Preload, Frank–Starling Relationships, and Ventricular Filling

A ventricle needs adequate filling to eject blood. Beyond a certain point, however, more filling may produce little additional flow and a substantial rise in pressure. Understanding this balance requires a clear distinction between myocardial stretch, filling pressure, and the response to a change in loading.

8.1 What Preload Actually Means

Preload describes the mechanical loading of the myocardium at the end of diastole, before contraction begins. Sarcomeres are the repeating contractile units within muscle cells. Their length before contraction helps describe preload at the muscle level. For the whole ventricle, end-diastolic wall stress describes the force borne per unit area of myocardium. A routine pressure transducer measures neither sarcomere length nor wall stress directly.

End-diastolic volume is often used as an estimate of preload because ventricular filling generally lengthens myocardial fibers. The amount of stretch also depends on chamber shape and wall thickness. A given volume within a small thick-walled ventricle therefore does not impose the same muscle loading as that volume within a dilated ventricle.

Filling pressure is another estimate used at the bedside. The pressure needed to fill a chamber depends on how completely the muscle relaxes and how readily the chamber distends. Pressure around the heart and pressure from the neighboring ventricle also contribute. Filling pressure can therefore rise substantially without a similar increase in the myocardial stretch that supports ejection.

Total blood volume does not specify ventricular preload either. Venous constriction or dilation can redistribute the same blood volume and change cardiac filling. An obstruction to return can prevent blood from reaching the right heart. Poor right ventricular ejection can prevent blood from reaching the left heart. A patient can therefore retain substantial fluid while one ventricle remains inadequately filled.

Keep the measurement separate from its interpretation. “Central venous pressure is 12 mm Hg” reports a pressure. “Preload is adequate” claims that the current myocardial loading is sufficient for the patient’s needs. That conclusion requires information about ventricular function, surrounding pressure, flow, and perfusion.

8.2 The Frank–Starling Mechanism

The Frank–Starling mechanism allows the heart to develop greater force when myocardial fibers begin contraction at a greater length within their physiological working range. This intrinsic response helps the heart match ejection to the blood arriving during filling.

The mechanism involves more than improved overlap between actin and myosin. At a greater initial length, the contractile apparatus can generate more force in response to a given calcium signal. This is length-dependent activation. Contractile proteins and structural elements influence that response. Activation of neighboring parts of the thin filament can also reinforce one another, a process called cooperative activation.

Terui and colleagues studied length-dependent activation in permeabilized porcine ventricular muscle. Their experiments demonstrated an important contribution from cooperative activation of the thin filament. This is experimental evidence about myocardial mechanisms, rather than a clinical rule specifying how much fluid a patient should receive. 1

A stronger beat can arise by two different routes. If the same myocardium contracts more strongly because it starts at a greater length, that is the Frank–Starling response. If it contracts more strongly at a comparable initial length because its activation state changes, contractility has increased. Fluid and vasoactive treatment may change both loading and activation in a patient, so an observed stroke-volume increase does not automatically isolate one mechanism.

The mechanism contributes to balancing right and left ventricular output. If one ventricle temporarily receives more blood, increased filling can support greater ejection on subsequent beats. The adjustment has limits. Severe dysfunction, excessive afterload, or mechanical constraint can prevent the expected increase in forward flow.

The clinical implication is conditional. Increasing filling can increase stroke volume when sufficient functional reserve remains and the added filling reaches the relevant ventricle. It does not follow that increasing filling is always necessary, safe, or beneficial.

8.3 Reading a Ventricular Function Curve

A ventricular function curve plots a measure of output against a measure of filling. Stroke volume is often placed on the vertical axis. The horizontal axis may show end-diastolic volume, filling pressure, or another estimate of preload. These are different curves if the relationship between pressure and volume changes.

On a steep portion of the curve, a small filling increase produces a clear stroke-volume increase. On a flatter portion, the same filling increase produces little extra ejection. The slope describes the response near the patient’s current position on the curve. Treatment or illness can change that position and the curve itself.

The curve changes when other determinants change. Greater contractility can increase stroke volume at the same filling. Greater afterload can reduce it. Rate, rhythm, valve function, and the other ventricle also affect ejection. If one of these changes during a fluid challenge, the observed result may combine a filling response with a change in the ventricular function curve.

The plateau on a teaching diagram does not identify one filling pressure that applies to everyone. A stiff ventricle can reach a high pressure after a small volume increase. A more compliant ventricle may accept more volume with a smaller rise. When the horizontal axis uses pressure, equal positions on the graph can therefore represent different chamber volumes and myocardial loading.

Preload responsiveness means that a defined increase in filling produces a measurable increase in stroke volume or cardiac output. It does not mean that the circulation lacks fluid. A healthy resting heart can be preload responsive. A critically ill patient may remain responsive despite clinically important pulmonary or systemic congestion.

Responsiveness also depends on timing. Sepsis, vasopressors, ventilation, ischemia, and rhythm changes can move the operating point. A result obtained earlier in resuscitation should not be assumed to describe the present circulation.

8.4 How the Ventricle Fills During Diastole

Ventricular filling requires a pressure gradient from the atrium to the ventricle while the atrioventricular valve is open. The gradient changes throughout diastole. Filling therefore depends on events before and during valve opening, rather than simply on the pressure measured at the end.

Relaxation requires active cellular work. Cytosolic calcium must fall, and actin–myosin cross-bridges must detach. ATP supports calcium handling and cross-bridge detachment. Ischemia or disturbed calcium regulation can therefore delay relaxation. The ventricle may still eject a preserved fraction of its volume while taking too long to relax for the next filling period.

As left ventricular pressure falls below left atrial pressure, the mitral valve opens. The pressure difference moves blood into the ventricle during early filling. Ventricular recoil helps lower ventricular pressure and sustain this gradient. As atrial and ventricular pressures approach each other, filling slows during diastasis. Atrial contraction then pushes additional blood into the ventricle before systole.

The contribution of each filling phase varies. When early relaxation is impaired, late filling from atrial contraction may become more important. Atrial fibrillation removes that organized contraction and makes the interval available for each beat unequal. A ventricle that depends on late filling may then lose stroke volume, especially at a rapid ventricular rate.

Tachycardia shortens the cardiac cycle, often leaving substantially less time for diastolic filling. It can also prevent relaxation from completing before the next contraction. A higher heart rate may then fail to increase output because stroke volume falls. This is particularly relevant when the ventricle is stiff or relaxation is already impaired.

In a prospective study of 262 patients with severe sepsis or septic shock, Landesberg and colleagues found that reduced early mitral annular diastolic velocity was associated with mortality. The finding supports the clinical importance of diastolic function. It does not show that changing a particular Doppler value or filling pressure improves survival. 2

8.5 Relaxation, Compliance, and the Diastolic Pressure–Volume Relationship

Relaxation and compliance describe different properties. Relaxation is the time-dependent decline in active myocardial tension. Compliance describes how much chamber volume changes for a given change in distending pressure. A ventricle may have impaired relaxation, reduced compliance, or both.

Over a specified range, an approximate chamber compliance can be written as:

If volume is measured in mL and pressure in mm Hg, compliance is expressed in mL/mm Hg. Stiffness is the inverse relationship: the pressure change required for a given volume change. A single compliance value should not be assumed to apply across the entire filling range.

The passive end-diastolic pressure–volume curve is not a straight line. As volume increases, each further volume increment generally requires a larger pressure increase. The chamber becomes harder to distend. Chamber shape and pressure around the heart also influence the curve. A higher intracavitary pressure at the same volume therefore does not prove that the muscle tissue itself has become stiffer.

Consider an illustrative increase in chamber volume of 20 mL. If distending pressure rises by 2 mm Hg:

Compliance = 20 ÷ 2 = 10 mL/mm Hg

If the same 20 mL increase requires a pressure rise of 10 mm Hg:

Compliance = 20 ÷ 10 = 2 mL/mm Hg

The second chamber accepts less volume for each unit pressure increase. Its average compliance over the tested interval is lower. These assumed values are neither normal ranges nor treatment targets.

In a poorly compliant ventricle, a small volume increase can therefore cause a large rise in filling pressure. If left atrial and pulmonary venous pressures rise with it, pulmonary microvascular pressure can increase and drive more fluid into lung tissue. Higher right-sided filling pressure can worsen systemic venous congestion. These effects can occur before stroke volume increases enough to improve perfusion.

Measure the flow response and assess the consequences of the pressure rise. Additional filling is useful when it produces needed forward flow without causing clinically important pulmonary edema or systemic congestion. The fact that more volume entered the chamber does not establish that treatment helped.

8.6 Why Filling Pressures Do Not Predict the Response to Fluid

Central venous pressure estimates right atrial pressure under suitable measurement conditions. Pulmonary artery occlusion pressure, when correctly obtained and physiologically interpretable, can estimate left atrial pressure. Neither measurement directly identifies the slope of the ventricular function curve.

The same filling pressure can arise in several ways. A low pressure may reflect little chamber volume, but a compliant chamber can also contain more blood at low pressure. Low surrounding pressure can lower the measured value further. A high pressure may reflect more volume, delayed relaxation, poor compliance, external compression, or an obstacle to onward flow. Identify which mechanism fits the chamber and the rest of the circulation.

Pulmonary artery occlusion pressure and left ventricular end-diastolic pressure also represent different measurements. Mitral stenosis creates a pressure difference between the left atrium and ventricle, so an elevated atrial pressure may not represent the pressure within the filling ventricle. Mitral regurgitation can produce large atrial pressure waves that raise mean occlusion pressure. Even with normal valves, an atrial pressure averaged over the cycle is not the same as ventricular pressure measured at the end of diastole.

Kumar and colleagues examined normal subjects during volume infusion. Baseline central venous and pulmonary artery occlusion pressures did not reliably predict ventricular filling volume or the subsequent cardiac response. These were healthy volunteers, so the study should not be presented as an ICU outcome trial. 3

Osman and colleagues evaluated 150 volume challenges in 96 patients with sepsis. Filling pressures had poor ability to predict the hemodynamic response. This clinical study reinforces the distinction between a static pressure and the response to an intervention. 4

A filling pressure still provides useful information. Its waveform, trend, and relation to respiratory pressure can help identify congestion, valve abnormalities, right ventricular dysfunction, or external constraint. The error is asking it to answer a different question: how much output will increase after more filling.

8.7 Transmural Pressure and External Constraint

A pressure transducer normally reports pressure relative to atmosphere. Chamber distension depends on the pressure difference across the wall:

Pressure outside the chamber need not be atmospheric. Pleural pressure and the pericardium affect it. Abdominal pressure transmitted into the thorax and expansion of the neighboring chamber can also constrain filling. A transducer inside the chamber does not separate these external influences from the pressure stretching its wall.

For an illustrative intracavitary diastolic pressure of 16 mm Hg and surrounding pressure of 10 mm Hg:

Transmural pressure = 16 − 10 = 6 mm Hg

Now raise both pressures by 4 mm Hg. Intracavitary pressure becomes 20 mm Hg and surrounding pressure becomes 14 mm Hg:

Transmural pressure = 20 − 14 = 6 mm Hg

The displayed intracavitary pressure has risen, but the pressure difference distending the chamber has not. This simplified example assumes that surrounding pressure is known.

Tyberg and colleagues measured pericardial surface pressure during volume loading in nine cardiac surgical patients. Their observations showed that the normal pericardium can contribute appreciably to measured filling pressure. The controlled operative setting limits direct application of its numerical relationships to all ICU patients. 5

Positive end-expiratory pressure can raise pressure around the heart. How much airway pressure reaches the pleural space depends on lung and chest wall mechanics, lung volume, and regional conditions. Pleural and pericardial pressures are not simply equal to the displayed airway pressure. Subtracting the same fixed fraction of PEEP from every filling pressure therefore cannot reliably calculate transmural pressure.

Measuring at end expiration compares readings at a similar point in the respiratory cycle. It does not make the pressure around the heart zero. Applied PEEP, intrinsic PEEP, and active expiratory muscles can still contribute. Examine the respiratory effort and pressure waveform before interpreting the reading.

This distinction is especially important when a ventilator adjustment changes the displayed pressure. A new pressure value should be interpreted with the simultaneous change in cardiac output, chamber appearance, and the respiratory pressure environment.

8.8 The Two Ventricles Compete and Cooperate During Filling

The ventricles are connected in series, share a septum, and lie within a common pericardial space. Their filling behavior is therefore interdependent. A change on one side can affect the other before any change has traveled through the pulmonary circulation.

Marked right ventricular expansion can move the shared septum toward the left ventricle. The pericardium also limits how much the whole heart can expand. Less space is then available for left ventricular filling. Left ventricular volume can fall even while right-sided pressure and systemic venous congestion increase.

This produces an important apparent contradiction. The left ventricle can look small while the circulation is congested. Its small cavity does not necessarily establish that the whole circulation needs additional fluid. The immediate problem may be right ventricular pressure overload and competition for space.

In an experimental study of acute pulmonary embolism in dogs, Belenkie and colleagues showed that volume loading under severe right ventricular loading conditions could reduce left ventricular filling and stroke work. The experiment demonstrates a mechanism of adverse ventricular interaction. It does not establish that fluid is harmful in every patient with pulmonary embolism. 6

There is also a series effect. Right ventricular output supplies left ventricular inflow after passage through the lungs. Improving right ventricular ejection can therefore improve left ventricular filling without adding fluid. Conversely, an intervention that increases right ventricular afterload may reduce left-sided filling over subsequent beats.

Assessment should include both ventricles, septal configuration, the pericardium, and the likely pulmonary vascular load. The response to loading depends on which part of the system is limiting flow. Looking at the left ventricular cavity alone leaves much of that system unexplained.

8.9 What Echocardiography Can Say About Filling

Echocardiography shows chamber size, wall thickness, systolic function, valve disease, and pericardial abnormalities. Doppler adds information about filling velocities and myocardial motion. These observations help explain filling, but they answer several different physiological questions.

The mitral E wave measures early diastolic inflow velocity as blood moves from the atrium into the ventricle. The A wave measures late inflow associated with atrial contraction. Their ratio, E/A, compares those velocities. Tissue Doppler measures movement of the mitral annulus itself. Its early diastolic velocity, e′, provides information related to ventricular relaxation.

Early inflow depends on the atrial-to-ventricular pressure gradient and ventricular relaxation. Annular e′ adds information about relaxation. Combining them as E/e′ can help estimate filling pressure in suitable circumstances. The ratio still does not measure pressure directly, and it does not show how stroke volume will change with more filling. A result suggesting high filling pressure therefore cannot establish fluid responsiveness.

ICU conditions complicate interpretation. Tachycardia can fuse filling waves. Atrial fibrillation removes a consistent A wave and creates beat-to-beat variation. Valve disease, regional annular abnormalities, ventilation, and rapid loading changes can affect the measurements. Ejection fraction also depends on loading and does not establish normal relaxation.

Brault and colleagues pooled three prospective cohorts containing 98 mechanically ventilated patients. An algorithm based on the 2016 echocardiographic recommendations could not classify filling pressure in 49% of them. Agreement with measured pulmonary artery occlusion pressure was moderate. The study illustrates the uncertainty of applying a pressure-estimation algorithm in this ICU population; it does not directly test later versions of echocardiographic guidance. 7

The useful interpretation integrates several observations with the clinical setting. Serial studies can help establish how the circulation is changing, provided the acquisition is comparable. When the images and hemodynamics conflict, investigate the reason rather than forcing one isolated ratio to settle the diagnosis.

8.10 Testing Whether More Filling Increases Flow

A functional test changes filling and measures whether stroke volume or cardiac output rises. It observes the response near the current point on the Frank–Starling relationship. One baseline pressure cannot provide that same information because it does not show what happens when filling changes.

Passive leg raising transfers blood toward the central circulation through a reversible postural change. The response is transient, so flow must be measured promptly. Heart rate, vasoactive treatment, ventilation, and measurement technique should remain sufficiently stable to interpret the change.

Monnet and colleagues studied passive leg raising in critically ill patients, including patients with spontaneous breathing activity or arrhythmias. The change in measured aortic flow predicted the response to subsequent volume expansion. The flow response was more informative than relying on the arterial pulse-pressure change alone. 8

The test still needs adequate execution. A flow measurement that arrives after the transient response has faded can miss it. A change smaller than the measurement method can reliably distinguish should be treated as uncertain. Pain, movement, and pressure changes caused by the maneuver can also complicate interpretation.

A fluid challenge similarly tests the circulation but leaves the administered volume in the patient. It should therefore have a defined purpose, a measurable flow endpoint, and a reason to expect that greater flow may help. A pressure rise alone is not a positive flow response.

Three judgments remain necessary. The patient must have a clinical reason to consider increasing flow. The circulation must be capable of increasing flow through the proposed intervention. The pressure and congestion consequences must be acceptable. A positive responsiveness test addresses the second judgment; it does not settle the other two.

8.11 Bringing the Filling Assessment to the Bedside

Consider an illustrative patient with septic shock, a history of hypertension, concentric left ventricular hypertrophy, and a preserved ejection fraction. The patient has persistent hypoperfusion after initial treatment. Lung ultrasound shows bilateral B lines, and a central venous pressure is moderately elevated.

The ejection fraction does not establish normal filling. Hypertrophy can coexist with impaired relaxation and a steep diastolic pressure–volume relationship. B lines indicate an interstitial ultrasound pattern and need a clinical explanation; they are not a specific measurement of left atrial pressure. The venous pressure also requires waveform and respiratory interpretation.

The next assessment examines both ventricles, valve function, pericardial conditions, and the current respiratory support. It reviews heart rate, rhythm, perfusion findings, and the adequacy of the flow measurement. No one finding provides the whole explanation.

A technically adequate passive leg raise produces no convincing stroke-volume increase beyond measurement variability. Under these conditions, another fluid bolus has little support as a way to increase output. A small apparent change should not be upgraded into a definite response simply because the arterial pressure remains low.

Attention then returns to the mechanism of shock. Vasodilation, impaired myocardial performance, excessive ventricular load, or another process may require treatment. Congestion and diastolic constraints influence that treatment, but they do not replace diagnosis. The assessment must also continue to look for the cause of persistent hypoperfusion.

If flow had increased clearly during leg raising, that would support the prediction that more filling could increase output at that time. Next determine whether the additional flow is likely to improve the persisting perfusion problem. Also assess whether a sustained infusion could worsen lung edema or systemic venous congestion. The response test informs the decision, while mechanism, clinical need, and tolerance complete it.

References

1. Terui T, Shimamoto Y, Yamane M, et al. Regulatory mechanism of length-dependent activation in skinned porcine ventricular muscle: role of thin filament cooperative activation in the Frank-Starling relation. J Gen Physiol. 2010;136(4):469–482. doi:10.1085/jgp.201010502.

2. Landesberg G, Gilon D, Meroz Y, et al. Diastolic dysfunction and mortality in severe sepsis and septic shock. Eur Heart J. 2012;33(7):895–903. doi:10.1093/eurheartj/ehr351.

3. Kumar A, Anel R, Bunnell E, et al. Pulmonary artery occlusion pressure and central venous pressure fail to predict ventricular filling volume, cardiac performance, or the response to volume infusion in normal subjects. Crit Care Med. 2004;32(3):691–699. doi:10.1097/01.CCM.0000114996.68110.C9.

4. Osman D, Ridel C, Ray P, et al. Cardiac filling pressures are not appropriate to predict hemodynamic response to volume challenge. Crit Care Med. 2007;35(1):64–68. doi:10.1097/01.CCM.0000249851.94101.4F.

5. Tyberg JV, Taichman GC, Smith ER, Douglas NW, Smiseth OA, Keon WJ. The relationship between pericardial pressure and right atrial pressure: an intraoperative study. Circulation. 1986;73(3):428–432. doi:10.1161/01.CIR.73.3.428.

6. Belenkie I, Dani R, Smith ER, Tyberg JV. Effects of volume loading during experimental acute pulmonary embolism. Circulation. 1989;80(1):178–188. doi:10.1161/01.CIR.80.1.178.

7. Brault C, Marc J, Mercado P, et al. Estimation of pulmonary artery occlusion pressure using Doppler echocardiography in mechanically ventilated patients. Crit Care Med. 2020;48(10):e943–e950. doi:10.1097/CCM.0000000000004512.

8. Monnet X, Rienzo M, Osman D, et al. Passive leg raising predicts fluid responsiveness in the critically ill. Crit Care Med. 2006;34(5):1402–1407. doi:10.1097/01.CCM.0000215453.11735.06.