m10 · Arrhythmia mechanisms: focal, macro-reentry, rotors, atrial fibrillation
Education only. Not for diagnosis, treatment or any clinical decision.
In 60 seconds
Most tachycardias that mapping systems are asked to explain come down to three mechanisms. A focus fires faster than the sinus node and its waves spread outward from one point. Reentry is a wave that keeps finding excitable tissue ahead of it around a circuit, either fixed by anatomy or scar (macro-reentry) or formed by the wave itself. A rotor is a spiral wave turning around a phase singularity. Whether reentry can last depends on the wavelength, velocity times refractory period, against the length of the circuit. Whether rhythms stay stable depends on restitution: if action potential duration changes faster than the rest interval, beats start to alternate and waves break. Atrial fibrillation mixes these ideas: focal triggers, often in the pulmonary veins, start it, and short wavelengths and wave breaks sustain it. A map shows which mechanism fits: activation spreading outward from an earliest point, or activation covering the whole cycle with early meeting late.
From the field
- Short wavelengths go with flutter and fibrillation. In conscious dogs the atrial wavelength was 14–18 cm during normal rhythm; flutter was induced below 9.7 cm and fibrillation below 7.8 cm. The wavelength predicted which arrhythmia was induced in 75% of cases, better than refractory period (48%) or velocity (38%) alone (Rensma et al. 1988).
- Atrial fibrillation often starts from a focus. In 45 patients, 65 of 69 ectopic foci that initiated atrial fibrillation (94%) were in the pulmonary veins, firing in bursts of 340 per minute (Haïssaguerre et al. 1998).
- Reentry needs no anatomical obstacle. Allessie and colleagues described circus movement in atrial muscle without one, the “leading circle” (Allessie et al. 1977).
- Rotor-guided ablation is contested. One trial reported 82.4% against 44.9% freedom from atrial fibrillation when rotors and focal sources were ablated (Narayan et al. 2012); a later two-center series reported 37% freedom from atrial fibrillation at 18 months and called for randomized studies (Buch et al. 2016).
Derivations
Symbols used throughout. CV conduction velocity (mm/ms = m/s); ERP effective refractory period (ms); \(\lambda\) wavelength (mm); \(L\) circuit length (mm); CL cycle length (ms); APD action potential duration, DI diastolic interval (ms); \(f\) the restitution function \(\text{APD} = f(\text{DI})\); \(\tau_{\text{open}}\), \(\tau_{\text{close}}\), \(h_{\min}\) the m08 model constants; \(\mathbf x_0\) the position of a focus; \(T\) activation time.
Wavelength and reentry
A wave leaves refractory tissue behind it for one refractory period, so it occupies a length of tissue:
\[ \lambda = \text{CV}\times\text{ERP} \tag{1}\]
Rensma and colleagues measured both factors in conscious dogs and found the product a better predictor of induced arrhythmias than either factor alone (Rensma et al. 1988). Drugs that shorten refractoriness or slow conduction shorten the wavelength. A wave can circulate around a fixed circuit only if the circuit is longer than its wavelength:
\[ \text{CL} = \frac{L}{\text{CV}}, \qquad L > \lambda, \qquad \text{gap} = L - \lambda = \text{CV}\,(\text{CL} - \text{ERP}) \tag{2}\]
The excitable gap is the stretch of recovered tissue ahead of the front; pacing that reaches it can enter the circuit, which is what entrainment uses (m11). In a Mitchell–Schaeffer ring whose pulse leaves 75 mm of tissue above \(v_{\text{gate}}\), reentry continued on rings of 90 mm and more and failed on 80 mm; on shorter rings the front ran into less-recovered tissue and slowed, from 0.31 m/s on a 200-mm ring to 0.25 m/s on a 90-mm ring (lab part B).
Exact
Assumptions
- Model: Equation 1 and Equation 2 for a single circuit with uniform velocity and refractory period. Tested against
epml.arrhythmia. - Parameters: circuit length, velocity and refractory period are free choices for illustration.
- What it ignores: velocity and refractory period that change with the rest interval (lab part B), curvature, anatomy and heterogeneity.
Key takeaways
- The wavelength \(\text{CV}\times\text{ERP}\) is the tissue a wave occupies; a circuit must be longer than it.
- The excitable gap is the circuit’s spare room, in millimeters or milliseconds.
Restitution and alternans
When pacing at cycle length CL, each beat’s APD depends on the rest before it, and that rest is what the previous beat left:
\[ \text{APD}_{n+1} = f(\text{CL} - \text{APD}_n), \qquad \delta_{n+1} = -f'(\text{DI}^*)\,\delta_n \tag{3}\]
A small disturbance \(\delta\) around the steady state changes sign every beat and grows when the restitution slope \(f'\) exceeds 1: APDs alternate long and short. Nolasco and Dahlen studied this graphically with the restitution curve and the pacing line (Nolasco and Dahlen 1968). With the m08 closed form, the slope reaches 1 at
\[ \text{DI}_c = \tau_{\text{open}}\ln\!\Big[(1 - h_{\min})\Big(1 + \frac{\tau_{\text{close}}}{\tau_{\text{open}}}\Big)\Big] \tag{4}\]
that is DI = 54.7 ms, APD = 95.8 ms and CL = 150.5 ms with the m08 parameters. A paced model cell alternates at a longer cycle length, 210 ms, and drops to 2:1 capture at 190 ms, because its APDs run longer than the closed form (lab part C). In guinea pig hearts, repolarization alternated out of phase between neighboring sites above a critical rate, and a small further acceleration produced unidirectional block, reentry and ventricular fibrillation (Pastore et al. 1999).
Exact
Assumptions
- Model: the restitution map Equation 3 with the m08 closed-form restitution (eq-m08-apd), starting 5 ms away from the steady state. Tested against
epml.arrhythmia. - Parameters: Mitchell–Schaeffer constants as in m08 (Nieto Ramos et al. 2023); the cycle length is adjustable.
- What it ignores: memory beyond one beat, calcium-driven alternans, coupling to neighboring cells, and the gap between the closed form and simulated APDs (m08).
Key takeaways
- Restitution slope above 1 makes the steady rhythm unstable: alternans.
- The threshold depends on the whole restitution curve, not on one APD value.
Rotors and phase singularities
A wavefront that breaks, on an obstacle or on a refractory island, curls at its free end into a spiral. The spiral turns around a point where all phases meet, a phase singularity, and the topological charge of m29 (eq-m29-charge) is \(\pm1\) around it, the sign giving the direction of rotation (Bray et al. 2001). In a Barkley-model sheet (illustrative, dimensionless) started with cross-field conditions, one singularity with charge \(-1\) forms at the center; the mirror image has charge \(+1\) (lab part D). The spiral’s tissue footprint is set by the same wavelength as above, so short wavelengths allow more rotors in a chamber.
Key takeaways
- A rotor is reentry around a phase singularity, with no anatomical obstacle needed (Allessie et al. 1977).
- Singularities carry charge \(\pm1\); they appear and vanish in pairs.
Focal or reentrant activation
From a focus at \(\mathbf x_0\), activation spreads outward:
\[ T(\mathbf x) = T_0 + \frac{\lvert\mathbf x - \mathbf x_0\rvert}{\text{CV}}, \qquad \text{span} = \frac{\max T - \min T}{\text{CL}} < 1 \tag{5}\]
The activation map has one earliest point and covers only the time to reach the farthest point: 11% of a 300-ms cycle for a focus in a 40 × 40 mm area at 1.0 m/s. Around a circuit the activation is continuous and covers the whole cycle, so the latest site borders the earliest (“early meets late”) (lab part E). The span depends on how much of the chamber is mapped, so a partial map of a circuit can look focal. A consensus statement classifies atrial flutter and regular atrial tachycardia by such mechanisms and their anatomical bases (Saoudi et al. 2001).
Key takeaways
- Focal: centrifugal spread from an earliest point; the activation span is short.
- Reentry: activation covers the cycle; the map must include the whole circuit to show it.
Clinical correlate
This section describes published findings only. It is not guidance for patient care, and it has not yet been reviewed by an EP physician.
Where this shows up in EP. Mapping is a search for mechanism. Focal sources show centrifugal activation; in one series most foci that initiated atrial fibrillation were inside the pulmonary veins, 2 to 4 cm from the ostium, with local activity preceding the ectopic beat on the surface ECG by 106 ± 24 ms (Haïssaguerre et al. 1998). Macro-reentry is classified by its circuit and anatomical basis (Saoudi et al. 2001), and its activation covers the cycle (Equation 5). Whether rotors sustain human atrial fibrillation is contested: a trial of rotor-guided ablation reported higher freedom from atrial fibrillation (Narayan et al. 2012), and a later series reported poor long-term efficacy (Buch et al. 2016). Wavelength ties these mechanisms to tissue properties (Rensma et al. 1988), and electrical remodeling (m08) shortens it (Wijffels et al. 1995).
What the evidence is. Animal studies (Allessie et al. 1977; Rensma et al. 1988; Wijffels et al. 1995), a clinical mapping series (Haïssaguerre et al. 1998), an isolated-heart study (Pastore et al. 1999), a consensus classification (Saoudi et al. 2001), a controlled clinical study (Narayan et al. 2012) and a retrospective multicenter series (Buch et al. 2016).
What a reader should ask:
- Does the map cover the whole circuit? A partial map of reentry can look focal (Equation 5).
- How were singularities detected, with what electrode spacing? See m29.
- Was outcome assessed against a control group, with what monitoring? As the rotor studies show, mechanism claims and outcomes need separate evidence.
Visual summary
Schematic
Assumptions
- Schematic: not to scale, no quantitative claim. Labels follow Equation 5, Equation 2 and m29.
How experts think
When a map shows a tachycardia, ask:
- Where is the earliest activation, and does activation cover the cycle? Focal or reentrant (Equation 5).
- If reentrant, where is the circuit, and is it longer than the wavelength? (Equation 2)
- Is there an excitable gap? It decides whether pacing can enter the circuit (m11).
- Is the rhythm stable from beat to beat? Alternation points to steep restitution (Equation 3).
- For rotors: how were they detected, and would the detector find them in noise? (m29)
One sentence to keep: wavelength sets what can fit, restitution sets what stays stable, and the map has to cover enough of the chamber to tell.
Cheat sheet
| Quantity | Result |
|---|---|
| Wavelength (Equation 1) | \(\lambda = \text{CV}\times\text{ERP}\) |
| Circuit (Equation 2) | \(\text{CL} = L/\text{CV}\); needs \(L > \lambda\); gap \(= L - \lambda\) |
| Restitution map (Equation 3) | \(\delta_{n+1} = -f'\delta_n\); alternans when \(f' > 1\) |
| Alternans onset, closed form (Equation 4) | \(\text{DI}_c = \tau_{\text{open}}\ln[(1 - h_{\min})(1 + \tau_{\text{close}}/\tau_{\text{open}})]\) |
| Focal spread (Equation 5) | \(T = T_0 + \lvert\mathbf x - \mathbf x_0\rvert/\text{CV}\); span below 1 |
| Rotor (eq-m29-charge) | charge \(\pm1\) around a phase singularity |
Values used in this module
| Value | Source |
|---|---|
| Dog atrial wavelength 14–18 cm; flutter below 9.7 cm, fibrillation below 7.8 cm; 75% prediction | Rensma et al. (1988) |
| 94% of initiating foci in the pulmonary veins; bursts of 340 per minute; 106 ± 24 ms lead | Haïssaguerre et al. (1998) |
| Rotor-guided ablation 82.4% vs 44.9% freedom from atrial fibrillation | Narayan et al. (2012) |
| 37% freedom from atrial fibrillation at 18 ± 7 months | Buch et al. (2016) |
| Ring lengths, velocities, alternans thresholds, singularities, spans | our own code |
Interview questions
Try each question before you open its model answer. In a real interview, say which assumptions you are making.
Q1. Define the wavelength of excitation and why it matters.
Model answer
\(\text{CV}\times\text{ERP}\), the length of tissue a wave keeps refractory. A circuit must be longer than it for reentry; short wavelengths let more and smaller reentrant waves fit (Equation 1) (Rensma et al. 1988).
Q2. A drug slows conduction and prolongs refractoriness. What happens to the wavelength?
Model answer
It depends on the balance; the product can stay the same. Rensma and colleagues saw drugs with opposite effects on the two factors change the wavelength little (Rensma et al. 1988).
Q3. What is the excitable gap, and why does an EP study care?
Model answer
The recovered tissue ahead of the front, \(L - \lambda\) or \(\text{CL} - \text{ERP}\); pacing captures into the circuit through it, which entrainment uses (Equation 2; m11).
Q4. Why can’t reentry continue on a ring shorter than the wavelength?
Model answer
The front reaches tissue still refractory from the same wave and blocks. Near the limit it also slows, because it travels into less recovered tissue (lab part B).
Q5. Explain alternans with the restitution map.
Model answer
\(\text{APD}_{n+1} = f(\text{CL} - \text{APD}_n)\); deviations are multiplied by \(-f'\) each beat, so with slope above 1 they alternate and grow (Equation 3) (Nolasco and Dahlen 1968).
Q6. Why might alternans lead to fibrillation?
Model answer
When repolarization alternates out of phase in neighboring regions, steep gradients of refractoriness can block a wave in one direction and start reentry, as shown in guinea pig hearts (Equation 3) (Pastore et al. 1999).
Q7. What is a phase singularity?
Model answer
A point around which phase takes every value, so the phase winds by \(\pm 2\pi\) around it; a spiral wave turns around one (eq-m29-charge) (Bray et al. 2001).
Q8. Does reentry need an anatomical obstacle?
Model answer
No: functional reentry, the leading circle or a rotor, turns around tissue kept refractory by the wave itself (Allessie et al. 1977).
Q9. How do you tell focal from reentrant activation on a map?
Model answer
Focal: one earliest site with outward spread and a short activation span. Reentry: activation covering the cycle, early meeting late. Check that the map covers the circuit (Equation 5).
Q10. Where do atrial fibrillation triggers usually come from?
Model answer
In the series that showed it, 94% of initiating foci were inside the pulmonary veins (Haïssaguerre et al. 1998).
Q11. A vendor claims rotor maps guide atrial fibrillation ablation. What evidence would you want?
Model answer
How rotors were detected and validated, and randomized outcome data with careful monitoring; results so far conflict (Narayan et al. 2012; Buch et al. 2016).
Q12. Why does electrical remodeling make atrial fibrillation more likely to persist?
Model answer
It shortens refractoriness, which shortens the wavelength, so more reentrant waves fit (Equation 1) (Wijffels et al. 1995; Rensma et al. 1988).
Code lab
The lab builds each mechanism in a model.
| Part | What you do | Data |
|---|---|---|
| A | The wavelength of excitation (Equation 1) | model values |
| B | Reentry on a ring: the shortest circuit (Equation 2) | synthetic |
| C | Restitution and alternans (Equation 3, Equation 4) | synthetic |
| D | A spiral wave and its phase singularity | synthetic |
| E | Focal or reentrant? The activation span (Equation 5) | synthetic |
- Python notebook: labs/python/m10_lab.ipynb. Run it from the course repository after
pip install -r requirements.txt, or open it in Google Colab. - Data: none to download.
Each part ends with checks: lines that pass with the code as given. If you change something and a check fails, that is information, not an error to hide.
Self-check
Short checks you can do in your head. Open each answer after you have tried it.
S1. CV 0.6 m/s, ERP 150 ms. Wavelength?
Answer
90 mm (Equation 1).
S2. A 150-mm circuit conducting at 0.5 m/s. Cycle length?
Answer
300 ms (Equation 2).
S3. Same circuit, ERP 200 ms. Excitable gap in ms and mm?
Answer
100 ms and 50 mm (Equation 2).
S4. Restitution slope 0.5 at the steady state. A 4-ms disturbance after two beats?
Answer
1 ms, with the same sign as at the start (Equation 3).
S5. Total charge around a loop enclosing two rotors of opposite rotation?
Answer
Zero (eq-m29-charge).
S6. A focus in the middle of a 60-mm region at 1 m/s, CL 400 ms. Activation span?
Answer
About 30 mm / 1 m/s = 30 ms to the edge (more to the corners): under 15% of the cycle (Equation 5).
Literature figures
These figures are worth looking at alongside this module. We link to them instead of copying them: some licenses do not allow reuse here, and others we could not confirm. Each entry says what to look for and where this module explains it.
- Wavelength and arrhythmia type. Rensma and colleagues plot wavelength for rapid responses, flutter and fibrillation (Rensma et al. 1988). This module: wavelength and reentry.
- Pulmonary vein foci. Haïssaguerre and colleagues map the sites of initiating ectopic beats (Haïssaguerre et al. 1998). This module: clinical correlate.
- Rotor maps. Narayan and colleagues show phase maps of rotors in human atrial fibrillation (Narayan et al. 2012). This module: rotors and phase singularities.
Further reading
- Reentry. Leading circle (Allessie et al. 1977); wavelength (Rensma et al. 1988); classification (Saoudi et al. 2001).
- Stability. Alternans (Nolasco and Dahlen 1968; Pastore et al. 1999); phase singularities (Bray et al. 2001).
- Atrial fibrillation. Triggers (Haïssaguerre et al. 1998); remodeling (Wijffels et al. 1995); rotor-guided ablation (Narayan et al. 2012; Buch et al. 2016).
- Models. Barkley’s two-variable model (Barkley 1991); the cell model of m08 (Nieto Ramos et al. 2023).
- Related modules. m08 (action potentials), m09 (propagation), m11 (the EP study), m24 (activation maps), m29 (complex maps), m41 (locating arrhythmia origins).