Fundamental electrophysiology
Fundamental electrophysiology
In the last section, we looked at the architecture of the conduction system used by the heart to regularly generate and efficiently disperse electrical signals throughout the myocardium. But how are these signals actually generated, and how do they propagate through this biological wiring? The answer to this comes to us from the field of cardiac electrophysiology: the study of how cardiomyocytes control ion flow using specialized channels, exchangers, and membrane pumps to generate and propagate electrical impulses.
Having a good grasp of electrophysiology can help demystify ECG interpretation and make it much more intuitive. But for many, that’s easier said than done. Learners might face several hurdles trying to learn electrophysiology:
Our goal is to present electrophysiology in a clear and practical way, and have it be a running theme woven into the ECG lessons on this site. The concepts introduced here will come up repeatedly to help reinforce understanding. The goal is to build a robust, functional understanding of electrophysiology that you can easily apply to ECG interpretation.
Every cardiac cell has ions both inside and outside the cell membrane, and specialized ion pumps and channels that closely regulate their concentrations and movement. This creates an electrical charge across the membrane. At baseline, all cells are slightly negative in charge (because of the difference in ion concentrations on either side of the cell membrane) — this is known as the resting membrane potential. Note: “potential” is interchangeable with “voltage” here. We say that cells at resting membrane potential are “polarized”.
For most heart cells, this resting state is stable, maintained by ion pumps. But pacemaker cells are the exception. They have “leaky” ion channels that slowly let in positive ions (i.e., cations), making the membrane potential creep upwards. This is known as DEpolarization (i.e., making the cell less polarized). Once the membrane potential increases past a certain threshold voltage, the cellular floodgates open and a massive surge of positive ions influxes into the cell. The rapidly upward spike of the membrane potential that ensues is called an action potential. Afterwards, ion channels and pumps work to displace cations back out, bringing the cellular voltage back down, towards the resting potential. This is called REpolarization (i.e., restoring polarization). In pacemaker cells, this cycle repeats automatically, thanks to their leaky ion currents constantly nudging them back towards the threshold voltage.
But then what? Because cardiomyocytes are connected by gap junctions, they form a functional syncytium so ions can freely flow between cells. So, when a pacemaker cell fires, it fills with positive ions, which then spread to adjacent cells that are still at their more negative resting potential (since opposite charges attract). This movement of ions from a concentrated, positively-charged area to a low-concentration, negatively-charged area is known as passive or electrotonic spread. As these neighbouring contractile cells flood with positive ions, their membrane potential rises until they, too, reach threshold and fire. This process continues in a chain reaction, with each depolarized cell triggering the next to depolarize, allowing a single action potential generated in one cell to propagate across the entire heart.
Now, how does an action potential translate to muscle contraction? Well, one of the cations rushing in during depolarization is calcium. Calcium ions bind to proteins in the sarcomere (the basic contractile unit of the heart), triggering the sarcomeric actin and myosin filaments to slide past each other, causing the cell to contract (like a microscopic telescoping tube). During repolarization, calcium is pumped back out, allowing the sarcomere shortening to reverse and causing cellular relaxation. This entire process, where action potentials cause physical shortening of the sarcomeres, is known as excitation-contraction coupling.
Some terms are worth clarifying since different sources may use various terms interchangeably. A depolarized cell is often described as having fired, been activated, discharged, or excited. When a cell is repolarizing, it may be referred to as recovering, relaxing, or resetting as its resting potential is restored. If, for whatever reason, the membrane potential drops below the normal resting potential, we say that the cell is hyperpolarized. If one cell triggers an action potential in a neighboring cell, we say the first cell has captured the second, meaning it successfully stimulated it to depolarize.
A propagating action potential may also be called an impulse, a signal, or a depolarization wavefront. It’s also important to note that the term “action potential” can refer not only to the rapid upward spike in membrane potential, but also to the entire cycle of depolarization and repolarization that occurs before a cell returns to its resting state. The length of this cycle is measured by the action potential duration (APD). This will make more sense when you see graphical representations of the action potential.
There are likely more synonyms out there for these concepts, but we’ll stick to these core terms on this site.
Let’s talk about how cells maintain their resting membrane potential (RMP). The RMP, normally around -86 mV, exists thanks to two key factors: the concentration gradient of ions across the membrane, and how easily those ions can pass through the membrane (a.k.a., membrane ion conductance).
Below are some ballpark values for the intracellular and extracellular concentrations of key ions involved in establishing the RMP. Of these, potassium (K+) is the most impactful.
| Ion | Intracellular concentration (mmol/L) | Extracellular concentration (mmol/L) |
| Potassium (K+) | 150 | 5 |
| Sodium (Na+) | 15 | 140 |
| Chloride (Cl–) | 4 | 100 |
| Calcium (Ca2+) | 0.0001 | 2 |
Let’s focus on K+. There’s ~150 mmol/L of K+ in the cell, and only ~5 mmol/L outside. These K+ ions are bound to various negative ions (anions). Naturally, K+ wants to follow its concentration gradient and exit the cell. The cell membrane has channels that allow K+ ions to leave, but the anions they’re bound to cannot pass. So, as K+ exits, the anions that are stuck behind will lower the intracellular voltage, establishing what’s called an electrostatic gradient across the membrane.
This sets up a tug-of-war: the concentration gradient pushes K+ out of the cell, while the electrostatic gradient pulls it back in. The membrane potential at these forces balance out is called the equilibrium potential, or alternatively the Nernst potential.
How do you determine the Nernst potential? Using the Nernst equation (yay math).
Based on K+ alone, the equilibrium potential would be -90 mV. But we can do the same calculation for the other ions, getting the following Nernst potentials: +59 mV for Na+, -85 mV for Cl–, and +131 mV for Ca2+. Now, despite ions constantly moving across the membrane, only a small number need to shift to establish these equilibrium potentials, so the overall ionic concentrations stay relatively stable.
But here’s the question: if the Nernst potentials for the different ions vary so much, how do we end up with a final RMP of -86 mV?
You might think that averaging them would work, but that would land us around +4 mV, which is clearly wrong. The real answer lies in ion conductance. Thinking about it, if an ion can’t freely move across the membrane, there won’t be the charge movement necessary to set up an electrostatic gradient. Naturally, ions with the most permeability will have the biggest influence.
Since K+ has more prevalent membrane channels, it crosses the membrane far more easily than the others, making it the dominant influence on the RMP. When we account for these conductance differences, K+ contributes about 90%, Cl– about 8%, and Na+/Ca2+ a mere ~1% each. Doing a proper weighted average with these values gets us the correct RMP of -86 mV.
Different cells have varying ionic concentrations and membrane ion conductance, meaning the Nernst potential isn’t a fixed value — it can differ between individuals, change within the same person over time, and even vary across different regions of the heart at any given moment.
Several key proteins work to keep the heart’s RMP stable. We’ve noted the main ones worth noting:
We’ve discussed the flow of ions across the cell membrane through leaky channels, which allow ions to move down their concentration gradient, establishing an electrostatic potential. We’ve also covered how different ions have varying permeabilities, meaning they contribute differently to the overall equilibrium potential of the cell. To account for these differences, we used a weighted average.
Another way to think about ion movement, as commonly done in literature, is in terms of the rate of ion movement—that is, how fast ions move. Greater conductance allows ions to move at higher rates. The rate of charge movement is known as current (scientifically represented by the capital letter I). By thinking of transmembrane ion movement as membrane currents, we can quantify how much movement is occurring.
We’ve mentioned several important channels and pumps above. The ion currents associated with them have specific names:
Note: The SERCA pump is not a membrane protein but rather a sarcoplasmic reticulum protein, so it doesn’t have a designated membrane current.
On the last page, we learned that conductive cells can have automaticity and generate impulses on their own (i.e., pacemaker cells), with external input, at a set frequency that can be altered by autonomic stimuli. Normally, the P cells in the sinus node are responsible for generating the impulse that drives the rhythm of the heart. But how does this actually work on a microscopic scale?
First, let’s examine some baseline characteristics of pacemaker cells. As mentioned earlier, pacemaker cells are exceptions in which there is no “stable” resting membrane potential – instead, there is a constant trickling in of cations that gradually nudges the cell towards threshold voltage, eventually causing depolarization. The minimum potential that the cell reaches is usually around -60 mV (also called the maximum diastolic potential), immediately after repolarization finishes. The threshold voltage sits at around -40 mV. Once threshold is reached, a massive influx of cations are allowed in, causing depolarization and ushering in the start of the action potential. The voltage goes up to +30 mV, after which point the cell starts letting out cations to repolarize the cell. Different parts of this cycle are called phases (of the action potential), and range from 0-4. Now, there are some pecularities about how phases are numbered, which makes more sense after we talk about the contractile myocardium action potentials; right now it seems very arbitrary and like the numbering system was poorly retrofitted to work with pacemaker cell action potentials. Important to note that the ion channels that allow this to be achieved are called voltage-gated ion channels, because they are not always open, but they open at certain voltages and close at other voltages. This allows the cell to switch between different phases of the action potential.
Let’s examine this cyclical process step-by-step, starting from a fully recovered pacemaker cell, sitting at -60 mV.
This phase starts with the recovered pacemaker cell, at -60 mV. During this phase, we start seeing a lot of things happen that allow for the gradual inward leakage of cations that drives the cellular voltage towards threshold, which is a process known as diastolic depolarization and is the defining characteristic of pacemaker cells that gives them their automaticity. This is allowed by several mechanisms:
While this is happening, recall that there are a few processes in the background, such as the inward rectifier potassium current (IK1) and the electrogenic Na+/K+ pump current (INaK), which are both repolarizing currents, and provide some sort of counterbalance to the diastolic depolarization.
There is an additional inward-rectifying potassium channel (IKACh) that can be found in sinus node cells, which is another hyperpolarizing current but one that specifically responds to acetylcholine. Therefore, when you have parasympathetic system activation, the IKACh works to lower the cell voltage, essentially hyperpolarizing the membrane potential at the end of repolarization (i.e., lower than -60 mV) and opposing diastolic depolarization, making it harder and making it take longer to reach threshold. The end outcome would be reducing the automaticity and hence the heart rate.
Similarly, sympathetic nervous system can enhance diastolic depolarization and increase automaticity/heart rate through a few mechanisms, such as increasing the sodium conductance in the HCN channel (thereby increasing I_f), increasing ICa,T, acting on the RYR2 to increase the frequency of calcium sparks.
When diastolic depolarization edges the membrane voltage over the threshold voltage of -40 mV, a few things happen.
The HCN and T-type calcium channels close to formally stop diastolic depolarization.
Then, a different kind of voltage-gated channel opens, the L-type (“long-lasting”) calcium channel, which is associated with ICa,L current. This channel lets a lot more calcium in, and at a rapid rate, so the membrane voltage rapidly increases to +30 mV. The ICa,T can trigger even more depolarization through mechanisms outside itself in the following ways:
As rapid as this stage of depolarization is, it is slower than the contractile cell correlate, which involves a much faster sodium channel as you’ll come to see. Sympathetic system activation can increase ICa,L leading to faster, steeper depolarization.
Weirdly enough, there is no Phase 1 or 2 for the conductive cell action potentials. This is probably because the naming system was adopted from the convention used for the contractile myocardium action potential, but it doesn’t necessarily directly apply because there are fewer phases in the conductive cells.
This phase is responsible for bringing the membrane voltage down from the peak voltage of +30 mV to the baseline voltage of -60 mV. There are several mechanisms involved in organizing this:
Sympathetic and parasympathetic stimuli also impact repolarization. The SNS increases IKr and IKs to speed up repolarization and recovery (to shorten time before the cell is excitable again). The PSNS again activates the IKACh to more quickly repolarize, and even hyperpolarize, the cell. Interestingly, both SNS and PSNS can speed up repolarization in the pacemaking cells.
To simplify, the pacemaker cells owe their automaticity to a concept called diastolic depolarization — instead of having a static resting membrane potential, the membrane potential slowly creeps up due to various mechanisms (such as leaky sodium channels). Eventually a threshold voltage is reached and rapid depolarization occurs due to a large influx of calcium ions. Then, after peak voltage, calcium ion influx is shut off and potassium ions flood out of the cell, bringing the voltage back to baseline. This repeats cyclically, causing a cell capable of automatically generating action potentials. While this happens, the cell recycles all the other ions with various pumps and exchangers that reestablish the ionic concentration gradients. The rate of automaticity is largely dependent on the rate of diastolic depolarization. The autonomic system can impact this rate.
Once an impulse is generated by a pacemaker cell in the sinus node, how does it propagate through the heart? Well, to understand that, you need to understand how action potentials propagate.
Before we think about how action potentials travel between cells, we have to think about how action potentials travel within cells. Again, at rest, the inside of the cell is negative compared to the outside. When an action potential fires off in one part of the cell membrane (for whatever reason – automaticity, or external stimulus), it lets a rush of cations into the cell at that location. When cations flood into one area of the membrane, the cations spread outward towards the more negative parts of the cell. This is called electrotonic conduction, because ions move down their concentration and charge gradients. When the cations diffuse into parts of the cell that were previously negatively charged, they raise the membrane voltage at those regions, perhaps above threshold, and then another action potential fires at the new location and spreads outwards from there. It becomes this chain reaction situation that eventually traverses down the length of the cell in either direction from the original spot where it originated.
When this chain reaction reaches the border between two adjacent cells (i.e., the intercalating disc), it passes through the gap junctions into the resting cell. The more or larger the gap junctions, the faster the electrotonic movement of ions from one cell to the next, influencing the overall conduction velocity. Once the positive ions that were the results of action potentialling in one cell travel to an adjacent cell, they can again trigger this chain reaction again until it propagates to the next cell.
Since the heart is a form of functional syncytium, all of the cardiomyocytes in the heart are theoretically connected to one another through gap junctions, meaning an action potential in one part of the heart can make its way through to all the other cells of the heart. Of course, the exception are non-cardiomyocyte cells, such as fibroblasts, which are not electrically coupled to the cardiomyocytes.
Insulation around a cardiomyocyte helps enhance the speed at which signals propagate. Why? Let’s think about what happens in the ECM as a cell undergoes an action potential. The outside of the cell becomes locally more negative at the site of action potential, as the cations flood into the cell. The potential difference between negatively charged outside and positively charged inside is a form of potential (or stored) electrical energy. However, just like how there is electrotonic spread of positive ions to more negative spaces inside the cell, there will be electrotonic spread of positive ions from neighbouring ECM to the locally negative ECM at the site of action potential. This will reduce the local negative ECM charge, and by doing so, dissipate some of that potential energy. Insulation around the cell reduces the ability for dissipation of electrical energy laterally outside of the cardiomyocytes. This energy is preserved for propagating action potentials.
We’ve seen what the action potential of pacemaking cells, such as those found in the sinus node or AV node, look like in the previous section. The action potentials of contractile cardiomyocytes, such as those found in the working muscle of the atria or ventricles, work differently. There are also other cells, such as transitional cells, that share properties with both conductive and contractile cardiomyocytes. However, we’ll focus on the contractile cardiomyocytes, which make up the bulk of cardiac tissue and cardiac action potentials.
Unlike pacemaker cells, contractile cardiomyocytes do not normally have automaticity. They do not exhibit diastolic depolarization. If left to their own devices, they will simply lay dormant at the resting membrane potential, as opposed to firing spontaneously. They require input from external sources (i.e. electrotonic spread from neighbouring cells that have fired, or external electrical sources such as an implantable pacemaker). In other words, the cells need to be TRIGGERED, or CAPTURED, by something external to those cells themselves, in order to fire. Once something does trigger them, the impulses can travel through the cell as mentioned above through electrotonic spread, and then these cells can go on to become the trigger for the next cells in the syncytium.
Let’s again break down the function of these contractile cells, and their action potentials, phase by phase.
The action potential begins (and ends) with phase 4. This is when the cell is at RMP, roughly -86 mV. As mentioned earlier, this RMP is maintained by currents such as IK1, INCX, INaK, as well as SERCA activity. The cell at this point waits for a stimulus in the form of electrotonic spread of positive ions from a neighbouring cell (due to a propagating impulse). Once triggered, the cell voltage will increase. If it hits the threshold voltage of -70 mV, we move on to the next phase. If not, the electrical energy will dissipate, no action potential will be fired, the gradients will get reestablished, and the cell will remain in rest.
With a sufficieicnt enough stimulus, the threshold voltage will be reached. At this point, the IK1 will turn off, and the voltage-gated fast sodium channels (associated with the INa current) will open. This causes a rapid flood of sodium ions into the cell, causing depolarization. This ion channel has a high conductance, so ions will very rapidly move into the cell. This causes a sharper spike in the action potential compared to the pacemaker cells, which rely on slower ICa,L currents to depolarize. The cell will depolarize up to around +30 mV.
In terms of details on the fast sodium channel, it’s a bit unique from other voltage-gated channels. Other channels may only have one physical “gate” that swings open when you enter a range of voltages that it deems acceptable, and closes shut once you get out of that range. This channel interestingly has TWO gates. It has an m gate (a.k.a. a voltage-dependent activation gate), and an h gate (a.k.a., a time-dependent inactivation gate). The way it works is as follows:
This can be succinctly and amusingly represnted by the diagram below, that captures as much nuance as is needed on this site.
Notably, there is a caveat. Even if the h gate closes and inactivates the sodium channel, there is often a smlall amount of leakage current that continues to leak small amounts of sodium into the cell through phases 1, 2, and sometimes 3. This is known as the late sodium current (INa,late), also sometimes called the persistent sodium current (IpNa). There is unfortunately some value in knowing it because it does relate to pathology.
After the rapid depolarization phase, at the peak voltage, the fast sodium channels will shut off, and you get opening of the voltage-gated transient outward potassium current channel, which corresponds with the Ito current. This briefly allows a small amount of K+ to exit the cell early on, causing a small drop in the membrane voltage and a downward notch on the graphical representation of the action potential. This is the dominant current.
At this point, the L-type calcium channels open too, allowing a small amount of ICa,L into the cell. However, at this stage, this is a much smaller current and not as relevant compared to Ito.
Recall that this phase does not exist in pacemaker cells.
This phase is unique to contractile cells. At this phase, you get opening of the L-type calcium channels, and ICa,L becomes the dominant inward current. The inward calcium shift also triggers CICR from the sarcoplasmic reticulum, and activates the INCx, which also contributes a depolarizing current.
However, in this phase, you also get gradual opening of the delayed potassium rectifier channels, IKr and IKs, which contribute a repolarizing current.
Overall, the inward calcium current opposes the outgoing potassium current, and you get a relative “plateau” of the action potential, where the membrane potential doesn’t change much. You may get a initial positive deflection on the action potential graph at the beginning of phase 2, because at this point the depolarizing effects of ICa,L are dominant, but then you may see that the graph develops a subtle downslope towards the end of the phase, signalling the dominance of the potassium currents at this point.
Recall that this phase doesn’t exist in pacemaker cells.
In this phase, you get more dominant action of various outgoing potassium currents (IKr, IKs, IK1) and closure of the ICa,L current. These work in tandem to bring the membrane potential back down to resting. We then come back towards phase 4 and the cell will again wait for another impulse.
The cardiac action potential is not homogeneous. In fact, because of changes in ion concentration, ion conductances, and channel expression in different parts of the heart, the morphology of the action potential varies widely, as shown in the image below. As you can see, the SA nodal and AV nodal action potentials are similar, with the slower depolarization velocity due to reliance on ICa,L channels.
In contrast, the action potentials of the working myocardium in the atria and ventricles are similar, with the spike and plateau shape. This is because of depolarization related to fast sodium channels. However, the morphology of atrial APs differs from that of ventricular APs, and even different parts of the ventricular wall have different AP morphologies.
Important to note here a few points which will become salient later: the atrial AP is relatively short, and the epicardial AP is a lot shorter than the endocardial or midmyocardial APs. The midmyocardial AP is actually the longest. In fact, the epicardial AP is so short in comparison to the endocardial AP, that even if we account for the slight delay in activating the epicardium (compared to the time taken to activate the endocardium), usually the epicardium will FINISH repolarizing first.
Contractile myocardium does not have the same density of autonomic innervation as the nodes. So, they don’t express as much channels that are directly responsive to this stimuli. This is why parasympathetic tone, while it may slow down automaticity and conduction at the SA or AV nodes, it will not similarly affect the conduction in the atrial or ventricular myocardium.
References