventricular fibrillation. He has a high-flow nasal cannula running at 60 L/min and 100 per cent oxygen, a central venous catheter in situ, and an implantable cardioverter-defibrillator (ICD) in the left infraclavicular pocket that has not fired. Three staff members are at the bedside. The defibrillator is biphasic. You are leading the arrest.
Red flags
Clinical pearls
A quick-revision summary table
Normal ECG values to commit to memory
| Quantity | Normal range |
|---|---|
| Paper speed | 25 mm/s (1 small square = 0.04 s, 1 large square = 0.20 s); 1 mV = 10 mm |
| Sinus rate | 60-100 bpm |
| PR interval | 0.12-0.20 s (3-5 small squares) |
| QRS duration | 0.06-0.10 s (1.5-2.5 small squares) |
| QTc (Bazett) | < 0.44 s (men), < 0.46 s (women) |
| Frontal QRS axis | -30 to +90 degrees |
| P wave | < 0.12 s wide, < 0.25 mV tall |
| ST segment | Isoelectric (±1 mm limb leads, ±1-2 mm precordial) |
| R-wave progression | Dominant R by V4 |
The electrolyte and drug signatures at a glance
| Disturbance / drug | Hallmark ECG finding |
|---|---|
| Hyperkalaemia | Peaked narrow T → PR prolong → loss of P → wide QRS → sine wave → arrest |
| Hypokalaemia | Flattened/inverted T + prominent U + ST depression + long QT |
| Hypercalcaemia | Short QT (short ST segment) |
| Hypocalcaemia | Long QT (long ST segment with normal T) |
| Digoxin (therapeutic) | Scooped downsloping ST depression, T flattening, short QT, PR prolongation |
| Digoxin (toxicity) | Atrial tachycardia with block, bidirectional VT, AV block |
| Quinidine / class IA | Long QT, prominent U, torsades |
| Amiodarone / sotalol | Long QT (torsades rarer with amiodarone, dose-dependent with sotalol) |
| Tricyclic overdose | Wide QRS, tall R in aVR, right-axis of terminal 40 ms, long QT |
| Hypothermia | Osborne (J) waves, bradycardia, long QT |
Microshock versus macroshock — the defining ICU contrast
| Feature | Macroshock | Microshock |
|---|---|---|
| Pathway | Skin surface contact | Direct to myocardium via intracardiac catheter/pacing wire |
| Skin resistance | ~1000-100,000 ohm (protective) | Bypassed — direct endocardial contact |
| VF threshold | ~100 mA | ~10 microamperes (10,000x lower) |
| Perception threshold | ~1 mA | None — current is below perception even at lethal levels |
| Typical victim | Any person touching faulty mains | ICU patient with a central line / pacing wire |
| Protection | RCD, isolation, dry skin, insulation | CF-rated equipment, isolated supply, equipotential bonding, LIM |
| Standards reference | IEC 60601-1 Type B/BF (100 µA normal) | IEC 60601-1 Type CF (10 µA normal) |
STEMI, STEMI-equivalent, and high-risk ischaemic patterns at a glance
| Pattern | Leads | Hallmark | Action |
|---|---|---|---|
| Inferior STEMI | II, III, aVF | ST elevation; check V4R for RV infarct | Cath lab; avoid nitrates if RV infarct |
| Anterior STEMI | V1-V4 | ST elevation; proximal LAD if with RBBB and ST elevation in V1 | Cath lab |
| Lateral STEMI | I, aVL, V5, V6 | ST elevation | Cath lab |
| Posterior MI | V1-V3 (mirror) | Tall R, ST depression, upright T; confirm V7-V9 | Cath lab |
| Wellens | V2-V3 | Biphasic/deeply inverted T, pain-free | Urgent catheterisation; NO stress test |
| de Winter | V1-V6 | Upsloping ST depression, tall symmetric T | Cath lab (STEMI-equivalent) |
| aVR sign | aVR | ST elevation in aVR > V1, diffuse ST depression | Left main / three-vessel; urgent catheterisation |
| Hyperacute T | territory of occlusion | Broad, tall, asymmetric T, early | Serial ECGs; prepare reperfusion |
Defibrillation and electrical-counter-shock essentials
| Parameter | Value |
|---|---|
| Waveform (modern) | Biphasic truncated exponential |
| First-shock energy (VF/pVT) | 150-200 J biphasic (manufacturer-specific) |
| Monophasic (if only option) | 360 J |
| Transthoracic impedance (typical) | 70-80 ohms; falls ~8 per cent per successive shock |
| Pad position (default) | Anterolateral (below right clavicle + V6) |
| Pad position (cardioversion) | Anteroposterior (lower impedance) |
| Before shock | Remove/displace oxygen; ensure all staff clear; check gel does not bridge pads |
| ICD internal shock | ~30 J directly to endocardium |
Electrical safety — protection hierarchy and what each layer achieves
| Layer | Protects against | Mechanism | Limitation |
|---|---|---|---|
| Dry intact skin | Macroshock | High resistance (~100 kilo-ohms) limits current | Bypassed by wet skin, gel, needles, catheters |
| Class I equipment (earthed) | Macroshock | Fault current flows to earth, blowing a fuse/RCD | Useless if earth is broken; no microshock protection |
| Class II (double-insulated) | Macroshock | No earthed chassis; double insulation | No microshock protection |
| Residual current device (RCD) | Macroshock | Trips on live-neutral imbalance | Trips at 5-30 mA — far above microshock threshold; disconnects power |
| Isolated power supply + LIM | Macroshock (and continuity) | Secondary not earth-referenced; LIM alarms before danger | Two faults can still deliver a shock |
| Equipotential bonding | Macroshock (two-surface path) | All surfaces at one potential | Does not protect against microshock |
| CF-rated applied part | Microshock | Leakage < 10 \u00b5A normal, 50 \u00b5A single fault | Only as good as its maintenance and correct connection |
Key trials and evidence
Thygesen 2018 — Fourth Universal Definition of Myocardial Infarction (PMID 30165617)
Source
European Heart Journal 40(3):237-269 — joint ESC/ACC/AHA/WHF Task Force consensus
Design
Expert consensus document updating the diagnostic criteria for MI
Key finding
ST-elevation MI defined as ST elevation at the J point > 1 mm in any two contiguous leads EXCEPT V2-V3, where thresholds are > 2 mm in men over 40 and > 1.5 mm in women and men under 40 (sex- and age-adjusted). Contiguous = same vascular territory.
Clinical bottom line
The current diagnostic standard for STEMI — the sex- and age-adjusted V2-V3 thresholds matter; using a flat 1 mm across all leads over-calls anterior MI
Sgarbossa 1996 — ECG diagnosis of acute MI in LBBB (PMID 8559200)
Source
New England Journal of Medicine 334(8):481-487 — derivation cohort from GUSTO-1 trial
Design
Retrospective analysis of ECGs in patients with LBBB and acute MI confirmed by enzyme rise
Key finding
Three independent criteria diagnostic of MI in LBBB: (1) concordant ST elevation > 1 mm; (2) concordant ST depression > 1 mm in V1-V3; (3) excessively discordant ST elevation (originally > 5 mm, later refined by Smith to ST/QRS ratio > 0.25). Highly specific (> 90 per cent) but modestly sensitive.
Clinical bottom line
LBBB's appropriate discordant ST-T changes are normal; only CONCORDANT changes (or excessive discordance by ratio) indicate infarction. The Smith-modified criterion (ratio > 0.25) improves sensitivity while preserving specificity.
Long 2018 — ECG features of atrioventricular block (PMID 29801849)
Source
American Journal of Emergency Medicine 36(11):2073-2079 — narrative review
Key finding
Mobitz I (progressive PR prolongation then dropped beat) is usually nodal and benign; Mobitz II (constant PR then sudden dropped beat, often wide QRS) is infranodal and carries high risk of progression to complete block, typically warranting pacing. Complete heart block (atrial and ventricular rates independent, no constant PR) requires permanent pacing.
Clinical bottom line
The PR behaviour before the dropped beat distinguishes Mobitz I from II — and determines whether to observe or pace
Camm 2009 — Atrial fibrillation and heart failure (PMID 19451352)
Source
Circulation 119(18):2516-2525 — clinical review
Key finding
AF and heart failure coexist in ~30 per cent of patients and each worsens the prognosis of the other; AF in decompensated heart failure increases mortality and thromboembolism. Rate-control is usually first-line; rhythm-control considered for first episodes, haemodynamic compromise from AF, or failed rate control.
Clinical bottom line
In the ICU, AF is commonly precipitated by sepsis, electrolyte disturbance, hypoxia, and sympathetic surge — treat the trigger first; rate-control with amiodarone if structurally abnormal heart or decompensated HF
Camm 2017 — ECG features of hyperkalaemia (PMID 28296349)
Source
Cardiology Journal 24(2):133-140 — narrative review
Key finding
Hyperkalaemia produces a predictable ECG progression — peaked T waves and shortened QT (early), PR prolongation and P-wave loss (K ~6.5), widened QRS (K ~7-8), sine wave (K > 8), and asystole/PEA/VF (K > 9-10). However, ECG changes correlate imperfectly with the absolute potassium level and can be absent even at dangerous levels — treat the ECG, not the number.
Clinical bottom line
The ECG is more urgent than the biochemistry — a hyperkalaemic ECG gets calcium gluconate immediately, then insulin-dextrose, salbutamol, and definitive removal
de Zwan, Bar & Wellens 1982 — the Wellens syndrome ECG (PMID 7069946)
Source
American Heart Journal 103(4):730-736 — observational case series
Design
Case series of patients with angina at rest and characteristic T-wave changes in V2-V3, correlated with coronary angiography
Key finding
Deeply inverted or biphasic T waves in V2-V3 with preserved R-wave progression, recorded during pain-free intervals, identified a critical (> 50 per cent, often subtotal) stenosis in the proximal LAD. The ECG could normalise between episodes, masking the severity of the underlying lesion.
Clinical bottom line
Wellens syndrome is a pre-infarction state. Stress testing provokes occlusion; the correct response is urgent coronary angiography and revascularisation, not functional testing
de Winter 2008 — a new ECG sign of proximal LAD occlusion (PMID 19005195)
Source
New England Journal of Medicine 359(19):2071-2073 — observational ECG series (Image Challenge)
Design
Descriptive series of patients whose ECGs showed upsloping ST depression with tall symmetric T waves in the precordial leads rather than ST elevation, all with angiographic proximal LAD occlusion
Key finding
A persistent pattern of 1-2 mm upsloping ST depression at the J point in V1-V6 with tall, symmetric, positive T waves and no STEMI-pattern evolution was associated with acute proximal LAD occlusion. The pattern does not progress to classic ST elevation.
Clinical bottom line
De Winter T waves are a STEMI-equivalent occlusion-MI pattern; recognise them as proximal LAD occlusion and activate the catheter lab
Brugada & Brugada 1992 — the Brugada syndrome (PMID 1309182)
Source
Journal of the American College of Cardiology 20(6):1391-1396 — case series
Design
Eight patients with RBBB pattern, persistent ST elevation in V1-V3, and aborted sudden death, despite a structurally normal heart
Key finding
A distinct clinical-electrocardiographic syndrome: right bundle branch block morphology with coved ST elevation in V1-V3 (type 1) in patients with a structurally normal heart, associated with ventricular fibrillation and sudden cardiac death. Later linked to loss-of-function mutations in the cardiac sodium channel (SCN5A).
Clinical bottom line
Brugada pattern is not ischaemic. Fever unmasks or worsens the ST changes and the arrhythmic risk — treat fever aggressively and refer for electrophysiological evaluation and ICD risk stratification
Standards and guidance
- IEC 60601-1 (Medical electrical equipment — Part 1): the international standard defining basic safety and essential performance, including the equipment/applied-part classifications (Class I/II/III; Type B/BF/CF) and leakage-current limits (CF 10 µA normal / 50 µA single fault).[1]
- AS/NZS 3003 (Electrical installations — Patient areas): the Australian/New Zealand standard defining cardiac-protected and body-protected treatment areas, equipotential bonding, and isolated power-supply requirements.[1]
- Thygesen et al., 2018 (Fourth Universal Definition of MI): the current diagnostic standard for STEMI, NSTEMI, and the troponin-based definition of myocardial infarction.[1]
- Surawicz & Knilans, Chou's Electrocardiography in Clinical Practice: the canonical reference for ECG morphology, conduction, and arrhythmia recognition.[1]
References
- [1]Thygesen K, Alpert JS, Jaffe AS, et al. Fourth universal definition of myocardial infarction (2018) Eur Heart J, 2019.PMID 30165617
- [2]Long B, Oliver J, Tagliareni F, Koyfman A Structural changes and digestibility of waxy maize starch debranched by different levels of pullulanase Carbohydr Polym, 2018.PMID 29801849
- [3]Sgarbossa EB, Pinski SL, Barbagelata A, et al. Electrocardiographic diagnosis of evolving acute myocardial infarction in the presence of left bundle-branch block. GUSTO-1 (Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries) Investigators N Engl J Med, 1996.PMID 8559200
- [4]Anter E, Jessup M, Callans DJ In vivo plaque composition and morphology in coronary artery lesions in adolescents and young adults long after Kawasaki disease: a virtual histology-intravascular ultrasound study Circulation, 2009.PMID 19451352
- [5]Camm CF, Corrado S, Kearney M, et al. Dipeptidyl peptidase-4 inhibitors as preferable oral hypoglycemic agents in terms of treatment satisfaction: Results from a multicenter, 12-week, open label, randomized controlled study in Japan (PREFERENCE 4 study) J Diabetes Investig, 2018.PMID 28296349
- [6]de Zwan C, Bar FW, Wellens HJJ Pacemaker implantation in a neonate with congenital complete heart block--a case report Jpn J Surg, 1982.PMID 7069946
- [7]de Winter RJ, Verouden NJ, Wellens HJJ, Wilde AAM General and abdominal adiposity and risk of death in Europe N Engl J Med, 2008.PMID 19005195
- [8]Brugada P, Brugada J Right bundle branch block, persistent ST segment elevation and sudden cardiac death: a distinct clinical and electrocardiographic syndrome. A multicenter report J Am Coll Cardiol, 1992.PMID 1309182